research paper over effects of alcohol

• Introduction
• Conclusions
• Article Information

eTable 1. Health Deficits of the Frailty Index in the UK Biobank Cohort

eTable 2. Association of Wine Preference and Drinking During Meals With Mortality in Older Drinkers From the UK Biobank Cohort

eTable 3. Association of Average Alcohol Intake Status With Mortality in Older Drinkers From the UK Biobank Cohort, Excluding Participants With Prevalent Cancer at Baseline for Cancer Mortality, or Those With Prevalent CVD at Baseline for CVD Mortality

eTable 4. Association of Wine Preference or Drinking During Meals With Mortality in Older Drinkers From the UK Biobank Cohort, Excluding Participants With Prevalent Cancer at Baseline for Cancer Mortality, or Those With Prevalent CVD at Baseline for CVD Mortality

eTable 5. Association of Wine Preference and Drinking During Meals With Mortality in Older Drinkers From the UK Biobank Cohort, Excluding Participants With Prevalent Cancer at Baseline for Cancer Mortality, or Those With Prevalent CVD at Baseline for CVD Mortality

eTable 6. Association of Average Alcohol Intake Status With Mortality in Older Drinkers From the UK Biobank Cohort, by Drinking Patterns, Excluding Participants With Prevalent Cancer at Baseline for Cancer Mortality, or Those With Prevalent CVD at Baseline for CVD Mortality

Data Sharing Statement

See More About

Sign up for emails based on your interests, select your interests.

Customize your JAMA Network experience by selecting one or more topics from the list below.

• Academic Medicine
• Acid Base, Electrolytes, Fluids
• Allergy and Clinical Immunology
• American Indian or Alaska Natives
• Anesthesiology
• Anticoagulation
• Art and Images in Psychiatry
• Artificial Intelligence
• Assisted Reproduction
• Bleeding and Transfusion
• Caring for the Critically Ill Patient
• Challenges in Clinical Electrocardiography
• Climate and Health
• Climate Change
• Clinical Challenge
• Clinical Decision Support
• Clinical Implications of Basic Neuroscience
• Clinical Pharmacy and Pharmacology
• Complementary and Alternative Medicine
• Consensus Statements
• Coronavirus (COVID-19)
• Critical Care Medicine
• Cultural Competency
• Dental Medicine
• Dermatology
• Diabetes and Endocrinology
• Diagnostic Test Interpretation
• Drug Development
• Electronic Health Records
• Emergency Medicine
• End of Life, Hospice, Palliative Care
• Environmental Health
• Equity, Diversity, and Inclusion
• Facial Plastic Surgery
• Gastroenterology and Hepatology
• Genetics and Genomics
• Genomics and Precision Health
• Global Health
• Guide to Statistics and Methods
• Hair Disorders
• Health Care Delivery Models
• Health Care Economics, Insurance, Payment
• Health Care Quality
• Health Care Reform
• Health Care Safety
• Health Care Workforce
• Health Disparities
• Health Inequities
• Health Policy
• Health Systems Science
• History of Medicine
• Hypertension
• Images in Neurology
• Implementation Science
• Infectious Diseases
• Innovations in Health Care Delivery
• JAMA Infographic
• Law and Medicine
• Leading Change
• Less is More
• LGBTQIA Medicine
• Lifestyle Behaviors
• Medical Coding
• Medical Devices and Equipment
• Medical Education
• Medical Education and Training
• Medical Journals and Publishing
• Mobile Health and Telemedicine
• Narrative Medicine
• Neuroscience and Psychiatry
• Notable Notes
• Nutrition, Obesity, Exercise
• Obstetrics and Gynecology
• Occupational Health
• Ophthalmology
• Orthopedics
• Otolaryngology
• Pain Medicine
• Palliative Care
• Pathology and Laboratory Medicine
• Patient Care
• Patient Information
• Performance Improvement
• Performance Measures
• Perioperative Care and Consultation
• Pharmacoeconomics
• Pharmacoepidemiology
• Pharmacogenetics
• Pharmacy and Clinical Pharmacology
• Physical Medicine and Rehabilitation
• Physical Therapy
• Physician Leadership
• Population Health
• Primary Care
• Professional Well-being
• Professionalism
• Psychiatry and Behavioral Health
• Public Health
• Pulmonary Medicine
• Regulatory Agencies
• Reproductive Health
• Research, Methods, Statistics
• Resuscitation
• Rheumatology
• Risk Management
• Scientific Discovery and the Future of Medicine
• Shared Decision Making and Communication
• Sleep Medicine
• Sports Medicine
• Stem Cell Transplantation
• Substance Use and Addiction Medicine
• Surgical Innovation
• Surgical Pearls
• Teachable Moment
• Technology and Finance
• The Art of JAMA
• The Arts and Medicine
• The Rational Clinical Examination
• Tobacco and e-Cigarettes
• Translational Medicine
• Trauma and Injury
• Treatment Adherence
• Ultrasonography
• Users' Guide to the Medical Literature
• Vaccination
• Venous Thromboembolism
• Veterans Health
• Women's Health
• Workflow and Process
• Wound Care, Infection, Healing

Get the latest research based on your areas of interest.

Others also liked.

• Download PDF
• X Facebook More LinkedIn

Ortolá R , Sotos-Prieto M , García-Esquinas E , Galán I , Rodríguez-Artalejo F. Alcohol Consumption Patterns and Mortality Among Older Adults With Health-Related or Socioeconomic Risk Factors. JAMA Netw Open. 2024;7(8):e2424495. doi:10.1001/jamanetworkopen.2024.24495

Manage citations:

© 2024

• Permissions

Alcohol Consumption Patterns and Mortality Among Older Adults With Health-Related or Socioeconomic Risk Factors

• 1 Department of Preventive Medicine and Public Health, Universidad Autónoma de Madrid, Madrid, Spain
• 2 Center for Biomedical Research in Epidemiology and Public Health, Madrid, Spain
• 3 Department of Environmental Health and Nutrition, Harvard T.H. Chan School of Public Health. Boston, Massachusetts
• 4 Madrid Institute for Advanced Studies Food Institute, Campus of International Excellence Universidad Autónoma de Madrid + Spanish National Research Council, Madrid, Spain
• 5 Department of Chronic Diseases, National Center for Epidemiology, Carlos III Health Institute, Madrid, Spain

Question   Do health-related or socioeconomic risk factors modify the associations of alcohol consumption patterns with mortality among older drinkers?

Findings   This cohort study in 135 103 older drinkers found that even low-risk drinking was associated with higher mortality among older adults with health-related or socioeconomic risk factors. Wine preference and drinking only with meals were associated with attenuating the excess mortality associated with alcohol consumption.

Meaning   This cohort study identified inequalities in the detrimental health outcomes associated with alcohol that should be addressed to reduce the high disease burden of alcohol use.

Importance   Alcohol consumption is a leading cause of morbidity and mortality that may be more important in older adults with socioeconomic or health-related risk factors.

Objective   To examine the association of alcohol consumption patterns with 12-year mortality and its modification by health-related or socioeconomic risk factors.

Design, Setting, and Participants   This prospective cohort study used data from the UK Biobank, a population-based cohort. Participants were current drinkers aged 60 years or older. Data were analyzed from September 2023 to May 2024.

Exposure   According to their mean alcohol intake in grams per day, participants’ drinking patterns were classified as occasional: ≤2.86 g/d), low risk (men: >2.86-20.00 g/d; women: >2.86-10.00 g/d), moderate risk (men: >20.00-40.00 g/d; women: >10.00-20.00 g/d) and high risk (men: >40.00 g/d; women: >20.00 g/d).

Main Outcomes and Measures   Health-related risk factors were assessed with the frailty index, and socioeconomic risk factors were assessed with the Townsend deprivation index. All-cause and cause-specific mortality were obtained from death certificates held by the national registries. Analyses excluded deaths in the first 2 years of follow-up and adjusted for potential confounders, including drinking patterns and preferences.

Results   A total of 135 103 participants (median [IQR] age, 64.0 [62.0-67.0] years; 67 693 [50.1%] women) were included. In the total analytical sample, compared with occasional drinking, high-risk drinking was associated with higher all-cause (hazard ratio [HR], 1.33; 95% CI, 1.24-1.42), cancer (HR, 1.39; 95% CI, 1.26-1.53), and cardiovascular (HR, 1.21; 95% CI, 1.04-1.41) mortality; moderate-risk drinking was associated with higher all-cause (HR, 1.10; 95% CI, 1.03-1.18) and cancer (HR, 1.15; 95% CI, 1.05-1.27) mortality, and low-risk drinking was associated with higher cancer mortality (HR, 1.11; 95% CI, 1.01-1.22). While no associations were found for low- or moderate-risk drinking patterns vs occasional drinking among individuals without socioeconomic or health-related risk factors, low-risk drinking was associated with higher cancer mortality (HR, 1.15; 95% CI, 1.01-1.30) and moderate-risk drinking with higher all-cause (HR, 1.10; 95% CI, 1.01-1.19) and cancer (HR, 1.19; 95% CI, 1.05-1.35) mortality among those with health-related risk factors; low-risk and moderate-risk drinking patterns were associated with higher mortality from all causes (low risk: HR, 1.14; 95% CI, 1.01-1.28; moderate risk: HR, 1.17; 95% CI, 1.03-1.32) and cancer (low risk: HR, 1.25; 95% CI, 1.04-1.50; moderate risk: HR, 1.36; 95% CI, 1.13-1.63) among those with socioeconomic risk factors. Wine preference (>80% of alcohol from wine) and drinking with meals showed small protective associations with mortality, especially from cancer, but only in drinkers with socioeconomic or health-related risk factors and was associated with attenuating the excess mortality associated with high-, moderate- and even low-risk drinking.

Conclusions and Relevance   In this cohort study of older drinkers from the UK, even low-risk drinking was associated with higher mortality among older adults with health-related or socioeconomic risk factors. The attenuation of mortality observed for wine preference and drinking only during meals requires further investigation, as it may mostly reflect the effect of healthier lifestyles, slower alcohol absorption, or nonalcoholic components of beverages.

Alcohol consumption is a leading cause of morbidity and mortality, accounting for approximately 5.1% of the global burden of disease and 5.3% of all deaths and being responsible for significant social and economic losses, thus representing a major public health problem. 1 Additionally, the assumed benefits of drinking low amounts of alcohol, especially on cardiovascular disease (CVD) mortality, 2 - 4 are being questioned due to selection biases, reverse causation, and residual confounding, 5 supporting health messaging that the safest level of drinking is no drinking at all or less is better. 6 , 7 Selection biases are often overlooked, but they can lead to a systematic underestimation of alcohol-related burden. That is the case of the abstainer bias, whereby the apparently lower mortality of light drinkers compared with abstainers could be explained by the higher death risk of the abstainers because they include former drinkers who quit alcohol due to poor health, as well as lifetime abstainers, 5 who often have worse lifestyle and health characteristics than regular drinkers. 8 Also, the healthy drinker/survivor bias, caused by overrepresentation of healthier drinkers who have survived the deleterious effects of alcohol, can distort comparisons, especially in older age. 5 In addition, drinking habits may influence the association between the amount of alcohol consumed and health. In this context, wine preference has been associated with lower risk of death, 9 CVD morbimortality, 10 and diabetes, 11 attributing the beneficial associations of wine to its high content in polyphenols. 12 Furthermore, drinking with meals has been associated with lower risk of all-cause, non-CVD, and cancer deaths 13 and frailty, 14 so this might be a safer option for alcohol drinkers along with moderate consumption. 15

The health impact of alcohol consumption may be greater in individuals with socioeconomic or health-related risk factors. On one hand, older adults with health-related risk factors are more susceptible to the harmful outcomes associated with alcohol due to their greater morbidity, higher use of alcohol-interacting drugs, and reduced tolerance. 16 , 17 However, some studies have observed benefits of alcohol on unhealthy aging or frailty, especially of light alcohol intake 18 , 19 and of a Mediterranean alcohol drinking pattern, defined as moderate alcohol consumption, preferably wine and accompanying meals, 14 , 20 suggesting that the protective associations of these potentially beneficial drinking patterns might be greater in individuals with ill health, although they might be due to the aforementioned methodological issues. 5 Therefore, it would be of interest to examine whether health-related risk factors modify the associations between alcohol consumption patterns and mortality.

On the other hand, there is evidence that socioeconomically disadvantaged populations have higher rates of alcohol-related harms for equivalent and even lower amounts of alcohol, probably due to the coexistence of other health challenges, including less healthy lifestyles, and lower social support or access to health care. 21 , 22 Also, the potentially beneficial associations of wine preference and drinking during meals might be more important in individuals with socioeconomic risk factors. However, to our knowledge, no previous research has examined whether socioeconomic status modifies the associations between these potentially beneficial drinking patterns and health.

Therefore, the aim of our study is to examine the associations of several potentially beneficial alcohol consumption patterns, that is, consumption of low amounts of alcohol, wine preference, and drinking only during meals, with all-cause, cancer, and CVD mortality in older adults and their modification by health-related or socioeconomic risk factors, while addressing the main methodological issues deemed to bias such associations. Thus, we restrict analyses to current drinkers and use occasional drinkers instead of abstainers as the reference group to prevent selection biases, exclude deaths in the first 2 years of follow-up to reduce reverse causation, and adjust analyses for many sociodemographic, lifestyle, and clinical variables to palliate residual confounding. We also restrict analyses to older adults because most deaths occur in this population group, which also has a high prevalence of health-related risk factors and because the protective associations of alcohol consumption have been specifically observed in older adults, 6 which is consistent with our aim to study potentially beneficial drinking patterns.

This cohort study was approved by the North West Multi-Centre Research Ethics Committee, and all participants provided written informed consent before enrollment. This study is reported following the Strengthening the Reporting of Observational Studies in Epidemiology ( STROBE ) reporting guideline.

We used data from the UK Biobank cohort, a multicenter, prospective, population-based study with more than 500 000 participants aged 40 to 69 years identified from National Health Service primary care registers and enrolled at 22 assessment sites across England, Scotland, and Wales between 2006 and 2010. At the baseline assessment visit, they completed a computer-assisted interview and a touch-screen questionnaire on sociodemographic, lifestyle, and clinical characteristics, provided biological samples, and underwent physical and medical examinations. They were followed-up for mortality through linkage to national death registries. Additional information on the UK Biobank study has been reported elsewhere. 23 , 24

At the baseline assessment visit, study participants were asked about the frequency and mean amount of the main types of alcoholic beverages that they consumed, and alcohol content was estimated by multiplying the volume ingested (in milliliters) by the volume percentage of alcohol (4.5% for beer and cider, 11.5% for white and sparkling wine, 13% for red wine, 20% for fortified wine, and 40% for spirits) and by the specific gravity of ethanol (0.789 g/mL). According to their mean alcohol intake, drinking patterns were classified into occasional (≤2.86 g/d), low risk (men: >2.86-20.00 g/d; women: >2.86-10.00 g/d), moderate risk (men: >20.00-40.00 g/d; women: >10-20.00 g/d), and high risk (men: >40.00 g/d; women: >20.00 g/d), a categorization based on the recommendations from health authorities that we have used in previous studies. 25 - 27 When more than 80% of alcohol came from a certain type of beverage, drinkers were classified as with preference for wine, with preference for other drinks, or with no preference. 27 Participants were also classified as drinkers only during meals and as drinkers either only outside of meals or at any time. Finally, participants were classified as drinkers with no wine preference nor drinking only during meals, drinkers with wine preference or drinking only during meals, and drinkers with wine preference and drinking only during meals.

Health-related risk was assessed at baseline using the frailty index (FI) developed specifically for the UK Biobank 28 based on the procedure used by Rockwood et al. 29 A total of 49 health deficits were considered, most dichotomously (1 point if present and 0 points otherwise), and a few according to severity (0 points for no deficit, 0.25-0.75 points for mild to moderate deficits, and 1 point for severe deficit). The FI score was calculated as the total sum of points assigned to each health deficit divided by the number of deficits considered and ranged from 0.00 to 0.57. The complete list of health deficits and associated scores can be found in eTable 1 in Supplement 1 . Participants were considered to have health-related risk factors if they were prefrail or frail (FI > 0.12). 28

Socioeconomic risk was assessed at baseline using the Townsend deprivation index (TDI), 30 which measures the level of an area’s socioeconomic deprivation. TDI ranges from −6.26 to 10.16, with higher score indicating greater deprivation. Participants were considered to have socioeconomic risk factors if they lived in more deprived areas (TDI > 0) and not if they lived in more affluent areas (TDI ≤ 0).

Information on mortality was obtained from death certificates held by the National Health Service (NHS) Information Centre (NHS England) up to September 30, 2021, for participants in England and Wales, and by the NHS Central Register Scotland (National Records of Scotland) up to October 31, 2021, for participants in Scotland. 31 , 32 Length of follow-up was estimated as the time from the baseline assessment visit to the date of death or administrative censoring, whichever came first. Cause-specific mortality was ascertained with the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision ( ICD-10 ) classification 33 : codes C00 to C97 as primary cause of death for cancer and codes I00 to I99 for CVD.

We also used baseline information on sociodemographic, lifestyle, and clinical characteristics, including sex, age, self-reported race and ethnicity, education (college or university degree; A levels, AS levels, or equivalent; O levels, General Certificate of Secondary Education, or equivalent; Certificate of Secondary Education or equivalent; National Vocational Qualification, Higher National Diploma, Higher National Certificate, or equivalent; other professional qualifications; and no qualifications), tobacco smoking (never, former, or current), leisure-time physical activity (metabolic equivalents of task-hours per week), time spent watching television (hours per day), and prevalent morbidities (diabetes, CVD, and cancer) that could have a potential effect on the amount of alcohol consumed. In the UK Biobank, race and ethnicity are classified as Asian (Indian, Pakistani, Bangladeshi, any other Asian background), Black (Caribbean, African, any other Black background), Chinese, multiple (White and Black Caribbean, White and Black African, White and Asian, any other mixed background), White (British, Irish, any other White background), and other (any group not specified, eg, Arab).

From 217 462 participants aged at least 60 years in the UK Biobank cohort, we excluded 36 284 with incomplete information on alcohol consumption, 10 456 never drinkers, 8295 former drinkers, and 20 167 known binge drinkers (those who consumed ≥6 units of alcohol in 1 session) to avoid classifying binge drinkers with low mean alcohol intake as low-risk drinkers. We additionally excluded 1140 participants who died in the first 2 years of follow-up and 6017 participants with missing information on the FI (194 participants), the TDI (116 participants), and potential confounders (5707 participants). Thus, the analytical sample included 135 103 individuals.

The associations of alcohol consumption patterns (mean alcohol intake status, wine preference, and drinking during meals) at baseline with all-cause and cause-specific mortality were summarized with hazard ratios (HRs) and their 95% CIs obtained from Cox regression; the models included interactions between alcohol consumption patterns and health-related or socioeconomic risk factors and adjusted for baseline sociodemographic (sex, age, race and ethnicity, education, and TDI [except when stratifying by socioeconomic risk factors]), lifestyle (tobacco smoking, leisure-time physical activity, and time spent watching television), and clinical characteristics (diabetes, CVD, cancer, and FI score [except when stratifying by health-related risk factors]) of study participants. Analyses of alcohol intake were further adjusted for wine preference and drinking during meals, whereas analyses of wine preference and drinking during meals were further adjusted for mean alcohol intake and the other drinking pattern.

To characterize whether wine preference and drinking during meals modified the association of mean alcohol intake with mortality, we tested interaction terms defined as the product of the categories of mean alcohol intake by 3 categories of drinking patterns (no wine preference nor drinking only during meals, wine preference or drinking only during meals, and wine preference and drinking only during meals).

Additionally, we assessed whether sociodemographic and lifestyle variables modified the study associations by testing interaction terms defined as the product of alcohol consumption patterns by categories of such variables (except mean alcohol intake status by sex, as sex was included in the definition of alcohol intake status). Since no interactions were found, the results are presented for the total sample. Finally, we performed additional sensitivity analyses excluding participants with prevalent cancer at baseline for cancer mortality or those with prevalent CVD at baseline for CVD mortality.

Statistical significance was set at 2-sided P  < .05. Analyses were performed with Stata software version 17 (StataCorp). Data were analyzed from September 2023 to May 2024.

A total of 135 103 participants (median [IQR] age, 64.0 [62.0-67.0] years; 67 693 [50.1%] women) were included. Occasional drinkers less often identified as White; were more frequently residents in England, women, and never smokers; were less physically active; had a lower educational level, a lower prevalence of CVD; and had a higher prevalence of diabetes, cancer, and health-related risk factors. Having socioeconomic risk factors was less frequent in low- and moderate-risk drinkers ( Table 1 ).

Over a median (range) follow-up of 12.4 (2.0 to 14.8) years, 15 833 deaths were recorded, including 7871 cancer deaths and 3215 CVD deaths. Compared with occasional drinking, low-risk drinking was associated with higher cancer mortality (HR, 1.11; 95% CI, 1.01-1.22); moderate-risk drinking was associated with higher all-cause (HR, 1.10; 95% CI, 1.03-1.18) and cancer (HR, 1.15; 95% CI, 1.05-1.27) mortality; and high-risk drinking was associated with higher all-cause (HR, 1.33; 95% CI, 1.24-1.42), cancer (HR, 1.39; 95% CI, 1.26-1.53), and CVD (HR, 1.21; 95% CI, 1.04-1.41) mortality ( Table 2 ). Hazards were greater in individuals with health-related or socioeconomic risk factors vs those without across categories of alcohol intake. Interestingly, while no associations with mortality were found in participants without health-related or socioeconomic risk factors among low- or moderate-risk drinkers, low-risk drinkers with health-related risk factors had higher cancer mortality (HR, 1.15; 95% CI, 1.01-1.30) and moderate-risk drinkers with health-related risk factors had higher all-cause (HR, 1.10; 95% CI, 1.01-1.19) and cancer (HR, 1.19; 95% CI, 1.05-1.35) mortality ( Table 2 ). Likewise, both low-risk and moderate-risk drinkers with socioeconomic risk factors showed higher mortality from all causes (low risk: HR, 1.14; 1.01-1.28; moderate risk: 1.17; 95% CI, 1.03-1.32) and cancer (low-risk: HR, 1.25; 95% CI, 1.04-1.50; moderate risk: HR, 1.36; 95% CI, 1.13-1.63) ( Table 2 ).

Wine preference and drinking only during meals were associated with lower all-cause mortality only in participants with health-related risk factors (wine preference: HR, 0.92; 95% CI, 0.87-0.97; drinking only during meals: HR, 0.93; 95% CI, 0.89-0.97), as well as in participants with socioeconomic risk factors (wine preference: HR, 0.84; 95% CI, 0.78-0.90; drinking only during meals: HR, 0.83; 95% CI, 0.78-0.89) ( Table 3 ). Drinking only during meals was also associated with lower cancer mortality in participants with health-related risk factors (HR, 0.92; 95% CI, 0.86-0.99) or socioeconomic risk factors (HR, 0.85; 95% CI, 0.78-0.94) ( Table 3 ). Furthermore, in individuals with socioeconomic risk factors, wine preference was associated with lower cancer mortality (HR, 0.89; 95% CI, 0.80-0.99) and drinking only during meals with lower CVD mortality (HR, 0.86; 95% CI, 0.75-1.00) ( Table 3 ). Adhering to both drinking patterns was associated with lower all-cause, cancer, and CVD mortality in drinkers with health-related or socioeconomic risk factors, and to a lesser extent, with lower all-cause death in drinkers without health-related risk factors (eTable 2 in Supplement 1 ). Importantly, wine preference and drinking during meals modified the association of mean alcohol intake with mortality: the excess risk of all-cause, cancer, and CVD death for high-risk drinkers, of all-cause and cancer death for moderate-risk drinkers, and of cancer death for low-risk drinkers vs occasional drinkers was attenuated and even lost among individuals with these drinking patterns ( Table 4 ). Analyses excluding participants with prevalent cancer at baseline for cancer mortality, or those with prevalent CVD at baseline for CVD mortality showed consistent results (eTables 3-6 in Supplement 1 ).

This cohort study in older alcohol drinkers from the UK found that compared with occasional drinkers, low-risk drinkers had higher cancer mortality, moderate-risk drinkers had higher all-cause and cancer mortality, and high-risk drinkers had higher all-cause, cancer, and CVD mortality. The excess mortality associated with alcohol consumption was higher in individuals with health-related and socioeconomic risk factors, among whom even low-risk drinkers had higher mortality, especially from cancer. Wine preference and drinking only with meals showed small protective associations with mortality, especially from cancer, among drinkers with health-related and socioeconomic risk factors, and these 2 drinking patterns attenuated the excess mortality associated with high-, moderate-, and even low-risk drinking.

In line with recent research on the associations between alcohol use and health, 6 , 34 , 35 our results corroborate the detrimental outcomes associated with heavy drinking in older adults. However, we also found higher risk for all-cause and cancer deaths in moderate-risk drinkers, unlike most previous research, which has reported protective associations of low to moderate alcohol consumption, mainly for all-cause 2 - 4 , 36 and CVD 3 , 36 , 37 mortality, ischemic heart disease, 3 , 6 , 34 and diabetes, 6 or null associations with all-cause mortality, 38 CVD, 39 and unhealthy aging. 20 This discrepancy may be due to the implementation of an important methodological improvement in our analyses, that is, using occasional drinkers as the reference group instead of lifetime abstainers, to prevent selection bias caused by misclassification of former drinkers as abstainers, and to palliate residual confounding because they are more like light drinkers than are never drinkers. 40 , 41 In fact, another analysis of the UK Biobank cohort that also avoided selection biases found an increased CVD risk in the general population for drinking up to 14 units per week. 42

To our knowledge, there are no studies examining the potential modification of health-related risk factors on the association between alcohol use and health. The stronger associations between mean alcohol intake and mortality observed in older adults with health-related risk factors make sense, since they have more morbid conditions potentially aggravated by alcohol and greater use of alcohol-interacting medications than their counterparts without health-related risk factors. 16 , 17 The fact that even low-risk drinkers with these risk factors had higher risk of cancer death is an important finding, which is consistent with the reported increased risk of several types of cancer and cancer mortality even with very low amounts of alcohol. 6 , 36 , 37 , 43

Our results also suggest that socioeconomic status acts as a modifier of the association between the amount of alcohol consumed and mortality, as mortality hazard was much greater in individuals with socioeconomic risk factors than in individuals without, in line with previous research. 21 , 22 , 44 , 45 We even found a detrimental association of low amounts of alcohol with all-cause and cancer mortality in this group, unlike the MORGAM study by DiCasetnuovo et al 44 reporting a lower mortality associated with consuming no more than 10 g/d of alcohol, which was clearer in individuals with higher vs lower education. 44 These discrepant results could again be explained by the different reference groups used: occasional drinkers in our study and never drinkers in the MORGAM study. Importantly, although older adults with socioeconomic risk factors have a higher risk of ill health and death, probably due to the coexistence of other health challenges, especially poorer lifestyles, 21 , 22 the observed associations in our study were independent of lifestyles, suggesting that other factors should account for them.

Regarding the potentially beneficial drinking patterns, that is, wine preference and drinking during meals, the literature is inconsistent. A 2018 pool of studies 34 reported a nondifferential association of specific types of alcoholic drinks with all-cause mortality and several CVD outcomes, whereas other studies have found protective health associations for wine but not other beverages. 15 , 46 Drinking with meals has also shown protective associations with several health outcomes. 15 In our analysis, these drinking patterns modified the association between alcohol intake and death risk. On one hand, the protective association for mortality of these patterns was only observed in individuals with socioeconomic or health-related risk factors, independently of the amount of alcohol consumed. On the other hand, the detrimental association of alcohol intake was more evident in individuals without these patterns. These findings suggest that the less detrimental associations of alcohol intake from wine or during meals are not due to alcohol itself, but to other factors, including nonalcoholic components of wine, such as antioxidants, slower absorption of alcohol ingested with meals and its consequent reduced alcoholaemia, as well as spacing drinks when drinking only with meals, or more moderate attitudes in individuals who choose to adhere to these drinking patterns.

Our study has several strengths, such as the large sample size, the long follow-up, and the methodological improvements implemented to prevent selection biases and reduce reverse causation. However, it also has some limitations. First, alcohol intake was self-reported, and therefore prone to some degree of misclassification. Also, alcohol intake was measured only at baseline and not at multiple time points over the life span, not allowing us to take into account changes in alcohol intake before the baseline assessment or to redistribute former drinkers among categories of current drinkers to reduce selection bias; this may have led to an underestimation of the true effects of alcohol consumption. 5 Second, as in any observational study, we cannot entirely rule out residual confounding, despite adjusting for many potential confounders. And third, this study was conducted in older adults in the UK with a high proportion of White participants, so our results may not be generalizable to other racial ethnic groups or populations with different lifestyles, drinking patterns, or socioeconomic development.

This cohort study among older drinkers from the UK did not find evidence of a beneficial association between low-risk alcohol consumption and mortality; however, we observed a detrimental association of even low-risk drinking in individuals with socioeconomic or health-related risk factors, especially for cancer deaths. The attenuation of the excess mortality associated with alcohol among individuals who preferred to drink wine or drink only during meals requires further investigation to elucidate the factors that may explain it. Finally, these results have important public health implications because they identify inequalities in the detrimental health outcomes associated with alcohol that should be addressed to reduce the high burden of disease of alcohol use.

Accepted for Publication: May 30, 2024.

Published: August 12, 2024. doi:10.1001/jamanetworkopen.2024.24495

Open Access: This is an open access article distributed under the terms of the CC-BY License . © 2024 Ortolá R et al. JAMA Network Open .

Corresponding Author: Rosario Ortolá, MD, PhD, Department of Preventive Medicine and Public Health, School of Medicine, Universidad Autónoma de Madrid, Calle del Arzobispo Morcillo 4, 28029 Madrid, Spain ( [email protected] ).

Author Contributions: Dr Ortolá had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Ortolá.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Ortolá.

Critical review of the manuscript for important intellectual content: Sotos-Prieto, García-Esquinas, Galán, Rodríguez-Artalejo.

Statistical analysis: Ortolá.

Obtained funding: Sotos-Prieto, Rodríguez-Artalejo.

Administrative, technical, or material support: Rodríguez-Artalejo.

Supervision: García-Esquinas, Galán.

Conflict of Interest Disclosures: None reported.

Funding/Support: This work was supported by the Plan Nacional sobre Drogas, Ministry of Health of Spain (grant No. 2020/17), Instituto de Salud Carlos III, State Secretary of R+D+I and Fondo Europeo de Desarrollo Regional/Fondo Social Europeo (Fondo de Investigación en Salud grants No. 19/319, 20/896, and 22/1111), Agencia Estatal de Investigación (grant No. CNS2022-135623), Carlos III Health Institute and the European Union “NextGenerationEU (grant No. PMP21/00093), and the Fundación Francisco Soria Melguizo (Papel de la Disfunción Mitocondrial en la Relación Entre Multimorbilidad Crónica y Deterioro Funcional en Ancianos project grant). Mercedes Sotos-Prieto holds a Ramón y Cajal contract (contract No. RYC-2018-025069-I) from the Ministry of Science, Innovation and Universities.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 2 .

• Register for email alerts with links to free full-text articles
• Access PDFs of free articles
• Manage your interests
• Save searches and receive search alerts

• - Google Chrome

Intended for healthcare professionals

• My email alerts
• BMA member login
• Username * Password * Forgot your log in details? Need to activate BMA Member Log In Log in via OpenAthens Log in via your institution

Search form

• Advanced search
• Search responses
• Search blogs
• Moderate alcohol...

Moderate alcohol consumption as risk factor for adverse brain outcomes and cognitive decline: longitudinal cohort study

• Related content
• Peer review
• Anya Topiwala , clinical lecturer in old age psychiatry 1 ,
• Charlotte L Allan , academic clinical lecturer in old age psychiatry 1 ,
• Vyara Valkanova , specialist registrar in old age psychiatry 1 ,
• Enikő Zsoldos , postdoctoral scientist 1 ,
• Nicola Filippini , postdoctoral scientist 1 ,
• Claire Sexton , postdoctoral scientist 2 ,
• Abda Mahmood , research assistant 1 ,
• Peggy Fooks , medical student 3 ,
• Archana Singh-Manoux , professor of epidemiology and public health 4 ,
• Clare E Mackay , associate professor 1 ,
• Mika Kivimäki , professor 4 ,
• Klaus P Ebmeier , professor of old age psychiatry 1
• 1 Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford OX3 7JX, UK
• 2 FMRIB Centre, Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, OX3 9DU, UK
• 3 University of Oxford, Warneford Hospital, Oxford, OX3 9DU, UK
• 4 Department of Epidemiology and Public Health, University College London, London, WC1E 6BT, UK
• Correspondence to: A Topiwala anya.topiwala{at}psych.ox.ac.uk
• Accepted 11 May 2017

Objectives  To investigate whether moderate alcohol consumption has a favourable or adverse association or no association with brain structure and function.

Design  Observational cohort study with weekly alcohol intake and cognitive performance measured repeatedly over 30 years (1985-2015). Multimodal magnetic resonance imaging (MRI) was performed at study endpoint (2012-15).

Setting  Community dwelling adults enrolled in the Whitehall II cohort based in the UK (the Whitehall II imaging substudy).

Participants  550 men and women with mean age 43.0 (SD 5.4) at study baseline, none were “alcohol dependent” according to the CAGE screening questionnaire, and all safe to undergo MRI of the brain at follow-up. Twenty three were excluded because of incomplete or poor quality imaging data or gross structural abnormality (such as a brain cyst) or incomplete alcohol use, sociodemographic, health, or cognitive data.

Main outcome measures  Structural brain measures included hippocampal atrophy, grey matter density, and white matter microstructure. Functional measures included cognitive decline over the study and cross sectional cognitive performance at the time of scanning.

Results  Higher alcohol consumption over the 30 year follow-up was associated with increased odds of hippocampal atrophy in a dose dependent fashion. While those consuming over 30 units a week were at the highest risk compared with abstainers (odds ratio 5.8, 95% confidence interval 1.8 to 18.6; P≤0.001), even those drinking moderately (14-21 units/week) had three times the odds of right sided hippocampal atrophy (3.4, 1.4 to 8.1; P=0.007). There was no protective effect of light drinking (1-<7 units/week) over abstinence. Higher alcohol use was also associated with differences in corpus callosum microstructure and faster decline in lexical fluency. No association was found with cross sectional cognitive performance or longitudinal changes in semantic fluency or word recall.

Conclusions  Alcohol consumption, even at moderate levels, is associated with adverse brain outcomes including hippocampal atrophy. These results support the recent reduction in alcohol guidance in the UK and question the current limits recommended in the US.

Introduction

Alcohol use is widespread and increasing across the developed world. 1 2 3 It has historically been viewed as harmless in moderation, 4 defined variably from 9-18 units (72-144 g) a week. 5 6 Recent evidence of associations with risk of cancer 7 has prompted revision of UK government alcohol guidance, though US Federal Dietary guidelines (2015-20) allow up to 24.5 units a week for men. 8 Even light drinking (midpoint <12.5g daily/8 units a week) has been associated with increased risk of oropharnygeal, oesophageal, and breast cancer. 7 9 While chronic dependent drinking is associated with Korsakoff syndrome and alcoholic dementia, 10 the long term effects of non-dependent alcohol consumption on the brain are poorly understood. Robust evidence of adverse associations would have vital implications for public health.

Some authors have suggested an inverted U shaped relation between alcohol use and brain outcomes, similar to that seen with cardiovascular disease. Light-to-moderate drinking has been associated with a lower risk of dementia 11 12 and a reduced incidence of myocardial infarction 13 and stroke. 14 Brain imaging studies, however, have thus far failed to provide a convincing neural correlate that could underpin any protective effect. Results of research into the effects of moderate alcohol on the brain are inconsistent. 15 Moderate alcohol consumption in older people has been associated with reduced total brain volume, 16 increased ventricle size, 17 grey matter atrophy, 18 and reduced density of frontal and parietal grey matter, 19 20 but others have not found such associations 15 or only at higher consumptions. 21 Associations between moderate alcohol consumption and white matter findings are also inconsistent. De Bruin and colleagues reported increased white matter volume in moderate drinkers compared with abstainers, 22 whereas Anstey and colleagues found the inverse relation. 23 Similarly, whereas increased white matter hyperintensities have been described in moderate drinkers compared with abstainers, 24 others found no association. 17 23 25

Unresolved questions persist because of design limits to existing studies of non-dependent drinking and brain imaging. Alcohol consumption cannot be randomised over long periods. Most studies to date have been cross sectional or with limited prospectively gathered data on alcohol. People typically underestimate their alcohol intake, 26 a problem likely to be worse in a retrospective study. Studies have also included elderly people, in whom sub-threshold presymptomatic cognitive impairment might already have an impact on drinking patterns.

We used data on alcohol consumption gathered prospectively over 30 years to investigate associations with brain structural and functional outcomes in 550 non-alcohol dependent participants. Our hypotheses were twofold: light drinking (<7 units weekly) is protective against adverse brain outcomes and cognitive decline and heavier drinking (above recommended guidelines) is associated with adverse brain and cognitive outcomes.

Study design and participants

Five hundred and fifty people were randomly selected for the current Whitehall II imaging substudy (2012-15) from the Whitehall II cohort study. 27 The Whitehall II study was established in 1985 at University College London, with the aim of investigating the relation between socioeconomic status, stress, and cardiovascular health. It recruited 10 308 non-industrial civil servants across a range of employment grades. Sociodemographic, health, and lifestyle variables (including alcohol use) were measured over a follow-up period of about 30 years, at about five year intervals (phase 1: 1985-88, phase 3: 1991-93, phase 5: 1997-99, phase 7: 2003-04, phase 9: 2007-09, phase 11: 2011-12). To make the sample as representative as possible of the cohort at baseline, we drew a random list of 1380 participants from those who took part in the Whitehall II phase 11 clinical examination or phase 10 pilot examination and had consented. Participants were sampled from high, intermediate, and low socioeconomic groups.

Alcohol variables collected in each phase included units drunk a week, frequency of drinking a week over the previous year, and results of the CAGE screening questionnaire. 28 We used weekly consumption in this analysis as there is less likelihood of a ceiling effect in comparison with drinking frequency. We calculated average alcohol use across the study as mean consumption a week averaged across all study phases. Participants were deemed “abstinent” if they consumed less than 1 unit of alcohol a week. “Light” drinking was defined as between 1 and <7 units a week and “moderate” drinking as 7 to <14 units a week for women and 7 to <21 units for men, based on use in the existing literature and government guidelines (fig 1 ⇓ ). “Unsafe drinking” was defined according to pre-2016 (21 units (168 g) a week for men and 14 units (112 g) for women) and newly revised UK Department of Health guidelines (>14 units (112 g) for men and women) and further categorised (14-20, 21-30, >30 units weekly) for the purposes of the logistic regression analysis. 29 Non-dependent drinkers were defined as those scoring <2 on the CAGE questionnaire.

Fig 1  UK 2016 guidelines on alcohol consumption (see www.alcoholconcern.org.uk/help-and-advice/help-and-advice-with-your-drinking/unit-calculator/ ) (redrawn from Alcohol Concern, 2016)

• Download figure
• Open in new tab
• Download powerpoint

Age, sex, education, smoking, social activity—such as attendance at clubs and visits with family/friends, physical activity, voluntary work—and component measures of the Framingham stroke risk score—such as blood pressure, smoking, history of cardiovascular events, cardiovascular drugs—were assessed by self report questionnaire. Social class was determined according to occupation at phase 3 (highest class=1, lowest=4). Drugs (number of psychotropic drugs reported as taken) and lifetime history of major depressive disorder (assessed by structured clinical interview for DSM IV) were assessed at the time of the scan. Information about personality traits was determined by questionnaire at phase 1 and included trait impulsivity (question: “Are you hot-headed?”).

Cognitive function was assessed longitudinally at phases 3, 5, 7, 9, and 11 and at the time of scanning with lexical (how many words beginning with a specific letter can be generated in one minute) and semantic (how many words in a specific category can be named in one minute) fluency tests. Short term memory recall (20 words) was tested at phases 3, 5, 7, 9, and 11. Cross sectional cognitive performance was measured at the time of the scan with the Montreal cognitive assessment (MoCA, education adjusted), trail making test (TMT-A and B), Rey-Osterrieth complex figure (RCF) test (copy, immediate, delay, recognition), Hopkins verbal learning test (HVLT-R; immediate, delay), Boston naming test (BNT), and digit span and digit substitution test (DSST). Full scale IQ (FSIQ) was estimated at the time of the scan with the test of premorbid functioning-UK version (TOPF-UK), with adjustment for sex and education.

Participants were included in the imaging substudy if they were safe to undergo MRI and able to give informed consent. Exclusions were due to incomplete or poor quality imaging data or gross structural abnormality (such as a brain cyst), incomplete data on alcohol use (>2 study phases data missing), and missing sociodemographic, health, or cognitive data (fig 2 ⇓ ).

Fig 2  Flow chart of participants included in analysis alcohol consumption and brain function

MRI analysis

All MRI scans were acquired at the functional magnetic resonance imaging of the brain (FMRIB) centre, University of Oxford, with a 3 Tesla Siemens Verio scanner (2012-15). We used T1-weighted and diffusion tensor (DTI) 3T MRI sequences for these analyses. 30

Full technical details are in the appendix. In brief, we initially examined associations between alcohol use and grey matter using voxel based morphometry, an objective method to compare grey matter density between individuals in each voxel (smallest distinguishable image volume) of the structural image. For each participant for subsequent analyses we additionally extracted hippocampal volumes (adjusted for total intracranial volume) using an automated segmentation/registration tool. Automated segmentation of the amygdala was less reliable in this sample so we did not use extracted volumes in this analysis. Three clinicians independently defined hippocampal atrophy according to visual rating (Scheltens score 31 ) and reached a consensus.

Diffusion tensor images indicate the directional preference of water diffusion in neural tissue and allow inferences about the structural integrity of white matter tracts. In healthy myelinated fibres diffusion is restricted perpendicular to the longitudinal axis of the fibre—that is, it is anisotropic. We carried out voxel-wise statistical analysis of diffusion tensor data (fractional anisotropy (FA), axial diffusivity (AD), radial diffusivity (RD), and mean diffusivity (MD)) using tract based spatial statistics (TBSS). 32

Primary outcomes were continuous measures of grey matter density in the voxel based morphometry analysis and white matter integrity in the tract based spatial statistics analysis (fractional anisotropy, mean, radial, and axial diffusivity).

Visual ratings of hippocampal atrophy were dichotomised into atrophy versus no atrophy based on 0/1 on the (4 point) Scheltens scale to reflect clinical use (“abnormal” versus “normal”). 31 Hippocampal volume (%intracranial volume) was used as a continuous variable in a multiple linear regression analysis.

As cognitive outcomes we used decline in short term memory, semantic and lexical fluency, and cross sectional performance on Montreal cognitive assessment, trail making test, Rey-Osterrieth complex figure test, Hopkins verbal learning test, Boston naming test, digit span, and digit substitution test.

Statistical analysis

All analyses were done with R, 33 unless otherwise stated. To assess representativeness of included participants we examined differences between included and excluded participants using t tests of means (continuous variables) or χ 2 tests of independence (categorical variables). According to variable type, we used means (standard deviations), medians (interquartile ranges), or numbers (percentages) to summarise sociodemographic and clinical measures for included participants who were split by safe versus unsafe average alcohol use averaged over all phases, on the basis of UK contemporary (pre-2016) guidelines. Significant differences between safe and unsafe drinkers in continuous variables were tested with t tests of means (normally distributed) or Wilcoxon rank sum tests (non-normally distributed), and in binary categorical variables (and mini-mental state examination, Montreal cognitive assessment, and Framingham stroke risk score, which have lower and upper bounds) with Fisher’s exact test of proportions. In view of small group numbers (<5) for social class, we performed a simulation test to estimate group differences. 34 Weekly consumption of alcohol (units and grams) was described with means, standard deviations, medians, and interquartile ranges.

We examined alcohol trends over time using mixed effects modelling, with time from study baseline (phase 1) as the independent variable and alcohol consumption (units/week) as the dependent variable. This method accounts for missing data and correlation of repeated measures (in this case alcohol use). We calculated intercepts (baseline consumption) and slopes (trends over study) for each participant. The ability of other variables to predict longitudinal trends of alcohol consumption was tested by inclusion of the following in the mixed effects model: age, sex, education, premorbid IQ, social class, Framingham risk score (a composite measure including smoking, cardiovascular disease or diabetes, cardiovascular drugs), exercise frequency, club attendance, voluntary work, visits with friends and family, lifetime history of major depressive disorder on the structured clinical interview for DSM IV (SCID) (yes-2/no-1), and current psychotropic drugs (yes-2/no-1).

We included mean alcohol consumption (units/week) across all study phases as an independent variable in voxel based morphometry (grey matter density as dependent variable) and tract based spatial statistics analyses (FA/MD/RD/AD as dependent variable). Voxel-wise, we applied a generalised linear model (GLM) using permutation based non-parametric testing (randomise), 35 correcting for multiple comparisons across space (threshold-free cluster enhancement, TFCE).

We used two post hoc tests to confirm the associations between alcohol consumption and hippocampal size after the voxel based morphometry analysis. Firstly, we used logistic regression to calculate odds ratios for left and right hippocampal atrophy versus no atrophy (visual atrophy ratings based on a cut off of 0/1 on the Scheltens scale), 31 given average alcohol consumption across study phases. The latter was categorised as abstinent (<1 unit, reference group), 1 to <7 units, 14 to <21 units, 21 to <30 units, and >30 units a week. Secondly, we performed multiple linear regression with hippocampal volume (extracted from FIRST (an automated segmentation/registration tool), adjusted for intracranial volume and transformed by squaring to normalise the residuals) as the dependent variable and alcohol consumption as an independent variable.

In all analyses with a brain measure as the dependent variable, we included the following potential confounding variables (identified from knowledge of the literature) as independent variables: age, sex, premorbid IQ, education, social class, Framingham risk score, current psychotropic drugs (number), lifetime history of major depressive disorder (structured clinical interview: yes-2/no-1), exercise frequency, club attendance, voluntary work, and visits with friends and family. In the subset with data on personality traits (n=179), analyses were additionally adjusted for impulsiveness.

We used mixed effects models to model longitudinal cognitive data. For count data (word recall from list of 20: “memory”) we used a binomial regression and for lexical and semantic fluency (performed within a certain time) we used Poisson regression. The following fixed effects were included: time from study baseline, average alcohol consumption across the study (abstinent (reference group, <1 unit weekly), 1- <7, 7- <14, 14- <21, >21), age, sex, education, social class, premorbid IQ, and Framingham stroke risk score. To test whether cognitive decline significantly differed between abstainers and those with higher alcohol intakes, we added interaction terms between time and alcohol category. Contrasts between other categories of drinking were also checked to test for significant differences in cognitive decline—for example, those drinking 1-<7 versus >21 units. We used Wald tests, 36 37 estimating the overall effect of all interactions between alcohol and time on the models, to test the null hypothesis that rates of cognitive decline did not differ between alcohol categories. Learning effects have been well demonstrated when the same cognitive test is presented more than once to a participant, which in our study could obscure true cognitive decline. In an attempt to control for this we added a dummy variable to code for the first time the test was taken (First). We also dummy coded for the test being performed at Oxford (Oxford), as there was an atypically short time interval between phase 11 and the last measurement point, which we hypothesised could result in an increased learning effect. We included interaction terms for FSIQ*First and FSIQ*Oxford to check if learning effects differ with premorbid IQ. Participant identification was included as a random effect. Usual diagnostic checks were performed on the models. The resulting coefficients from binomial regression equate to log(odds) and from Poisson regression to log(Poisson mean count). Exponentiated estimates are reported in the appendix. Regression coefficients were converted into interpretable differences in lexical decline per year compared with abstainers by: 100*(1−(exp (estimates)). Models were visually presented with graphs to predict trends in cognitive test scores over the study for a “typical” participant: male, mean age 70, 15 years’ education, social class I, IQ 118, and Framingham stroke risk score 10%.

We fitted regression models to check whether average alcohol consumption over the study (independent variable) predicted cross sectional performance on a range of memory tests (dependent variable) performed at the study end point. Age, sex, education, and premorbid IQ were included as covariates. When the test score represented a continuous variable, we used multiple linear regression. For count data (such as digit coding), we initially fitted Poisson regression and checked for over-dispersion. If this was found, we used a negative binomial model. For the remainder of the tests, where the upper score is bounded, we initially fitted regression models using binomial distributions. If over-dispersion was in evidence we performed a folded transformation and checked for approximate normality using Q-Q plots of residuals. The same models were re-fitted with and without alcohol consumption, and a hypothesis test (likelihood ratio) was performed. Calculated P values were used to test whether alcohol made a significant difference to the model.

Structural equation modelling (SEM; Amos 24 for Windows) was used post hoc for hypothesis testing and to generate fit statistics for models of relations between alcohol use, brain measures, and cognitive decline. This modelling allows simultaneous analysis of multiple variables in one model, and time series with auto-correlated errors. The hypothesised underlying structure of the model was constructed following the voxel based morphometry, tract based spatial statistics, and mixed effects analyses, with average alcohol consumption as an exogenous variable, hippocampal volume, corpus callosum mean diffusivity (generally the most sensitive measure of loss of white matter integrity), and decline in lexical fluency (slopes from mixed effects model) included as endogenous variables (with latent variables to account for measurement error). We modelled covariance of alcohol with sex and IQ and between brain measures. The model was improved by iteratively eliminating paths with P>0.1 and monitoring of the successive improvement of the model fits statistics (χ 2 , comparative fit index, root mean square error of approximation, and the Tucker-Lewis index) until we identified the most parsimonious model.

In all analyses, results were judged significant if the adjusted P value was <0.05. Bootstrapping was performed to derive 95% confidence intervals for estimates.

Patient involvement

Participants were from the Whitehall II cohort. No patients were involved in setting the research question or the outcome measures, nor were they involved in the design, recruitment, or conduct of the study. No patients were asked to advise on interpretation or writing up of results. Results were disseminated to the study participants in abstract format and as presentations at the 30th anniversary day for the Whitehall II cohort.

Participants/descriptive data

Sociodemographic, health, and lifestyle data are reported for the 527 included participants, separated into alcohol consumption groups (table 1 ⇓ ). Twenty three participants were excluded from the voxel based morphometry and visual ratings analyses on the basis of structural brain abnormalities, poor quality images, or missing confounder data (fig 2 ⇑ ). A further 16 were excluded from the tract based spatial statistics analysis because of missing or poor quality diffusion tensor images. Excluded participants did not significantly differ from those included on any of the reported characteristics (data available on request). There was a higher proportion of men, and participants were slightly less educated, with higher blood pressure and lower measures of depressive symptoms compared with the larger Whitehall II cohort (see appendix table A). Mean age was 43.0 (SD 5.4) at the start of the study (appendix table B). Unsafe drinkers differed from safe drinkers by having a higher premorbid IQ, a higher percentage of men and smokers, and higher Framingham risk scores (table 1 ⇓ ).

Baseline (phase 1 unless otherwise indicated) summary characteristics of 527 participants (unless marked) included in analysis by safe (<14 units/week for women, <21 units/week for men) and unsafe alcohol consumption, defined by contemporaneous (pre-2016) UK Department of Health guidelines, on average over study duration

• View inline

Median alcohol consumption across study phases (fig 3 ⇓ and appendix table B) was 11.5 units (85.8 g) a week (interquartile range 6.2-18.8 units (51.7-154.3 g)) for men and 6.4 units (51.4 g) a week (2.8-11.9 units (22.7-103.6 g)) for women. Weekly alcohol intake did not significantly increase over the phases of the study for the group as a whole (change in weekly alcohol units per 10 years of follow-up 0.15, 95% confidence interval −0.21 to 0.51; P=0.4), but trends over time correlated with baseline intake (intercepts and slopes correlated negatively ( r = −0.43, 95% confidence interval −0.50 to −0.36)—that is, those drinking more at baseline tended to lower their consumption more over the course of the study, a finding consistent with regression to the mean. Male sex (difference in weekly alcohol units compared with women 4.89, 2.54 to 7.19; P<0.001) and higher premorbid IQ (change in weekly alcohol units for every 1 IQ point 0.18, 0.06 to 0.30; P=0.004) predicted higher baseline consumption but not changes in consumption with time. Other sociodemographic and clinical factors were not related to consumption. Average alcohol use over the study was over “safe limits” in 13.6% women and 20.0% men, as judged by pre-2016 UK guidelines (>21 units (168 g)/week for men, >14 units (112 g)/week for women), and 40.3% as judged by the 2016 revised UK guidelines (>14 units (112 g)/week for men and women) (see appendix for consumption data for single phases). Scores on the CAGE questionnaire were below the sensitive screening cut off of 2 28 for all participants at all Whitehall II phases (appendix table C).

Fig 3  Frequency distribution of alcohol consumption on average across study by sex

Alcohol and brain structure

Higher alcohol use was associated with reduced grey matter density, hippocampal atrophy, and reduced white matter microstructural integrity.

Grey matter

Average alcohol consumption over the study (units/week) was negatively correlated with grey matter density in the voxel based morphometry analyses, especially in hippocampi (fig 4 ⇓ ), even after adjustment for multiple potential confounders. Associations also extended anteriorly into the amygdalae. Frontal regions were unaffected.

Fig 4  Results of voxel based morphometry (corrected for threshold-free cluster enhancement (TFCE)): significant negative correlation between weekly alcohol units (average of all phases across study) and grey matter density in 527 participants. Adjusted for age, sex, education, premorbid IQ, social class, physical exercise, club attendance, social activity, Framingham stroke risk score, psychotropic drugs, and history of major depressive disorder

Compared with abstinence, higher alcohol consumption was also associated with increased odds of abnormally rated hippocampal atrophy (defined as score >0 on Scheltens visual rating scale; table 2 ⇓ ). This was a dose dependent effect. The highest odds were in those drinking in excess of 30 units a week (odds ratio 5.8, 95% confidence interval 1.8 to 18.6; P≤0.001), but odds of atrophy were higher compared with abstinence even in those drinking at moderate levels of 7-<14 units a week (3.4, 1.4 to 8.1; P=0.007). There was no protective effect (that is, reduced odds of atrophy) with light drinking (1-<7 units a week) over abstinence. Findings were similar in subanalyses of men alone but not in the smaller subgroup of women. The risk of right sided hippocampal atrophy was significantly greater at >14 alcohol units a week compared with abstinence, but for left sided atrophy at only >30 units a week.

Adjusted * odds ratios for left and right sided hippocampal atrophy on Scheltens visual rating score (reference based on abstainers), with average alcohol consumption (abstinence (<1 unit) is reference category) in 527 participants. Figures are numbers with hippocampal atrophy and total numbers in drinking category with odds ratios (95% confidence interval), and P values

Mean hippocampal volumes (raw and adjusted for intracranial volume) were within the range cited in the literature (appendix table D) 38 39 40 and correlated with visual ratings of hippocampal atrophy (Spearman’s r =−0.4; P<0.001). Consistent with voxel based morphometry and visual ratings findings, alcohol consumption independently predicted FIRST-extracted hippocampal volume (%ICV) (table 3 ⇓ ). Exclusion of the three individual highest drinkers (>60 units weekly) did not substantially change the results (appendix table E). In the subset of participants for whom personality trait data were available from phase 1 (n=179), additionally adjustment for the analysis for trait impulsivity did not alter the findings.

Multiple linear regression results, with squared hippocampal volume (% of intracranial volume) as dependent variable and average weekly alcohol consumption across study as independent variable

White matter

Higher average alcohol consumption across the study was inversely associated with white matter integrity (fig 5 ⇓ ), reflected by lower corpus callosum fractional anisotropy and higher radial, axial and mean diffusivity. These associations were focused on the anterior corpus callosum (genu and anterior body, fig 5 ⇓ ).

Fig 5  Tract based spatial statistics results (corrected for threshold-free cluster enhancement, TFCE) showing negative correlation between average alcohol across study (all phases) and fractional anisotropy, and positive correlations with radial diffusivity, mean diffusivity, and axial diffusivity in 511 participants. Adjusted for age, sex, education, premorbid IQ, social class, physical exercise, club attendance, social activity, Framingham stroke risk score, psychotropic drugs, and history of major depressive disorder

Alcohol and cognitive function

Higher alcohol consumption over the study predicted faster decline on lexical fluency but not semantic fluency or word recall (fig 6 ⇓ ). Those drinking 7-<14, 14-<21, and >21 units a week declined faster in terms of lexical scores than abstainers. This effect was independent of age, sex, premorbid IQ, education, social class, and Framingham stroke risk score. The size of the difference can be interpreted as follows: people drinking 7-<14 units experienced a 0.5% greater reduction from their baseline in lexical fluency per year (14% over 30 years), those drinking 14-<21 units 0.8% greater per year (17% over 30 years), and those drinking >21 units 0.6% per year (16% over 30 years) than abstainers (appendix table F). Though the three categories of higher consumption (7-<14, 14-<21, and >21 units/week) showed significantly greater decline than abstainers, the only significant difference in trends between these three groups was between those drinking 14-21 units and those drinking 7-14 units (14-21 units experience 0.3% faster decline per year; P=0.02). There was no evidence to support light drinkers being relatively protected from cognitive decline compared with abstainers. Overall results of tests examining the question of whether rates of cognitive decline are linked to alcohol were significant (after multiple comparisons correction) for lexical fluency (χ 2 =14.4; P=0.006) but not semantic fluency (χ 2 =10.0; P=0.04) or memory recall (χ 2 =9.8; P=0.04).

Fig 6  Predicted longitudinal change in cognitive test scores (lexical and semantic fluency, word recall “memory”) for man of mean age (70) and premorbid IQ (118), median education (15 years), social class I and Framingham stroke risk score (10%) according to average alcohol consumption (weekly units). Predictions made on basis of mixed effects models with cognitive testing performed at phases 3, 5, 7, 9, and 11 and time of scan

We found evidence of learning effects on lexical and categorical fluency tests (P≤0.01), such that the second time a participant was presented with a test they performed better. This learning effect was predicted by premorbid IQ (First*premorbid IQ P=0.002-0.02).

There was a trend towards higher baseline performance on lexical fluency and memory recall in those drinking compared with abstainers (appendix table F), but these findings did not reach significance after correction for multiple testing.

We did not find any significant relations between alcohol consumption and cross sectional performance on cognitive tests performed at the time of scanning (a summary of cognitive test performance and its relation to alcohol is given in appendix table H).

Modelling alcohol consumption and brain structure and function

To see how alcohol consumption and the associated brain regions interacted with cognitive decline, we used structural equation modelling. Hippocampal volume and corpus callosum mean diffusivity were included as exogenous variables. Age, sex, and premorbid FSIQ were also incorporated.

Removal of regression arrows from age, sex, premorbid IQ, and hippocampal volume to lexical fluency decline improved the model fit. Alcohol consumption independently predicted decline in lexical fluency. The final parsimonious model explained 21% of corpus callosum mean diffusivity, 14% of right hippocampal volume, and 2% of lexical fluency decline variance (fig 7 ⇓ , table 4 ⇓ ), with good model fit. Alcohol consumption (in addition to age) predicted smaller hippocampal volume and greater corpus callosum mean diffusivity. Through its relation with corpus callosum mean diffusivity, and through a direct path, increased alcohol consumption predicted faster decline of lexical fluency.

Fig 7  Final parsimonious structural equation model illustrating relations among alcohol consumption (average across study phases, as fraction of 100 units weekly), hippocampal volume (average, %intracranial volume), corpus callosum mean diffusivity (as multiplicative of 1000), decline in lexical fluency (slopes), and age in 511 participants. Values on arrows represent unit changes in dependent variable for 1 unit increase in predictor. Model explained 21% of corpus callosum mean diffusivity, 14% of hippocampal variance, and 2% of lexical fluency decline variance (R 2 ). Model fit: χ 2 =5.6, df=4, P=0.23, root mean square error of approximation=0.03, comparative fit index=0.99, Tucker-Lewis index=0.97

Parameter estimates for paths in final structural equation model (fig 6 ⇑ ), with their bias corrected 95% confidence intervals and P values, in 511 participants

Principal findings

We have found a previously uncharacterised dose dependent association between alcohol consumption over 30 years of follow-up and hippocampal atrophy, as well as impaired white matter microstructure. Additionally, higher alcohol consumption predicted greater decline in lexical fluency but not in semantic fluency or word recall. There was no evidence of a protective effect of light drinking over abstinence on brain structure or function. The hippocampal findings were consistent between the brain-wide voxel based approach, automatically extracted volumes, and clinical visual ratings of hippocampal atrophy. The relation was dose dependent, and increased odds of hippocampal atrophy were found even in moderate drinkers (14-<21 units/week in men). The association between alcohol consumption and white matter microstructure in non-dependent drinkers is also novel and seemed to be driven by greater radial relative to axial diffusivity.

Strengths and limitations

Strengths of this study are the 30 year longitudinal data on alcohol consumption and the detailed available data on confounders. Additional strengths include the availability of a large amount of MRI data and the advanced methods of imaging analysis. Grey matter findings were replicated with a voxel based approach, automated hippocampal volumes, and visual ratings. Visual atrophy ratings are known to correlate closely with automated methods (own data) and are more applicable to clinical settings. 41 In large neuroimaging studies, automatic segmentation is widespread. 42 43 The automated approach we use (FIRST) has been shown to give accurate and robust results. 44

When interpreting these results, some caveats are necessary. While the sample comprised people living in the community, it might not be representative of the wider UK population. Most participants were educated and middle class men. The hippocampal atrophy associations we found in the total sample were replicated in men alone but not in women. This could reflect a lower power to detect an effect in women, in part because the sample was dominated by men (a reflection of the sex disparity in the civil service in the 1980s) and in part because few of the included women drank heavily. This is an observational study as long term alcohol use cannot be randomised. The Rosenthal effect could have influenced participants to lead healthier lifestyles as they were enrolled in the Whitehall II “stress and health” study. Data on alcohol use were self reported, and participants could have underestimated their drinking, though the longitudinal rather than cross sectional approach often taken in other reported studies might minimise this, 27 and the percentage of people drinking “unsafely” was comparable with that reported elsewhere. 45 46 47 We used the CAGE screening instrument to identify alcohol dependence as it is well validated. 28 48 There were 75 (14.2%) individuals with missing CAGE data from at least one phase, and we cannot exclude the possibility that we have included some people who were alcohol dependent at points during the study period. All included individuals, however, had at least three (out of a total of five) CAGE measurements, and individuals with incomplete CAGE data on average drank significantly less than those with complete data (on a t test of means of 13.1 (SD 10.3) v 8.5 (SD 8.8) (P<0.001). Additionally, some participants reported drinking high levels of alcohol while screening negative on the CAGE, indicating a further possible inclusion of people with an alcohol use disorder in the sample. Increased odds of hippocampal atrophy and faster lexical fluency decline, however, were found even in those drinking moderate amounts. Although the alcohol and cognitive data were longitudinal, the analyses with MRI measures were cross sectional, raising the possibility that the associations between brain structure and alcohol were the result of a confounding variable. Longitudinal imaging over more than a couple of years adds further confounders as the physical scanner and imaging sequences are unlikely to be the same because of developments in MRI science, making results difficult to interpret. While efforts have been made to control for multiple potential sources of confounding, residual confounding from unmeasured sources is conceivable. To produce the adjusted associations we found, however, any uncontrolled confounders would need to be associated with both alcohol consumption and risk of brain abnormalities and unrelated to the multiple factors we controlled for. We cannot exclude the possibility, of unlikely face validity, that those with hippocampal atrophy at study baseline were more likely to drink more. Multiple testing and the possibility of a false positive is a concern when cognitive decline on three tests is performed. The small P values (range 0.015-0.004) for lexical decline according to differing alcohol consumption, which reach significance with a strict Bonferroni correction (that is, a reduced significance threshold of P<0.017), however, make this unlikely. In contrast, we cannot be as confident about the differences in baseline cognition for drinkers compared with abstainers (P=0.03).

Finally, we fitted a structural equation model for alcohol, brain, and cognitive data that was defined post hoc. As such, results of previous analyses affected the choice of included variables meaning that the fit of the model might be overoptimistic.

Comparison with other studies

On average, 20% of men and 14% of women were drinking above pre-2016 UK guidelines (>21 units/>14 units/week, respectively). Other studies vary in reported rates of heavy drinking, but our rates are comparable. 45 46 Alcohol consumption might vary with country, as highlighted by a study using the WHO global alcohol database. 47

Hippocampal atrophy is a sensitive and relatively specific marker of Alzheimer’s disease, 49 though it has also been reported in chronic alcoholics. 19 50 The brain regions most vulnerable to alcohol abuse are said to be the frontal lobes. 21 In our sample, higher but non-dependent alcohol use was not associated with subsequent frontal brain atrophy or impaired cognition. Only the study by Den Heijer and colleagues has reported hippocampal findings in non-dependent drinkers. 51 This used a manual tracing rather than voxel based or visual rating approach to estimate hippocampal size. They reported a protective effect of moderate alcohol intake compared with abstinence, which conflicts with our results. 19 Alcohol consumption, however, was determined cross sectionally, making it difficult to exclude reverse causation. In contrast, because of the longitudinal cognitive component of our study we could show an association between higher alcohol consumption and cognitive decline. Additionally, several known confounders of hippocampal size, such as depression, were not controlled for in the Den Heijer study. 51 Other studies in non-dependent drinkers have reported either no effect 52 53 or a negative correlation with global grey matter but not hippocampal atrophy. 17 18 In contrast with our first hypothesis and the findings of some other studies, 11 12 19 54 we observed no evidence of a protective effect of light drinking compared with abstinence on brain structure or cognitive function. Previous studies did not control for (premorbid) IQ, 11 12 and only a few for socioeconomic class. 55 56 57 The observed protective effect could be due to confounding as we and others found a positive association between alcohol intake and IQ. 58 These factors separately predict better performance on cognitive tests. Supporting our second hypothesis, we found heavier alcohol consumption to be associated with adverse brain outcomes. The biological mechanism for this is unclear. Ethanol and acetaldehyde (a metabolite) are neurotoxic 59 and cause reduced numbers 60 61 and morphological changes in hippocampal neurones in animal models. 62 Associated thiamine and folate deficiency, 63 repeated head trauma, cerebrovascular events, liver damage, and repeated intoxication and withdrawal have also been implicated in more severe drinkers. The risk of hippocampal atrophy might be stronger and at lower levels of alcohol consumption for the right side. More severe hippocampal atrophy on the right has been described in those at higher risk of Alzheimer’s disease (asymptomatic ApoE4 homozygotes), 64 as well as in those with mild cognitive impairment or Alzheimer’s disease. 65 We found no structural laterality in associations with cognitive function. The literature on this is scarce and conflicting. Stronger associations between right hippocampal volume and visuospatial memory have been reported. 66

The voxel based morphometry analysis also showed associations between increased alcohol consumption and reduced grey matter density in the amygdala. This result could not be confirmed with other methods as automated segmentation of these regions was unreliable, and we are unaware of any reliable visual atrophy rating scales. Amygdala atrophy has been described in those with Alzheimer’s disease 67 and is implicated in preclinical models of alcohol misuse, 68 alcohol abuse relapse, 69 and in abstinent alcoholics, 70 though others have found no association with lower levels of consumption. 53

In animals, radial diffusivity reflects differences in myelination. 71 72 Previous studies have highlighted the corpus callosum as an area affected in fetal alcohol syndrome 73 and in chronic alcoholism in Marchiafava-Bignami disease. 74 75 One study reported increased mean diffusivity in the amygdala in a post hoc analysis of female non-dependent drinkers. 25 We are not aware of any studies investigating microstructural changes in white matter in moderate drinkers using a data driven skeletonised tract approach to diffusion tensor images, such as tract based spatial statistics. Alternative voxel-wise methods could compromise optimal analysis of multiple participants as there are alignment problems causing potential difficulties with interpretation of voxel-wise statistics. 32

Participants drinking higher levels of alcohol over the study experienced a faster decline of lexical fluency compared with abstainers. Lexical fluency involves selecting and retrieving information based on spelling (orthography) and has characteristically been associated with frontal executive function, 76 in contrast with semantic fluency, which could depend more on temporal lobe integrity. 77 The distinction might not be as clear cut, however, as functional networks overlap. 78 The inverse relation between alcohol consumption and lexical decline was perhaps unsurprising given the frontal predominance of the negative associations with white matter integrity. We suggest two possibilities for the lack of more widespread associations with cognition, particularly with semantic fluency and short term memory decline, given the structural brain findings (hippocampal atrophy). Firstly, there are clear practice effects over the study—that is, at least some participants improve their performance after repeated testing, and this is positively associated with premorbid IQ. This might be greater for the semantic compared with lexical fluency tests. Variables predicting the ability to learn could be different from those protecting against cognitive impairment because of a neurodegenerative process. Though we attempted to control for both IQ and learning effects, this might be insufficient to remove the confounding effect if a third variable, such as diet, mediates the relation between IQ and learning but is not in the model. Secondly, the brain changes might reflect an intermediate phenotype, and cognitive change is not yet evident. It is now well documented that hippocampal atrophy precedes symptoms in those with Alzheimer’s dementia by several years, 79 so a similar phenomenon in alcohol related changes is plausible.

Conclusions and policy implications

Prospective studies of the effects of alcohol use on the brain are few, and replication of these findings in other populations will be important. Alcohol consumption for individuals was remarkably stable across the study phases. This sample was therefore underpowered to detect differences in those considerably changing their intake from others who drink consistently. Investigations with larger numbers are needed to clarify whether there are graded risks between short versus long periods of higher alcohol consumption.

The finding that alcohol consumption in moderate quantities is associated with multiple markers of abnormal brain structure and cognitive function has important potential public health implications for a large sector of the population. For example, in our sample nearly half of the men and a quarter of the women were currently drinking in this range. Additionally, drinking habits were remarkably stable over a 30 year period, suggesting that risky drinking habits might be embarked on in midlife. Recommended guidelines for drinking remained unchanged in the UK from 1987 until 2016. Our findings support the recent reduction in UK safe limits and call into question the current US guidelines, which suggest that up to 24.5 units a week is safe for men, as we found increased odds of hippocampal atrophy at just 14-21 units a week, and we found no support for a protective effect of light consumption on brain structure. Alcohol might represent a modifiable risk factor for cognitive impairment, and primary prevention interventions targeted to later life could be too late.

What is already known on this topic

Heavy drinking is associated with Korsakoff’s syndrome, dementia, and widespread brain atrophy

While smaller amounts of alcohol have been linked to protection against cognitive impairment, few studies have examined the effects of moderate alcohol on the brain

Previous studies have methodological limitations especially regarding the lack of prospective alcohol data, have been conflicting, and have failed to provide a convincing neural correlate

What this study adds

Compared with abstinence, moderate alcohol intake is associated with increased risk of adverse brain outcomes and steeper cognitive decline in lexical fluency

The hippocampus is particularly vulnerable, which has not been previously linked negatively with moderate alcohol use

No protective effect was found for small amounts of alcohol over abstinence, and previous reports claiming a protective effect of light drinking might have been subject to confounding by associations between increased alcohol and higher social class or IQ

We thank the Whitehall II cohort participants for their time and D Lunn for statistical advice to the corresponding author.

Contributors: AT and CLA planned the study and acquired and analysed data. VV and PF analysed data. EZ and AM acquired data. NF and CS acquired and analysed data. AS-M, CEM, and MK also planned the study. KPE planned the study and analysed data. All authors contributed towards writing the paper. AT, KPE, and MK are guarantors .

Funding: The study was funded by UK Medical Research Council (G1001354; KPE), the Gordon Edward Small’s Charitable Trust (SC008962; KPE), and the HDH Wills 1965 charitable trust (charity No: 1117747; KPE). MK was supported by the Medical Research Council (K013351) and NordForsk.

Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf and declare: grant support for the submitted work is detailed above; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

Ethical approval: This study was approved as part of a larger study (Predicting MRI abnormalities with longitudinal data of the Whitehall II sub- study; MSD/IDREC/C1/2011/71) by the University of Oxford medical sciences interdivisional research ethics committee.

Data sharing: Policy referenced on: https://www.psych.ox.ac.uk/research/neurobiology-of-ageing/research-projects-1/whitehall-oxford . Data will be shared a period of two years after collection to allow the research group and collaborators time for analysis and publication. Reference to Whitehall II Data sharing policy here: http://www.ucl.ac.uk/whitehallII/data-sharing .

Transparency: The lead authors affirm that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: http://creativecommons.org/licenses/by/4.0/ .

• ↵ Organization WH. Global status report on alcohol and health. World Health Organization, 2014 .
• ↵ Connor JP, Haber PS, Hall WD. Alcohol use disorders. Lancet 2015 ; 387 : 988 - 98 . pmid:26343838 . OpenUrl
• ↵ Leon DA, McCambridge J. Liver cirrhosis mortality rates in Britain from 1950 to 2002: an analysis of routine data. Lancet 2006 ; 367 : 52 - 6 . doi:10.1016/S0140-6736(06)67924-5   pmid:16399153 . OpenUrl
• ↵  HM Government. The government’s alcohol strategy. Stationary Office, 2012 .
• ↵ Stampfer MJ, Colditz GA, Willett WC, Speizer FE, Hennekens CH. A prospective study of moderate alcohol consumption and the risk of coronary disease and stroke in women. N Engl J Med 1988 ; 319 : 267 - 73 . doi:10.1056/NEJM198808043190503   pmid:3393181 . OpenUrl
• ↵ Rimm EB, Williams P, Fosher K, Criqui M, Stampfer MJ. Moderate alcohol intake and lower risk of coronary heart disease: meta-analysis of effects on lipids and haemostatic factors. BMJ 1999 ; 319 : 1523 - 8 . doi:10.1136/bmj.319.7224.1523   pmid:10591709 . OpenUrl
• ↵ Cao Y, Willett WC, Rimm EB, Stampfer MJ, Giovannucci EL. Light to moderate intake of alcohol, drinking patterns, and risk of cancer: results from two prospective US cohort studies. BMJ 2015 ; 351 : h4328 . OpenUrl
• ↵ 2015-2020 Dietary Guidelines for Americans. 8th ed. 2015. https://health.gov/dietaryguidelines/2015/guidelines/
• ↵ Bagnardi V, Rota M, Botteri E, et al. Alcohol consumption and site-specific cancer risk: a comprehensive dose-response meta-analysis. Br J Cancer 2015 ; 112 : 580 - 93 . doi:10.1038/bjc.2014.579   pmid:25422909 . OpenUrl
• ↵  American Psychiatric Association. Diagnostic and statistical manual of mental disorders . 4th ed . 2000 .
• ↵ Mukamal KJ, Kuller LH, Fitzpatrick AL, Longstreth WT Jr, , Mittleman MA, Siscovick DS. Prospective study of alcohol consumption and risk of dementia in older adults. JAMA 2003 ; 289 : 1405 - 13 . doi:10.1001/jama.289.11.1405   pmid:12636463 . OpenUrl
• ↵ Ruitenberg A, van Swieten JC, Witteman JC, et al. Alcohol consumption and risk of dementia: the Rotterdam Study. Lancet 2002 ; 359 : 281 - 6 . doi:10.1016/S0140-6736(02)07493-7   pmid:11830193 . OpenUrl
• ↵ Cleophas TJ. Wine, beer and spirits and the risk of myocardial infarction: a systematic review. Biomed Pharmacother 1999 ; 53 : 417 - 23 . doi:10.1016/S0753-3322(99)80121-8   pmid:10554677 . OpenUrl
• ↵ Berger K, Ajani UA, Kase CS, et al. Light-to-moderate alcohol consumption and the risk of stroke among U.S. male physicians. N Engl J Med 1999 ; 341 : 1557 - 64 . doi:10.1056/NEJM199911183412101   pmid:10564684 . OpenUrl
• ↵ Gu Y, Scarmeas N, Short EE, et al. Alcohol intake and brain structure in a multiethnic elderly cohort. Clin Nutr 2014 ; 33 : 662 - 7 . doi:10.1016/j.clnu.2013.08.004   pmid:24011900 . OpenUrl
• ↵ Paul CA, Au R, Fredman L, et al. Association of alcohol consumption with brain volume in the Framingham study. Arch Neurol 2008 ; 65 : 1363 - 7 . doi:10.1001/archneur.65.10.1363   pmid:18852353 . OpenUrl
• ↵ Ding J, Eigenbrodt ML, Mosley TH Jr, et al. Alcohol intake and cerebral abnormalities on magnetic resonance imaging in a community-based population of middle-aged adults: the Atherosclerosis Risk in Communities (ARIC) study. Stroke 2004 ; 35 : 16 - 21 . doi:10.1161/01.STR.0000105929.88691.8E   pmid:14657449 . OpenUrl
• ↵ Mukamal KJ, Longstreth WT Jr, , Mittleman MA, Crum RM, Siscovick DS. Alcohol consumption and subclinical findings on magnetic resonance imaging of the brain in older adults: the cardiovascular health study. Stroke 2001 ; 32 : 1939 - 46 . doi:10.1161/hs0901.095723   pmid:11546878 . OpenUrl
• ↵ den Heijer T, Vermeer SE, van Dijk EJ, et al. Alcohol intake in relation to brain magnetic resonance imaging findings in older persons without dementia. Am J Clin Nutr 2004 ; 80 : 992 - 7 . pmid:15447910 . OpenUrl
• ↵ Sachdev PS, Chen X, Wen W, Anstey KJ. Light to moderate alcohol use is associated with increased cortical gray matter in middle-aged men: a voxel-based morphometric study. Psychiatry Res 2008 ; 163 : 61 - 9 . doi:10.1016/j.pscychresns.2007.08.009   pmid:18407470 . OpenUrl
• ↵ Kubota M, Nakazaki S, Hirai S, Saeki N, Yamaura A, Kusaka T. Alcohol consumption and frontal lobe shrinkage: study of 1432 non-alcoholic subjects. J Neurol Neurosurg Psychiatry 2001 ; 71 : 104 - 6 . doi:10.1136/jnnp.71.1.104   pmid:11413273 . OpenUrl
• ↵ de Bruin EA, Hulshoff Pol HE, Schnack HG, et al. Focal brain matter differences associated with lifetime alcohol intake and visual attention in male but not in female non-alcohol-dependent drinkers. Neuroimage 2005 ; 26 : 536 - 45 . doi:10.1016/j.neuroimage.2005.01.036   pmid:15907310 . OpenUrl
• ↵ Anstey KJ, Mack HA, Cherbuin N. Alcohol consumption as a risk factor for dementia and cognitive decline: meta-analysis of prospective studies. Am J Geriatr Psychiatry 2009 ; 17 : 542 - 55 . doi:10.1097/JGP.0b013e3181a2fd07   pmid:19546653 . OpenUrl
• ↵ Fukuda K, Yuzuriha T, Kinukawa N, et al. Alcohol intake and quantitative MRI findings among community dwelling Japanese subjects. J Neurol Sci 2009 ; 278 : 30 - 4 . doi:10.1016/j.jns.2008.11.007   pmid:19059611 . OpenUrl
• ↵ Sasaki H, Abe O, Yamasue H, et al. Structural and diffusional brain abnormality related to relatively low level alcohol consumption. Neuroimage 2009 ; 46 : 505 - 10 . doi:10.1016/j.neuroimage.2009.02.007   pmid:19233298 . OpenUrl
• ↵ Poikolainen K. Underestimation of recalled alcohol intake in relation to actual consumption. Br J Addict 1985 ; 80 : 215 - 6 . doi:10.1111/j.1360-0443.1985.tb03276.x   pmid:3860245 . OpenUrl
• ↵ Marmot MG, Smith GD, Stansfeld S, et al. Health inequalities among British civil servants: the Whitehall II study. Lancet 1991 ; 337 : 1387 - 93 . doi:10.1016/0140-6736(91)93068-K   pmid:1674771 . OpenUrl
• ↵ Mayfield D, McLeod G, Hall P. The CAGE questionnaire: validation of a new alcoholism screening instrument. Am J Psychiatry 1974 ; 131 : 1121 - 3 . pmid:4416585 . OpenUrl
• ↵ Health Do. How to keep health risks from drinking alcohol to a low level: public consultation on proposed new guidelines. Williams Lea for the Department of Health, 2016 .
• ↵ Filippini N, Zsoldos E, Haapakoski R, et al. Study protocol: The Whitehall II imaging sub-study. BMC Psychiatry 2014 ; 14 : 159 . doi:10.1186/1471-244X-14-159   pmid:24885374 . OpenUrl
• ↵ Scheltens P, Leys D, Barkhof F, et al. Atrophy of medial temporal lobes on MRI in “probable” Alzheimer’s disease and normal ageing: diagnostic value and neuropsychological correlates. J Neurol Neurosurg Psychiatry 1992 ; 55 : 967 - 72 . doi:10.1136/jnnp.55.10.967   pmid:1431963 . OpenUrl
• ↵ Smith SM, Jenkinson M, Johansen-Berg H, et al. Tract-based spatial statistics: voxelwise analysis of multi-subject diffusion data. Neuroimage 2006 ; 31 : 1487 - 505 . doi:10.1016/j.neuroimage.2006.02.024   pmid:16624579 . OpenUrl
• ↵ R Core Team. R: A Language and Environment for Statistical Computing. 2015.
• ↵ Mooney CZ. Monte Carlo simulation. Sage Publications, 1997 doi:10.4135/9781412985116 .
• ↵ Winkler AM, Ridgway GR, Webster MA, Smith SM, Nichols TE. Permutation inference for the general linear model. Neuroimage 2014 ; 92 : 381 - 97 . doi:10.1016/j.neuroimage.2014.01.060   pmid:24530839 . OpenUrl
• ↵ Bolker BM, Brooks ME, Clark CJ, et al. Generalized linear mixed models: a practical guide for ecology and evolution. Trends Ecol Evol 2009 ; 24 : 127 - 35 . doi:10.1016/j.tree.2008.10.008   pmid:19185386 . OpenUrl
• ↵ Molenberghs G, Verbeke G. Likelihood ratio, score, and Wald tests in a constrained parameter space. Am Stat 2007 ; 61 : 22 - 7 doi:10.1198/000313007X171322 . OpenUrl
• ↵ Brown ES, Hughes CW, McColl R, Peshock R, King KS, Rush AJ. Association of depressive symptoms with hippocampal volume in 1936 adults. Neuropsychopharmacology 2014 ; 39 : 770 - 9 . doi:10.1038/npp.2013.271   pmid:24220026 . OpenUrl
• ↵ Bis JC, DeCarli C, Smith AV, et al. Enhancing Neuro Imaging Genetics through Meta-Analysis Consortium Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium. Common variants at 12q14 and 12q24 are associated with hippocampal volume. Nat Genet 2012 ; 44 : 545 - 51 . doi:10.1038/ng.2237   pmid:22504421 . OpenUrl
• ↵ Thomas AG, Dennis A, Rawlings NB, et al. Multi-modal characterization of rapid anterior hippocampal volume increase associated with aerobic exercise. Neuroimage 2016 ; 131 : 162 - 70 . doi:10.1016/j.neuroimage.2015.10.090   pmid:26654786 . OpenUrl
• ↵ Harper L, Fumagalli GG, Barkhof F, et al. MRI visual rating scales in the diagnosis of dementia: evaluation in 184 post-mortem confirmed cases. Brain 2016 ; 139 : 1211 - 25 . doi:10.1093/brain/aww005   pmid:26936938 . OpenUrl
• ↵ Stein JL, Medland SE, Vasquez AA, et al. Alzheimer’s Disease Neuroimaging Initiative EPIGEN Consortium IMAGEN Consortium Saguenay Youth Study Group Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium Enhancing Neuro Imaging Genetics through Meta-Analysis Consortium. Identification of common variants associated with human hippocampal and intracranial volumes. Nat Genet 2012 ; 44 : 552 - 61 . doi:10.1038/ng.2250   pmid:22504417 . OpenUrl
• ↵ Hibar DP, Stein JL, Renteria ME, et al. Alzheimer’s Disease Neuroimaging Initiative CHARGE Consortium EPIGEN IMAGEN SYS. Common genetic variants influence human subcortical brain structures. Nature 2015 ; 520 : 224 - 9 . doi:10.1038/nature14101   pmid:25607358 . OpenUrl
• ↵ Patenaude B, Smith SM, Kennedy DN, Jenkinson M. A Bayesian model of shape and appearance for subcortical brain segmentation. Neuroimage 2011 ; 56 : 907 - 22 . doi:10.1016/j.neuroimage.2011.02.046   pmid:21352927 . OpenUrl
• ↵ Djoussé L, Levy D, Benjamin EJ, et al. Long-term alcohol consumption and the risk of atrial fibrillation in the Framingham Study. Am J Cardiol 2004 ; 93 : 710 - 3 . doi:10.1016/j.amjcard.2003.12.004   pmid:15019874 . OpenUrl
• ↵ Fuchs CS, Stampfer MJ, Colditz GA, et al. Alcohol consumption and mortality among women. N Engl J Med 1995 ; 332 : 1245 - 50 . doi:10.1056/NEJM199505113321901   pmid:7708067 . OpenUrl
• ↵ Rehm J, Rehn N, Room R, et al. The global distribution of average volume of alcohol consumption and patterns of drinking. Eur Addict Res 2003 ; 9 : 147 - 56 . doi:10.1159/000072221   pmid:12970583 . OpenUrl
• ↵ Beresford TP, Blow FC, Hill E, Singer K, Lucey MR. Comparison of CAGE questionnaire and computer-assisted laboratory profiles in screening for covert alcoholism. Lancet 1990 ; 336 : 482 - 5 . doi:10.1016/0140-6736(90)92022-A   pmid:1974998 . OpenUrl
• ↵ McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 2011 ; 7 : 263 - 9 . doi:10.1016/j.jalz.2011.03.005   pmid:21514250 . OpenUrl
• ↵ Sullivan EV, Marsh L, Mathalon DH, Lim KO, Pfefferbaum A. Anterior hippocampal volume deficits in nonamnesic, aging chronic alcoholics. Alcohol Clin Exp Res 1995 ; 19 : 110 - 22 . doi:10.1111/j.1530-0277.1995.tb01478.x   pmid:7771636 . OpenUrl
• ↵ Videbech P, Ravnkilde B. Hippocampal volume and depression: a meta-analysis of MRI studies. Am J Psychiatry 2004 ; 161 : 1957 - 66 . doi:10.1176/appi.ajp.161.11.1957   pmid:15514393 . OpenUrl
• ↵ Preti A, Muscio C, Boccardi M, Lorenzi M, de Girolamo G, Frisoni G. Impact of alcohol consumption in healthy adults: a magnetic resonance imaging investigation. Psychiatry Res 2014 ; 224 : 96 - 103 . doi:10.1016/j.pscychresns.2014.06.005   pmid:25172407 . OpenUrl
• ↵ Anstey KJ, Jorm AF, Réglade-Meslin C, et al. Weekly alcohol consumption, brain atrophy, and white matter hyperintensities in a community-based sample aged 60 to 64 years. Psychosom Med 2006 ; 68 : 778 - 85 . doi:10.1097/01.psy.0000237779.56500.af   pmid:17012533 . OpenUrl
• ↵ Stampfer MJ, Kang JH, Chen J, Cherry R, Grodstein F. Effects of moderate alcohol consumption on cognitive function in women. N Engl J Med 2005 ; 352 : 245 - 53 . doi:10.1056/NEJMoa041152   pmid:15659724 . OpenUrl
• ↵ Britton A, Singh-Manoux A, Marmot M. Alcohol consumption and cognitive function in the Whitehall II Study. Am J Epidemiol 2004 ; 160 : 240 - 7 . doi:10.1093/aje/kwh206   pmid:15257997 . OpenUrl
• ↵ Sabia S, Elbaz A, Britton A, et al. Alcohol consumption and cognitive decline in early old age. Neurology 2014 ; 82 : 332 - 9 . doi:10.1212/WNL.0000000000000063   pmid:24431298 . OpenUrl
• ↵ McGuire LC, Ajani UA, Ford ES. Cognitive functioning in late life: the impact of moderate alcohol consumption. Ann Epidemiol 2007 ; 17 : 93 - 9 . doi:10.1016/j.annepidem.2006.06.005   pmid:17027288 . OpenUrl
• ↵ Touvier M, Druesne-Pecollo N, Kesse-Guyot E, et al. Demographic, socioeconomic, disease history, dietary and lifestyle cancer risk factors associated with alcohol consumption. Int J Cancer 2014 ; 134 : 445 - 59 . doi:10.1002/ijc.28365   pmid:23824873 . OpenUrl
• ↵ Arendt T, Allen Y, Sinden J, et al. Cholinergic-rich brain transplants reverse alcohol-induced memory deficits. Nature 1988 ; 332 : 448 - 50 . doi:10.1038/332448a0   pmid:3352743 . OpenUrl
• ↵ Bengoechea O, Gonzalo LM. Effects of alcoholization on the rat hippocampus. Neurosci Lett 1991 ; 123 : 112 - 4 . doi:10.1016/0304-3940(91)90170-X   pmid:2062446 . OpenUrl
• ↵ Cadete-Leite A, Tavares MA, Pacheco MM, Volk B, Paula-Barbosa MM. Hippocampal mossy fiber-CA3 synapses after chronic alcohol consumption and withdrawal. Alcohol 1989 ; 6 : 303 - 10 . doi:10.1016/0741-8329(89)90087-6   pmid:2765199 . OpenUrl
• ↵ McMullen PA, Saint-Cyr JA, Carlen PL. Morphological alterations in rat CA1 hippocampal pyramidal cell dendrites resulting from chronic ethanol consumption and withdrawal. J Comp Neurol 1984 ; 225 : 111 - 8 . doi:10.1002/cne.902250112   pmid:6539344 . OpenUrl
• ↵ Langlais PJ, Savage LM. Thiamine deficiency in rats produces cognitive and memory deficits on spatial tasks that correlate with tissue loss in diencephalon, cortex and white matter. Behav Brain Res 1995 ; 68 : 75 - 89 . doi:10.1016/0166-4328(94)00162-9   pmid:7619308 . OpenUrl
• ↵ Tohgi H, Takahashi S, Kato E, et al. Reduced size of right hippocampus in 39- to 80-year-old normal subjects carrying the apolipoprotein E ϵ4 allele. Neurosci Lett 1997 ; 236 : 21 - 4 . doi:10.1016/S0304-3940(97)00743-X   pmid:9404942 . OpenUrl
• ↵ Shi F, Liu B, Zhou Y, Yu C, Jiang T. Hippocampal volume and asymmetry in mild cognitive impairment and Alzheimer’s disease: Meta-analyses of MRI studies. Hippocampus 2009 ; 19 : 1055 - 64 . doi:10.1002/hipo.20573   pmid:19309039 . OpenUrl
• ↵ Soininen HS, Partanen K, Pitkänen A, et al. Volumetric MRI analysis of the amygdala and the hippocampus in subjects with age-associated memory impairment: correlation to visual and verbal memory. Neurology 1994 ; 44 : 1660 - 8 . doi:10.1212/WNL.44.9.1660   pmid:7936293 . OpenUrl
• ↵ Cuénod C-A, Denys A, Michot J-L, et al. Amygdala atrophy in Alzheimer’s disease. An in vivo magnetic resonance imaging study. Arch Neurol 1993 ; 50 : 941 - 5 . doi:10.1001/archneur.1993.00540090046009   pmid:8363448 . OpenUrl
• ↵ Koob GF. Neuroadaptive mechanisms of addiction: studies on the extended amygdala. Eur Neuropsychopharmacol 2003 ; 13 : 442 - 52 . doi:10.1016/j.euroneuro.2003.08.005   pmid:14636960 . OpenUrl
• ↵ Wrase J, Makris N, Braus DF, et al. Amygdala volume associated with alcohol abuse relapse and craving. Am J Psychiatry 2008 ; 165 : 1179 - 84 . doi:10.1176/appi.ajp.2008.07121877   pmid:18593776 . OpenUrl
• ↵ Fein G, Landman B, Tran H, et al. Brain atrophy in long-term abstinent alcoholics who demonstrate impairment on a simulated gambling task. Neuroimage 2006 ; 32 : 1465 - 71 . doi:10.1016/j.neuroimage.2006.06.013   pmid:16872844 . OpenUrl
• ↵ Budde MD, Kim JH, Liang HF, et al. Toward accurate diagnosis of white matter pathology using diffusion tensor imaging. Magn Reson Med 2007 ; 57 : 688 - 95 . doi:10.1002/mrm.21200   pmid:17390365 . OpenUrl
• ↵ Song S-K, Sun S-W, Ramsbottom MJ, Chang C, Russell J, Cross AH. Dysmyelination revealed through MRI as increased radial (but unchanged axial) diffusion of water. Neuroimage 2002 ; 17 : 1429 - 36 . doi:10.1006/nimg.2002.1267   pmid:12414282 . OpenUrl
• ↵ Bookstein FL, Streissguth AP, Sampson PD, Connor PD, Barr HM. Corpus callosum shape and neuropsychological deficits in adult males with heavy fetal alcohol exposure. Neuroimage 2002 ; 15 : 233 - 51 . doi:10.1006/nimg.2001.0977   pmid:11771992 . OpenUrl
• ↵ Pfefferbaum A, Lim KO, Desmond JE, Sullivan EV. Thinning of the corpus callosum in older alcoholic men: a magnetic resonance imaging study. Alcohol Clin Exp Res 1996 ; 20 : 752 - 7 . doi:10.1111/j.1530-0277.1996.tb01682.x   pmid:8800395 . OpenUrl
• ↵ Harper C, Kril J, Daly J. Does a “moderate” alcohol intake damage the brain? J Neurol Neurosurg Psychiatry 1988 ; 51 : 909 - 13 . doi:10.1136/jnnp.51.7.909   pmid:3204399 . OpenUrl
• ↵ Birn RM, Kenworthy L, Case L, et al. Neural systems supporting lexical search guided by letter and semantic category cues: a self-paced overt response fMRI study of verbal fluency. Neuroimage 2010 ; 49 : 1099 - 107 . doi:10.1016/j.neuroimage.2009.07.036   pmid:19632335 . OpenUrl
• ↵ Baldo JV, Schwartz S, Wilkins D, Dronkers NF. Role of frontal versus temporal cortex in verbal fluency as revealed by voxel-based lesion symptom mapping. J Int Neuropsychol Soc 2006 ; 12 : 896 - 900 . doi:10.1017/S1355617706061078   pmid:17064451 . OpenUrl
• ↵ Costafreda SG, Fu CH, Lee L, Everitt B, Brammer MJ, David AS. A systematic review and quantitative appraisal of fMRI studies of verbal fluency: role of the left inferior frontal gyrus. Hum Brain Mapp 2006 ; 27 : 799 - 810 . doi:10.1002/hbm.20221   pmid:16511886 . OpenUrl
• ↵ Jack CR Jr, , Knopman DS, Jagust WJ, et al. Tracking pathophysiological processes in Alzheimer’s disease: an updated hypothetical model of dynamic biomarkers. Lancet Neurol 2013 ; 12 : 207 - 16 . doi:10.1016/S1474-4422(12)70291-0   pmid:23332364 . OpenUrl

• Open access
• Published: 13 February 2020

Long-term effects of alcohol consumption on cognitive function: a systematic review and dose-response analysis of evidence published between 2007 and 2018

• Sue E. Brennan   ORCID: orcid.org/0000-0003-1789-8809 1 ,
• Steve McDonald 1 ,
• Matthew J. Page 1 ,
• Jane Reid 1 ,
• Stephanie Ward 1 ,
• Andrew B. Forbes 1 &
• Joanne E. McKenzie 1  

Systematic Reviews volume  9 , Article number:  33 ( 2020 ) Cite this article

22k Accesses

29 Citations

6 Altmetric

Metrics details

Understanding the long-term health effects of low to moderate alcohol consumption is important for establishing thresholds for minimising the lifetime risk of harm. Recent research has elucidated the dose-response relationship between alcohol and cardiovascular outcomes, showing an increased risk of harm at levels of intake previously thought to be protective. The primary objective of this review was to examine (1) whether there is a dose-response relationship between levels of alcohol consumption and long-term cognitive effects, and (2) what the effects are of different levels of consumption.

The review was conducted according to a pre-specified protocol. Eligible studies were those published 2007 onwards that compared cognitive function among people with different levels of alcohol consumption (measured ≥ 6 months prior to first follow-up of cognition). Major cognitive impairment was excluded. Searches were limited to MEDLINE, Embase and PsycINFO (January 2007 to April 2018). Screening, data extraction, and risk of bias assessment (ROBINS-I) were piloted by three authors, then completed by a single author and checked by a second. Analyses were undertaken to identify and characterise dose-response relationships between levels of alcohol consumption and cognition. Certainty of evidence was assessed using GRADE.

We included 27 cohort studies (from 4786 citations). Eighteen studies examined the effects of alcohol consumption at different levels (risk of bias 16 serious, 2 critical). Ten studies provided data for dose-response analysis. The pooled dose-response relationship showed a maximum standardised mean difference (SMD) indicating slightly better cognition among women with moderate alcohol consumption compared to current non-drinkers (SMD 0.18, 95%CI 0.02 to 0.34, at 14.4 grams/day; 5 studies, very low certainty evidence), and a trivial difference for men (SMD 0.05, 95% CI 0.00 to 0.10, at 19.4 grams/day; 6 studies, very low certainty evidence).

Conclusions

Major limitations in the design and reporting of included studies made it impossible to discern if the effects of ‘lower’ levels of alcohol intake are due to bias. Further review of the evidence is unlikely to resolve this issue without meta-analysis of individual patient data from cohort studies that address biases in the selection of participants and classification of alcohol consumption.

Peer Review reports

Alcohol consumption is an established risk factor for a large number of health conditions, contributing to morbidity and premature death from cancers, cardiovascular disease, and liver disease [ 1 , 2 ]. Governments have attempted to mitigate these health impacts by providing guidelines for lower risk consumption of alcohol. However, uncertainty around the effects of light to moderate alcohol consumption has made it challenging to establish thresholds for minimising the lifetime risk of harm [ 2 ]. While light to moderate alcohol consumption has been associated with a protective effect on some outcomes (e.g. all-cause mortality, cardiovascular disease, and dementia), there is mounting evidence that these findings are an artefact of study design [ 2 , 3 , 4 ]. Recent research has helped elucidate the dose-response relationship between alcohol consumption and some of these outcomes showing that, rather than having protective effect, light to moderate alcohol intake is associated with an increased risk of stroke, other cardiovascular disease subtypes (excluding myocardial infarction), and all-cause mortality [ 1 , 2 ]. Comparable studies examining the dose-response relationship between alcohol consumption and long-term cognitive outcomes are lacking [ 5 , 6 ].

Rehm and colleagues recently reported an overview of twenty systematic reviews (published 2000-March 2018) that had examined the relationship between alcohol use and dementia or cognitive impairment [ 6 ]. Only one of the twenty reviews reported a dose-response analysis. The analysis showed an elevated risk of dementia when 38 g of alcohol or more is consumed per day, and a lower risk of dementia with ‘modest’ alcohol consumption (between 6 and 12.5 g per day) compared to other levels of intake [ 7 ]. Studies measuring other cognitive outcomes were excluded from Xu et al. Although the findings from Xu et al. are consistent with earlier systematic reviews (e.g. [ 8 , 9 , 10 ]), the recent evidence against any protective effect of alcohol on cardiovascular outcomes signals the need to closely examine the association between light to moderate alcohol intake and cognition. In particular, a dose-response analysis considering other cognitive outcomes is needed, together with a detailed assessment of the extent to which observed results may be explained by bias.

The current systematic review aims to address evidence gaps, examining the dose-response relationship between alcohol and mild cognitive impairment. We focus on the cumulative effects of lower levels of alcohol exposure on cognitive function—those effects arising from drinking over time (not a single occasion). Although these effects may be most evident after a longer period of exposure (typically, later in life), there is also a need to examine the potential for long-term effects on cognition arising from drinking alcohol early in life (up to the age of 25). This is because of the concerns that exposure to alcohol during this period of brain development may bring an increased risk of cognitive impairment [ 11 , 12 ]. The review was commissioned to inform an update of the 2009 Australian Guidelines to Reduce Health Risks from Drinking Alcohol (the Alcohol Guidelines) [ 13 ]. As such, it considers evidence published from 2007 onwards (i.e. subsequent to the evidence review conducted for the 2009 Alcohol Guideline).

The objectives of the review are to address the following questions.

Is there a dose-response relationship between levels of alcohol consumption and long-term cognitive effects for women and men? If so, what are the effects at different levels of consumption?

The different levels of alcohol consumption defined for the review were based on increments of a single standard Australian drink (10 g of alcohol). This standard is common to a number of other countries (e.g. France, Netherlands, New Zealand, Spain), with some countries having slightly lower (e.g. United Kingdom) or higher (e.g. Canada, United States) standards. The levels were the following:

Never drinking or very low-level drinking (0 to < 10 g/week)

≥ 10 g/week and < 10 g/day

≥ 10 g/day and < 20 g/day

≥ 20 g/day and < 30 g/day

≥ 30 g/day and < 40 g/day

≥ 40 g/day and < 50 g/day

Secondary objectives

Is the effect of alcohol consumption on long-term cognition modified by age, co-morbidities, or drug use?

What studies are available comparing the long-term effects of different patterns of alcohol consumption on cognition for women and men? What questions are addressed by these studies (in terms of populations, alcohol consumption patterns, and outcomes)?

Different patterns of consumption were defined inclusively for the review. Examples include different levels of per-occasion consumption of alcohol (e.g. infrequent “heavy” or “binge” drinking versus regular drinking within lower risk levels), different frequency of drinking, and different patterns of consumption over time. Since the literature on the effects of different patterns of alcohol consumption covers diverse questions, examining non-comparable patterns of intake, among different populations, these studies were summarised to map available evidence.

Methods for the review were pre-specified in a protocol, which was peer-reviewed prior to conducting the review (Additional file 1 ; Changes to protocol, Additional file 2 , Appendix 1). The review was not registered on PROSPERO due to plans for public consultation prior to wider dissemination. The methods reported in this review are based on the Cochrane Handbook for Systematic Reviews of Interventions [ 14 ], with modifications for undertaking a review of exposures. The GRADE approach is used to summarise and assess the certainty of evidence arising from the review (see ‘Summary of findings tables and assessment of certainty of the body of evidence’ section for details). GRADE methods are widely used in guideline development to ensure a systematic, transparent and common approach to interpreting results [ 15 ]. The review is reported in accordance with the PRISMA statement [ 16 , 17 ], with additional methods description based on the PRISMA-P statement [ 18 , 19 ].

Criteria for considering studies for this review

Types of participants.

General population

Studies that were limited to one or more of the following subgroups were eligible for inclusion:

People in specific age groups identified in the 2009 Alcohol guideline as potentially having a higher risk of harm from alcohol exposure than the general population. For example, children and young people (less than 18 years), young adults (18–25), older people (65 and over)

Women or men

We planned to report data and analyses from studies that met other eligibility criteria for the following subgroups.

People with existing health conditions (physical, mental or both)

People using licit and/or illicit drugs

People with a family history of alcohol dependence.

Studies restricted to one or more of these three subgroups were eligible only if the study explicitly aimed to examine the association between alcohol consumption and long-term cognition.

Types of exposure

Eligible studies were those examining different levels of alcohol consumption, patterns of alcohol consumption, or both.

Measurement methods and quantification

Studies were eligible irrespective of the methods used to measure alcohol exposure. We anticipated that these methods would vary across studies, but would include retrospective survey involving recall of alcohol consumption over different periods of life or intake diaries to measure current alcohol consumption. Single or repeated measures of exposure were eligible. Studies had to report alcohol consumption in units that allowed quantification of the average amount of alcohol consumed (e.g. grams or millilitres of pure alcohol) over a period of time (e.g. per day, week, month).

Timing of alcohol exposure measurement

The timing of measurement needed to match the study design features listed in ‘Types of studies’ section for a prospective design. Data collected on alcohol consumption, and used in analyses, had to be collected at least 6 months prior to the first follow-up measure of cognition. Concurrent measures of alcohol were accepted only in studies with multiple measures of alcohol over time, where the final measure was taken concurrently with a baseline (not follow-up) measure of cognition.

To account for differences in the methods used to measure alcohol exposure, we extracted data on the measurement methods and assessed potential biases that may arise through the method used.

Types of comparator exposure

For inclusion in the review, the comparator group must have been a different level or pattern of alcohol consumption.

For inclusion in the meta-analysis of different levels of alcohol consumption and the dose-response analysis, studies had to report results for either a ‘never’ drinker group or a ‘very low-level’ drinker group. We broadly defined ‘never’ drinkers as individuals that had never consumed a serve of alcohol (lifetime abstainers) or had consumed very little alcohol across their lifetime. Where lifetime consumption was not measured, we accepted current non-drinkers (e.g. based on consumption over the preceding 12 months), noting in data extraction and risk of bias assessment the potential for misclassification and contamination of a non-drinking group with former drinkers. A similar approach was taken to misclassification of occasional drinkers, where the recall period was such that occasional drinkers might be missed and incorrectly categorised as non-drinkers. We defined very low-level drinkers as those whose average alcohol consumption was 0 to < 10 g/week. The latter threshold reflects consumption of a single Australian standard drink (10 g of alcohol), and is common to a number of other countries (e.g. France, Netherlands, New Zealand, Spain).

We anticipated diversity across studies in the definition and composition of potentially eligible comparator groups (which may or may not be the referent group to which other categories of alcohol consumption were compared in each study) [ 20 ]. For example, across studies referent groups have been defined as never drinking [ 21 ], not drinking above a certain threshold (e.g. less than 1 unit of alcohol per week [ 22 ]), and not drinking over a defined period of time (e.g. less than 1 unit over the preceding 12 months [ 23 ]). Studies reporting a group with these or similarly low levels of alcohol consumption were eligible, irrespective of whether the group was used as the referent in the study.

Types of outcomes

Eligible studies were those that reported at least one measure of cognitive function (or performance), which is the primary outcome for this review. Studies must have assessed cumulative long-term effects of alcohol consumption on cognitive function (e.g. decline in function over time). We excluded studies that only examined acute effects (during intoxication or withdrawal), long-term effects arising from injury on a single drinking occasion (e.g. a traumatic brain injury sustained while intoxicated), and those where there was insufficient length of follow-up to examine the longer-term effects of cumulative exposure (< 6 months). While we did not set a minimum threshold for ‘long-term’, we considered the extent to which studies provided evidence of a sustained effect, and the duration of this effect, when interpreting results (see ‘Timing of outcome measurement’ section). We also excluded studies that only examined cognitive function as a predictor of alcohol-use behaviours (e.g. studies examining whether prior cognitive function led to heavy alcohol use).

Eligible outcomes were broadly categorised as follows.

Cognitive function

Global cognitive function

Domain-specific cognitive function (especially domains that reflect specific alcohol-related neuropathologies, such as psychomotor speed and working memory)

Clinical diagnoses of cognitive impairment

Mild cognitive impairment (also referred to as mild neurocognitive disorders)

These conditions were ‘characterised by a decline from a previously attained cognitive level’ ([ 5 ], p2675).

Major cognitive impairment (also referred to as major neurocognitive disorders; including dementia) was excluded.

We expected that definitions and diagnostic criteria would vary across studies, so we accepted a range of definitions as noted under ‘Methods of outcome assessment’ section. Table 1 provides an example of specific domains of cognitive function used in the diagnosis of mild and major cognitive impairment in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) [ 24 ]).

Methods of outcome assessment

Any measure of cognitive function was eligible for inclusion. The tests or diagnostic criteria used in each study should have had evidence of validity and reliability for the assessment of mild cognitive impairment, but studies were not excluded on this basis.

We anticipated that many different methods would be used to assess cognitive functioning across studies. These include the following.

Clinical diagnoses of

Mild cognitive impairment using explicit criteria (e.g. [ 25 ], National Institute on Aging and the Alzheimer’s Association (United States; NIA-AA) criteria [ 26 ]; any of the definitions of mild cognitive impairment described in [ 27 ])

Neuropsychological tests used to assess global cognitive function, for example the:

Mini-Mental State Examination (MMSE)

Addenbrooke’s Cognitive Examination-Revised (ACE-R) which “incorporates the MMSE and assesses attention, orientation, fluency, language, visuospatial function, and memory, yielding subscale scores for each domain” [ 28 ]

Montreal Cognitive Assessment (MOCA), which provides measures for specific cognitive abilities and may be more suitable for assessing mild cognitive impairment than the MMSE [ 28 ]

Neuropsychological tests for assessing domain-specific cognitive function, for example, tests of:

Attention and processing speed, for example, the Trail making test (TMT-A)

Memory, for example, the Hopkins verbal learning test (HVLT-R; immediate, delay)

Visuospatial ability, for example the Block design test

Executive function, for example, the Controlled Oral Word Association Test (COWAT)

Results could be reported as an overall test score that provides a composite measure across multiple areas of cognitive ability (i.e. global cognitive function), sub-scales that provide a measure of domain-specific cognitive function or cognitive abilities (e.g. processing speed, memory), or both.

Timing of outcome measurement

Studies with a minimum follow-up of 6 months were eligible, a time frame chosen to ensure that studies were designed to examine more persistent effects of alcohol consumption. This threshold was based on previous reviews examining the association between long-term cognitive impairment and alcohol consumption (e.g. Anstey 2009 specified 12 months [ 29 ]) and guidance from the Cochrane Dementia and Cochrane Improvement Group, which suggests a minimum follow-up of 9 months for studies examining progression from mild cognitive impairment to dementia [ 28 ]. We deliberately specified a shorter period to ensure studies reporting important long-term effects were not missed.

No restrictions were placed on the number of points at which the outcome was measured, but the length of follow-up and number of measurement points (including a baseline measure of cognition) was considered when interpreting study findings and in deciding which outcomes were similar enough to combine for synthesis. Since long-term cognitive impairment is characterised as a decline from a previous level of cognitive function and implies a persistent effect, studies with longer-term outcome follow up at multiple time points should provide the most direct evidence.

Selection of cognitive outcomes where multiple are reported

We anticipated that individual studies would report data for multiple cognitive outcomes.

Specifically, a single study may report results:

For multiple constructs related to cognitive function, for example, global cognitive function and cognitive ability on specific domains (e.g. memory, attention, problem-solving, language);

Using multiple methods or tools to measure the same or similar outcome, for example reporting measures of global cognitive function using both the MMSE and the MOCA;

At multiple time points, for example, at 1, 5, and 10 years.

Where multiple cognition outcomes were reported, we selected one outcome for inclusion in analyses and for reporting the main outcomes (e.g. for GRADEing), choosing the result that provided the most complete information for analysis. Where multiple results remained, we listed all available outcomes (without results) and asked our content expert to independently rank these based on relevance to the review question, and the validity and reliability of the measures used. Measures of global cognitive function were prioritised, followed by measures of memory, then executive function. Methods for selecting results when there were multiple effect estimates and/or analyses are described in ‘Measures of association’ and ‘Summary of findings tables and assessment of certainty of the body of evidence’ sections.

Secondary outcomes

We planned to include studies that reported brain structure outcomes (as measured by neuroimaging) only if the study also reported a cognitive function outcome (i.e. studies reporting only a brain structure outcome with no measure of cognitive function were excluded).

Excluded outcomes

In line with recommendations from the Cochrane Dementia and Cognitive Improvement Group [ 30 ], surrogate outcomes were ineligible, for example:

Brain structure and function, in the absence of a measure of cognitive function

Types of studies

Cohort studies and nested case-control studies were eligible for inclusion in the review.

Broadly, these types of designs can be described as follows.

Cohort: “a study in which a defined group of people (the cohort) is followed over time, to examine associations between different … [exposures] and subsequent outcomes” [ 31 ].

Nested case-control: a study in which “Individuals experiencing an outcome of interest are identified from within a defined cohort (for which some data have already been collected) and form a group of ‘cases’. Individuals, often matched to the cases, who did not experience the outcome of interest are also identified from within the defined cohort and form the group of ‘controls’.” Data characterising prior exposure “are collected retrospectively” [ 31 ]. Data on alcohol exposure should be collected from existing records, since those experiencing cognitive decline may not be able to provide sufficiently valid and reliable information about their prior exposure.

In line with current Cochrane guidance, decisions about study eligibility were based on the assessment of the study design features listed in Table 2 rather than labels (‘cohort’ or ‘case-control’) or broad definitions of each type of study.

Definition of study ‘baseline’

Prospective assessment of alcohol consumption (Table 2 , design feature 3b) was judged to have occurred if data on alcohol consumption was collected at least 6 months prior to the first ‘follow-up’ measure of cognition. We defined the last point at which alcohol was measured as the ‘baseline’ for the study (an important consideration for studies with alcohol consumption data collected at multiple time points). A ‘baseline’ assessment of cognition may have been made at this point, but was not a requirement for inclusion in the review (Table 2 , design feature 3c). Studies that collected alcohol data concomitantly with follow-up measures of cognition (i.e. beyond ‘baseline’) were excluded unless they reported an analysis based only on the alcohol measures taken prospectively. To avoid ambiguity when describing data collection points, we used a standardised nomenclature for each point (T0 being the first measurement point, then each subsequent point numbered sequentially: T1, T2, T3, etc.).

While eligible for this review, randomised trials examining the effects of different levels and/or patterns of alcohol exposure are unlikely to be conducted because of ethical concerns and the length of follow-up required to measure long-term cognitive outcomes.

Excluded designs

Case-control studies were excluded, except for nested case-controls. Case-control studies compare “people with a specific outcome of interest (‘cases’) with people from the same source population but without that outcome (‘controls’), to examine the association between the outcome and prior exposure” [ 31 ]. This design is unsuitable for addressing the objectives of this review since it is unlikely to be possible to obtain valid and reliable estimates of prior exposure to alcohol from individuals with the outcome of interest (cognitive impairment).

Studies using other designs (before-after comparisons, cross-sectional studies) were excluded since it is difficult (if not impossible) to attribute observed changes in outcomes to the exposure [ 31 ]. Studies that collected longitudinal data, but only presented analyses based on concomitant measures of alcohol and cognition, were also excluded on this basis.

Date and language restrictions

Studies published from 2007 onwards were eligible for inclusion. Studies published in languages other than English were excluded. A recent study has shown that the exclusion of studies in languages other than English rarely impacts the results and conclusion of a review [ 32 ], a finding that is consistent with an earlier study that found no evidence that English-language restriction introduces systematic bias in meta-analytic results [ 33 ].

Search methods for identification of studies

Our approach combined searching for systematic reviews as well as primary studies. Searches were limited to bibliographic databases and checking the reference lists of eligible studies.

Systematic reviews

An independent evidence evaluation on the health effects of alcohol consumption commissioned by NHMRC [ 34 ] listed 13 systematic reviews (published between 2007 and 2016) that related to alcohol and cognitive impairment, and a further two systematic reviews were identified from an overview by Rehm et al [ 6 ]. From these reviews, we retrieved all primary studies that met the eligibility criteria. In addition, we searched MEDLINE and Embase for systematic reviews published since 2016 and ensured that any relevant primary studies included in these reviews were considered for inclusion.

Primary studies

The primary studies we identified from existing systematic reviews served as the initial source of studies. We used information about how these studies were indexed (i.e. thesaurus terms, text words) to help develop and validate the search strategy for primary studies. This technique (referred to as relative recall) is particularly useful when there are a reasonable number of studies (~20).

Independently of the search for systematic reviews, we searched for primary studies relevant to the review question published since January 2007. No language or geographic limitations were applied to the search. Searches were limited to MEDLINE, Embase, and PsycINFO.

The search strategy for Ovid MEDLINE was based on an assessment of the 2009 systematic review by Anstey [ 29 ] and the more recent 2017 meta-analysis by Xu [ 7 ]. The searches conducted for the Anstey review were very broad, generating over 33,000 citations, of which 15 were ultimately included in the meta-analysis. The MEDLINE search (see Additional file 2 , Appendix 2) retrieved all the studies included in the Anstey review but is considerably more precise. This search also retrieved all seven additional studies included in the meta-analysis by Xu.

We decided not to include the text word ‘impairment’ as a stand-alone term since records retrieved using this text word (not already retrieved by the text words ‘cognition’ or ‘cognitive’) were mostly concerned with kidney or liver impairment, or some other impairment, and unrelated to cognition.

The MEDLINE search was translated for Embase and PsycINFO, incorporating each database’s relevant thesaurus terms for alcohol, dementia/cognitive impairment, and study design (see Additional file 2 , Appendix 2).

Beyond database searching, we checked the reference lists of eligible studies for additional relevant publications.

Data collection and analysis

Selection of studies.

Citations identified from the literature searches and reference list checking were imported to EndNote and duplicates were removed. Three reviewers independently screened a sample of 109 citations to pre-test and refine coding guidance based on the inclusion criteria. Disagreements about eligibility were resolved through discussion. One reviewer (SB, JR, or SM) then each screened about a third of the remaining citations (grouped by year of publication) for inclusion in the review using the pre-tested coding guidance.

Full-text of all potentially eligible studies were retrieved. A sample of full-text studies was independently screened by two reviewers (SB and JR) until concordance was achieved (~15%; 37/228 of full-text studies screened). The remaining full-text studies were screened by one reviewer (SB or JR). All included studies, and those for which eligibility was uncertain, were screened by a second reviewer (JR or SB). Disagreements or uncertainty about eligibility were resolved through discussion, with advice from the review biostatisticians (JM, AF, or both) to confirm eligibility based on study design and analysis methods. Further information was sought from the authors of two studies (Piumatti 2018, Wardzala 2018) to clarify methods and interpretation of the analysis.

Citations that did not meet the inclusion criteria were excluded and the reason for exclusion was recorded at the full-text screening.

Cohort names, author names, and study locations, dates and samples characteristics were used to identify multiple reports arising from the same study (deemed to be a ‘cohort’). These reports were matched, and data extracted only from the report that provided the most relevant analysis and complete information for the review. In most cases, the decision was based on the outcome reported (global function was prioritised).

Data extraction and management

For each included study, one review author (SB, JR or JM) extracted data relating to study characteristics using a pre-tested data extraction and coding form. A second author (SB, JR, or JM) independently verified data relating to alcohol consumption categories (including conversions to grams per day) and outcome measures. One author extracted quantitative data (JM). Discrepancies were resolved through discussion, and advice sought from the review content expert (SW) or biostatistician (AF) if the agreement could not be reached or for more complex scenarios.

Pre-testing of the data extraction and coding form was done on two studies purposefully selected from the included studies to cover the diversity of data types anticipated in the review. Advice was sought from the review content expert (SW) and biostatisticians (JM or AF) to ensure data were extracted as planned. Revisions to the data extraction form were made as required to maximise the quality and consistency of data collection.

We extracted information relating to the characteristics of included studies and results as follows.

Study identifiers and characteristics of the study design

Study references (multiple publications arising from the same study were matched to an index reference, which is the study from which results were selected for analysis or summary)

Study or cohort name, location, and commencement date

Study design (categorised as ‘prospective cohort study’, ‘nested case-control study’, or ‘other’ using the checklist of study design features developed by Reeves and colleagues, [ 31 ])

Funding sources and funder involvement in the study

Characteristics of the exposure and comparator groups

Levels of alcohol consumption as defined in the study, including details of how consumption was measured and categorised, and information required to convert data for reporting and analysis

Qualitative descriptors of each category, if used (e.g. never or non-drinker, abstainer, former drinker, low/moderate/heavy consumption)

Upper and lower boundaries of each category (e.g. 1 to 29 g per day; 5.1 to 10 units per week based on a standard drink in the UK)

Group used as referent category (comparator) in analyses and how defined

Units of measurement (e.g. standard units of alcohol per day and definition of unit)

Method of collecting alcohol consumption data (e.g. retrospective survey involving recall of alcohol consumption over different periods of life; intake diaries to measure current alcohol consumption); time points at which exposure data were collected

Sample size for each exposure group at each measurement point and included in analysis; number lost to follow up [these data were used in the analysis and risk of bias assessment]

Any additional parameters used to derive each category or exposure measure (e.g. alcohol consumption at each drinking occasion; frequency of drinking; recall period)

Patterns of exposure

Any additional data not listed above that characterises and quantifies different patterns of alcohol exposure (e.g. consumption on heaviest drinking day; diagnosis of an alcohol-use disorder such as dependence or harmful drinking, and the method of assessment; definition of other frequency-based categories used to characterise patterns of drinking such as occasional drinking or infrequent consumption)

Duration/length of exposure period at study baseline and follow-up (directly reported or data that can be used to calculate)

Age at commencement of drinking (initial exposure)

Characteristics of participants

Age at baseline and follow up, sex, ethnicity, co-morbidities, socio-economic status (including education), use of licit or illicit drugs, family history of alcohol dependence

Other characteristics of importance within the context of each study

Eligibility criteria used in the study

Outcomes assessed and results

Outcomes domains (e.g. cognition, brain structure, function in daily life). We categorised specific domains of cognitive function by the domains used in the DSM-5 for diagnosis of cognitive impairment (Table 1 ).

For cognition outcomes:

Measurement method (e.g. Montreal cognitive assessment) and time points

Potential confounders, co-exposures and other sources of bias mentioned in the paper [ 35 ]. Baseline statistics of the confounders to allow assessment of the comparability of the exposure groups.

Results including: summary statistics (means and standard deviations, or number of events for cognitive outcomes that have been dichotomised, and sample size) in each exposure category, unadjusted and adjusted estimates of the associations (e.g. mean differences, confidence intervals, t-values, p-values, or risk ratios/odds ratios for binary outcomes) overall and stratified by the specified subpopulations, where possible. For adjusted estimates, we extracted information on the analysis method, how confounding was adjusted, and which confounders were adjusted for.

Data required to assess risk of bias (see ‘Assessment of risk of bias of included studies’ section) and report the methods that influenced judgements [ 35 ]. In particular, we collected and summarised information about study design features that potentially introduced selection bias (e.g. a lag time between initiating drinking and enrolment to the study), or bias through misclassification of alcohol consumption status (e.g. measures that do not capture variation in patterns of drinking over time).

Assessment of risk of bias of included studies

One author (MP) assessed risk of bias for each included study using ROBINS-I (Risk Of Bias In Non-randomized Studies of Interventions) tool [ 36 ], and a second author (SB) independently verified the assessments and summarised study design features on which judgements were made. Discrepancies were resolved through discussion, with advice from a third reviewer (JM) if the agreement could not be reached, for more complex scenarios or judgements of critical risk of bias (see below). To ensure concordance, the assessment process was piloted by all assessors (JM, SB, and MP) on two included studies.

ROBINS-I was developed for “evaluating risk of bias in estimates of the comparative effectiveness (harm or benefit) of interventions” from non-randomised studies (i.e. where randomisation was not used to allocate individuals to comparison groups) [ 36 ]. While alcohol is generally considered an exposure, ROBINS-I has been successfully applied to equivalent studies (e.g. those examining the association between change in body size and mortality) and has advantages over checklist approaches in that it facilitates an overall judgement of RoB that can be incorporated in the analysis and the GRADE assessment [ 36 , 37 ].

ROBINS-I requires assessment of the following seven domains:

Bias due to confounding (see below ‘Pre-specification of confounding factors and co-exposures’)

Bias in selection of participants into the study (e.g. we considered whether any lag between initiating drinking and enrolment into the study was likely to introduce bias)

Bias in classification of interventions (e.g. we considered whether the method of measuring alcohol consumption could lead to misclassification of the level of consumption due to problems with recall, underreporting, and not capturing variation in consumption over time)

Bias due to deviations from intended interventions (exposures)

Bias due to missing data

Bias in measurement of outcomes

Bias in selection of the reported result

It is recommended that users applying ROBINS-I should consider in advance the confounding factors and co-interventions that have the potential to lead to bias in included studies. These are listed at the end of this section.

Within each domain, we judged risk of bias as “low” (comparable to a well-performed randomised trial), “moderate” (sound for a non-randomised study), “serious” (there are some important problems) or “critical” (the study is too problematic to provide useful evidence).

We rated the overall risk of bias for each result based on the most serious risk of bias judgement across any of the seven domains (i.e. overall risk of bias is “serious” if at least one domain is rated “serious”). If we judged a result to be at “critical” risk of bias on the first domain (bias due to confounding), we did not assess other domains, since the overall risk of bias for the result would be “critical” by default. Studies that were judged to be at “critical” risk of bias overall were excluded from the summary and syntheses of results, and they do not contribute to our conclusions. For each study and result (outcome) assessed, we report our judgement of risk of bias by domain and provide a rationale for the judgment with supporting information about study methods.

Pre-specification of confounding factors and co-exposures

Confounding domains are “prognostic variables (factors that predict the outcome of interest)” that also predict the exposure at baseline [ 36 ]. ROBINS-I defines important confounding domains as those “for which, in the context of [a specific] study, adjustment is expected to lead to a clinically important change in the estimated effect of the [exposure]”. We considered the following confounding domains as important for most or all studies since they have been shown to be associated with alcohol consumption and are prognostic factors for cognitive impairment: age, sex, socioeconomic factors (especially education), smoking, and co-morbidities (especially diabetes, and obesity). Co-exposures were assessed on a study-by-study basis.

For GRADE assessments, it was necessary to summarise the risk of bias assessments across studies for each outcome. We followed recent GRADE guidance for making these judgements [ 37 ]. These summary assessments of risk of bias were used in determining the overall certainty of the body of evidence using GRADE, and the basis for each is reported as footnotes to the summary of findings tables.

Measures of association

Cognition was assessed using continuous measures with varying scales and neurocognitive tests across the studies. The standardised mean difference (SMD) was therefore used to standardise the associations so that they were comparable across studies. In some studies, the measures of cognition were dichotomised and analysed as binary outcomes. These studies reported odds ratios along with 95% confidence intervals. For these studies, we converted the odds ratios (ORs) and their confidence limits to SMDs using a simple approximation proposed by Chinn [ 38 ]. The accuracy of the resulting SMD variances was assessed, and where necessary, adjustments were made to these variances so that when they were back-transformed to the (log) OR scale, they yielded equivalent variances to the observed (log) OR variances. In the circumstance where results from multiple multivariable models were presented, we extracted associations from the most fully adjusted model, except in the case where an analysis adjusted for a possible intermediary along the causal pathway (i.e. post-baseline measures of prognostic factors (e.g. smoking, drug use, hypertension)) [ 39 ].

Unit of analysis issues

In this review, the unit of analysis issue that arose was multiple estimates of association calculated for different levels of alcohol consumption within the same study. These estimates are correlated since each level of alcohol consumption is compared against the same group of participants (i.e. current non-drinkers). Methods used to adjust for the correlation between the estimated associations are described in the ‘Data synthesis’ section.

Assessment of heterogeneity

We assessed heterogeneity through visual inspection of the study-specific dose-response curves, formal testing for heterogeneity using the Χ 2 test (using a significance level of α=0.1), and quantified heterogeneity in the study-specific dose-response coefficients using the I 2 statistic.

Assessment of reporting biases

We had planned to investigate the potential for small-study effects using contour-enhanced funnel plots and formal statistical tests for funnel plot asymmetry if there were at least 10 studies included in a synthesis. However, all syntheses included fewer than 10 studies.

Data synthesis

Investigation of the association between levels of alcohol consumption and cognition.

In planning the review, we anticipated that there may be too little data to conduct a dose-response analysis. We, therefore, planned to undertake pair-wise comparisons of the effects of never drinking or very low-level drinking (0 to < 10 g/week) with different levels of alcohol consumption (≥ 10 g/week and < 10 g/day; ≥ 10 g/day and < 20 g/day; ≥ 20 g/day and < 30 g/day; ≥ 30 g/day and < 40 g/day; ≥ 40 g/day and < 50 g/day; ≥ 50 g/day).) We did not undertake these analyses since all studies that contributed data suitable for synthesis were able to be included in the dose-response analyses. The dose-response analyses provide a more complete understanding of the relationship between alcohol consumption and the size of the SMDs since all data are modelled in a single synthesis. Further, from these models, the size of any effect on cognition (SMDs) can be predicted at any level of alcohol consumption (within the observed range).

Investigation of the dose-response relationship between levels of alcohol consumption and cognition

Analyses were undertaken to identify and characterise dose-response relationships between levels of alcohol consumption and cognition. For each study, the relationship between the SMD of cognition (compared with abstainers) and alcohol consumption was modelled using a restricted cubic spline with three knots (at the 10th, 50th, and 90th percentiles of alcohol consumption), accounting for correlation amongst the SMDs. The estimated study-specific dose-response coefficients and their covariance matrices were combined using a random-effects multivariate model [ 40 ]. The between-study variance of the dose-response coefficients was obtained using restricted maximum likelihood. Studies assessed as at a critical risk of bias were not included in the dose-response analysis.

In studies that reported alcohol consumption in different units (e.g. millilitres or standard drinks per days), we converted these to grams per day using the relevant country’s standards [ 41 ]. For each category of alcohol consumption, we used the median or mean of alcohol consumption in grams per day when presented. When not presented, we assigned the midpoint of the category as the dose value. When the largest dose category was reported without an upper bound, the dose value assigned was calculated as the lower bound of the largest dose category plus the width of the previous (second-to-largest) category [ 42 ].

The combined dose-response curves, along with 95% confidence intervals, were presented graphically and in tabular form (presenting predicted standardised mean differences of cognition for different alcohol consumption levels).

We examined the robustness of the combined dose-response model to different locations of the knots. We had also planned to examine the robustness of the combined dose-response model to different numbers of knots, but we did not do this. For each dose-response analysis, we were limited to a maximum of three knots due to some studies only reporting three levels of alcohol consumption.

The dose-response models were fitted using the package dosresmeta in the statistical program R [ 43 ].

Subgroup analyses

We present the dose-response relationships for females and males separately where possible (i.e. where the study was undertaken with only one sex, or the results were reported separately by sex within a study). For other potential modifying factors (age, co-morbidities, drug-taking, or a family history of alcohol use), no studies were limited to a particular subpopulation, nor did they report associations separately by particular subpopulations within a study.

Sensitivity analyses

We had planned to undertake sensitivity analyses examining the robustness of the results to the method of alcohol measurement (intake over multiple time points versus once) and limiting to studies that reported results for ‘never’ drinkers. We did not undertake these sensitivity analyses due to only a small number of studies available for any of the dose-response analyses (i.e. a maximum of six studies).

Summary of results from single studies

For studies that were not able to be included in the dose-response analyses, we summarised the risk of bias assessment, the study characteristics, the reported associations (including 95% confidence intervals and p values where reported), and provided an interpretation. We had planned to present reported associations using forest plots, but because of incomplete reporting and the variability in the measures of association (e.g. linear trends, quadratic trends, hazard ratios, odds ratios) used across the studies, this was not possible.

Summary of findings tables and assessment of certainty of the body of evidence

We assessed the certainty of the evidence for results from the dose-response analysis using the GRADE approach. In accordance with the detailed GRADE guidance [ 15 , 37 ], the following domains were assessed (as briefly summarised below) and a judgement made about whether there were serious, very serious or no concerns in relation to each domain.

Risk of bias. Based on the summary assessment across studies for each outcome reported for a comparison (see ‘Risk of bias’ section). The assessment was based on guidance for ROBINS-I [ 35 ] and GRADE [ 37 ].

Inconsistency. We assessed (1) whether there was heterogeneity in the observed effects across studies that suggested important differences in the effect of the exposure (based on visual inspection of data and statistical tests of heterogeneity), and (2) whether this could be explained (e.g. by variance in effects across subgroups if data were available).

Imprecision. We assessed whether the interpretation of the upper and lower confidence limits leads to conflicting interpretations about the effect of the exposure (e.g. benefit and appreciable harm).

Indirectness. We assessed whether there were differences between the characteristics of included studies (PECO of included studies) and the review question (in terms of the review PECO) such that the effects observed in the included studies were unlikely to apply directly to the review question. For example, studies with multiple measures of alcohol over time, and longer-term outcome follow up at multiple time points, were considered to provide the most direct evidence of the cognitive effects of life-long alcohol-use patterns. In general, this information was used to interpret results, rather than downgrade.

Publication bias. Our judgement of suspected publication bias was based on the assessment of reporting bias as described in ‘Assessment of reporting biases’ section. Evidence of small-study effects and the absence of a plausible alternative explanation for these effects indicate that publication bias should be suspected.

Upgrading domains (large effect size, dose-response gradient, opposing plausible residual confounding). Recent GRADE guidance is that observational studies may start as high certainty evidence when ROBINS-I is used for the risk of bias assessment [ 37 ]. Doing so alters the assessment of GRADE upgrading domains since these domains examine the likelihood that any observed association could be explained by residual confounding, and are typically used to upgrade observational studies from low to moderate or high certainty. In line with one of the options presented in recent GRADE guidance, we considered the upgrading domains when assessing confounding and selection bias using ROBINS-I.

GRADEpro GDT software ( www.gradepro.org ) was used to record decisions and derive an overall GRADE (high, moderate, low, or very low) for the certainty of evidence for each outcome, using the GRADE rules in which observational studies assessed using ROBINS-I begin as ‘high’ certainty evidence (score=4) and can be downgraded by −1 for each domain with serious concerns or −2 for very serious concerns [ 37 ].

A summary of findings table (using the evidence profile format for guidelines) was prepared using the GRADEpro GDT software. For each result from the dose-response analysis, the evidence profile includes estimates of the effects of alcohol exposure reported as standardised mean differences, and the overall GRADE (rating of certainty). The evidence profile also includes (1) the study design(s), number of studies contributing data (the type and size of the evidence base), (2) our assessment of each of the domains (risk of bias, inconsistency, indirectness, imprecision, publication bias), and (3) a statement interpreting the evidence (clinical impact) for each outcome (by population subgroup). Footnotes are included to explain judgements made about downgrading the rating of the certainty of the evidence.

Results of the search

The search of MEDLINE and Embase for systematic reviews published since the NHMRC evidence evaluation was conducted on 13 February 2018 and retrieved 251 records after duplicates were removed. Eleven systematic reviews were potentially eligible and we screened the included studies of these reviews, together with those from relevant systematic reviews from the 13 identified in the NHMRC overview report, to identify relevant primary studies. We did not identify any additional potentially eligible studies from these sources.

The searches of MEDLINE, Embase, and PsycINFO for primary studies were conducted on 9 April 2018. After removing duplicates, we screened 4786 records. Figure 1 shows the flow of references through the review. (See Additional file 2 , Appendix 2 for the search results for each source.) The full-text of 228 papers were screened, from which 195 were excluded.

Study flow diagram

After screening and full-text review, we included 27 studies (reported in 33 papers). Of these, 15 studies examined the effects of different levels of alcohol consumption, three examined both different levels and patterns of alcohol consumption, and nine examined patterns only. Sixteen of 18 studies that examined the effects of different levels of alcohol intake were included in the summary and synthesis of quantitative results. Two of the 18 were assessed as at a critical risk of bias (Hassing 2018, McGuire 2007), excluding them from the summary and synthesis of quantitative results. Study characteristics are reported for these studies, and the nine studies examining patterns.

Included studies were assigned a unique identifier (first author family name and year of publication) which is used throughout the review. A list of included studies and references to all linked papers is in Additional file 2 , Appendix 3.

Description of studies

Included studies, studies examining the effects of different levels of alcohol consumption.

Characteristics of the 18 included studies that examined the effects of different levels of alcohol consumption are summarised in Table 3 and reported in more detail in Table 4 .

Six of the 18 studies were conducted in the United States (Downer 2015, Lang 2007 [also UK], McGuire 2007, Richard 2017, Samieri 2013, Wardzala 2018), four were in the United Kingdom (Lang 2007, Piumatti 2018, Sabia 2014, Stott 2008), and two each in Sweden (Hassing 2018, Hogenkamp 2014) and France (Kesse-Guyot 2012, Sabia 2011). Other studies were in Australia (Heffernan 2016), Eastern Europe (Horvat 2015), Japan (Kitamura 2017) and Norway (Arntzen 2010).

Ascertainment of alcohol exposure

The first point at which alcohol consumption was measured was at mid-life in seven studies (Arntzen 2010, Downer 2015, Hassing 2018, Horvat 2015, Kesse-Guyot 2012, Sabia 2011, Sabia 2014), late-life in eight studies (Heffernan 2016, Hogenkamp 2014, Lang 2007, McGuire 2007, Samieri 2013, Solfrizzi 2007, Stott 2008, Wardzala 2018) and spanned from mid-life (~age 40 to 60) to late-life (~age 65 to > 80) in three studies (Kitamura 2017, Piumatti 2018, Richard 2017).

Only three studies measured alcohol at multiple time points. McGuire 2007 measured alcohol twice, 2 years apart (McGuire 2007). In Sabia 2011 and Sabia 2014, multiple measures of alcohol consumption were taken over 10 years; ten annual measures were taken in Sabia 2011 (a minimum of 1 measure in each 5-year period was required) and in Sabia 2014, three measures were taken at 5-year intervals. Details of the measurement methods and how these were used to categorise consumption are reported in Table 4 .

Measurement of cognition outcomes

Baseline measures of cognition were taken in eight of 18 studies (Heffernan 2016, Hogenkamp 2014 Horvat 2015, McGuire 2007, Piumatti 2018, Solfrizzi 2007, Stott 2008 and Wardzala 2018). Multiple follow-up measures of cognition were taken in eight studies (Downer 2015, Hassing 2018, Heffernan 2016, Sabia 2011, Sabia 2014, Samieri 2013a, Stott 2008, Wardzala 2018). Richard 2017 took multiple measures of cognition, but only to exclude those with cognitive impairment prior to age 85.

One of 18 studies reported a diagnosis of mild cognitive impairment, based on clinical exam and validated diagnostic criteria (Solfrizzi 2007). Eleven of 18 studies reported a measure of global cognitive function. Of these, six reported outcomes based on the MMSE (Downer 2015, Hassing 2018, Kitamura 2017, Richard 2017, Stott 2008, Wardzala 2018; see Table 3 and Table 4 for the metrics derived from the MMSE), and five reported composite measures of global cognitive function derived for tests of one or more specific cognitive domains (Kesse-Guyot 2012, Lang 2007, McGuire 2007, Sabia 2014, Samieri 2013). Six studies reported measures of function on specific cognitive domains, most reporting results for multiple domains from a battery of neurocognitive tests. The results selected for review from these studies were measures of learning and memory in three studies (Arntzen 2010, Heffernan 2016, Horvat 2015), executive function in one study (Hogenkamp 2014) and complex attention in two studies (Piumatti 2018, Sabia 2011).

Studies examining the effects of different patterns of alcohol consumption

Characteristics of the 12 included studies that examined the effects of different patterns of alcohol consumption are summarised in the Additional file 2 , Appendix 4, Table 4.1. Six of these studies were among adolescents or university students, while the other six involved participants at mid- to late-life. The studies varied considerably in terms of the types of patterns considered. Three of 12 examined heavy drinking episodes (“binge” drinking), six examined changes in the pattern of consumption over time (levels and frequency) of which two focused on changes in binge drinking patterns, one examined the age of onset of first and weekly drinking, and two examined frequency of consumption only. Importantly, the analysis methods used in these studies have not been carefully reviewed, so it is possible that some studies may not meet the eligibility criterion for using only prospective measures of alcohol in the analysis.

Ongoing studies and studies awaiting assessment

We did not identify any ongoing studies, although many of the identified cohorts are ongoing, so may generate analyses eligible for updates of this review. There are no studies awaiting assessment.

Excluded studies

Reasons for excluding the 195 studies are described in the Additional file 2 , Appendix 5 (Characteristics of excluded studies). An alphabetically sorted reference list of all studies excluded after full-text review is provided in the Additional file 2 , Appendix 10.

Of the 195 studies, eight were coded as “near miss” because they met all eligibility criteria but measures of alcohol were collected concomitantly with measures of cognition and the authors modelled the association between alcohol consumption and cognition over time (Additional file 2 , Appendix 5, Table 5.1). In many cases, this was done to provide a more reliable measure of alcohol intake over time; however, the approach rendered the studies ineligible because the analysis was not limited to prospective measures of alcohol, and hence do not enable causal inferences to be made about the effect of alcohol on cognition. For this dataset, it would have been possible for the study authors to have examined the association between alcohol consumption at a fixed time and future cognition.

A further 19 studies were excluded to narrow the scope of the review to a priority question that could be addressed within the required timeframe and resources. Since a recent systematic (Xu 2017) examined the effects of different levels of alcohol on dementia and presented a dose-response analysis, we excluded 15 studies for which the only eligible outcome was dementia or major cognitive impairment (Additional file 2 , Appendix 5, Table 5 .2). In addition, we excluded studies that examined the effects of alcohol among specific subgroups (two studies: alcohol use disorder or diabetes) or that only examined the effects of high levels of alcohol intake (Additional file 2 , Appendix 5, Table 5.3).

The remaining 176 excluded studies were excluded based on one or more of the pre-specified eligibility criteria, as reported in Tables 5.4-5.12 of the Additional file 2 (Appendix 5).

Risk of bias

The complete risk of bias assessment for each study, including the rationale for the judgement of each domain, is reported in the Additional file 2 , Appendix 6 (Risk of bias assessment of included studies). Study methods that influenced each judgement are also summarised. The overall judgement is noted in Table 4 (Study characteristics).

All studies were assessed as being at serious risk of bias, except for two (Hassing 2018, McGuire 2007), which were judged to be at critical risk of bias. In addition to concerns identified across all studies about selection bias and bias arising from misclassification of alcohol consumption, these two studies were judged to be at a critical risk of bias due to missing outcome data. Neither study reported whether missing data were balanced across groups, nor did the analysis approach address potential biases arising from missing data.

Across all studies, there were serious concerns about the risk of selection bias. Most studies enrolled participants at mid-life (~40 to 60 years of age) or late-life (~65 to 80 years). The lag time between initiating drinking and the first measurement of alcohol intake means that those who previously experienced harmful outcomes associated with drinking may be excluded (because they died or were inaccessible, declined or were unable to participate). Further, some studies excluded less healthy people (e.g. those with pre-existing cognitive impairment). While difficult to avoid, these design features are likely to result in the exclusion of drinkers with poorer health caused or exacerbated by alcohol (including those with alcohol-related cognitive impairment or alcohol-related risk factors for impairment). This risks biasing the sample through the inclusion of healthy drinkers, potentially attenuating differences between drinking and non-drinking groups.

There were also serious concerns about the risk of bias arising from methods used to categorise participants’ alcohol consumption and the resulting potential for misclassification. All but three studies (Sabia 2011, Sabia 2014, McGuire 2007) used a single assessment of alcohol consumption to estimate consumption, so most studies are unlikely to capture drinking patterns over time. Related to this, almost all studies categorised alcohol intake based on current consumption (recall over the last 12 months or less), so contamination of non-drinking groups with former drinkers is likely. To account for this, some studies used a low- or moderate-level drinking group as the referent, and two studies included 10-year abstainers only (Sabia 2011, Sabia 2014). However, the problems with measurement of lifetime consumption, together with underestimation (through poor recall) or conscious under-reporting of intake, mean that misclassification is likely across most included studies.

Since former drinkers have been shown to have poorer self-reported health and higher levels of depression than current drinkers (both associated with cognition), misclassification has implications for the comparability of groups and confounding [ 20 , 21 ]. Most studies adjusted for important confounding domains pre-specified for the review, but some residual confounding was likely.

No important conflicts of interest were identified for authors of any of the 18 included studies (Additional file 2 , Appendix 4, Table 4.2). One study (Kesse-Guyot 2012) received partial funding from a food catering company, in addition to government and non-food industry funding (the proportion of funding from each source was not reported). The authors reported that the funders had no involvement in the study; however, a conflict of interest could not be completely ruled out. Of the 17 remaining studies, 14 appeared free of any conflict of interest (funding or other), and three appeared free of financial conflicts but provided insufficient information to judge other conflicts. Ethics approval was reported for 14 of 18 studies (Additional file 2 , Appendix 4, Table 4.2).

Effects of different levels of alcohol on cognition

Dose-response syntheses.

In the following sections (‘Females’, ‘Males’, and, ‘Females and males’), the results from dose-response analyses are presented. For most studies, assumptions were required to calculate the doses of alcohol and the statistics used to compute the standardised mean differences (SMDs) (see Additional file 2 , Appendix 7 for details). Therefore, while the estimated dose-response relationships may be indicative of the shape of the relationship, the presented estimates should be cautiously interpreted.

Five of 15 eligible studies for this analysis were able to be included in the investigation of the dose-response relationship between levels of alcohol consumption and cognition. Study-specific dose-response curves of standardised mean differences (SMDs) of cognition (compared with current non-drinkers) and alcohol consumption (grams/day) are displayed in Fig. 2 . Three of the five studies reported measures of global cognitive function, derived by averaging standardised scores on tests of specific cognitive domains (Kesse-Guyot 2012; Sabia 2014), or from an MMSE score (Stott 2008). The other two studies reported measures of learning and memory (Arntzen 2010; Horvat 2015).

Study-specific standardised mean differences (SMDs) of cognition for increasing doses of alcohol (grams/day) for females. The relationship between the SMDs and cognition was modelled using a restricted cubic spline with three knots (located at 10th, 50th, and 90th percentiles of alcohol consumption observed across the studies). Black squares indicate a standardised mean difference and the whiskers indicate its 95% confidence interval. Solid lines represent the estimated dose-response curves, and the dashed lines the corresponding 95% confidence intervals. The current non-drinker served as the referent group

The pooled dose-response relationship is displayed in Fig. 3 and tabulated in Table 5 . For alcohol consumption less than 25.9 g alcohol/day (the point at which the predicted lower bound of the confidence interval crosses zero), cognition was slightly better in those consuming alcohol than current non-drinkers. However, the SMDs were small, with a maximum SMD of 0.18 (95%CI 0.02, 0.34), occurring at an intake of 14.4 g alcohol/day. Further, there was evidence of heterogeneity in the study-specific dose-response coefficients ( I 2 = 69.5%, Q test for heterogeneity p value = 0.001).

Pooled dose-response relationship between alcohol consumption (grams/day) and the standardised mean difference in cognition (solid line) for females. The study-specific relationships were modelled using restricted cubic splines and combined in a multivariate random-effects meta-analysis. The dashed lines represent the 95% confidence intervals for the combined spline model. The current non-drinker served as the referent group. Circles indicate study-specific observed SMDs, with the size of the bubbles proportional to precision (inverse of the variance) of the SMDs

Results from the sensitivity analyses revealed that the shape of the dose-response model was not robust to different locations of the knots for higher levels of alcohol consumption (Additional file 2 , Appendix 8, Figure 8.1). This was perhaps unsurprising since only one study (Kesse-Guyot 2012) contributed data for high levels of alcohol consumption. A further sensitivity analysis removing two SMDs associated with alcohol consumption greater than 30 g alcohol/day from Kesse-Guyot showed that the dose-response relationship at lower alcohol consumption levels was robust to the outlying observations (Additional file 2 , Appendix 8, Figure 8.2).

Six of 14 eligible studies for this analysis were able to be included in the investigation of the dose-response relationship between levels of alcohol consumption and cognition. Study-specific dose-response curves of standardised mean differences (SMDs) of cognition (compared with current non-drinkers) and alcohol consumption (grams/day) are displayed in Fig. 4 . Three of the six studies reported measures of global cognitive function, derived by averaging standardised scores on tests of specific cognitive domains (Kesse-Guyot 2012; Sabia 2014), or from an MMSE score (Stott 2008). The other three studies reported measures of a specific cognitive domain; learning and memory (Arntzen 2010; Horvat 2015) or complex attention (Sabia 2011).

Study-specific standardised mean differences (SMDs) of cognition for increasing doses of alcohol (grams/day) for males. The relationship between the SMDs and cognition was modelled using a restricted cubic spline with three knots (located at 10th, 50th, and 90th percentiles of alcohol consumption observed across the studies). Black squares indicate a standardised mean difference and the whiskers indicate its 95% confidence interval. Solid lines represent the estimated dose-response curves, and the dashed lines the corresponding 95% confidence intervals. The current non-drinker served as the referent group

The pooled dose-response relationship is displayed in Fig. 5 and tabulated in Table 5 . The shape of the dose-response relationship for males was similar to that observed for females; however, the maximum SMD of 0.05 (95%CI 0.00, 0.10), occurring at an intake of 19.4 g alcohol/day, was very small. For all levels of alcohol consumption, the predicted lower bound of the confidence interval of the SMD indicated that cognition was similar or poorer as compared to current non-drinkers, but the SMDs were small for alcohol intakes less than 55 g/day (Table 5 ). There was evidence of heterogeneity in the study-specific dose-response coefficients ( I 2 = 56.6%, Q test for heterogeneity p value = 0.011).

Pooled dose-response relationship between alcohol consumption (grams/day) and the standardised mean difference in cognition (solid line) for males. The study-specific relationships were modelled using restricted cubic splines and combined in a multivariate random-effects meta-analysis. The dashed lines represent the 95% confidence intervals for the combined spline model. The current non-drinker served as the referent group. Circles indicate study-specific observed SMDs, with the size of the bubbles proportional to precision (inverse of the variance) of the SMDs

Results from the sensitivity analyses revealed that the shape of the dose-response model was not robust to different locations of the knots for higher levels of alcohol consumption (Additional file 2 , Appendix 8, Figure 8.3). This was perhaps unsurprising since only one study (Kesse-Guyot 2012) contributed data for high levels of alcohol consumption. A further sensitivity analysis removing two SMDs associated with alcohol consumption greater than 70 g alcohol/day from Kesse-Guyot showed that the dose-response relationship at lower alcohol consumption levels was robust to the outlying observations (Additional file 2 , Appendix 8, Figure 8.4).

Females and males

Four of 16 eligible studies for this analysis were able to be included in the investigation of the dose-response relationship between levels of alcohol consumption and cognition. Study-specific dose-response curves of standardised mean differences (SMDs) of cognition (compared with current non-drinkers) and alcohol consumption (grams/day) are displayed in Fig. 6 . Three of the four studies reported measures of global cognitive function, derived by averaging standardised scores on tests of specific cognitive domains (Downer 2015), or from an MMSE score (Kitamura 2017; Richard 2017). The other study reported a measure of a specific cognitive domain, learning and memory (Heffernan 2016).

Study-specific standardised mean differences (SMDs) of cognition for increasing doses of alcohol (grams/day) for females and males. The relationship between the SMDs and cognition was modelled using a restricted cubic spline with three knots (located at 10th, 50th, and 90th percentiles of alcohol consumption observed across the studies). Black squares indicate a standardised mean difference and the whiskers indicate its 95% confidence interval. Solid lines represent the estimated dose-response curves, and the dashed lines the corresponding 95% confidence intervals. The current non-drinker served as the referent group

The pooled dose-response relationship is displayed in Fig. 7 and tabulated in Table 5 . The shape of the dose-response relationships for females only and males only was similar to the dose-response shape for females and males. The maximum SMD of 0.24 (95%CI −0.03, 0.51) occurred at an intake of 25 g alcohol/day. For higher levels of alcohol consumption (e.g. > 55 g alcohol/day), there may be detrimental effects on cognition; however, this is where there is most uncertainty in the predictions (see sensitivity analyses). There was some evidence of heterogeneity in the study-specific dose-response coefficients ( I 2 = 47.2%, Q test for heterogeneity p value = 0.078).

Pooled dose-response relationship between alcohol consumption (grams/day) and the standardised mean difference in cognition (solid line) for females and males. The study-specific relationships were modelled using restricted cubic splines and combined in a multivariate random-effects meta-analysis. The dashed lines represent the 95% confidence intervals for the combined spline model. The current non-drinker served as the referent group. Circles indicate study-specific observed SMDs, with the size of the bubbles proportional to precision (inverse of the variance) of the SMDs.

Results from the sensitivity analyses revealed that the shape of the dose-response model was not robust to different locations of the knots for higher levels of alcohol consumption (Additional file 2 , Appendix 8, Figure 8.5). This is likely due to only one study (Kitamura 2017) contributing data for high levels of alcohol consumption. A further sensitivity analysis removing one SMD associated with alcohol consumption greater than 55 g alcohol/day from Kitamura showed that the dose-response relationship at lower alcohol consumption levels was robust to the outlying observation (Additional file 2 , Appendix 8, Figure 8.6).

Six studies (Solfrizzi 2007, Lang 2007a, Hogenkamp 2014, Samieri 2013a, Piumatti 2018, Wardzala 2018) that examined the association between levels of alcohol consumption and cognition were not able to be included in the dose-response analyses (see Additional file 2 , Appendix 9 for reasons for exclusion). Study characteristics, reported associations, and interpretations are presented in Table 6 . The results are briefly summarised here. The study authors’ interpretations seemed often to be based on statistical significance. In combination, results were often incompletely reported (e.g. missing effect estimates, no information about the range of a scale) precluding clinical interpretation of the observed associations.

Solfrizzi 2007 found no evidence of an association between alcohol consumption and cognition using two different analysis methods. The authors reported that the associations were not modified by sex. Lang 2007a found the odds of poor cognition were greater for non-drinkers compared with those drinking > 0 to ≤ 1 drink/day (referent category). The odds of poor cognition in higher drinking categories (> 1 to ≤ 2 drinks/day; > 2 drinks/day) were less (i.e. ORs < 1) than the referent category, but were not statistically significantly different. The authors reported that the relationship was not modified by sex. Hogenkamp 2014 examined the linear association between alcohol consumption and executive function and found that the decline in executive function over time was less as the dosage of alcohol increased per day; however, the linear association was not statistically significant. Samieri 2013a found no evidence of a mean difference in global cognitive function between different levels of alcohol consumption compared with the non-drinker referent category. Piumatti 2018 examined the relationship between log alcohol and log reaction time using restricted cubic splines and found that cognitive performance improved up to 16 g alcohol/day but started to decline beyond 16 g. The authors concluded that the relationship was modified by age (for the non-linear effect), but was not modified by sex. Wardzala 2018 found that in females, the annual decline in global cognitive function was not found to be statistically significantly different between alcohol consumption categories and rare/never drinkers (referent category). In men, the annual decline in global cognitive function was not found to be statistically significantly different between the heavy drinkers and rare/never drinkers; however, it was found to be statistically significantly different between the moderate drinkers and the rare/never drinkers, with the rate of cognitive decline being less than in moderate drinkers.

Summary of findings table and assessment of certainty of the evidence

The summary of findings table (using the evidence profile format) is presented in Table 7 .

Summary of main results

This review included 18 studies that examined the effects of different levels of alcohol consumption on cognitive function, 16 of which contributed to the summary or synthesis of quantitative results. Ten studies were included in dose-response analyses (5 in the analysis for women, 6 in the analysis for men, and 4 in the analysis for men and women).

The pooled dose-response relationship for women showed that for alcohol consumption less than 25.9 g alcohol/day, cognition was slightly better in those consuming alcohol than current non-drinkers (very low certainty evidence). However, the effect sizes (reported as SMDs) were small, with the largest effect (SMD 0.18 (95%CI 0.02, 0.34) at an intake of 14.4 g alcohol/day (< 2 standard drinks per day, based on standards in Australia, France, the Netherlands, the United Kingdom and several other countries). For men, the pooled dose-response relationship was similar in shape to that observed for women; however, the maximum SMD of 0.05 (95%CI 0.00, 0.10), occurring at an intake of 19.4 g alcohol/day, was very small (very low certainty evidence). Limitations in the design of studies contributing to these analyses are such that the observed effects may be biased.

Overall completeness and applicability of the evidence

The studies included for review of the effects of different levels of alcohol consumption included participants at mid- to late-life, limiting applicability to other age groups (discussed below). Many of the studies reported single measures of cognition and had short-term follow-up (some without baseline assessment), so do not provide evidence about the persistence of observed effects. Only one study measured mild cognitive impairment using validated diagnostic criteria. Several studies reported measures of global cognitive function derived from a comprehensive battery of neurocognitive tests; however, the majority reported more limited measures that may be less suited to detecting mild cognitive impairment (e.g. MMSE scores).

None of the included studies examined the effects of different levels of alcohol intake on cognition among young people (up to age 25) or had measures of alcohol consumption among these age groups. Potentially eligible studies among this age group examined patterns of consumption but did not report analyses of the effects of different levels of consumption, or data that could be used in dose-response analyses. The absence of data following people from or close to the initiation of drinking in studies on the effects of average consumption has multiple ramifications. First, evidence about the effects of different levels of alcohol consumption on cognitive function among young people is lacking. Second, those who experience alcohol-related harm early in life may be missing from studies that begin in mid- to late-life, potentially leading to underrepresentation of the least healthy drinkers, and those who may be at most risk of cognitive impairment. Third, without measures of alcohol consumption early in life, studies are unable to reliably assess variation in average alcohol consumption or consumption patterns over the life-course. Consequently, studies may fail to differentiate between those who have very different historic patterns of consumption. All three issues limit the completeness and applicability of evidence in this review.

None of the studies included in the dose-response analysis examined whether the effects of alcohol were modified by co-morbidities or the use of licit or illicit drugs. We identified one eligible study that examined the effects of different levels of alcohol consumption among people with diabetes, and no studies involving people with other co-morbidities.

Our consideration of studies examining the effects of different patterns of consumption was limited to summarising study characteristics. Quite different patterns were examined across studies, and it is unlikely that studies examined sufficiently similar patterns to be meta-analysed, although more detailed review of this evidence is warranted.

Quality of the evidence

Overall, the evidence contributing to the dose-response analyses reported in this review is of very low quality. This is partly due to inconsistent findings across studies, but the main reason for uncertainty is the serious risk of bias arising from limitations in the design of all studies. Many of the study design limitations are difficult to address, largely because of ethical issues that prevent randomised trials of alcohol consumption. Whereas in a randomised trial known and unknown risk factors for cognitive impairment would be balanced across groups through randomisation, this is not the case in a cohort study. In observational studies of alcohol, participants have ‘selected’ to drink alcohol or not. Decisions to drink—or not drink—may be associated with a range of characteristics that may, in turn, be risk factors for cognitive function (e.g. those with ill health may be less likely to drink or may quit drinking as their health declines). Although most studies attempted to control for these factors, residual confounding is likely. Issues with confounding were exacerbated because very few studies controlled for biases arising from the misclassification of drinkers as non-drinkers. Consequently, those with potentially elevated risk for cognitive impairment were likely to have been included in non-drinking groups. Finally, the evidence contributing to the review derives entirely from cohort studies involving participants at mid- to late-life, potentially excluding less healthy drinkers, at higher risk of cognitive impairment related to alcohol consumption.

Potential biases in the review process

The review was conducted according to a pre-specified protocol with the aim of minimising biases in the review process. We conducted a comprehensive search of literature published from 2007 onwards. To minimise bias and error, we performed independent screening on samples of citations and full-text articles to ensure concordance, and a second person checked extraction of quantitative data (including that used to quantify alcohol intake) and risk of bias assessments. However, this is a rapid review, which inherently requires some methodological compromises that may introduce bias.

Due to the size of the reviewed literature, we were unable to perform double screening of all references, and we performed checks rather than independent assessment of the risk of bias and data extraction. However, we were over-inclusive in decisions to screen the full text of studies (retrieving full text of 5% of all citations, i.e. 228 papers from 4786 citations), reasons for exclusion were recorded when screening citations to facilitate verification of decisions, and the final list of included studies was cross-checked against a recent overview of systematic reviews [ 6 ] to ensure no studies were missed. At least three authors read all included studies (SB, JM, MP, JR), and a second author reviewed all papers for which there was uncertainty over inclusion or interpretation. All quantitative data were extracted and analysed by an experienced biostatistician (JM).

We did not contact authors for further information or data (with two exceptions, as documented in the methods). This meant that we may have missed subsequent publications of some studies published only as conference abstracts. It also meant that we relied on published data for our assessment of study design, risk of bias and for analysis.

Limitations of the review

For most studies, assumptions were required to standardise alcohol consumption (i.e. to calculate doses of alcohol in grams per day) and to calculate the statistics required to standardise effect measures (i.e. compute the standardised mean differences, SMDs). While these assumptions are not expected to bias results of the systematic review, limitations arise from making such assumptions. For example, where the authors did not specify the number of grams of alcohol in a standard drink, we standardised using published definitions of a standard drink for the country in which the study was conducted. It is possible that a different standard (or no standard) was used in these studies, which might have led to a slight over- or under-estimate of the level of alcohol intake. However, the alternative would have been not to standardise, making comparison across studies impossible. Importantly, standardising alcohol consumption and effect measures was a necessary step for enabling comparisons of findings across studies, irrespective of whether results were then pooled in a statistical analysis or not. Hence, any limitations arising from standardisation would have applied whether we reported standardised results from single studies, pooled results in pairwise meta-analyses (i.e. examining whether cognitive function differs for one level of alcohol consumption compared to another, for example < 10 g/week compared to ≥ 20 g/day to < 30 g/day), or pooled results in a dose-response analysis (i.e. examining whether cognitive function differs with increasing levels of alcohol consumption).

A further limitation of the review is that we did not report or synthesise results from studies that examined the effect of patterns of alcohol consumption. While dose-response analyses based on the average level of alcohol consumption provide important information, they do not account for the potentially harmful effects of different patterns of consumption and may mask such effects. In particular, the effects of irregular consumption above lower risk levels (e.g. weekly or monthly “binge” drinking) and the effects of drinking early on the life-course (e.g. less than 25 years of age) need to be examined. A simple, yet questionable, approach to considering results from studies examining different patterns of alcohol consumption would have been to report conclusions from the abstracts of included studies. However, given the known biases in the reporting of conclusions in the abstracts of non-randomised studies (see, for example [ 65 ]), and the number of analyses reported in each included study, it is unlikely that this would provide a valid summary of the evidence.

Authors’ conclusions

Implications for policy.

We found that there is currently very low certainty evidence showing a very small, probably unimportant, beneficial effects on cognition at levels of alcohol consumption at or below those currently indicated as lower risk for women and men in the 2009 Australian Guidelines, and those of New Zealand, and a number of European countries including the United Kingdom (i.e. two standard drinks or < 20 g of alcohol per day). The extent to which this reflects a true effect or bias arising from limitations of studies included in the systematic review cannot be determined.

Implications for research

Published research examining the effects of different levels of alcohol consumption on cognition has a number of limitations, some of which could be addressed through adherence to the STROBE guidelines for reporting observational studies [ 66 , 67 ]. The reporting of key elements of study design was particularly problematic, with many studies omitting information, or reporting ambiguous information, about the timing of data collection for alcohol exposure and cognition outcomes. In several studies, it was impossible to determine whether cross-sectional or longitudinal data were collected and whether the alcohol data used in analyses were entirely prospective or collected concomitantly with follow-up measures of cognition. Other problematic reporting practices included not presenting baseline characteristics (including covariate data) for each of the alcohol categories for which results were reported (needed to examine baseline imbalance), and not summarising information about missing data by alcohol categories (needed to examine whether there was a differential loss to follow-up across groups). Collectively, these problematic reporting practices may have led to an unnecessary exclusion of some studies based on design or a more serious rating of risk of bias than necessary.

More challenging to address are study design limitations that may bias the observed effects of alcohol on cognition in observational studies. The methodological literature on alcohol epidemiology identifies numerous recommendations for the study design that were not widely implemented in the studies included in this review. For example, methodological studies have identified and provided empirical evidence about methods for measuring alcohol, and dealing with potential bias and confounding arising from misclassification of alcohol consumption (see for example, [ 20 , 68 , 69 , 70 , 71 , 72 ]). These practices were rarely implemented in studies included in this review. Greater attention to applying these and other best-practice methods may increase the certainty of evidence arising from future research.

Availability of data and materials

Requests for data should be sent to the corresponding author.

Abbreviations

95% confidence interval

Addenbrooke’s Cognitive Examination ‐ Revised

Alcohol Working Committee

Cognitive impairment

Controlled Oral Word Association Test

Cardiovascular disease

Digit symbol coding test

Digit symbol substitution test

Graduated frequency questionnaire

Grading of Recommendations Assessment, Development and Evaluation

Hazard ratio

Hopkins verbal learning test

Mild cognitive impairment

Mean difference (usually based on a scale score or test)

Mini Mental State Examination

Montreal Cognitive Assessment

milliseconds

National Health and Medical Research Council

National Institute on Alcohol Abuse and Alcoholism (United States)

National Institute on Aging and the Alzheimer’s Association (United States)

Office of the National Health and Medical Research Council

Population, Exposure, Comparator, Outcome

Preferred Reporting Items for Systematic Reviews and Meta‐Analyses

Preferred Reporting Items for Systematic Reviews and Meta‐Analyses ‐ protocols

Risk Of Bias In Non‐randomized Studies of Interventions

Specific cognitive domain

Standard deviation

Standard error

Standardised mean difference

• Systematic review

Strengthening the Reporting of Observational Studies in Epidemiology

Time 0: 1 st measurement point; time 1: 2 nd measurement point

Telephone Interview for Cognitive Status

Trail making test part A; Trail making test part B

Griswold MG, Fullman N, Hawley C, Arian N, Zimsen SRM, Tymeson HD, Venkateswaran V, Tapp AD, Forouzanfar MH, Salama JS, et al. Alcohol use and burden for 195 countries and territories, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. The Lancet. 2018;392:1015–35.

Article   Google Scholar  

Wood AM, Kaptoge S, Butterworth AS, Willeit P, Warnakula S, Bolton T, Paige E, Paul DS, Sweeting M, Burgess S, et al: Risk thresholds for alcohol consumption: combined analysis of individual-participant data for 599&#x2008;912 current drinkers in 83 prospective studies. The Lancet 2018, 391:1513-1523.

Chikritzhs T, Stockwell T, Naimi T, Andreasson S, Dangardt F, Liang W. Has the leaning tower of presumed health benefits from 'moderate' alcohol use finally collapsed? Addiction. 2015;110:726–7.

Article   PubMed   Google Scholar  

Chikritzhs TN, Naimi TS, Stockwell TR, Liang W. Mendelian randomisation meta-analysis sheds doubt on protective associations between 'moderate' alcohol consumption and coronary heart disease. Evid Based Med. 2015;20:38.

Livingston G, Sommerlad A, Orgeta V, Costafreda SG, Huntley J, Ames D, Ballard C, Banerjee S, Burns A, Cohen-Mansfield J, et al. Dementia prevention, intervention, and care. Lancet. 2017;390:2673–734.

Rehm J, Hasan OSM, Black SE, Shield KD, Schwarzinger M. Alcohol use and dementia: a systematic scoping review. Alzheimer's Research & Therapy. 2019;11:1.

Xu W, Wang H, Wan Y, Tan C, Li J, Tan L, Yu J-T. Alcohol consumption and dementia risk: a dose-response meta-analysis of prospective studies. European journal of epidemiology. 2017;32:31–42.

Anstey KJ. Alcohol exposure and cognitive development: an example of why we need a contextualized, dynamic life course approach to cognitive ageing--a mini-review. Gerontology. 2008;54:283–91.

Lafortune L, Martin S, Kelly S, Kuhn I, Remes O, Cowan A, Brayne C. Behavioural Risk Factors in Mid-Life Associated with Successful Ageing, Disability. Dementia and Frailty in Later Life: A Rapid Systematic Review. PLoS One. 2016;11:e0144405.

PubMed   Google Scholar  

Neafsey EJ, Collins MA. Moderate alcohol consumption and cognitive risk. Neuropsychiatr Dis Treat. 2011;7:465–84.

Article   PubMed   PubMed Central   Google Scholar  

Nguyen-Louie TT, Matt GE, Jacobus J, Li I, Cota C, Castro N, Tapert SF. Earlier Alcohol Use Onset Predicts Poorer Neuropsychological Functioning in Young Adults. Alcoholism: Clinical and Experimental Research. 2017;41:2082–92.

Jacobus J, Squeglia LM, Bava S, Tapert SF. White matter characterization of adolescent binge drinking with and without co-occurring marijuana use: a 3-year investigation. Psychiatry Research. 2013;214:374–81.

Australian guidelines to reduce health risks from drinking alcohol. 2009.

Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ. Welch VA (Eds.): Cochrane Handbook for Systematic Reviews of Intervention. Version 6. London: Cochrane; 2019.

Book   Google Scholar  

Schunemann HJ, Brozek J, Guyatt G, Oxman AD (Eds.): Handbook for grading the quality of evidence and the strength of recommendations using the GRADE approach. Accessed. Hamilton, Canada: McMaster University. 5 July 2016:2013.

Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JPA, Clarke M, Devereaux PJ, Kleijnen J, Moher D: The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009, 339:b2700-.

Moher D, Liberati A, Tetzlaff J, Altman DG, for the PRISMA Group: Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009, 339:b2535-.

Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, Shekelle P, Stewart LA, Group P-P. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4:1.

Shamseer L, Moher D, Clarke M, Ghersi D, Liberati A, Petticrew M, Shekelle P, Stewart LA, Group P-P. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ. 2015;349:g7647.

Naimi TS, Stockwell T, Zhao J, Xuan Z, Dangardt F, Saitz R, Liang W, Chikritzhs T. Selection biases in observational studies affect associations between ‘moderate’ alcohol consumption and mortality. Addiction. 2017;112:207–14.

Knott CS, Coombs N, Stamatakis E, Biddulph JP. All cause mortality and the case for age specific alcohol consumption guidelines: pooled analyses of up to 10 population based cohorts. BMJ. 2015;350:h384.

Topiwala A, Allan CL, Valkanova V, Zsoldos E, Filippini N, Sexton C, Mahmood A, Fooks P, Singh-Manoux A, Mackay CE, et al. Moderate alcohol consumption as risk factor for adverse brain outcomes and cognitive decline: Longitudinal cohort study. BMJ (Online). 2017;357.

Australian Institute of Health and Welfare: National Drug Strategy Household Survey 2016: detailed findings. Drug Statistics series no. 31. Cat. no. PHE 214. Canberra: AIHW; 2017.

Black DW. DSM-5 guidebook : Chapter 17 Neurocognitive Disorders. In DSM-5 guidebook : the essential companion to the Diagnostic and statistical manual of mental disorders, fifth edition. First edition. edition. Edited by Grant JE. Arlington, VA: American Psychiatric Publishing; 2014.

Google Scholar  

Winblad B, Palmer K, Kivipelto M, Jelic V, Fratiglioni L, Wahlund LO, Nordberg A, Backman L, Albert M, Almkvist O, et al. Mild cognitive impairment--beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med. 2004;256:240–6.

Article   CAS   PubMed   Google Scholar  

Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, Gamst A, Holtzman DM, Jagust WJ, Petersen RC, et al. The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement. 2011;7:270–9.

Matthews FE, Stephan BC, McKeith IG, Bond J, Brayne C. Medical Research Council Cognitive Function and Ageing Study: Two-year progression from mild cognitive impairment to dementia: to what extent do different definitions agree? J Am Geriatr Soc. 2008;56:1424–33.

Davis DH, Creavin ST, Noel-Storr A, Quinn TJ, Smailagic N, Hyde C, Brayne C, McShane R, Cullum S. Neuropsychological tests for the diagnosis of Alzheimer's disease dementia and other dementias: a generic protocol for cross-sectional and delayed-verification studies. Cochrane Database Syst Rev. 2013.

Anstey KJ, Mack HA, Cherbuin N. Alcohol consumption as a risk factor for dementia and cognitive decline: meta-analysis of prospective studies. Am J Geriatr Psychiatry. 2009;17:542–55.

Harrison JK, Noel-Storr AH, Demeyere N, Reynish EL, Quinn TJ. Outcomes measures in a decade of dementia and mild cognitive impairment trials. Alzheimers Res Ther. 2016;8:48.

Reeves B, Deeks J, Higgins J, Wells G: Chapter 13: Including non-randomized studies. . In Cochrane Handbook for Systematic Reviews of Interventions Version 510 [updated March 2011] Available from wwwcochrane-handbookorg. Edited by Higgins J, Green S: The Cochrane Collaboration; 2011

Hartling L, Featherstone R, Nuspl M, Shave K, Dryden DM, Vandermeer B. Grey literature in systematic reviews: a cross-sectional study of the contribution of non-English reports, unpublished studies and dissertations to the results of meta-analyses in child-relevant reviews. BMC Medical Research Methodology. 2017;17:64.

Morrison A, Polisena J, Husereau D, Moulton K, Clark M, Fiander M, Mierzwinski-Urban M, Clifford T, Hutton B, Rabb D. The effect of English-language restriction on systematic review-based meta-analyses: a systematic review of empirical studies. Int J Technol Assess Health Care. 2012;28:138–44.

Clinical Trials Centre NHMRC. Evaluating the evidence on the health effects of alcohol consumption: evidence evaluation report commission by the Office of the National Health and Medical Research Council Sydney: The University of Sydney; 2017.

Risk Of Bias In Non-randomized Studies of Interventions (ROBINS-I): detailed guidance, updated 12 October 2016. Available from http://www.riskofbias.info Accessed 25 March 2018

Sterne JA, Hernan MA, Reeves BC, Savovic J, Berkman ND, Viswanathan M, Henry D, Altman DG, Ansari MT, Boutron I, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;355:i4919.

Schunemann HJ, Cuello C, Akl EA, Mustafa RA, Meerpohl JJ, Thayer K, Morgan RL, Gartlehner G, Kunz R, Katikireddi SV, et al: GRADE Guidelines: 18. How ROBINS-I and other tools to assess risk of bias in non-randomized studies should be used to rate the certainty of a body of evidence. J Clin Epidemiol 2018.

Chinn S. A simple method for converting an odds ratio to effect size for use in meta-analysis. Stat Med. 2000;19:3127–31.

Karahalios A, English DR, Simpson JA. Change in body size and mortality: a systematic review and meta-analysis. Int J Epidemiol. 2017;46:526–46.

Crippa A, Orsini N. Dose-response meta-analysis of differences in means. BMC Med Res Methodol. 2016;16:91.

Article   PubMed   PubMed Central   CAS   Google Scholar  

Stockwell T, Chikritzhs T. International guide for monitoring alcohol consumption and related harm. Department of Mental Health and Substance Dependence, Noncommunicable Diseases and Mental Health Cluster, World Health Organization: Geneva, Switzerland; 2000.

Il'yasova D, Hertz-Picciotto I, Peters U, Berlin JA, Poole C. Choice of exposure scores for categorical regression in meta-analysis: a case study of a common problem. Cancer Causes Control. 2005;16:383–8.

Crippa A, Orsini N: Multivariate Dose-Response Meta-Analysis: The dosresmeta R Package. 2016 2016, 72:15.

Arntzen KA, Schirmer H, Wilsgaard T, Mathiesen EB. Moderate wine consumption is associated with better cognitive test results: a 7 year follow up of 5033 subjects in the Tromso Study. Acta Neurologica Scandinavica. 2010;Supplementum:23–9.

Downer B, Jiang Y, Zanjani F, Fardo D. Effects of alcohol consumption on cognition and regional brain volumes among older adults. American Journal of Alzheimer's Disease & Other Dementias. 2015;30:364–74.

Hassing LB. Light alcohol consumption does not protect cognitive function: A longitudinal prospective study. Frontiers in Aging Neuroscience. 2018;10:81.

Heffernan M, Mather KA, Xu J, Assareh AA, Kochan NA, Reppermund S, Draper B, Trollor JN, Sachdev P, Brodaty H. Alcohol Consumption and Incident Dementia: Evidence from the Sydney Memory and Ageing Study. Journal of Alzheimer's Disease. 2016;52:529–38.

Hogenkamp PS, Benedict C, Sjogren P, Kilander L, Lind L, Schioth HB. Late-life alcohol consumption and cognitive function in elderly men. Age. 2014;36:243–9.

Horvat P, Richards M, Kubinova R, Pajak A, Malyutina S, Shishkin S, Pikhart H, Peasey A, Marmot MG, Singh-Manoux A, Bobak M. Alcohol consumption, drinking patterns, and cognitive function in older Eastern European adults. Neurology. 2015;84:287–95.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Kesse-Guyot E, Andreeva VA, Jeandel C, Ferry M, Touvier M, Hercberg S, Galan P. Alcohol consumption in midlife and cognitive performance assessed 13 years later in the SU.VI.MAX 2 cohort. PLoS ONE [Electronic Resource]. 2012;7:e52311.

Article   CAS   Google Scholar  

Kitamura K, Watanabe Y, Nakamura K, Takahashi A, Takachi R, Oshiki R, Kobayashi R, Saito T, Tsugane S, Sasaki A. Weight loss from 20 years of age is associated with cognitive impairment in middle-aged and elderly individuals. PLoS ONE [Electronic Resource]. 2017;12:e0185960.

Lang I, Guralnik J, Wallace RB, Melzer D. What level of alcohol consumption is hazardous for older people? Functioning and mortality in U.S. and English national cohorts. Journal of the American Geriatrics Society. 2007;55:49–57.

McGuire LC, Ajani UA, Ford ES. Cognitive functioning in late life: the impact of moderate alcohol consumption. Annals of Epidemiology. 2007;17:93–9.

Piumatti G, Moore SC, Berridge DM, Sarkar C, Gallacher J. The relationship between alcohol use and long-term cognitive decline in middle and late life: a longitudinal analysis using UK Biobank. Journal of Public Health. 2018;09.

Richard EL, Kritz-Silverstein D, Laughlin GA, Fung TT, Barrett-Connor E, McEvoy LK. Alcohol intake and cognitively healthy longevity in community-dwelling adults: the rancho bernardo study. Journal of Alzheimer's Disease. 2017;59:803–14.

Sabia S, Gueguen A, Berr C, Berkman L, Ankri J, Goldberg M, Zins M, Singh-Manoux A. High alcohol consumption in middle-aged adults is associated with poorer cognitive performance only in the low socio-economic group. Results from the GAZEL cohort study. Addiction. 2011;106:93–101.

Sabia S, Elbaz A, Britton A, Bell S, Dugravot A, Shipley M, Kivimaki M, Singh-Manoux A. Alcohol consumption and cognitive decline in early old age. Neurology. 2014;82:332–9.

Samieri C, Grodstein F, Rosner BA, Kang JH, Cook NR, Manson JE, Buring JE, Willett WC, Okereke OI. Mediterranean diet and cognitive function in older age. Epidemiology. 2013;24:490–9.

Solfrizzi V, D'Introno A, Colacicco AM, Capurso C, Del Parigi A, Baldassarre G, Scapicchio P, Scafato E, Amodio M, Capurso A, et al. Alcohol consumption, mild cognitive impairment, and progression to dementia. Neurology. 2007;68:1790–9.

Stott DJ, Falconer A, Kerr GD, Murray HM, Trompet S, Westendorp RGJ, Buckley B, de Craen AJM, Sattar N, Ford I. Does low to moderate alcohol intake protect against cognitive decline in older people? Journal of the American Geriatrics Society. 2008;56:2217–24.

Wardzala C, Murchison C, Loftis JM, Schenning KJ, Mattek N, Woltjer R, Kaye J, Quinn JF, Wilhelm CJ. Sex differences in the association of alcohol with cognitive decline and brain pathology in a cohort of octogenarians. Psychopharmacology. 2018;235:761–70.

Lang I, Wallace RB, Huppert FA, Melzer D. Moderate alcohol consumption in older adults is associated with better cognition and well-being than abstinence. Age & Ageing. 2007;36:256–61.

National Institute on Alcohol Abuse and Alcoholism. State of the science report on the effects of moderate drinking. Bethesda, MD: National Institutes of Health; 2003.

Petersen RC, Smith GE, Waring SC, Ivnik RJ, Tangalos EG, Kokmen E. Mild cognitive impairment: clinical characterization and outcome. Arch Neurol. 1999;56:303–8.

Lazarus C, Haneef R, Ravaud P, Boutron I. Classification and prevalence of spin in abstracts of non-randomized studies evaluating an intervention. BMC Med Res Methodol. 2015;15:85.

Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC, Mulrow CD, Pocock SJ, Poole C, Schlesselman JJ, Egger M. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. PLoS Med. 2007;4:e297.

von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. The Lancet. 2008;370:1453–7.

Fillmore KM, Stockwell T, Chikritzhs T, Bostrom A, Kerr W. Moderate alcohol use and reduced mortality risk: systematic error in prospective studies and new hypotheses. Ann Epidemiol. 2007;17:S16–23.

Liang W, Chikritzhs T. The association between alcohol exposure and self-reported health status: the effect of separating former and current drinkers. PLoS One. 2013;8:e55881.

Liang W, Chikritzhs T. Observational research on alcohol use and chronic disease outcome: new approaches to counter biases. ScientificWorldJournal. 2013;2013:860915.

PubMed   PubMed Central   Google Scholar  

Rehm J. Measuring quantity, frequency, and volume of drinking. Alcohol Clin Exp Res. 1998;22:4S–14S.

Stockwell T, Zhao J, Greenfield T, Li J, Livingston M, Meng Y. Estimating under- and over-reporting of drinking in national surveys of alcohol consumption: identification of consistent biases across four English-speaking countries. Addiction. 2016;111:1203–13.

Download references

Acknowledgements

We thank Sally Green for contributions to the protocol and comments on drafts if the original review report. We thank Emily Karahalios for her advice on the analysis plan (review protocol). We thank staff of the Office of the NHMRC, the NHMRC’s Alcohol Working Committee (AWC), and the independent methodological reviewers contracted by the NHMRC for their critical review of the systematic review protocol and the report on which this manuscript is based.

This review was commissioned and funded by the Australian Government National Health and Medical Research Council (NHMRC) to inform the update of the 2009 Australian Guidelines to Reduce Health Risks from Drinking Alcohol (the Alcohol Guidelines: https://nhmrc.gov.au/health-advice/alcohol ). The NHMRC and the NHMRC’s Alcohol Working Committee (AWC: https://nhmrc.gov.au/alcohol-working-committee ) conceived of the review, drafted the initial review questions (but not the detailed PECO), and provided critical review of the review protocol and report on which this manuscript is based. The NHMRC contracted independent methodological experts to undertake peer review of the protocol and the review report. SB, SM and SG are staff of Cochrane Australia which is funded by the Australian Government through the NHMRC. JEM is supported by an NHMRC Career Development Fellowship (1143429). MP was supported by an NHMRC Early Career Fellowship (1088535).

Author information

Authors and affiliations.

School of Public Health and Preventive Medicine, Monash University, Melbourne, Australia

Sue E. Brennan, Steve McDonald, Matthew J. Page, Jane Reid, Stephanie Ward, Andrew B. Forbes & Joanne E. McKenzie

You can also search for this author in PubMed   Google Scholar

Contributions

Design and conduct of the SR was led by SB (overall), SM (search methods and conduct) and JM (analysis and question specification), with input from SW (outcome specification/selection), MP (risk of bias assessment) and AF (analysis). Study selection was performed by SB and JR, and study eligibility confirmed by JM (design-related decisions). Data extraction was performed by SB (study characteristics, risk of bias), JR (study characteristics) and JM (quantitative data). Risk of bias assessments were performed by MP, SB and JM. SB and JM drafted the manuscript, with contributions from SM (search methods and results). All authors provided critical review of drafts of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Sue E. Brennan .

Ethics declarations

Ethics approval and consent to participate.

Not required

Consent for publication

Competing interests.

The authors declare that they have no competing interests (financial or non-financial).

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Additional file 1..

Protocol for the systematic review

Additional file 2.

Appendices 1 to 10

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License ( http://creativecommons.org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Cite this article.

Brennan, S.E., McDonald, S., Page, M.J. et al. Long-term effects of alcohol consumption on cognitive function: a systematic review and dose-response analysis of evidence published between 2007 and 2018. Syst Rev 9 , 33 (2020). https://doi.org/10.1186/s13643-019-1220-4

Download citation

Received : 27 June 2019

Accepted : 04 November 2019

Published : 13 February 2020

DOI : https://doi.org/10.1186/s13643-019-1220-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Cognition dose-response
• Meta-analysis

Systematic Reviews

ISSN: 2046-4053

• Submission enquiries: Access here and click Contact Us
• General enquiries: [email protected]

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 08 June 2023

Alcohol consumption and risks of more than 200 diseases in Chinese men

• Pek Kei Im   ORCID: orcid.org/0000-0002-2624-9766 1 ,
• Neil Wright   ORCID: orcid.org/0000-0002-3946-1870 1 ,
• Ling Yang   ORCID: orcid.org/0000-0001-5750-6588 1 , 2 ,
• Ka Hung Chan   ORCID: orcid.org/0000-0002-3700-502X 1 , 3 ,
• Yiping Chen   ORCID: orcid.org/0000-0002-4973-0296 1 , 2 ,
• Huaidong Du   ORCID: orcid.org/0000-0002-9814-0049 1 , 2 ,
• Xiaoming Yang 1 ,
• Daniel Avery   ORCID: orcid.org/0000-0002-9823-9575 1 ,
• Shaojie Wang 5 ,
• Canqing Yu   ORCID: orcid.org/0000-0002-0019-0014 6 , 7 ,
• Jun Lv 6 , 7 ,
• Robert Clarke   ORCID: orcid.org/0000-0002-9802-8241 1 ,
• Junshi Chen 8 ,
• Rory Collins 1 ,
• Robin G. Walters   ORCID: orcid.org/0000-0002-9179-0321 1 , 2 ,
• Richard Peto 1 ,
• Liming Li   ORCID: orcid.org/0000-0001-5873-7089 6 , 7   na1 ,
• Zhengming Chen   ORCID: orcid.org/0000-0001-6423-105X 1 , 2   na1 ,
• Iona Y. Millwood   ORCID: orcid.org/0000-0002-0807-0682 1 , 2   na1 &

China Kadoorie Biobank Collaborative Group

Nature Medicine volume  29 ,  pages 1476–1486 ( 2023 ) Cite this article

31k Accesses

25 Citations

824 Altmetric

Metrics details

• Epidemiology
• Genetics research
• Risk factors

Alcohol consumption accounts for ~3 million annual deaths worldwide, but uncertainty persists about its relationships with many diseases. We investigated the associations of alcohol consumption with 207 diseases in the 12-year China Kadoorie Biobank of >512,000 adults (41% men), including 168,050 genotyped for ALDH2 - rs671 and ADH1B - rs1229984 , with >1.1 million ICD-10 coded hospitalized events. At baseline, 33% of men drank alcohol regularly. Among men, alcohol intake was positively associated with 61 diseases, including 33 not defined by the World Health Organization as alcohol-related, such as cataract ( n  = 2,028; hazard ratio 1.21; 95% confidence interval 1.09–1.33, per 280 g per week) and gout ( n  = 402; 1.57, 1.33–1.86). Genotype-predicted mean alcohol intake was positively associated with established ( n  = 28,564; 1.14, 1.09–1.20) and new alcohol-associated ( n  = 16,138; 1.06, 1.01–1.12) diseases, and with specific diseases such as liver cirrhosis ( n  = 499; 2.30, 1.58–3.35), stroke ( n  = 12,176; 1.38, 1.27–1.49) and gout ( n  = 338; 2.33, 1.49–3.62), but not ischemic heart disease ( n  = 8,408; 1.04, 0.94–1.14). Among women, 2% drank alcohol resulting in low power to assess associations of self-reported alcohol intake with disease risks, but genetic findings in women suggested the excess male risks were not due to pleiotropic genotypic effects. Among Chinese men, alcohol consumption increased multiple disease risks, highlighting the need to strengthen preventive measures to reduce alcohol intake.

Similar content being viewed by others

A burden of proof study on alcohol consumption and ischemic heart disease

Effect of alcohol consumption on kidney function: population-based cohort study

Genomic prediction of alcohol-related morbidity and mortality

Alcohol consumption is a major risk factor for poor physical and mental health, accounting for about 3 million deaths and over 130 million disability-adjusted life years worldwide in 2016 (ref. 1 ). Since the 1990s, alcohol consumption has increased in many low- and middle-income countries, including China, where it almost exclusively involves men 2 , 3 . Among Chinese men, those who reported alcohol consumption in the past 12 months increased from 59% to 85% and yearly per-capita alcohol consumption increased from 7.1 to 11.2 l between 1990 and 2017 and these have been predicted to increase in future years 2 .

Previous epidemiological studies conducted in mainly western populations have provided consistent evidence about the hazards of alcohol drinking for several major diseases, including several types of cancers and cardiovascular diseases (CVDs), liver cirrhosis, infectious diseases (for example tuberculosis and pneumonia) and injuries 4 , 5 , 6 , 7 , 8 , 9 . Large western cohort studies with linkage to hospital records have also investigated the associations of alcohol with risks of several less-common or non-fatal disease outcomes (for example certain site-specific cancers 10 , 11 , 12 , dementia 13 , falls 14 and cataract surgery 15 ). For some (for example stomach cancer), there was suggestive evidence of weak positive associations with heavy drinking 10 , 11 , whereas for others (for example cataract) the limited available evidence has been contradictory 10 , 12 , 13 , 15 ; however, the evidence from western populations, even for diseases known to be associated with alcohol, may not be generalizable to Chinese populations, where the prevalence and types of alcohol drinking (mainly spirits), patterns of diseases (for example high stroke rates) and differences in the ability to metabolize alcohol 8 , 9 , 16 differ markedly from those in western populations 4 , 17 .

For many diseases, including those considered by the World Health Organization (WHO) 4 to be alcohol-related (for example ischemic heart disease (IHD) and diabetes), uncertainty remains about the causal relevance of these associations, which can be assessed in genetic studies using a Mendelian randomization (MR) approach 18 . In such studies, genetic variants can be used as instruments for alcohol consumption to investigate the potential causal relevance of alcohol drinking for diseases, which can limit the biases of confounding and reverse causality common in conventional observational studies 18 . Such studies are particularly informative in East Asian populations where two common genetic variants ( ALDH2 - rs671 and ADH1B - rs1229984 ), which are both rare in western populations, greatly alter alcohol metabolism and strongly affect alcohol intake 19 . Several studies have explored the causal relevance of alcohol consumption with CVD risk factors and morbidity 19 , 20 , 21 , 22 and cancer 16 using these genetic variants, yet findings remain inconclusive for certain diseases (for example IHD) and evidence for other diseases is sparse.

To address these questions, we conducted analyses using observational and genetic approaches to evaluate the associations between alcohol consumption and the risks of a wide range of disease outcomes in the prospective China Kadoorie Biobank (CKB).

Among the 512,724 participants (Supplementary Fig. 1 ), the mean age at baseline was 52 (s.d. 10.7) years, 41% were men and 56% lived in rural areas. Among men, 33% reported drinking alcohol regularly (at least once a week) at baseline (current drinkers), consuming on average 286 g of alcohol per week, mainly from spirits (Supplementary Tables 1 and 2 ). Non-drinkers and ex-drinkers were older and more likely to report poor self-rated health or previous chronic diseases, compared to occasional or current drinkers (Table 1 ). Compared to moderate drinkers (<140 g per week), heavier drinkers were more likely to be rural residents, had received lower education and had more unhealthy lifestyle factors (for example smoking and infrequent fresh fruit intake), higher mean blood pressure and longer duration of drinking (Supplementary Table 3 ). Among male current drinkers, 62% reported drinking daily and 37% engaging in heavy episodic drinking (Supplementary Table 2 ). Among women, only 2% drank alcohol at least weekly (mean intake 116 g per week), but there were similar associations with other baseline characteristics (Table 1 and Supplementary Tables 3 and 4 ) compared to those in men.

During a median of 12.1 (interquartile range 11.1–13.1) years of follow-up, 134,641 men (44,027 drinkers) and 198,430 women (4,420 drinkers) experienced at least one reported hospitalization event or death at age-at-risk 35–84 years, involving a total of 1,111,495 hospitalization episodes. Among men, there were 333,541 (107,857 in current drinkers) recorded events from 207 diseases across 17 International Classification of Diseases Tenth Revision (ICD-10) chapters studied that had at least 80 cases each among current drinkers (Table 2 ), while among women there were 476,986 (11,773) events from 48 diseases across 18 ICD-10 chapters (Supplementary Table 5 ).

Observational associations of alcohol with disease risks

Among men, alcohol drinking was significantly associated with higher risks of 61 disease outcomes from 15 ICD-10 chapters based on two separate analyses, (1) comparing ever-regular versus occasional drinkers and (2) dose–response among current drinkers (Table 2 and Extended Data Fig. 1 ). In each of the analyses in men, there were significant associations of alcohol consumption with 42 diseases (or outcomes), of which 23 were significant in both analyses and the remainder were directionally consistent with one exception (transient cerebral ischemic attacks, ICD-10 code G45) (Fig. 1 ). In further analyses covering all alcohol consumption categories, there were typical U-shaped or J-shaped associations, with excess risks in male ex-drinkers and non-drinkers compared to occasional or moderate drinkers for most of these diseases (Supplementary Table 6 ). Among male ex-drinkers, the overall excess morbidity risks were more considerable for alcohol-associated diseases than for other diseases, but these excess risks were lower with increasing duration after stopping drinking (Extended Data Fig. 2 ).

Cox models ( a ) comparing ever-regular drinkers with occasional drinkers or ( b ) assessing the dose–response per 280 g per week higher usual alcohol intake within current drinkers, were stratified by age at risk and study area and were adjusted for education and smoking. Each solid square represents HR with the area inversely proportional to the variance of the log HR. The horizontal lines indicate 95% CIs. Diseases considered to be alcohol-related by the WHO are indicated with ‘W’ under the ‘WHO’ column. The individual diseases listed included all that showed FDR-adjusted significant associations with alcohol (FDR-adjusted P  < 0.05, indicated with ‘Y’ under the ‘FDR sig.’ column) and WHO alcohol-related diseases that showed nominally significant associations with alcohol ( P  < 0.05). All P values are two-sided. † Included less-common ICD-10 codes within the corresponding ICD-10 chapter that were not individually investigated in the present study. ‘Less-common psychiatric and behavioral conditions’ consisted of ICD-10 codes F00–F99, excluding F32, F33 and F99. ‘Less-common circulatory diseases’ consisted of ICD-10 codes I00–I99, excluding I10, I11, I20, I21, I24, I25, I27, I42, I46, I48–I51, I60–I67, I69, I70, I80 and I83. ‘Less-common injury, poisoning and other external causes’ consisted of ICD-10 codes S00–T98, excluding S06, S09, S22, S32, S42, S52, S62, S72, S82, S92 and T14.

Of the 61 diseases positively associated with alcohol intake in male participants, 28 were considered by the WHO to be alcohol-related diseases, including tuberculosis (A15–A19 and B90), six site-specific cancers including cancers in the larynx (C32), esophagus (C15), liver (C22), colon (C18), rectum (C19 and C20) and lips, oral cavity and pharynx (C00–C14), diabetes (E10–E14), epilepsy (G40 and G41), several hypertensive diseases (I10 and I11) and cerebrovascular diseases (I61, I63, I65, I66, I67, I69 and G45), chronic IHD (I25), cardiomyopathy (I42), pneumonia (J12–J18), alcoholic liver disease (K70) and liver cirrhosis (K74), pancreatitis (K85 and K86) and external causes including self-harm (X60–X84), falls (W00–W19), transport accidents (V01–V99) and other external causes (rest of V–Y) (Fig. 1 and Extended Data Fig. 3 ). Of these 28 diseases, 22 showed significant dose–response associations with alcohol intake. The hazard ratios (HRs) per 280 g per week higher intake for the aggregated WHO alcohol-related diseases were 1.22 (95% confidence interval (CI) 1.19–1.25) (Supplementary Table 7 for detailed outcome classification), ranging from 1.12 (1.05–1.20) for pneumonia to 1.97 (1.80–2.15) for esophageal cancer.

The 33 other diseases showing false discovery rate (FDR)-adjusted significant positive associations with alcohol drinking in men included lung (C34) and stomach (C16) cancers, cataract (H25 and H26), six digestive diseases such as gastroesophageal reflux disease (K21) and gastric ulcer (K25), three musculoskeletal conditions, including gout (M10), three fracture types (S22, S42 and S72), and the aggregates of less-common psychiatric and behavioral conditions and circulatory diseases (Fig. 1 and Extended Data Fig. 4 ). Of these 33 diseases, 22 showed significant dose–response associations, with HRs per 280 g per week higher intake ranging from 1.16 (95% CI 1.04–1.30) for lung cancer to 1.94 (1.43–2.63) for purpura and other hemorrhagic conditions (D69) and 1.20 (1.16–1.24) for the aggregated CKB new alcohol-associated diseases. In contrast, three diseases showed FDR-adjusted significant inverse associations with alcohol drinking (other nontoxic goiter (E04), hyperplasia of prostate (N40) and inguinal hernia (K40)). Overall, for all-cause morbidity, the HR per 280 g per week higher intake was 1.12 (1.10–1.14) in male current drinkers.

Supplementary Figs. 2 – 4 show the dose–response associations for all disease outcomes investigated in male current drinkers. For alcohol-associated diseases and for total morbidity, the dose–response associations were unaltered after additional covariate adjustments or excluding participants with poor baseline health conditions (Supplementary Fig. 5 and Supplementary Table 8 ). Moreover, the associations were similar across various male population subgroups, but seemed to be stronger in younger men, urban residents and higher socioeconomic groups for new alcohol-associated diseases (Supplementary Fig. 6 ).

Among male current drinkers, drinking daily, heavy episodic drinking and drinking spirits were each associated with higher risks for alcohol-related diseases, but most of these associations were attenuated to the null after adjusting for total alcohol intake (Extended Data Fig. 5 ); however, for a given total alcohol intake among male current drinkers, drinking daily was associated with 30–40% higher risks of alcohol-related cancers (1.30, 1.17–1.45) and liver cirrhosis (1.39, 1.13–1.72), compared to non-daily drinking. Similarly, heavy episodic drinking was associated with higher risks of diabetes (1.23, 1.12–1.34) and IHD (1.11, 1.03–1.19), whereas drinking outside of meals was associated with 49% (1.49, 1.19–1.86) higher risk of liver cirrhosis than drinking with meals. The risks of all major alcohol-associated diseases were higher with longer duration of alcohol consumption in men (Extended Data Fig. 6 ).

Among women, due to few reported current drinkers there was a lack of statistical power to detect any associations of self-reported alcohol intake with disease risks (Supplementary Table 5 , Extended Data Fig. 7 and Supplementary Fig. 7 ).

Genetic associations of alcohol with disease risks

A genetic instrument for alcohol intake was derived using ALDH2 - rs671 (G > A) and ADH1B - rs1229984 (G > A) genotypes. The overall A-allele frequency was 0.21 for ALDH2 - rs671 and 0.69 for ADH1B - rs1229984 , with both A-alleles being more common in southern than northern study areas (Supplementary Table 9 ). Both ALDH2 - rs671 and, to a lesser extent, ADH1B - rs1229984 were strongly associated with alcohol drinking in men, but much less so in women (Supplementary Table 10 ). In men, the derived genetic instrument predicted a >60-fold difference (range 4–255 g per week, C1 to C6) in mean alcohol intake, whereas in women mean alcohol intake remained low (<10 g per week) across genetic categories (Supplementary Table 11 ). Both variants and the derived instrument were not associated with smoking or other major self-reported baseline characteristics, except for a small difference in fresh fruit intake by ALDH2 - rs671 genotype in men.

Among men, genotype-predicted mean alcohol intake was positively associated with higher risks of CKB WHO alcohol-related (HR per 280 g per week higher genotype-predicted mean male alcohol intake: 1.14, 95% CI 1.09–1.20) and CKB new alcohol-associated (1.06, 1.01–1.12) diseases (Fig. 2 ), both of which were slightly weaker than the conventional associations. For certain diseases, however, the genetic associations were stronger, with HRs of 1.38 (1.27–1.49) for stroke, 2.30 (1.58–3.35) for liver cirrhosis and 2.33 (1.49–3.62) for gout, in men (Fig. 3 and Extended Data Fig. 8 ). For individual genetic variants, the associations were directionally consistent (Extended Data Figs. 9 and 10 ). Conversely, there were no significant dose–response genotypic associations with IHD, inguinal hernia or hyperplasia of prostate in men. For other alcohol-associated diseases, higher genotype-predicted mean male alcohol intake was significantly associated with higher risks of esophageal cancer, cataract, occlusion and stenosis of cerebral arteries, sequelae of cerebrovascular disease, essential primary hypertension and fractures of ribs, sternum or thoracic spine. There were also suggestive positive genotypic associations with several digestive tract cancer types (liver, colon and stomach) and circulatory and digestive diseases, and significant inverse associations with lung cancer and other chronic obstructive pulmonary disease (J44) in men (Extended Data Figs. 8 – 10 ). Sensitivity analyses using different analytical methods to adjust for confounding by study area, or a two-stage least-squares MR approach, did not alter the main genetic findings in men (Supplementary Table 12 ). In contrast, genotypes that increased alcohol intake in men were not adversely associated with most alcohol-related disease risks among women (for example HR 1.00 (0.97–1.04) for all morbidity among female non-drinkers; Supplementary Fig. 7 and Extended Data Figs. 8 – 10 ).

Each box represents HR with the area inversely proportional to the variance of the group-specific log hazard within subplot. The vertical lines indicate group-specific 95% CIs. Conventional epidemiological analyses relate self-reported drinking patterns to risks of diseases (reference group is occasional drinkers), using Cox models stratified by age at risk and study area and adjusted for education and smoking. Within current drinkers, HRs were plotted against usual alcohol intake and were calculated per 280 g per week higher usual alcohol intake. Genetic epidemiological analyses relate genetic categories to risks of diseases (reference group is the genotype group with lowest genotype-predicted mean male alcohol intake), using Cox models stratified by age at risk and study area and adjusted for genomic principal components. The HR per 280 g per week higher genotype-predicted mean male alcohol intake was calculated from the inverse-variance-weighted mean of the slopes of the fitted lines in each study area. The corresponding slopes in women were summarized in text and the slopes of the fitted line by sex were compared and assessed for heterogeneity using chi-squared tests (indicated by P for heterogeneity by sex). All P values are two-sided. Analyses of these aggregated outcomes were based on first recorded event of the aggregate during follow-up and participants may have had multiple events of different types of diseases. ‘All alcohol-related diseases’ includes the first recorded event from ‘CKB WHO alcohol-related diseases’ or ‘CKB new alcohol-associated diseases’ during follow-up.

Each box represents HR with the area inversely proportional to the variance of the group-specific log hazard within subplot. The vertical lines indicate group-specific 95% CIs. Conventional epidemiological analyses relate self-reported drinking patterns to risks of diseases (reference group is occasional drinkers), using Cox models stratified by age at risk and study area and adjusted for education and smoking. Within current drinkers, HRs were plotted against usual alcohol intake and were calculated per 280 g per week higher usual alcohol intake. Genetic epidemiological analyses relate genetic categories to risks of diseases (reference group is the genotype group with lowest genotype-predicted mean male alcohol intake), using Cox models stratified by age at risk and study area and adjusted for genomic principal components. The HR per 280 g per week higher genotype-predicted mean male alcohol intake was calculated from the inverse-variance-weighted mean of the slopes of the fitted lines in each study area. The corresponding slopes in women were summarized in text and the slopes of the fitted line by sex were compared and assessed for heterogeneity using chi-squared tests (indicated by P for heterogeneity by sex). All P values are two-sided. Corresponding ICD-10 codes, IHD (I20–I25); stroke (I60, I61, I63 and I64); liver cirrhosis (K70 and K74); gout (M10); inguinal hernia (K40); hyperplasia of prostate (N40).

Hospitalizations associated with alcohol drinking

Among men, ever-regular drinkers had higher numbers of hospitalizations for any causes than occasional drinkers, particularly for cancer hospitalizations, and these differences increased with increasing age at risk, except for CVD hospitalizations (Supplementary Fig. 8 ).

This prospective study provides a comprehensive assessment of the impact of alcohol consumption on a very wide range of disease outcomes in Chinese adults. Among men, alcohol consumption was associated with significantly higher risks of 61 diseases, including 33 not previously reported as alcohol-related diseases by the WHO, and higher risks of hospitalizations for any causes. For a given total amount, drinking daily, heavy episodic drinking and drinking outside of meals exacerbated the risks of four major diseases in Chinese men. Moreover, most of these associations in Chinese men were confirmed in genetic analyses, at least when assessed collectively, and are likely to reflect the effects alcohol consumption itself rather than any pleiotropic effects of the genetic instruments.

Based primarily on observational findings in western populations, alcohol consumption has been considered by the WHO 4 and the Global Burden of Disease (GBD) study 23 to be related to about 20 distinct disease categories, involving chronic diseases and cancers largely in the gastrointestinal system, several CVD types, infectious diseases and injuries. The observational analyses largely confirmed these known associations (Supplementary Table 13 ), but also provided insights into additional hazards of certain drinking patterns suggested by previous studies 8 , 9 , 24 , 25 . Moreover, this study discovered 33 additional alcohol-associated diseases across various body systems in Chinese men that had not been previously reported by the WHO. For these 33 disease outcomes, their associations with alcohol intake were confirmed in genetic analyses, at least collectively as well as for certain specific diseases (for example gout), as was the case for a similar number of WHO alcohol-related diseases. The somewhat smaller relative (but not absolute) risks of alcohol drinking with major diseases at older than younger age in men from observational analyses were consistent with previous studies of other risk factors (for example blood pressure 26 and smoking 27 ), which could be driven by a number of factors such as selection bias 27 and comorbidities.

For certain major WHO alcohol-related diseases, particularly IHD and ischemic stroke, observational studies, including this study, have consistently reported J-shaped associations, with those who drank moderately (for example 1–2 units a day) having the lowest risks 6 , 28 ; however, these apparent protective effects of moderate drinking probably largely reflect residual confounding (for example non-drinkers having worse health and socioeconomic profiles than occasional drinkers) and uncontrolled reverse causation (for example sick-quitter effect where pre-existing poor health or changes in health conditions lead to alcohol cessation), including the difficulty in defining abstainers (for example ex-drinkers may be reported as non-drinkers) as the reference group in many previous studies 3 , 29 . In this study, we used occasional drinkers rather than non-drinkers as the reference group, which, together with separate dose–response analyses among current drinkers, helped to reduce but not eliminate any such biases, which could largely be mitigated in genetic analyses using an MR approach.

To date the existing MR studies for alcohol have focused mainly on CVD types 30 , 31 , 32 and cancers 33 , 34 , 35 , with limited data for other diseases. Moreover, previous studies mainly involved European-ancestry populations and hence were constrained by availability of relatively weak genetic instruments. Using genetic instruments specific to East Asian populations that predicted >60-fold difference in alcohol consumption, we previously reported evidence for the causal relevance and apparent dose–response effects of alcohol consumption on upper-aerodigestive tract cancers 16 and stroke 19 . These findings were further corroborated by subsequent European ancestry-based MR studies 30 , 32 , 36 and the analyses presented in this study with additional follow-up data. In contrast to stroke, we found no reliable genetic evidence for a cardioprotective, nor harmful, effect of moderate drinking on risk of IHD in men, consistent with findings in other MR studies 30 , 32 . The present study also demonstrated a log-linear genetic association of alcohol with liver cirrhosis and suggestive positive associations for several WHO alcohol-related digestive tract cancers in men. Moreover, separate genetic analyses among women suggests that the excess risks observed among men were due chiefly to alcohol per se rather than to potential pleiotropic effects of the alcohol-related genotypes. Further larger genetic studies are required to confirm and elucidate the potential causal relevance for each of the other WHO alcohol-related diseases individually.

For the new alcohol-associated diseases identified in this study, the available prospective epidemiological evidence has been sparse and mostly confined to western populations. For gout, previous western prospective studies have reported positive associations 37 , 38 and an MR study of 8,000 Korean men has also reported positive associations of alcohol consumption with hyperuricemia, a risk factor for gout 39 . The present study provides genetic evidence that alcohol drinking increases the risk of gout. Consistent with the present study, previous European-ancestry-based observational studies 40 , 41 and one MR study 42 also reported positive associations of alcohol intake with risks of several fracture types. The available prospective evidence on associations between alcohol drinking and risk of cataract has been conflicting 15 , 43 and one European-ancestry-based MR study reported no genetic associations 44 . We found a significant dose–response association between alcohol and risk of cataract among Chinese men in observational analyses, which was supported by the present genetic analyses.

For several other diseases (for example gastroesophageal reflux disease and gastric ulcer), the observational findings provide additional evidence to the existing literature 5 , 45 , 46 , 47 , but the supporting genetic evidence is still constrained by limited statistical power. Similarly, our observational findings for lung and stomach cancers were generally consistent with evidence provided by previous prospective studies 7 , 11 , 48 , 49 ; however, the causal relevance of these associations remains to be elucidated in future larger MR studies with appropriate consideration of the potential gene–environment interactions between ALDH2 - rs671 and alcohol intake (the effect of alcohol intake on cancer risks being modified by ALDH2 - rs671 genotype due to excessive acetaldehyde) 16 and other aldehyde exposures 50 in cancer risks, which might similarly affect the genetic associations for respiratory diseases and other potential acetaldehyde-related diseases. In observational analyses, we found significant inverse associations for inguinal hernia, prostate hyperplasia and other nontoxic goiter, but not for several other diseases previously inversely associated with alcohol drinking, including non-Hodgkin lymphoma 48 , kidney cancer 48 , thyroid cancer 48 and gallstones 51 . The genetic analyses, albeit with limited power, did not provide reliable evidence supporting the inverse associations with these outcomes. Future well-powered genetic investigations are warranted for less-common diseases in different populations.

The strengths of this study include the prospective design, large sample size, detailed information on alcohol consumption and drinking patterns, completeness of follow-up and a wide range of morbidity outcomes analyzed. We were also able to assess the potential causal relevance of the associations using two powerful East Asian genetic variants. Moreover, the extremely low drinking prevalence in women (regardless of their genotypes) enabled assessment for potential pleiotropy, further supporting the genetic findings among men.

Nevertheless, the study also has limitations. First, it is still possible that heavy drinking was under-reported, which could have underestimated the hazards of heavy episodic drinking. Second, as in many population-based cohort studies, extreme problematic drinkers and certain alcohol-related disease events may be under-represented, but this should not affect the assessment of the associations of alcohol with most disease outcomes. Third, while the repeated measures of alcohol consumption available in the re-survey subsets allowed us to estimate long-term usual mean alcohol intake at the group level to account for regression dilution bias, we were unable to study the effects of longitudinal alcohol drinking trajectories on health. Fourth, we were unable or underpowered to study diseases that do not normally require hospitalization (for example dementia and depression), nor alcohol-related diseases only affecting women, given the low proportion of female drinkers (for example <70 cases of breast cancer in female drinkers). While the low female drinking prevalence in CKB was consistent with findings in a nationwide survey 52 , it is possible that women may be more likely to under-report drinking than men for cultural and social reasons. Hence our null findings in women should be interpreted with caution and not be taken as a lack of alcohol-related harms in women in general, especially in the context of rising alcohol consumption among Asian women 2 . Fifth, as spirits were the main beverage type and our genetic instrument did not distinguish between beverage types, we were unable to assess beverage-specific effects on disease risks, including wine consumption, which is uncommon in China 17 and has been proposed as potentially cardioprotective due to other non-alcoholic components in red wine 53 . Sixth, although our genetic analyses allowed comparison of the overall genetic effects of negligible, moderate and high mean alcohol intake levels for major and overall morbidities, we had limited power to confidently clarify any small threshold effects in the low consumption end, especially for individual diseases. Finally, the genetic analyses lacked statistical power to assess the associations with several individual alcohol-associated diseases so these findings should still be viewed as hypothesis-generating.

In recent decades, several studies have estimated the alcohol-attributable disease burden, involving predominantly WHO alcohol-related diseases. These estimates were based mainly on observational evidence and included the potentially biased U- or J-shaped associations with IHD and ischemic stroke 1 , 23 , 54 . We have demonstrated in both conventional and genetic analyses that alcohol drinking is associated with hazards in a dose–response manner with a much wider range of disease outcomes than previously considered by the WHO 4 and the GBD study 23 and do not find any evidence for protective effects for IHD or stroke, suggesting that the actual alcohol-attributable disease burden is likely to be much greater than widely believed.

Overall, the present study demonstrated substantial hazards of alcohol consumption with a wide range of disease outcomes among Chinese men. The findings reinforce the need to lower population mean levels of alcohol consumption as a public health priority in China. Future estimation of the alcohol-attributable disease burden worldwide and in specific regions should incorporate new genetic evidence from the present and any future studies about the likely causal relevance of alcohol consumption for a broad range of disease outcomes.

Study population

Details of the CKB study design and methods have been previously reported 55 . Briefly, 512,724 adults aged 30–79 years were recruited from ten geographically diverse (five rural and five urban) areas across China during 2004–2008. At local study assessment clinics, trained health workers administered a laptop-based questionnaire recording sociodemographic factors, lifestyle (for example alcohol drinking, smoking, diet and physical activity) and medical history; undertook physical measurements (for example blood pressure and anthropometry); and collected a blood sample for long-term storage. Two resurveys of ~5% randomly selected surviving participants were subsequently conducted in 2008 and 2013–2014 using similar procedures.

Ethics approval

Ethical approval was obtained from the Ethical Review Committee of the Chinese Centre for Disease Control and Prevention (Beijing, China, 005/2004) and the Oxford Tropical Research Ethics Committee, University of Oxford (UK, 025-04). All participants provided written informed consent.

Assessment of alcohol consumption

Detailed questionnaire assessment of alcohol consumption has been described previously 3 , 17 , 56 . In the baseline questionnaire, participants were asked how often they had drunk alcohol during the past 12 months (never or almost never, occasionally, only at certain seasons, every month but less than weekly or usually at least once a week). Those who had not drunk alcohol at least weekly in the past 12 months were asked whether there was a period of at least a year before that when they had drunk some alcohol at least once a week. Based on their past and current drinking history, participants were classified into: non-drinkers (had never drunk alcohol in the past year and had not drunk in most weeks in the past); ex-drinkers (had not drunk alcohol in most weeks in the past year but had done so in the past); occasional drinkers (had drunk alcohol but less than weekly in the past year and had not drunk alcohol in most weeks in the past); and current drinkers (had drunk alcohol on a weekly basis (regularly) in the past year).

Current drinkers were asked further questions about their drinking patterns, including frequency, beverage type (beer, grape wine, rice wine, weak spirits with <40% alcohol content and strong spirits with ≥40% alcohol content) and amount consumed on a typical drinking day, mealtime drinking habits, age started drinking in most week and their experience of flushing or dizziness after drinking.

Alcohol intake level was estimated based on the reported frequency (taken as the median of the reported frequency intervals; 1.5 for 1–2 d per week, 4 for 3–5 d per week, 6.5 for 6–7 d per week), beverage type and amount consumed, assuming the following alcohol content by volume (v/v) typically seen in China: beer 4%, grape wine 12%, rice wine 15%, weak spirits 38% and strong spirits 53% 57 . Among current drinkers, men were grouped into four consumption categories (<140, 140–279, 280–419 and 420+ g per week) and women into three categories (<70, 70–139 and 140+ g per week), broadly based on the recommended cutoffs for alcohol categories by the WHO 58 and national drinking guidelines. Heavy episodic drinking was defined as consuming >60 g of alcohol on a typical drinking occasion for men and >40 g per occasion for women 58 . Drinking outside of meals was defined as usually drinking between or after meals or having no regular patterns (versus usually drinking with meals). Duration of drinking was derived by the difference in years between age at baseline and age started drinking.

Ex-drinkers were asked how long (in years) ago they had stopped drinking in most weeks. Ex-drinkers were grouped with current drinkers as ‘ever-regular drinkers’.

Follow-up for mortality and morbidity

The vital status of participants was obtained periodically from local death registries, supplemented by annual active confirmation through local residential, health insurance and administrative records. Additional information on morbidity was collected through linkage with disease registries (for cancer, stroke, IHD and diabetes) and the national health insurance system, which record any episodes of hospitalization and almost has universal coverage. All events were coded with ICD-10 codes, blinded to the baseline information. By 1 January 2019, 56,550 (11%) participants had died, 311,338 (61%) were ever hospitalized, but only 4,028 (<1%) were lost to follow-up.

Outcome measures

To enable a ‘phenome-wide’ investigation, all recorded diseases and injuries (referred to as ‘diseases’ for simplicity) coded by three-character ICD-10 codes were reviewed. ICD-10 codes were combined (where appropriate) based on disease characteristics and their potential relationships with alcohol consumption 4 , 8 , 10 , 59 . Disease end points were curated based on diseases considered to be causally impacted by alcohol by the WHO 4 , 59 and major diseases previously shown to be related to alcohol in CKB and other large prospective cohort studies 8 , 10 , while retaining maximal granularity. Diseases with at least 80 cases recorded during follow-up among current drinkers, separately by sex, were analyzed individually to capture a wide range of specific conditions while ensuring reasonable statistical power (around 60–80% power to detect a HR of 2.00 per 280 g per week higher usual alcohol intake at P  < 0.01 and P  < 0.05, respectively). Within each ICD-10 chapter, diseases with <80 events were grouped into a ‘less-common’ category. Several ICD-10 chapters considered not directly relevant in this population (for example perinatal-origin diseases (chapter XVI) and congenital conditions (XVII); pregnancy-related diseases (XV) in men) were excluded.

Major diseases defined by the WHO as likely to be causally related with alcohol consumption 4 , including several cancers (mouth and throat, esophagus, colon-rectum, liver and female breast), diabetes mellitus, IHD, stroke, liver cirrhosis and external causes, were also selected a priori for detailed analyses of associations with drinking patterns (daily drinking, heavy episodic drinking, mealtime habit, spirit drinking and drinking duration). Similarly, diseases that were significantly and adversely associated with alcohol in the ‘phenome-wide’ investigations (either with ever-regular versus occasional drinking or in dose–response associations with amounts consumed) were further categorized as ‘CKB WHO alcohol-related diseases’ and ‘CKB new alcohol-associated diseases’ respectively for genetic investigation of causality. Detailed outcome classifications are reported in Supplementary Table 7 .

Genotyping and alcohol genetic instruments

The two East Asian genetic variants ( ALDH2 - rs671 and ADH1B - rs1229984 ) were genotyped in 168,050 participants (151,347 randomly selected, 16,703 selected as part of nested case–control studies of CVD and chronic obstructive pulmonary disease, which were only included in analyses of relevant outcomes; Supplementary Fig. 1 ) using Affymetrix Axiom ( n  = 100,396) or custom Illumina GoldenGate ( n  = 93,125) arrays at BGI (Shenzhen, China), with some overlap between them. Among 25,471 participants genotyped with both arrays, the concordance was >99.9% for both variants. Where discordant, genotypes obtained from the Affymetrix Axiom array were used.

The genetic instrument for alcohol was derived from ALDH2 - rs671 and ADH1B - rs1229984 and ten study areas from the random genotyped subset of male participants to avoid potential selection bias, using a previously developed method in CKB 19 . Briefly, nine genotype combinations were defined based on the genotypes for each of the two variants (each AA, AG or GG). As alcohol use varies greatly by study area, among men, mean alcohol intake was calculated for each of these nine genotype across ten study areas (that is a total of 90 genotype-area combinations) to reflect a wide range of alcohol consumption, assigning an intake of 5 g per week to occasional drinkers and excluding ex-drinkers from the calculation. Ex-drinkers were excluded from the calculation of mean alcohol intake as their baseline intake did not reflect their long-term intake; nevertheless, they were included in subsequent genetic analyses once they had been assigned a genetic group. These 90 combinations were then grouped into six categories (C1–C6) according to their corresponding mean intake values, at cutoff points of 10, 25, 50, 100 and 150 g per week, selected to facilitate investigation of the causal effects of alcohol across a wide range of mean alcohol intakes while allowing adequate sample size in each category for reliable comparisons. In this way participants (including ex-drinkers) were classified only based on their genotypes and study area, but not on individual self-reported drinking patterns. Comparisons of these six genetic categories can, where analyses are stratified by area, be used to estimate the genotypic effects on disease risks.

To facilitate the comparison of genotypic effects between sexes (pleiotropic effects), women were classified into the same six categories as men based on their genotypes and study area, regardless of female alcohol intake. This allowed comparison of genotypic effects between men (where genotype were strongly associated with alcohol intake) and women (where alcohol intake was low in all genotypic categories) (Supplementary Tables 10 and 11 ).

Statistical analysis

Given the extremely low alcohol use among women 3 , 17 , the analyses were conducted separately by sex but focused chiefly on men. All CKB participants and the genotyped subset with genomic principal components (PCs; derived from genome-wide genotyping array data and were informative for CKB population structure) 60 were included in conventional and genetic analyses, respectively (Supplementary Fig. 1 ). Means and percentages of baseline characteristics were calculated by self-reported alcohol consumption patterns and by genotype categories, adjusted for age (in 10-year intervals), ten study areas and (for genetic analysis) genomic PCs 60 to control for differences in genetic distribution due to population stratification, as appropriate.

For conventional observational analyses, Cox proportional hazard models were used to estimate HRs for individual diseases associated with different alcohol consumption categories (in three broad categories: occasional drinkers, ever-regular drinkers, non-drinkers; and in 6–7 detailed categories: occasional drinkers, ex-drinkers, non-drinkers, 3–4 further current drinker groups defined by alcohol intake level) and among current drinkers with continuous levels of alcohol intake (per 280 g per week in men, per 100 g per week in women) or with categories of alcohol intake (<140, 140–279, 280–419 and 420+ g per week in men; <70, 70–139 and 140+ g per week in women). The Cox models were stratified by age at risk (5-year groups between 35–84 years) and ten areas and adjusted for education (four groups: no formal school, primary school, middle or high school and technical school/college or above) and smoking status (six groups in men: never, occasional, ex-regular, current <15, current 15–24, current ≥25 cigarettes equivalent per day; four groups in women: never, occasional, ex-regular and current). Smoking data have been previously validated against exhaled carbon monoxide 61 . Competing risks from all-cause mortality for disease events were handled by censoring participants at death from any cause to estimate cause-specific HRs comparing event rates in participants who were alive and free of the disease of interest 62 . To reduce biases from residual confounding and uncontrolled reverse causation related to the choice of using non-drinkers (for example sick-quitter effect, pre-existing poor health or social disadvantages leading to alcohol cessation or abstinence) as the reference group 3 , 29 , we used occasional drinkers as the reference group, together with separate dose–response analyses among current drinkers. To account for within-person variation of alcohol intake over the follow-up period, repeat alcohol measures for participants who attended the two resurveys were used to estimate usual alcohol intake (Supplementary Table 1 ) and correct for regression dilution bias 9 , 63 . The shapes of dose–response associations between alcohol and disease risks were assessed among current drinkers by plotting the HRs of predefined baseline consumption categories against the corresponding mean usual alcohol intake. Log HR estimates and the corresponding standard errors for baseline alcohol intake, modeled as a continuous variable, were divided by the regression dilution ratio (0.53 for both men and women; calculated using the McMahon–Peto method 64 ) to obtain estimated HRs per 280 g per week higher usual alcohol intake among male current drinkers and HRs per 100 g per week among female current drinkers. For analyses involving drinking patterns, additional adjustments were conducted for total alcohol intake (continuous) and baseline age (continuous; for drinking duration analysis) where appropriate.

Sensitivity analyses were performed by (1) additional adjustments for further covariates (household income (<10,000, 10,000–19,999, 20,000–34,999 and ≥35,000 yuan per year), fresh fruit intake (4–7 d per week and ≤3 d per week), physical activity (continuous, in metabolic equivalent of task per hour per day), body mass index (<22, 22–24.9, 25–26.9 and ≥27 kg m 2 ); and (2) excluding individuals with poor self-reported health or previous major chronic diseases (including self-reported coronary heart diseases, stroke, transient ischemic attack, tuberculosis, emphysema or bronchitis, liver cirrhosis or chronic hepatitis, peptic ulcer, gallstone or gallbladder disease, kidney disease, rheumatoid arthritis, cancer and diabetes) at baseline. For all aggregated end points (for example CKB WHO alcohol-related, CKB new alcohol-associated and all morbidity), subgroup analyses were conducted by baseline age (<55, 55–64 and ≥65 years), area (urban and rural), education (primary school or below, middle school, high school or above), household income (<10,000, 10,000–19,999 and 20,000+ yuan per year) and smoking status (ever-regular and never-regular), with heterogeneity or trend assessed by chi-squared tests 65 . HRs for diseases associated with years of stopping among ex-drinkers compared to occasional drinkers were also estimated.

In genetic analyses, Cox regression, stratified by age at risk and study area and adjusted for 11 genomic PCs 60 , were used to estimate HRs for major alcohol-related diseases associated with the six genetic categories (C1–C6). Log HRs were plotted against the genotype-predicted mean male alcohol intake in the six categories. To control for potential confounding by population structure, similar analyses were repeated within each study area using age-at-risk-stratified and genomic PC-adjusted Cox models. A line of best fit was fitted through the log HRs against genotype-predicted mean male alcohol intake in the genetic categories present in the corresponding study area, using meta-regression. These within-area slopes (each reflecting purely genotypic effects) were combined by inverse-variance-weighted meta-analysis to yield the overall area-stratified genotypic associations, which controlled for any potential bias resulted from variations due to population structure, summarized as HR per 280 g per week higher genotype-predicted mean male alcohol intake. For total morbidity and aggregated alcohol-associated outcomes, sensitivity analyses were performed by (1) using age-at-risk- and area-stratified and genomic PC-adjusted Cox models to estimate HR per 280 g per week (area-adjusted genotypic associations); and (2) using a two-stage least-squares approach 66 .

Genotypic analyses in women were conducted not to assess the health effects of alcohol in women, but to investigate the extent to which the genotypes studied in men had pleiotropic effects (genotypic effects not mediated by drinking patterns). As few women consumed alcohol, any genotypic effects of the six genetic categories that are mediated by drinking alcohol should be much smaller in women than in men, but any other pleiotropic genotypic effects should be similar in both sexes. Hence, among women, we used the same genetic categories as in men and related the genotypic effects in women to the mean male alcohol intake in these six categories, which allows comparisons of genetic findings by sex and assessment of potential pleiotropy. To further remove the small genotypic effects on alcohol use in women (Supplementary Tables 10 and 11 ), we restricted the genetic analyses to female non-drinkers in sensitivity analyses.

The genotypic associations of individual genetic variants ( rs671 , rs1229984 ; GG versus AG genotype) with alcohol-related disease risks were also assessed using a similar area-stratified approach.

The proportional hazards assumption was tested using scaled Schoenfeld residuals for the pre-specified major diseases (no clear evidence of violation was found). For analyses involving more than two exposure categories, the floating absolute risks were used to estimate group-specific 95% CIs for all categories including the reference group 9 , 19 , 67 . All P values were two-sided. Statistical significance (at the 5% level) was evaluated using both FDR-adjusted P values applied within ICD-10 chapters to correct for multiple testing in the ‘phenome-wide’ investigation 68 , 69 , 70 and conventional P values for hypothesis testing for observational analyses of WHO alcohol-related diseases, analyses of drinking patterns and genetic analyses.

To assess the cumulative burden of alcohol consumption, the total number of hospitalizations were estimated for ever-regular versus occasional drinkers using the mean cumulative count, which does not assume independence between hospitalizations and all-cause mortality 71 , 72 , 73 . All analyses used R software (v.4.0.5).

Ethics and inclusion statement

In accordance with the Nature Portfolio journals’ editorial policies, the research has included local researchers from China throughout the research process, including study design, study implementation, data ownership and authorship. The roles and responsibilities were agreed among collaborators ahead of the research and capacity-building plans, including data collection and study implementation skills for local researchers, were discussed and delivered. This research is locally relevant to the studied country and included local collaborative partners in all aspects of the study, thus, will provide local and regional organizations with epidemiological evidence on the health impacts of alcohol consumption to inform public health policies.

This research was not restricted nor prohibited in the setting of the researchers. The study was approved by local ethics review committee. The research raised no risks related to stigmatization, incrimination, discrimination, animal welfare, the environment, health, safety, security or other personal or biorisks. No biological materials, cultural artifacts or associated traditional knowledge has been transferred out of the country. In preparing the manuscript, the authors have reviewed and cited local and regional relevant studies.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The CKB is a global resource for the investigation of lifestyle, environmental, blood biochemical and genetic factors as determinants of common diseases. The CKB study group is committed to making the cohort data available to the scientific community in China, the United Kingdom and worldwide to advance knowledge about the causes, prevention and treatment of disease. For detailed information on what data are currently available to open access users, how to apply for them and the timeline for data access (12–16 weeks), please visit the CKB website: https://www.ckbiobank.org/data-access . Researchers who are interested in obtaining the raw data from the CKB study that underlines this paper should contact [email protected]. A research proposal will be requested to ensure that any analysis is performed by bona fide researchers and, where data are not currently available to open access researchers, is restricted to the topic covered in this paper. Further information is available from the corresponding authors upon request.

Code availability

The codes used for the data analyses in this study can be made available by contacting the corresponding authors. Access to codes will be granted for requests for academic use within 4 weeks of application.

Shield, K. et al. National, regional, and global burdens of disease from 2000 to 2016 attributable to alcohol use: a comparative risk assessment study. Lancet Public Health 5 , e51–e61 (2020).

PubMed   Google Scholar  

Manthey, J., Shield, K. D., Rylett, M., Hasan, O. S. M., Probst, C. & Rehm, J. Global alcohol exposure between 1990 and 2017 and forecasts until 2030: a modelling study. Lancet 393 , 2493–2502 (2019).

Im, P. K. et al. Patterns and trends of alcohol consumption in rural and urban areas of China: findings from the China Kadoorie Biobank. BMC Public Health 19 , 217 (2019).

PubMed   PubMed Central   Google Scholar  

World Health Organization. Global Status Report on Alcohol and Health 2018 (World Health Organization, 2018).

Corrao, G., Bagnardi, V., Zambon, A., La & Vecchia, C. A meta-analysis of alcohol consumption and the risk of 15 diseases. Prev. Med. 38 , 613–619 (2004).

Ferrari, P. et al. Lifetime alcohol use and overall and cause-specific mortality in the European Prospective Investigation into Cancer and nutrition (EPIC) study. BMJ Open 4 , e005245 (2014).

Yang, L. et al. Alcohol drinking and overall and cause-specific mortality in China: nationally representative prospective study of 220,000 men with 15 years of follow-up. Int. J. Epidemiol. 41 , 1101–1113 (2012).

Im, P. K. et al. Alcohol drinking and risks of total and site-specific cancers in China: a 10-year prospective study of 0.5 million adults. Int J. Cancer 149 , 522–534 (2021).

CAS   PubMed   PubMed Central   Google Scholar  

Im, P. K. et al. Alcohol drinking and risks of liver cancer and non-neoplastic chronic liver diseases in China: a 10-year prospective study of 0.5 million adults. BMC Med. 19 , 216 (2021).

Allen, N. E. et al. Moderate alcohol intake and cancer incidence in women. J. Natl Cancer Inst. 101 , 296–305 (2009).

Jayasekara, H. et al. Lifetime alcohol intake, drinking patterns over time and risk of stomach cancer: a pooled analysis of data from two prospective cohort studies. Int J. Cancer 148 , 2759–2773 (2021).

Botteri, E. et al. Alcohol consumption and risk of urothelial cell bladder cancer in the European prospective investigation into cancer and nutrition cohort. Int J. Cancer 141 , 1963–1970 (2017).

CAS   PubMed   Google Scholar  

Sabia, S. et al. Alcohol consumption and risk of dementia: 23 year follow-up of Whitehall II cohort study. BMJ 362 , k2927 (2018).

Tan, G. J. al. The relationship between alcohol intake and falls hospitalization: results from the EPIC-Norfolk. Geriatr. Gerontol. Int. https://doi.org/10.1111/ggi.14219 (2021).

Chua, S. Y. L. et al. Alcohol consumption and incident cataract surgery in two large UK cohorts. Ophthalmology https://doi.org/10.1016/j.ophtha.2021.02.007 (2021).

Im, P. K. et al. Alcohol metabolism genes and risks of site-specific cancers in Chinese adults: an 11-year prospective study. Int. J. Cancer 150 , 1627–1639 (2022).

Millwood, I. Y. et al. Alcohol consumption in 0.5 million people from 10 diverse regions of China: prevalence, patterns and socio-demographic and health-related correlates. Int. J. Epidemiol. 42 , 816–827 (2013).

Davey Smith, G. & Hemani, G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum. Mol. Genet. 23 , R89–R98 (2014).

Millwood, I. Y. et al. Conventional and genetic evidence on alcohol and vascular disease aetiology: a prospective study of 500 000 men and women in China. Lancet 393 , 1831–1842 (2019).

Au Yeung, S. L. et al. Moderate alcohol use and cardiovascular disease from Mendelian randomization. PLoS ONE 8 , e68054 (2013).

Shin, M. J., Cho, Y., Davey & Smith, G. Alcohol consumption, aldehyde dehydrogenase 2 gene polymorphisms, and cardiovascular health in Korea. Yonsei Med J. 58 , 689–696 (2017).

Taylor, A. E. et al. Exploring causal associations of alcohol with cardiovascular and metabolic risk factors in a Chinese population using Mendelian randomization analysis. Sci. Rep. 5 , 14005 (2015).

Griswold, M. G. et al. Alcohol use and burden for 195 countries and territories, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 392 , 1015–1035 (2018).

Google Scholar  

Simpson, R. F. et al. Alcohol drinking patterns and liver cirrhosis risk: analysis of the prospective UK Million Women Study. Lancet Public Health 4 , e41–e48 (2019).

Roerecke, M. & Rehm, J. Alcohol consumption, drinking patterns, and ischemic heart disease: a narrative review of meta-analyses and a systematic review and meta-analysis of the impact of heavy drinking occasions on risk for moderate drinkers. BMC Med. 12 , 182 (2014).

Lacey, B. et al. Age-specific association between blood pressure and vascular and non-vascular chronic diseases in 0.5 million adults in China: a prospective cohort study. Lancet Glob. Health 6 , e641–e649 (2018).

Hernán, M. A., Alonso, A. & Logroscino, G. Cigarette smoking and dementia: potential selection bias in the elderly. Epidemiology 19 , 448–450 (2008).

Rehm, J., Rovira, P., Llamosas-Falcón, L. & Shield, K. D. Dose-response relationships between levels of alcohol use and risks of mortality or disease, for all people, by age, sex, and specific risk factors. Nutrients 13 , 2652 (2021).

Rehm, J., Irving, H., Ye, Y., Kerr, W. C., Bond, J. & Greenfield, T. K. Are lifetime abstainers the best control group in alcohol epidemiology? On the stability and validity of reported lifetime abstention. Am. J. Epidemiol. 168 , 866–871 (2008).

Lankester, J., Zanetti, D., Ingelsson, E. & Assimes, T. L. Alcohol use and cardiometabolic risk in the UK Biobank: a Mendelian randomization study. PLoS ONE 16 , e0255801 (2021).

Rosoff, D. B., Davey Smith, G., Mehta, N., Clarke, T.-K. & Lohoff, F. W. Evaluating the relationship between alcohol consumption, tobacco use, and cardiovascular disease: a multivariable Mendelian randomization study. PLoS Med. 17 , e1003410 (2020).

Larsson, S. C., Burgess, S., Mason, A. M. & Michaelsson, K. Alcohol consumption and cardiovascular disease: a Mendelian randomization study. Circ. Genom. Precis. Med. 13 , e002814 (2020).

Larsson, S. C. et al. Smoking, alcohol consumption, and cancer: a mendelian randomisation study in UK Biobank and international genetic consortia participants. PLoS Med. 17 , e1003178 (2020).

Zhou, X. et al. Alcohol consumption, DNA methylation and colorectal cancer risk: results from pooled cohort studies and Mendelian randomization analysis. Int. J. Cancer https://doi.org/10.1002/ijc.33945 (2022).

Zhou, X. et al. Alcohol consumption, blood DNA methylation and breast cancer: a Mendelian randomisation study. Eur. J. Epidemiol. 37 , 701–712 (2022).

Biddinger, K. J. et al. Association of habitual alcohol intake with risk of cardiovascular disease. JAMA Netw. Open 5 , e223849 (2022).

Wang, M., Jiang, X., Wu, W. & Zhang, D. A meta-analysis of alcohol consumption and the risk of gout. Clin. Rheumatol. 32 , 1641–1648 (2013).

Neogi, T., Chen, C., Niu, J., Chaisson, C., Hunter, D. J. & Zhang, Y. Alcohol quantity and type on risk of recurrent gout attacks: an internet-based case-crossover study. Am. J. Med 127 , 311–318 (2014).

Jee, Y. H., Jung, K. J., Park, Y. B., Spiller, W. & Jee, S. H. Causal effect of alcohol consumption on hyperuricemia using a Mendelian randomization design. Int J. Rheum. Dis. 22 , 1912–1919 (2019).

Berg, K. M. et al. Association between alcohol consumption and both osteoporotic fracture and bone density. Am. J. Med. 121 , 406–418 (2008).

Søgaard, A. J. et al. The association between alcohol consumption and risk of hip fracture differs by age and gender in Cohort of Norway: a NOREPOS study. Osteoporos. Int. 29 , 2457–2467 (2018).

Yuan, S., Michaëlsson, K., Wan, Z. & Larsson, S. C. Associations of smoking and alcohol and coffee intake with fracture and bone mineral density: a Mendelian randomization study. Calcif. Tissue Int. 105 , 582–588 (2019).

Gong, Y., Feng, K., Yan, N., Xu, Y. & Pan, C.-W. Different amounts of alcohol consumption and cataract: a meta-analysis. Optom. Vis. Sci. 92 , 471–479 (2015).

Yuan, S., Wolk, A. & Larsson, S. C. Metabolic and lifestyle factors in relation to senile cataract: a Mendelian randomization study. Sci. Rep. 12 , 409 (2022).

Pan, J., Cen, L., Chen, W., Yu, C., Li, Y. & Shen, Z. Alcohol consumption and the risk of gastroesophageal reflux disease: a systematic review and meta-analysis. Alcohol Alcohol. 54 , 62–69 (2019).

Strate, L. L., Singh, P., Boylan, M. R., Piawah, S., Cao, Y. & Chan, A. T. A prospective study of alcohol consumption and smoking and the risk of major gastrointestinal bleeding in men. PLoS ONE 11 , e0165278 (2016).

Yuan, S. & Larsson, S. C. Adiposity, diabetes, lifestyle factors and risk of gastroesophageal reflux disease: a Mendelian randomization study. Eur. J. Epidemiol. 37 , 747–754 (2022).

Bagnardi, V. et al. Alcohol consumption and site-specific cancer risk: a comprehensive dose–response meta-analysis. Br. J. Cancer 112 , 580–593 (2015).

Shen, C., Schooling, C. M., Chan, W. M., Xu, L., Lee, S. Y. & Lam, T. H. Alcohol intake and death from cancer in a prospective Chinese elderly cohort study in Hong Kong. J. Epidemiol. Community Health 67 , 813–820 (2013).

Kuroda, A. et al. Effects of the common polymorphism in the human aldehyde dehydrogenase 2 (ALDH2) gene on the lung. Respir. Res. 18 , 69 (2017).

Cha, B. H., Jang, M. J. & Lee, S. H. Alcohol consumption can reduce the risk of gallstone disease: a systematic review with a dose–response meta-analysis of case–control and cohort studies. Gut Liver 13 , 114–131 (2019).

Ma, G. S., Xhu, D. H., Hu, X. Q., Luan, D. C., Kong, L. Z. & Yang, X. Q. The drinking practice of people in China. Acta Nutrimenta Sin. 27 , 362–365 (2005).

Arranz, S., Chiva-Blanch, G., Valderas-Martínez, P., Medina-Remón, A., Lamuela-Raventós, R. M. & Estruch, R. Wine, beer, alcohol and polyphenols on cardiovascular disease and cancer. Nutrients 4 , 759–781 (2012).

Bryazka, D. et al. Population-level risks of alcohol consumption by amount, geography, age, sex, and year: a systematic analysis for the Global Burden of Disease Study 2020. Lancet 400 , 185–235 (2022).

Chen, Z. et al. China Kadoorie Biobank of 0.5 million people: survey methods, baseline characteristics and long-term follow-up. Int. J. Epidemiol. 40 , 1652–1666 (2011).

Im, P. K. et al. Problem drinking, wellbeing and mortality risk in Chinese men: findings from the China Kadoorie Biobank. Addiction 115 , 850–862 (2020).

Cochrane, J., Chen, H., Conigrave, K. M. & Hao, W. Alcohol use in China. Alcohol Alcohol. 38 , 537–542 (2003).

World Health Organization. International Guide for Monitoring Alcohol Consumption and Related Harm (World Health Organization, 2000).

Rehm, J. et al. The relationship between different dimensions of alcohol use and the burden of disease-an update. Addiction 112 , 968–1001 (2017).

Walters, R. G. et al. Genotyping and population structure of the China Kadoorie Biobank. Preprint at medRxiv https://doi.org/10.1101/2022.05.02.22274487 (2022).

Zhang, Q. et al. Exhaled carbon monoxide and its associations with smoking, indoor household air pollution and chronic respiratory diseases among 512,000 Chinese adults. Int. J. Epidemiol. 42 , 1464–1475 (2013).

Lau, B., Cole, S. R. & Gange, S. J. Competing risk regression models for epidemiologic data. Am. J. Epidemiol. 170 , 244–256 (2009).

Clarke, R., Shipley, M. & Lewington, S. et al. Underestimation of risk associations due to regression dilution in long-term follow-up of prospective studies. Am. J. Epidemiol. 150 , 341–353 (1999).

MacMahon, S., Peto, R. & Cutler, J. et al. Blood pressure, stroke, and coronary heart disease. Part 1, prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 335 , 765–774 (1990).

Early Breast Cancer Trialists’ Collaborative Group. Section 5: Statistical Methods. Treatment of Early Breast Cancer. Worldwide Evidence, 1985–1990 (Oxford University Press, 1990).

Burgess, S. et al. Guidelines for performing Mendelian randomization investigations. Wellcome Open Res. 4 , 186 (2019).

Easton, D. F., Peto, J. & Babiker, A. G. Floating absolute risk: an alternative to relative risk in survival and case-control analysis avoiding an arbitrary reference group. Stat. Med. 10 , 1025–1035 (1991).

Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. B 57 , 289–300 (1995).

Braun, J. M., Kalloo, G., Kingsley, S. L. & Li, N. Using phenome-wide association studies to examine the effect of environmental exposures on human health. Environ. Int. 130 , 104877 (2019).

Glickman, M. E., Rao, S. R. & Schultz, M. R. False discovery rate control is a recommended alternative to Bonferroni-type adjustments in health studies. J. Clin. Epidemiol. 67 , 850–857 (2014).

Cook, R. J. & Lawless, J. F. Marginal analysis of recurrent events and a terminating event. Stat. Med. 16 , 911–924 (1997).

Dong, H., Robison, L. L., Leisenring, W. M., Martin, L. J., Armstrong, G. T. & Yasui, Y. Estimating the burden of recurrent events in the presence of competing risks: the method of mean cumulative count. Am. J. Epidemiol. 181 , 532–540 (2015).

Ghosh, D. & Lin, D. Y. Nonparametric analysis of recurrent events and death. Biometrics 56 , 554–562 (2000).

Download references

Acknowledgements

The chief acknowledgment is to the participants, the project staff and the China National Centre for Disease Control and Prevention (CDC) and its regional offices for assisting with the fieldwork. We thank J. Mackay in Hong Kong; Y. Wang, G. Yang, Z. Qiang, L. Feng, M. Zhou, W. Zhao. and Y. Zhang in China CDC; L. Kong, X. Yu and K. Li in the Chinese Ministry of Health; and S. Clark, M. Radley and M. Hill in the CTSU, Oxford, for assisting with the design, planning, organization and conduct of the study. A complete list of members of the China Kadoorie Collaborative Group is provided in the Supplementary Information. The CKB baseline survey and the first re-survey were supported by the Kadoorie Charitable Foundation in Hong Kong. The long-term follow-up of the CKB study has been supported by Wellcome grants to Z.C. at Oxford University (212946/Z/18/Z, 202922/Z/16/Z, 104085/Z/14/Z, 088158/Z/09/Z) and grants to L.L. from the National Natural Science Foundation of China (82192901, 82192904 and 82192900) and from the National Key Research and Development Program of China (2016YFC0900500). DNA extraction and genotyping was supported by grants to Z.C. from GlaxoSmithKline and the UK Medical Research Council (MC-PC-13049, MC-PC-14135). The UK Medical Research Council (MC_UU_00017/1, MC_UU_12026/2, MC_U137686851), Cancer Research UK (C16077/A29186; C500/A16896) and the British Heart Foundation (CH/1996001/9454) provide core funding to the CTSU and Epidemiological Studies Unit at Oxford University for the project. P.K.I. is supported by an Early Career Research Fellowship from the Nuffield Department of Population Health, University of Oxford. K.H.C. acknowledges support from the British Heart Foundation Centre of Research Excellence, University of Oxford (RE/18/3/34214). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. For the purpose of Open Access, the author has applied a CC-BY public copyright license to any author accepted manuscript version arising from this submission.

Author information

These authors jointly supervised this work: Liming Li, Zhengming Chen, Iona Y. Millwood.

Authors and Affiliations

Clinical Trial Service Unit and Epidemiological Studies Unit (CTSU), Nuffield Department of Population Health, University of Oxford, Oxford, UK

Pek Kei Im, Neil Wright, Ling Yang, Ka Hung Chan, Yiping Chen, Huaidong Du, Xiaoming Yang, Daniel Avery, Robert Clarke, Rory Collins, Robin G. Walters, Richard Peto, Zhengming Chen, Iona Y. Millwood, Maxim Barnard, Derrick Bennett, Ruth Boxall, Johnathan Clarke, Ahmed Edris Mohamed, Hannah Fry, Simon Gilbert, Andri Iona, Maria Kakkoura, Christiana Kartsonaki, Hubert Lam, Kuang Lin, James Liu, Mohsen Mazidi, Sam Morris, Qunhua Nie, Alfred Pozarickij, Paul Ryder, Saredo Said, Dan Schmidt, Becky Stevens, Iain Turnbull, Baihan Wang, Lin Wang & Pang Yao

Medical Research Council Population Health Research Unit (MRC PHRU), Nuffield Department of Population Health, University of Oxford, Oxford, UK

Ling Yang, Yiping Chen, Huaidong Du, Robin G. Walters, Zhengming Chen, Iona Y. Millwood, Derrick Bennett, Ruth Boxall & Christiana Kartsonaki

Oxford British Heart Foundation Centre of Research Excellence, University of Oxford, Oxford, UK

Ka Hung Chan

Fuwai Hospital, Chinese Academy of Medical Sciences, Beijing, China

NCD Prevention and Control Department, Qingdao CDC, Qingdao, China

Shaojie Wang, Liang Cheng, Ranran Du, Ruqin Gao, Feifei Li, Shanpeng Li, Yongmei Liu, Feng Ning, Zengchang Pang, Xiaohui Sun, Xiaocao Tian, Yaoming Zhai & Hua Zhang

Department of Epidemiology and Biostatistics, School of Public Health, Peking University, Beijing, China

Canqing Yu, Jun Lv, Liming Li & Dianjianyi Sun

Peking University Center for Public Health and Epidemic Preparedness & Response, Beijing, China

Canqing Yu, Jun Lv, Liming Li, Xiao Han, Can Hou, Qingmei Xia, Chao Liu, Pei Pei & Dianjianyi Sun

China National Center for Food Safety Risk Assessment, Beijing, China

• Junshi Chen

WHO Collaborating Center for Tobacco Cessation and Respiratory Diseases Prevention, China-Japan Friendship Hospital, Beijing, China

Institute of Respiratory Medicine, Chinese Academy of Medical Sciences, Beijing, China

NCD Prevention and Control Department, Guangxi Provincial CDC, Nanning, China

Naying Chen, Duo Liu & Zhenzhu Tang

NCD Prevention and Control Department, Liuzhou CDC, Liubei, Liuzhou, China

Ningyu Chen, Qilian Jiang, Jian Lan, Mingqiang Li, Yun Liu, Fanwen Meng, Jinhuai Meng, Rong Pan, Yulu Qin, Ping Wang, Sisi Wang, Liuping Wei & Liyuan Zhou

NCD Prevention and Control Department, Gansu Provincial CDC, Lanzhou, China

Caixia Dong, Pengfei Ge & Xiaolan Ren

NCD Prevention and Control Department, Maijixiang CDC, Maijixiang, Tianshui, China

Zhongxiao Li, Enke Mao, Tao Wang, Hui Zhang & Xi Zhang

NCD Prevention and Control Department, Hainan Provincial CDC, Haikou, China

Jinyan Chen, Ximin Hu & Xiaohuan Wang

NCD Prevention and Control Department, Meilan CDC, Meilan, Haikou, China

Zhendong Guo, Huimei Li, Yilei Li, Min Weng & Shukuan Wu

NCD Prevention and Control Department, Heilongjiang CDC, Harbin, China

Shichun Yan, Mingyuan Zou & Xue Zhou

NCD Prevention and Control Department, Nangang CDC, Harbin, China

Ziyan Guo, Quan Kang, Yanjie Li, Bo Yu & Qinai Xu

NCD Prevention and Control Department, Henan Provincial CDC, Zhengzhou, China

Liang Chang, Lei Fan, Shixian Feng, Ding Zhang & Gang Zhou

NCD Prevention and Control Department, Huixian CDC, Huixian, China

Yulian Gao, Tianyou He, Pan He, Chen Hu, Huarong Sun & Xukui Zhang

NCD Prevention and Control Department, Hunan Provincial CDC, Changsha, China

Biyun Chen, Zhongxi Fu, Yuelong Huang, Huilin Liu, Qiaohua Xu & Li Yin

NCD Prevention and Control Department, Liuyang CDC, Liuyang, China

Huajun Long, Xin Xu, Hao Zhang & Libo Zhang

NCD Prevention and Control Department, Jiangsu Provincial CDC, Nanjing, China

Jian Su, Ran Tao, Ming Wu, Jie Yang, Jinyi Zhou & Yonglin Zhou

NCD Prevention and Control Department, Wuzhong CDC, Wuzhong, Suzhou, China

Yihe Hu, Yujie Hua, Jianrong Jin, Fang Liu, Jingchao Liu, Yan Lu, Liangcai Ma, Aiyu Tang & Jun Zhang

NCD Prevention and Control Department, Licang CDC, Qingdao, China

Wei Hou, Silu Lv & Junzheng Wang

NCD Prevention and Control Department, Sichuan Provincial CDC, Chengdu, China

Xiaofang Chen, Xianping Wu, Ningmei Zhang & Xiaoyu Chang

NCD Prevention and Control Department, Pengzhou CDC, Pengzhou, Chengdu, China

Xiaofang Chen, Jianguo Li, Jiaqiu Liu, Guojin Luo, Qiang Sun & Xunfu Zhong

NCD Prevention and Control Department, Zhejiang Provincial CDC, Hangzhou, China

Weiwei Gong, Ruying Hu, Hao Wang, Meng Wang & Min Yu

NCD Prevention and Control Department, Tongxiang CDC, Tongxiang, China

Lingli Chen, Qijun Gu, Dongxia Pan, Chunmei Wang, Kaixu Xie & Xiaoyi Zhang

You can also search for this author in PubMed   Google Scholar

• , Zhengming Chen
• , Robert Clarke
• , Rory Collins
• , Liming Li
• , Chen Wang
• , Richard Peto
• , Robin G. Walters
• , Daniel Avery
• , Maxim Barnard
• , Derrick Bennett
• , Ruth Boxall
• , Ka Hung Chan
• , Yiping Chen
• , Johnathan Clarke
• , Huaidong Du
• , Ahmed Edris Mohamed
• , Hannah Fry
• , Simon Gilbert
• , Pek Kei Im
• , Andri Iona
• , Maria Kakkoura
• , Christiana Kartsonaki
• , Hubert Lam
• , Kuang Lin
• , James Liu
• , Mohsen Mazidi
• , Iona Y. Millwood
• , Sam Morris
• , Qunhua Nie
• , Alfred Pozarickij
• , Paul Ryder
• , Saredo Said
• , Dan Schmidt
• , Becky Stevens
• , Iain Turnbull
• , Baihan Wang
• , Neil Wright
• , Ling Yang
• , Xiaoming Yang
• , Qingmei Xia
• , Dianjianyi Sun
• , Canqing Yu
• , Naying Chen
• , Zhenzhu Tang
• , Ningyu Chen
• , Qilian Jiang
• , Mingqiang Li
• , Fanwen Meng
• , Jinhuai Meng
• , Ping Wang
• , Sisi Wang
• , Liuping Wei
• , Liyuan Zhou
• , Caixia Dong
• , Pengfei Ge
• , Xiaolan Ren
• , Zhongxiao Li
• , Hui Zhang
• , Jinyan Chen
• , Xiaohuan Wang
• , Zhendong Guo
• , Huimei Li
• , Shukuan Wu
• , Shichun Yan
• , Mingyuan Zou
• , Ziyan Guo
• , Quan Kang
• , Yanjie Li
• , Liang Chang
• , Shixian Feng
• , Ding Zhang
• , Gang Zhou
• , Yulian Gao
• , Tianyou He
• , Huarong Sun
• , Xukui Zhang
• , Biyun Chen
• , Zhongxi Fu
• , Yuelong Huang
• , Huilin Liu
• , Qiaohua Xu
• , Huajun Long
• , Hao Zhang
• , Libo Zhang
• , Jinyi Zhou
• , Yonglin Zhou
• , Yujie Hua
• , Jianrong Jin
• , Jingchao Liu
• , Liangcai Ma
• , Aiyu Tang
• , Jun Zhang
• , Liang Cheng
• , Ranran Du
• , Ruqin Gao
• , Feifei Li
• , Shanpeng Li
• , Yongmei Liu
• , Feng Ning
• , Zengchang Pang
• , Xiaohui Sun
• , Xiaocao Tian
• , Shaojie Wang
• , Yaoming Zhai
• , Hua Zhang
• , Junzheng Wang
• , Xiaofang Chen
• , Xianping Wu
• , Ningmei Zhang
• , Xiaoyu Chang
• , Jianguo Li
• , Jiaqiu Liu
• , Guojin Luo
• , Qiang Sun
• , Xunfu Zhong
• , Weiwei Gong
• , Ruying Hu
• , Meng Wang
• , Lingli Chen
• , Dongxia Pan
• , Chunmei Wang
• , Kaixu Xie
•  & Xiaoyi Zhang

Contributions

P.K.I., I.Y.M., L.Y. and Z.C. contributed to the conception of this paper. P.K.I., N.W., K.H.C., I.Y.M. and Z.C. planned the statistical analysis. P.K.I. analyzed the data and drafted the manuscript. P.K.I., I.Y.M. and Z.C. contributed to the interpretation of the results and the revision of manuscript. R. Collins, R.P., J.C., L.L. and Z.C. designed the study. L.L., Z.C., I.Y.M., L.Y., Y.C., Y.G., H.D., S.W., C.Y., J.L., J.C., R. Collins, R. Clarke and R.G.W. contributed to data acquisition and general study management. X.Y. and D.A. provided administrative and technical support. All authors critically reviewed the manuscript and approved the final submission.

Corresponding authors

Correspondence to Zhengming Chen or Iona Y. Millwood .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Nature Medicine thanks Shiu Lun Au Yeung, Yan-Bo Zhang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editor: Ming Yang, in collaboration with the Nature Medicine team.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended data fig. 1 adjusted hrs for icd−10 chapter−specific morbidities associated with ever-regular drinking and with usual alcohol intake, in men..

Cox models comparing ever-regular drinkers with occasional drinkers, or assessing the dose–response per 280 g/week higher usual alcohol intake within current drinkers, were stratified by age-at-risk and study area and adjusted for education and smoking. Each solid square or diamond represents HR with the area inversely proportional to the variance of the log HR. The horizontal lines indicate 95% CIs. CI, confidence interval; HR hazard ratio; ICD-10, International Classification of Diseases, 10th Revision.

Extended Data Fig. 2 Adjusted HRs for different aggregated and all-cause morbidities associated with years after stopping drinking, in men.

Cox models comparing ex-drinker groups with occasional drinkers were stratified by age-at-risk and study area and were adjusted for education and smoking. Each box represents HR with the area inversely proportional to the variance of the group-specific log hazard within subplot. The vertical lines indicate group-specific 95% CIs for various ex-drinker groups. The shaded strip indicate the group-specific 95% CIs for occasional drinkers. The numbers above the error bars are point estimates for HRs. CI, confidence interval; HR hazard ratio; CKB, China Kadoorie Biobank; WHO, World Health Organization.

Extended Data Fig. 3 Associations of alcohol consumption with risks of 28 diseases previously defined as alcohol-related by the WHO, in male current drinkers.

Cox models were stratified by age-at-risk and study area and were adjusted for education and smoking. HRs were plotted against usual alcohol intake and were calculated per 280 g/week higher usual alcohol intake. All specific diseases displayed were significantly associated with alcohol intake (ever-regular drinking or per 280 g/week higher usual alcohol intake) after multiple testing correction (FDR-adjusted p<0.05), except transient cerebral ischemic attacks and related syndromes (ICD-10 code: G45), occlusion and stenosis of precerebral arteries (I65) and pancreatitis (K85-K86) which showed statistical significance at nominal level (p<0.05). Each box represents HR with the area inversely proportional to the variance of the group-specific log hazard within subplot. The vertical lines indicate group-specific 95% CIs. The numbers above the error bars are point estimates for HRs and the numbers below are number of events. All P values are two-sided. CI, confidence interval; HR hazard ratio; CKB, China Kadoorie Biobank; FDR, false discovery rate; ICD-10, International Classification of Diseases, 10th Revision; WHO, World Health Organization.

Extended Data Fig. 4 Associations of alcohol consumption with risks of 36 diseases not previously defined as alcohol-related, in male current drinkers.

Cox models were stratified by age-at-risk and study area and were adjusted for education and smoking. HRs were plotted against usual alcohol intake and were calculated per 280 g/week higher usual alcohol intake. All specific diseases displayed were significantly associated with alcohol intake (ever-regular drinking or per 280 g/week higher usual alcohol intake) after multiple testing correction (FDR-adjusted p<0.05). Each box represents HR with the area inversely proportional to the variance of the group-specific log hazard within subplot. The vertical lines indicate group-specific 95% CIs. The numbers above the error bars are point estimates for HRs and the numbers below are number of events. All P values are two-sided. CI, confidence interval; HR hazard ratio; CKB, China Kadoorie Biobank; FDR, false discovery rate.

Extended Data Fig. 5 Adjusted HRs for major diseases associated with drinking patterns, in male current drinkers.

Cox models were stratified by age-at-risk and study area and were adjusted for education and smoking and for total alcohol intake where indicated. Each solid square or diamond represents HR with the area inversely proportional to the variance of the log HR. The horizontal lines indicate 95% CIs. CI, confidence interval; HR hazard ratio; HED, heavy episodic drinking; CKB, China Kadoorie Biobank; WHO, World Health Organization.

Extended Data Fig. 6 Adjusted HRs for major diseases associated with duration of drinking, in male current drinkers.

Cox models were stratified by age-at-risk and study area and were adjusted for education, smoking, total alcohol intake and baseline age in (A). (B) had the same model specifications as (A) plus further adjustments for income, physical activity, fruit intake and body mass index. (C) had the same model specifications as (A) and excluded participants with poor self-reported health or prior chronic disease at baseline. Each box represents HR with the area inversely proportional to the variance of the group-specific log hazard. The horizontal lines indicate group-specific 95% CIs. All P values are two-sided. CI, confidence interval; HR hazard ratio; CKB, China Kadoorie Biobank; WHO, World Health Organization.

Extended Data Fig. 7 Adjusted HRs for ICD−10 chapter−specific morbidities associated with ever-regular drinking and with usual alcohol intake, in women.

Cox models comparing ever-regular drinkers with occasional drinkers, or assessing the dose–response per 100 g/week higher usual alcohol intake within current drinkers, were stratified by age-at-risk and study area and adjusted for education and smoking. Each solid square or diamond represents HR with the area inversely proportional to the variance of the log HR. The horizontal lines indicate 95% CIs. CI, confidence interval; HR hazard ratio; ICD-10, International Classification of Diseases, 10th Revision.

Extended Data Fig. 8 Adjusted HRs per 280 g/week higher genotype-predicted mean male alcohol intake for specific alcohol-associated diseases by ICD-10 chapters, in men and women.

Cox modes, stratified by age-at-risk and adjusted for genomic principal components, were used to relate genetic categories to risks of diseases within each study area. The HR per 280 g/week higher genotype-predicted mean male alcohol intake was calculated from the inverse-variance-weighted mean of the slopes of the fitted lines in each study area. Each solid square or diamond represents HR per 280 g/week higher genetically-predicted mean male alcohol intake, with the area inversely proportional to the variance of the log HR. The horizontal lines indicate 95% CIs. Diseases considered to be alcohol-related by the WHO are indicated with ‘Y’ under the ‘WHO’ column. The ‘RC’ column indicates the number of study areas that contributed to the overall area-stratified genotypic associations, as for certain less common diseases some study areas may not have enough number of cases to contribute to the inverse-variance-weighted meta-analysis. The ‘P het’ column indicates the p-value from a \(\chi\) 2 test for heterogeneity between sexes. All P values are two-sided. † Included less common ICD-10 codes within the corresponding ICD-10 chapter which were not individually investigated in the present study. CI, confidence interval; HR hazard ratio; ICD-10, International Classification of Diseases, 10th Revision; WHO, World Health Organization.

Extended Data Fig. 9 Adjusted HRs associated with GG versus AG genotype of ALDH2 - rs671 for specific alcohol-associated diseases by ICD-10 chapters, in men and women.

Area-specific genotypic effects (GG vs. AG genotype) were estimated within each study area (thus each reflecting the purely genotypic effects) using age-at-risk-stratified and genomic principal components-adjusted Cox models and were combined by inverse-variance-weighted meta-analysis to yield the overall area-stratified genotypic associations. Each solid square represents HR for GG vs. AG genotype, with the area inversely proportional to the variance of the log HR. The horizontal lines indicate 95% CIs. Diseases considered to be alcohol-related by the WHO are indicated with ‘Y’ under the ‘WHO’ column. The ‘RC’ column indicates the number of study areas that contributed to the overall area-stratified genotypic associations, as for certain less common diseases some study areas may not have enough number of cases to contribute to the inverse-variance-weighted meta-analysis. The ‘P het’ column indicates the P value from a \(\chi\) 2 test for heterogeneity between sexes. All P values are two-sided. † Included less common ICD-10 codes within the corresponding ICD-10 chapter which were not individually investigated in the present study. CI, confidence interval; HR hazard ratio; ICD-10, International Classification of Diseases, 10th Revision; WHO, World Health Organization.

Extended Data Fig. 10 Adjusted HRs associated with GG versus AG genotype of ADH1B - rs1229984 for specific alcohol-associated diseases by ICD-10 chapters, in men and women.

Supplementary information, supplementary information.

Supplementary Figs. 1–8 and Supplementary Tables 1–13.

Reporting Summary

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Im, P.K., Wright, N., Yang, L. et al. Alcohol consumption and risks of more than 200 diseases in Chinese men. Nat Med 29 , 1476–1486 (2023). https://doi.org/10.1038/s41591-023-02383-8

Download citation

Received : 16 December 2022

Accepted : 02 May 2023

Published : 08 June 2023

Issue Date : June 2023

DOI : https://doi.org/10.1038/s41591-023-02383-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Uncovering newly identified aldehyde dehydrogenase 2 genetic variants that lead to acetaldehyde accumulation after an alcohol challenge.

• Freeborn Rwere
• Joseph R. White
• Eric R. Gross

Journal of Translational Medicine (2024)

Alcohol consumption may be a risk factor for cerebrovascular stenosis in acute ischemic stroke and transient ischemic attack

• Xiaoyan Guo

BMC Neurology (2024)

Western diets and chronic diseases

• Timon E. Adolph
• Herbert Tilg

Nature Medicine (2024)

Mapping multimorbidity progression among 190 diseases

Communications Medicine (2024)

The aldehyde dehydrogenase 2 rs671 variant enhances amyloid β pathology

Nature Communications (2024)

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

• Study protocol
• Open access
• Published: 15 October 2022

Underlying mechanisms in the relationship between stress and alcohol consumption in regular and risky drinkers (MESA): methods and design of a randomized laboratory study

• Charlotte Wittgens   ORCID: orcid.org/0000-0001-6365-6662 1 , 2 ,
• Markus Muehlhan   ORCID: orcid.org/0000-0002-8855-8724 1 , 3 ,
• Anja Kräplin   ORCID: orcid.org/0000-0002-1612-3932 4 ,
• Max Wolff 5 &
• Sebastian Trautmann   ORCID: orcid.org/0000-0002-8976-3244 1 , 2  

BMC Psychology volume  10 , Article number:  233 ( 2022 ) Cite this article

10k Accesses

9 Citations

10 Altmetric

Metrics details

Excessive alcohol consumption and alcohol use disorders (AUD) are among the leading preventable causes of premature morbidity and mortality and are considered a major public health concern. In order to reduce the individual and societal burden of excessive alcohol use, it is crucial to identify high-risk individuals at earlier stages and to provide effective interventions to prevent further progression. Stressful experiences are important risk factors for excessive alcohol consumption and AUDs. However, the underlying biological and psychological mechanisms are still poorly understood.

The project “Underlying mechanisms in the relationship between stress and alcohol consumption in regular and risky drinkers (MESA)” is a randomized controlled study that started in December 2018 and is conducted in a laboratory setting, which aims to identify moderators and mediators of the relationship between acute stress and alcohol consumption among regular and risky drinkers. Regular and risky drinkers are randomly assigned to a stress induction or a control condition. Several processes that may mediate (emotional distress, endocrine and autonomic stress reactivity, impulsivity, inhibitory control, motivational sensitization) or moderate (trait impulsivity, childhood maltreatment, basal HPA-axis activity) the relation between stress and alcohol consumption are investigated. As primary dependent variable, the motivation to consume alcohol following psychosocial stress is measured.

The results of this study could help to provide valuable targets for future research on tailored interventions to prevent stress-related alcohol consumption.

Peer Review reports

Excessive alcohol consumption and alcohol use disorders (AUD) are among the leading preventable causes of premature morbidity and mortality [ 1 , 2 ]. They come along with an immense individual and societal burden and are considered a major public health problem [ 3 ]. The World Health Organization reported 3 million deaths due to harmful use of alcohol in their latest report [ 4 ]. In the age group 20–39 years, approximately 13.5% of the total deaths are attributable to alcohol [ 5 ]. In particular, men are considered at high risk to develop AUD [ 2 ] with global prevalence five times that in women with 8.6% and 1.7% for males and females, respectively [ 4 ]. However, latest data indicated that this gap is narrowing in recent years [ 6 , 7 ]. Treatments for excessive alcohol use and AUD are initiated at a very late stage of symptom progression when adverse somatic and mental consequences have already occurred [ 8 , 9 ]. It is therefore necessary to identify high-risk individuals at an earlier stage of alcohol consumption in order to reduce individual and societal burden and to implement effective interventions to prevent further progressions. Risk factors and underlying mechanisms promoting excessive alcohol use need to be identified for tailoring new preventive approaches.

Stress and alcohol use

Alcohol consumption is a commonly used coping strategy to reduce stress [ 10 ]. It is very well known that Stress increases the amount of alcohol consumed and the risk of relapse, but little is known about the psychological mechanisms that underlie these effects [ 11 ]. The experience of stressful events, defined as unpredictable or uncontrollable events that exceed the regulatory capacity of an organism and that could threaten an organism’s physical or psychosocial integrity [ 10 , 11 ] has been identified as a major risk factor for excessive alcohol use and AUD [ 12 , 13 ]. The impact of stress on alcohol use and the risk of AUDs depends on the type, age, duration, and severity of the stress experienced [ 14 ]. The consumption of alcohol is a habitual response to stressful situations in people with AUD [ 15 ]. Stress plays an important role at all levels of alcohol consumption, beginning with facilitation of initial use through early stages of transition to regular use and from regular to excessive use [ 16 , 17 , 18 ]. In AUD, alcohol use also represents a habitual response to stressful situations [ 15 ].

Mediators and moderators

Despite this well-established association between stress and alcohol use, the underlying mechanisms are complex and still not well understood. Studies trying to explain this association show inconsistent results. Stress does not necessarily lead to alcohol consumption in every person [ 19 ], which suggests the relevance of potential moderating factors. Several environmental, biological, and psychological factors that could moderate the relation between stress and alcohol consumption at different stages of alcohol use progression are discussed in the existing literature. The hypothalamic–pituitary–adrenal (HPA) axis plays an important role in this context as it is a major stress response pathway and has been studied extensively in relation to alcohol use [ 20 ]. Altered HPA axis regulation is associated with problematic alcohol use and dependence and the nature of this dysregulation varies with respect to the stages of progression toward AUD [ 21 ]. Glucocorticoid secretion upon activation of (HPA) axis by stressors is normally adaptive, and was discussed to promote coping after stressful events whereas excessive and prolonged HPA axis activation results in wear-and-tear on numerous physiological systems [ 22 ]. Furthermore, dysregulation in stress-related cortisol production is a risk factor for developing AUD [ 20 ]. Therefore, studies suggest that there might be a moderating effect on the relationship between stress and alcohol consumption by individual differences in basal cortisol secretion [ 18 , 19 ]. Further, there is evidence from observational studies that childhood maltreatment moderates the association between stressful experiences and the development of alcohol use problems [ 23 , 24 ]. Individuals with childhood trauma exposure, particularly abuse, neglect, or chaotic home environments, are at heightened risk for heavy alcohol consumption [ 24 ]. Further childhood maltreatment is associated with early alcohol use initiation, alcohol-related problem behaviors, and alcohol use disorders in adulthood [ 25 ]. Other possible moderators considered in this context are personality traits. Personality traits such as trait impulsivity reflect people’s characteristic patterns of thoughts, feelings, and behaviors and imply consistency over time and stability across situations [ 26 ]. Trait impulsivity was found to predict risk for alcohol use problems in general [ 27 , 28 , 29 ] and further moderates the association between stress and alcohol use [ 30 , 31 ]. Although the consideration of these moderating factors might help to elucidate previous inconsistent findings on the association between stress and alcohol use and develop more targeted interventions, they have barely been considered in studies on its underlying mechanisms.

Regarding the underlying mechanisms of the relationship between stress and alcohol consumption, the idea of alcohol use as a dysfunctional coping strategy to self-medicate aversive emotional states following stressful experiences has long been the predominant model [ 32 ]. Although there is considerable empirical support for the self-medication hypothesis [ 32 , 33 , 34 , 35 ], it is not able to fully explain the association between stressful experiences and alcohol use. Alcohol consumption does not necessarily reduce aversive emotional states [ 36 , 37 ], violating the negative reinforcement assumption underlying the self-medication hypothesis. Therefore, knowledge on additional mechanisms beyond self-medication at different stages of alcohol use progression is required to explain the association between stress and alcohol use. Several relevant psychological and biological factors that might affect this relationship have been described in the literature [ 20 , 38 ]. Acute stress activates an immediate reaction increasing cerebral and peripheral adrenalin and noradrenalin and a delayed endocrine response (via HPA axis) increasing glucocorticoids (mainly cortisol in humans) [ 39 ]. These systems affect different mechanisms relevant to alcohol use depending on the stage of alcohol use progression. At early stages of alcohol use progression, alcohol use leads to increased autonomic arousal and HPA axis activation. These effects potentiate both stress and alcohol-related effects on motivation and reinforcement learning [ 40 ] which can further facilitate alcohol use as a stress-related coping mechanism [ 38 ]. It further promotes the salience of drug-related cues known as attentional bias as these cues ‘grab the attention’ and further increase alcohol craving [ 41 ]. At later stages of alcohol use progression, binge and excessive alcohol consumption results in larger-scale adaptations in terms of a neuroendocrine tolerance response to stress and alcohol intake [ 10 , 42 ] which may be involved in the transition from controlled to compulsive alcohol consumption [ 10 , 43 ]. Also, a sensitization of motivational systems can manifest, again, in priority processing alcohol-related cues, i.e. attentional bias [ 44 , 45 ]. The stress-induced sensitization at later stages of alcohol use progression is assumed to be active in parallel to the noradrenalin-related mechanisms [ 18 ]. Taken together, stress and stress system alterations by alcohol consumption could be associated with biased information processing, increased impulsivity and impaired control functions; a pattern that is known to be a key mechanism in the development of excessive alcohol use [ 46 , 47 ].

Need for controlled laboratory studies

Most studies, addressing the association between stress and alcohol consumption are based on clinical populations with limited sample sizes and participants who already developed AUD. In this context different moderators and mediators leading to alcohol dependence are often center of the research question [ 48 , 49 ]. There is need for research that investigates the underlying mechanisms that lead to AUD before it is manifested. Therefore, especially laboratory settings with non-clinic samples are suitable to investigate mediators and moderators on this relationship as they allow the investigation of specific mechanisms through randomized manipulation of the factor of interest and at the same time allow to control for confounding variables [ 50 ].

Aims and hypotheses

The present and ongoing study aims to fill this research gap by conducting an experimental laboratory design to investigate the underlying mechanisms of the association between stress and alcohol consumption (MESA) in the at-risk population of young men. Since these mechanisms are expected to differ depending on the stage of alcohol use, they are examined in regular and risky drinkers. Therefore, several processes that could mediate the relation between stress and alcohol consumption at different stages of alcohol use progression are assessed.

The research questions are as follows:

Does acute stress increase alcohol consumption in a laboratory setting?

What are the mediators of the association between acute stress and alcohol use?

What are the moderators of the association between acute stress and alcohol use?

Are effects of acute stress on alcohol use as well as moderators and mediators of this association different in risky drinkers compared to regular drinkers?

The following a priori hypothesis were formulated:

Acute stress increases alcohol consumption in a laboratory setting.

This effect is stronger in risky compared to regular drinkers.

Emotional distress, endocrine and autonomic stress reactivity as well as impulsivity account for most of the effect of stress on alcohol use in regular drinkers (mediation).

Emotional distress, endocrine and autonomic stress reactivity, impulsivity, attentional bias and craving account for most of the effect of stress on alcohol use in risky drinkers (mediation).

A history of childhood maltreatment, basal HPA-axis activity and impulsivity are related to a stronger effect of acute stress on alcohol consumption in regular and risky drinkers (moderation).

Methods/design

Study design

The MESA study is a randomized controlled study that started in December 2018 and is being conducted in a laboratory setting at the Medical School Hamburg. The study is divided into an online screening and a main examination, with detailed description in the following (“ Procedure ” section). The study has a four-group design. Participants are stratified into equal groups of regular and risky drinkers (with regular drinking being defined as average daily alcohol consumption of less than 24 g over the past 30 days and risky drinking being defined as average daily alcohol consumption of more than 24 g over the past 30 days [ 51 ]). Regular and risky drinkers are then randomly assigned to either an experimental (acute stress) or a control condition (Fig.  1 ).

Research is conducted in accordance with national data protection acts, the revised declaration of Helsinki and Good Clinical Practice Guidelines. After complete description of the study, written informed consent is obtained from all participants. The study is approved by the Institutional Review Boards of Technische Universität Dresden (EK 522122016) and Medical School Hamburg (MSH-2020/114).

Inclusion and exclusion criteria

Males have a higher risk of developing drinking problems compared to females [ 52 ] and are more likely to report stress-induced drinking [ 6 ]. Therefore, only male individuals are included to reduce heterogeneity and potential of confounding factors (e.g. intake of oral contraceptives, menstrual cycle), especially in the biological measures. All participants have to be between 18–40 years old. The upper age limit results from the fact that most alcohol use problems develop during adolescence and young adulthood [ 53 , 54 , 55 ]. Further, eligible individuals have to drink alcohol at least occasionally and have beer as their favorite alcoholic drink since it is necessary for the success of the study that participants are familiar with alcoholic brands to recognize them in the attentional bias paradigm (“ Assessment ” section). In addition, it might be perceived as unethical to provide abstinent individuals with alcoholic beverages. Additionally, having a hair length of at least 2 cm is required to analyze hair cortisol concentrations [ 56 ] as a cumulative measure for basal cortisol secretion of the association between stress and alcohol use (for detailed description see 2.4 Biological measures). Exclusion criteria are lifetime psychotic symptoms, lifetime alcohol or any other substance use disorder, current psychological or psycho-pharmacological interventions and acute suicidality, current psychotropic or other medication or any somatic diseases that might confound the study measures, especially with regard to the endocrine measures, and alcohol consumption on the study day. All subjects meeting the inclusion criteria will be stratified into the reported groups.

A priori power analysis

A non-clinical target sample of 400 young men is aimed for the MESA study. A power analysis was conducted to calculate the needed sample size. A series of Monte Carlo simulations (each simulating 1000 ANOVA F tests) using the simpower program in STATA 12.1 [ 57 ] was run. The Monte Carlo simulations revealed that assuming a sample size of n = 200 per drinking stratum, statistical power ranges between 0.80 and 0.95 for different group size ratios. Given the stratified randomized design of the study, this results in the final group size of n = 100 (Fig.  1 ).

Recruitment and screening procedures

Participants are recruited via personal contacts, flyers and advertisement in university and public settings in Hamburg (cafés, bars, supermarkets, sports clubs; student dormitories) as well as via social media (e.g. Instagram, Facebook) and student job markets. In addition, advertising is made in lectures and on the university website.

All individuals willing to participate in the study have to complete an online screening in advance of the main assessment, where basic demographic variables as well as all in- and exclusion criteria are assessed. Further, the usual alcohol consumption is measured using a self-administered timeline follow-back consisting of a calendar on which participants provide retrospective reports of average daily alcohol intake for the past 30 days [ 58 ]. The information on daily alcohol consumption is used to allocate participants to the groups of regular and risky drinkers. All individuals meeting the inclusion criteria are then invited to participate in the main study.

Person-related measures

Participants complete a comprehensive baseline assessment (questionnaire package) including the measures of the proposed moderators (childhood maltreatment, trait impulsivity), mediators (attentional bias to alcohol related stimuli, inhibitory control and impulsivity, and stress reactivity during the acute stressor), and variables that might affect the associations of interest (usual alcohol consumption, drinking motives, perceived stress, trait anxiety, difficulties in emotion regulation, psychological flexibility) (Table 1 ).

Biological measures

Hair strands are taken to reflect cumulative long-term cortisol secretion within two months prior to the respective assessment point [ 56 ]. The cumulative cortisol secretion consisting of basal cortisol secretion as well as stress-induced cortisol secretion, has been shown to be an important moderator of stress-related adverse consequences including increase in alcohol use [ 85 , 86 ]. In addition, during the study four saliva samples are collected using Salivettes ® “code blue” (Sarstedt, Nümbrecht, Germany) with synthetic swabs to measure free cortisol levels and alpha-amylase activity as biological indicators of stress reactivity (Fig.  5 ). The first saliva sample is taken immediately before the stress induction (for detailed description for the stress induction see “ Sample storage, biochemical analyses and data preparation ” section). The second saliva sample is taken right after the stress induction as well as 12 (3rd salvia sample) and 24 (4th salvia sample) minutes after the stress induction. Cortisol is the final output of the hypothalamic pituitary adrenal (HPA) axis, and is among the most frequently used biological markers of psychological stress [ 87 , 88 ]. Moreover, given that the biologically active (free) fraction of cortisol is reflected in saliva, it can be a preferred measure relative to serum cortisol [ 87 , 89 ]. In addition, alpha-amylase is an enzyme component of saliva and has been proposed as a marker for stress-induced activity of the sympathetic nervous system (SNS). The advantage of a saliva-based measure of SNS activity is the convenience of assessing activity of both major stress systems (i.e. SNS and HPA-axis) in a single test tube, without the need for technically sophisticated instrumentation [ 63 ].

Behavioral measures

In addition to self-report measures, three behavioral tasks are conducted to measure attentional bias to alcohol related stimuli, inhibitory control and impulsivity as possible mediators in the association between stress and alcohol use [ 18 , 90 ]. Attentional bias towards alcohol-related cues is measured using a dot-probe task (Fig.  2 ), which was programmed based on previous tasks in similar settings [ 41 , 90 ]. Subjects are presented with pairs of matched alcoholic (beer) and non-alcoholic beverages for 500 ms (stimulus-onset asynchrony, SOA). Another SOA of 100 ms will be added to the paradigm in the proposed study to be able to capture automatic initial reactions (see [ 91 ]). Stimuli were chosen based on expert ratings regarding similarity in color, shape and recognition. Subjects respond to a probe that appears behind either the alcoholic or the non-alcoholic beverage. The difference in reaction time between alcoholic and non-alcoholic stimuli is a measure of attentional bias towards alcohol-related cues. Although the dot-probe task is a widely used paradigm to measure attentional biases, there is debate about its reliability [ 92 , 93 ]. A new trial-based conceptualization of attentional bias has been proposed, which can increase reliability [ 94 ].

Dot-probe task

Inhibitory control is measured using a go/no-go task (Fig.  3 ) where participants are presented with 320 trials (280 go and 40 nogo trials) of stimuli containing two dots. Each dot pair is displayed for 500 ms and is arranged horizontally or vertically. Horizontally arranged dots indicate go-trials where participants have to press the response key as fast as possible while participants are instructed to withhold when seeing vertically arranged dots. Since there is evidence that participants balance the speed-accuracy trade-off differently [ 95 ], the dependent measure of the go/nogo task is the balanced integration score (BIS). This score is calculated in two steps. First, the responsive times (RTs) as well as the proportions of correct responses (PCs) are standardized. Second, one standardized score is subtracted from the other [ 96 ].

Go-nogo task

The delay discounting task as measure of impulsivity was taken from a task battery developed by Pooseh et al. ([ 65 ]; MATLAB scripts available from https://github.com/spooseh/VBDM ) and is described in detail in Kräplin et al. [ 66 ]. The task consisted of 30 trials. Participants had to decide between a smaller financial gain delivered sooner and a larger financial gain delivered later. The two options were simultaneously presented on a computer screen using the Psychophysics Toolbox [ 97 ] in MATLAB R2018a (MathWorks Inc., Natick, MA). Between the shorter and later choice options, delays were 3, 7, 14, 31, 61, 180, and 365 days. Monetary gains ranged from 0.30 to 10 €. A Bayesian adaptive algorithm was implemented. This way, the parameter estimation is updated after each trial and serves as the basis for the calculation of the options in the next trial. The method was used to determine the most informative offers nearest to the individual’s point of indifference between two choice alternatives (i.e. indifference point). Thus, decision-making parameters can be efficiently inferred without the use of post-hoc parameter estimations. A hyperbolic value function was generated to describe the decline of subjective values of delayed reward according to the discounting rate k (Mazur 1987). Individuals with higher impulsivity are assumed to display higher k values (Fig.  4 ).

Schematic overview of the tasks in the decision-making battery. a Delay discounting task. b Probability discounting for gains. c Probability discounting for losses. d Mixed gambles task [ 66 ]

Stress induction

Stress is induced with the Trier Social Stress Test (TSST) [ 98 ] as one of the most frequently inserted research tools for the induction of acute psychosocial stress in experimental, laboratory research worldwide. The TSST is a standardized laboratory protocol, which provides a reliable and ecologically valid stressor [ 98 ]. The TSST contains elements of social evaluative threat and uncontrollability, which are associated with high cortisol responses [ 99 ]. The test is divided in three equal five-minute parts. It begins with a preparation period, followed by a free speech for a job interview and finishes with an arithmetic task. All tasks are held in front of a two-person audience. The TSST leads to robust changes in the hypothalamus–pituitary–adrenal (HPA) axis and the autonomic stress response compared to other stress induction paradigms [ 39 , 100 ].

In the control condition, subjects participate in a Placebo-TSST, which is comparable in time and task division but without any audience and stress exposure for the participants [ 101 ]. It starts with a preparation period, followed by a free speech about the last vacation and finishes with a simple task of counting forward. Furthermore, participants are standing during the two tasks. This creates a setting that is as close as possible to the TSST, but does not contain stressful components (evaluative threat and uncontrollability).

Ad-libitum taste test

After completion of the behavioral tasks and the intervention, participants are asked to take part at an ad-libitum taste test as a covert measure for alcohol consumption. The ad-libitum taste test is a widely used method, which provides an unobtrusive and indirect measure of participants’ motivation to drink alcohol [ 84 ]. All participants are given two 0.33 l glasses of beer (two brands each containing 5% alcohol) and two 0.33 l glasses containing different soft drinks. Participants are instructed that they have 15 min to taste each glass to rate qualities about each drink (e.g. gassy, bitter). Participants are told to drink whatever amount necessary to make accurate judgements. The dependent variable is the amount of alcoholic beverage (beer) consumed and can range between 0 and 666 ml (equals 26.64 g ethanol). Non-alcoholic drinks are presented to control for the potential effect of thirst. The ad-libitum taste test is a valid method for the assessment of alcohol intake in the laboratory supported by strong associations between ad-libitum consumption and typical alcohol consumption [ 84 ]. It is also robust against several potential confounders such as time of day or participant awareness [ 84 ]. The taste test has been used to investigate a number of potential influences on alcohol consumption, including alcohol cues [ 102 , 103 , 104 ], impulse control [ 105 , 106 ], and social influences [ 107 ], and it has been used to establish initial proof of concept for novel behavioral interventions [ 108 , 109 , 110 ].

The main assessments are conducted between 14–20 p.m. in order to reduce the variance in biological measures (e.g. saliva cortisol) due to diurnal rhythms [ 111 ]. It is also likely that the willingness to drink alcohol is smaller in the morning than in the evening while there is no influence of day time on alcohol consumption in the ad libitum taste test between 14p.m. and 20p.m. [ 84 ]. Figure  5 gives an overview of the main study procedure. First, participants are asked to provide written informed consent. Participants’ absence from alcohol is verified by taking a breathalyzer reading with any value above zero leading to the immediate end of the examination. Hair strands for basal cortisol secretion are taken scalp-near from a posterior vertex position to be able to reflect basal cortisol secretion within two months prior to the respective assessment point. Then participants complete the baseline questionnaires including the measures of the proposed moderators (childhood maltreatment, trait impulsivity) and variables that might affect the associations of interest (usual alcohol consumption, drinking motives, stressful life events, trait anxiety, difficulties in emotion regulation, psychological flexibility). Subsequently, participants either take part in the stress induction (experimental condition) or placebo intervention (control condition) followed by behavioral assessments. Deviating from previous TSST protocols, the stress condition is maintained during the behavioral assessments. Therefore, participants are instructed that the TSST panel remain observing and evaluating the given performance during the computer tasks and further the camera is still pointed on the participant. The ad libitum taste-test is the last assessment of the procedure. After the taste test, all participants are debriefed about the true study purposes including the TSST procedure. Moreover, repeated breathalyzer readings are taken until blood alcohol concentration reaches 0.0‰ in two consecutive measures. Participants willing to leave before blood alcohol concentration reaches 0.0‰ have to confirm that they do not drive when leaving the laboratory. Participants who insist to leave with a blood alcohol concentration still being higher than 0.4‰ (only expected in rare cases) are sent home with a taxi.

Study procedure. Note : IC Informed consent, TSST Trier social stress test 

Sample storage, biochemical analyses and data preparation

The saliva samples are taken using salivette ‘code blue’ devices (Sarstedt, Nümbrecht, Germany) directly before the intervention and at three time points after the intervention (Fig.  5 ). Saliva samples are stored at − 20 °C in a laboratory freezer. After thawing, saliva samples will be centrifuged for 10 min at 4000 rpm. Salivary cortisol concentrations will be determined using a commercially available chemiluminescence assay (CLIA, IBL-Hamburg, Germany). Concentrations of salivary alpha-amylase will be detected by using an in-house enzyme kinetic method according to the protocol described in [ 63 ]. Hair cortisol concentrations will be determined via liquid chromatography tandem mass spectrometry (for detailed information on analysis methods see [ 56 ]).

Dimensional variables that are not normally distributed (expected e.g. for hair cortisol concentration, salivary cortisol and alpha amylase) will be Box–Cox transformed towards normal distribution. For all biological and behavioral variables, participants with outlying values of more than three standard deviations above the mean will be excluded from the respective analysis. Besides, robust linear regressions will complement conventional linear regressions because they down weight observations with large residuals to meet the assumption of equal variances of residuals. Composite measures of the entire cortisol secretion during the TSST (area under the curve with respect to ground; AUC G ) and the cortisol stress reactivity (area under the curve with respect to increase; AUC i ) will be calculated [ 112 ]. Analysis with these variables will be adjusted for initial cortisol concentration to alleviate confounding risk as AUC variables may be comprised of variance due to stress reactivity and stress-unrelated HPA axis activity [ 113 ]. All analysis including cortisol secretion during the TSST will be run twice with all participants in the first and with exclusion of non-responders to the TSST (increase of 1.5 nmol/l compared to baseline [ 114 ]) in the second run. With regard to alpha-amylase, both AUC measures and peak minus baseline levels will be calculated.

Statistical analyses

Main effects of stress exposure (stress vs control group) on alcohol consumption (amount of alcoholic beverage consumed) will be determined using linear regressions adjusting for the amount of non-alcoholic beverages consumed (which reduces unspecific variance in outcome). However, in case of considerable by chance differences in baseline characteristics between the two groups despite randomization, these characteristics will be included in the regression model if they are associated with alcohol consumption. To address potential biases related to missing data, we will conduct sensitivity analyses using multiple imputation.

Moderation analyses: Moderators are defined causally [ 50 ]. Linear regressions with interaction terms will be applied to test whether stress effects on alcohol consumption are moderated by childhood maltreatment, hair cortisol concentration and trait impulsivity (with main effects terms and interaction term, e.g. group × childhood maltreatment). Significant interactions indicate that a respective factor (moderator) predicts different effects of stress on alcohol assumptions. To approach causal conclusions, we will fit these models again while adjusting for shared factors of moderators and outcomes (e.g. previous stressful events, previous alcohol use) [ 50 ].

Mediation analyses: Mediators are also defined causally according to the counterfactual definition of Robins and Greenland [ 115 ] that is implemented in the ‘paramed’ package in Stata. This module allows dividing the estimated total stress effect (stress vs. control group) on alcohol consumption into a direct effect and an indirect effect mediated through stress reactivity (saliva cortisol, alpha amylase, self-reported stress), impulsivity (delay discounting), inhibitory control, and motivational sensitization (attentional bias). Mediation analyses will be adjusted for putative sociodemographic (e.g. age) and other shared factors of a potential mediator and outcome (e.g. time of day, preference of beer) as well as for the mentioned factors for moderation. The alpha level will be specified at two-sided 0.05. If necessary, the analyses will be repeated with robust standard errors (via the sandwich estimator) and robust linear regressions [ 116 ]].

Study progress and preliminary feasibility data

The data collection of the presented MESA study started in December 2018. From December 2018 until March 2022 N = 623 persons participated in the online screening. A total of 213 complete data sets have been collected so far. All participants are male and between 18–40 ( M  = 25) years old. 97 of the 213 participants took part in the stress condition, stratified in 40 risky drinkers and 57 regular drinkers. Further, 117 participants took part in the control condition stratified in 36 risky drinkers and 81 regular drinkers. More than half reported they were university students.

Due to the Covid-19 pandemic, the study was paused in the beginning of March 2020 in order to protect the safety and health of all personnel involved in the study and to comply with legislative regulations. The laboratories reopened in September 2021 and data collection was continued.

The present MESA study was developed in response to the incomplete understanding of the underlying mechanisms of the relationship between stress and alcohol consumption. As pointed out, there is a significant amount of people suffering from AUDs with tremendous consequences for the individual as well as for society and health care systems. There is need for preventive interventions at the biological, psychological or social level for individuals at high risk of problematic alcohol consumption before the manifestation of AUD. Research to date has focused primarily on secondary prevention, which aims to prevent AUD progression and relapse, and tertiary prevention, which aims to minimize functional deterioration in chronic AUDs [ 117 ]. The present study focuses on the identification of targets for primary prevention, which is focused on the protection of healthy individuals, and may be provided on a universal, selective or indicated level. The various tasks designed to examine different, potential moderators and mediators can then be used to develop interventions and provide information for the at-risk population. The identification of specific mediators is of key importance as they help to elucidate what mechanisms underlie the association between stress and alcohol consumption. Knowledge about specific mechanisms are of high relevance as it can be used to allocate existing interventions. For example, there are already trainings for many of the investigated mediators (e.g. Attentional Bias Modification, Inhibitory Control trainings), which, should these factors prove to be relevant, could then be specifically adapted and applied in the context of stress [ 118 , 119 , 120 , 121 ]. It can also be used to develop novel interventions that might be useful to prevent stress-related alcohol consumption. Identifying specific moderators will help to tailor these preventive interventions to at high-risk individuals, which increases their potential efficacy and cost-effectiveness.

Given its focus on internal validity using a carefully controlled design in a laboratory setting, external validity will be a limitation of this study. Thus, findings will have to be complemented by investigations in real world settings to make definite conclusions about the association between stress and alcohol use and its underlying mechanisms. This could be achieved for example with ecological momentary assessments, which have shown good feasibility in a couple of promising recent studies on stress-related alcohol use and the role of craving, alterations in mood and inhibitory control [ 122 , 123 , 124 , 125 ].

Taken together, the presented study has a high potential to advance our understanding of stress-related alcohol use. In the long-term, it could stimulate the development of tailored preventive interventions and contribute to a reduction of problematic alcohol use.

Availability of data and materials

The data will be made available on the OSF after completion of the first data analyses.

Abbreviations

Area under the curve

Area under the curve with respect to ground

Area under the curve with respect to increase

Alcohol use disorder

Balanced integration score

Difficulties in emotion regulation scale

Drinking Motive questionnaire—revised

Hypothalamic–pituitary–adrenal axis

Multidimensional Mood State Questionnaire

Underlying mechanisms in the relationship between stress and alcohol consumption in regular and risky drinkers

Proportion of correct responses

Perceived Stress Scale

Response times

Sympathetic nervous system

State-trait-anxiety-inventory

Trier Social Stress Test

Carvalho AF, Heilig M, Perez A, Probst C, Rehm J. Alcohol use disorders. The Lancet. 2019;394(10200):781–92.

Article   Google Scholar  

Rehm J, Shield KD. Global burden of disease and the impact of mental and addictive disorders. Curr Psychiatry Rep. 2019;21(2):10.

Article   PubMed   Google Scholar  

Rehm J, Baliunas D, Borges GLG, Graham K, Irving H, Kehoe T, et al. The relation between different dimensions of alcohol consumption and burden of disease: an overview. Addiction. 2010;105(5):817–43.

Article   PubMed   PubMed Central   Google Scholar  

World Health Organization, World Health Organization, World Health Organization, Management of Substance Abuse Team. Global status report on alcohol and health 2018. 2018.

Bloomfield K, Kraus L, Soyka M, Koch-Institut R. Gesundheitsberichterstattung des Bundes. Berlin, Heidelberg: Robert-Koch-Institut; 2008 S. 34. Report No.: 40.

Peltier MR, Verplaetse TL, Mineur YS, Petrakis IL, Cosgrove KP, Picciotto MR, et al. Sex differences in stress-related alcohol use. Neurobiol Stress. 2019;10:100149.

White A, Castle IJP, Chen CM, Shirley M, Roach D, Hingson R. Converging patterns of alcohol use and related outcomes among females and males in the United States, 2002 to 2012. Alcohol Clin Exp Res. 2015;39(9):1712–26.

Rehm J, Allamani A, Elekes Z, Jakubczyk A, Manthey J, Probst C, et al. Alcohol dependence and treatment utilization in Europe—a representative cross-sectional study in primary care. BMC Fam Pract. 2015;16(1):90.

Trautmann S, Pieper L, Kuitunen-Paul S, Manthey J, Wittchen HU, Bühringer G, et al. Prävalenz und Behandlungsraten von Störungen durch Alkoholkonsum in der primärärztlichen Versorgung in Deutschland. SUCHT. 2016;62(4):233–43.

Sinha R. How does stress lead to risk of alcohol relapse? Alcohol Res Curr Rev. 2012;34(4):432.

Google Scholar  

McGrath E, Jones A, Field M. Acute stress increases ad-libitum alcohol consumption in heavy drinkers, but not through impaired inhibitory control. Psychopharmacology. 2016;233(7):1227–34.

Becker HC. Influence of stress associated with chronic alcohol exposure on drinking. Neuropharmacology. 2017;1(122):115–26.

Hasin D, Keyes KM. The epidemiology of alcohol and drug disorders. In: Addiction medicine. Berlin: Springer; 2011. p. 23–49.

Keyes KM, Hatzenbuehler ML, Grant BF, Hasin DS. Stress and alcohol: epidemiologic evidence. Alcohol Res Curr Rev. 2012;34:391–400.

Marlatt G. Taxonomy of high-risk situations for alcohol relapse: evolution and development of a cognitive-behavioral model. Addict (Abingdon Engl). 1997;91(1):S37-49.

Blomeyer D, Treutlein J, Esser G, Schmidt MH, Schumann G, Laucht M. Interaction between CRHR1 gene and stressful life events predicts adolescent heavy alcohol use. Biol Psychiatry. 2008;63(2):146–51.

Kingston S, Raghavan C. The relationship of sexual abuse, early initiation of substance use, and adolescent trauma to PTSD. J Trauma Stress. 2009;22(1):65–8.

Lijffijt M, Hu K, Swann AC. Stress modulates illness-course of substance use disorders: a translational review. Front Psychiatry. 2014. https://doi.org/10.3389/fpsyt.2014.00083/abstract .

Spanagel R, Noori HR, Heilig M. Stress and alcohol interactions: animal studies and clinical significance. Trends Neurosci. 2014;37(4):219–27.

Stephens MAC, Wand G. Stress and the HPA Axis. Alcohol Res Curr Rev. 2012;34(4):468–83.

Wand G. The influence of stress on the transition from drug use to addiction. Alcohol Res Health J Natl Inst Alcohol Abuse Alcohol. 2008;31(2):119–36.

McEwen BS, Gianaros PJ. Stress-and allostasis-induced brain plasticity. Annu Rev Med. 2011;62:431–45.

Keyes KM, Shmulewitz D, Greenstein E, McLaughlin K, Wall M, Aharonovich E, et al. Exposure to the Lebanon War of 2006 and effects on alcohol use disorders: the moderating role of childhood maltreatment. Drug Alcohol Depend. 2014;134:296–303.

Kim JH, Martins SS, Shmulewitz D, Santaella J, Wall MM, Keyes KM, et al. Childhood maltreatment, stressful life events, and alcohol craving in adult drinkers. Alcohol Clin Exp Res. 2014;38(7):2048–55.

Young-Wolff KC, Kendler KS, Prescott CA. Interactive effects of childhood maltreatment and recent stressful life events on alcohol consumption in adulthood. J Stud Alcohol Drugs. 2012;73(4):559–69.

Diener E, Lucas RE. Personality traits. Gen Psychol Required Read. 2019;278.

de Wit H. Impulsivity as a determinant and consequence of drug use: a review of underlying processes. Addict Biol. 2009;14(1):22–31.

de Wit H, Richards JB. Dual determinants of drug use in humans: reward and impulsivity. Neb Symp Motiv Neb Symp Motiv. 2004;50:19–55.

Stamates AL, Lau-Barraco C. Momentary patterns of impulsivity and alcohol use: A cause or consequence? Drug Alcohol Depend. 2020;217:108246.

Boileau I, Dagher A, Leyton M, Gunn RN, Baker GB, Diksic M, et al. Modeling sensitization to stimulants in humans: an [11C]raclopride/positron emission tomography study in healthy men. Arch Gen Psychiatry. 2006;63(12):1386–95.

Cuomo C, Sarchiapone M, Giannantonio MD, Mancini M, Roy A. Aggression, impulsivity, personality traits, and childhood trauma of prisoners with substance abuse and addiction. Am J Drug Alcohol Abuse. 2008;34(3):339–45.

Khantzian EJ. The self-medication hypothesis of substance use disorders: a reconsideration and recent applications. Harv Rev Psychiatry. 1997;4(5):231–44.

Crum RM, La Flair L, Storr CL, Green KM, Stuart EA, Alvanzo AAH, et al. Reports of drinking to self-medicate anxiety symptoms: longitudinal assessment for subgroups of individuals with alcohol dependence: research article: self-medication and alcohol dependence. Depress Anxiety. 2013;30(2):174–83.

Kushner M. The relationship between anxiety disorders and alcohol use disorders A review of major perspectives and findings. Clin Psychol Rev. 2000;20(2):149–71.

Lazareck S, Robinson JA, Crum RM, Mojtabai R, Sareen J, Bolton JM. A longitudinal investigation of the role of self-medication in the development of comorbid mood and drug use disorders: findings from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC). J Clin Psychiatry. 2012;73(5):e588-593.

Le Berre AP. Emotional processing and social cognition in alcohol use disorder. Neuropsychology. 2019;33(6):808–21.

Petit G, Luminet O, Maurage F, Tecco J, Lechantre S, Ferauge M, et al. Emotion regulation in alcohol dependence. Alcohol Clin Exp Res. 2015;39(12):2471–9.

Blaine SK, Sinha R. Alcohol, stress, and glucocorticoids: from risk to dependence and relapse in alcohol use disorders. Neuropharmacology. 2017;122:136–47.

Schommer NC, Hellhammer DH, Kirschbaum C. Dissociation between reactivity of the hypothalamus-pituitary-adrenal axis and the sympathetic-adrenal-medullary system to repeated psychosocial stress. Psychosom Med. 2003;65(3):450–60.

Lee S, Rivier C. Alcohol increases the expression of type 1, but not type 2α corticotropin-releasing factor (CRF) receptor messenger ribonucleic acid in the rat hypothalamus. Mol Brain Res. 1997;52(1):78–89.

Field M, Powell H. Stress increases attentional bias for alcohol cues in social drinkers who drink to cope. Alcohol Alcohol. 2007;42(6):560–6.

Koob GF, Le Moal M. Plasticity of reward neurocircuitry and the „dark side“ of drug addiction. Nat Neurosci. 2005;8(11):1442–4.

Sinha R. The clinical neurobiology of drug craving. Curr Opin Neurobiol. 2013;23(4):649–54.

Field M, Munafò MR, Franken IHA. A meta-analytic investigation of the relationship between attentional bias and subjective craving in substance abuse. Psychol Bull. 2009;135(4):589–607.

Franken IHA. Drug craving and addiction: integrating psychological and neuropsychopharmacological approaches. Prog Neuropsychopharmacol Biol Psychiatry. 2003;27(4):563–79.

Stamates AL, Lau-Barraco C. The dimensionality of impulsivity: Perspectives and implications for emerging adult drinking. Exp Clin Psychopharmacol. 2017;25(6):521–33.

Wiers RW, Bartholow BD, van den Wildenberg E, Thush C, Engels RCME, Sher KJ, et al. Automatic and controlled processes and the development of addictive behaviors in adolescents: a review and a model. Pharmacol Biochem Behav. 2007;86(2):263–83.

Blaine SK, Nautiyal N, Hart R, Guarnaccia JB, Sinha R. Craving, cortisol and behavioral alcohol motivation responses to stress and alcohol cue contexts and discrete cues in binge and non-binge drinkers. Addict Biol. 2019;24(5):1096–108.

Kim ST, Hwang SS, Kim HW, Hwang EH, Cho J, Kang JI, et al. Multidimensional impulsivity as a mediator of early life stress and alcohol dependence. Sci Rep. 2018;8(1):4104.

VanderWeele T. Explanation in causal inference: methods for mediation and interaction. Oxford: Oxford University Press; 2015. p. 729S.

Herrick C. Governing health and consumption: sensible citizens, behaviour and the city. Bristol: Policy Press; 2011. p. 265S.

Book   Google Scholar  

Institut Für Therapieforschung (IFT), München, Institut Für Klinische Psychologie Und Psychotherapie Der Technischen Universität Dresden, Bundesministerium Für Gesundheit Und Soziale Sicherung, Berlin. Epidemiological Survey on Substance Abuse in Germany 2012 (ESA)Repräsentativerhebung zum Gebrauch und Missbrauch psychoaktiver Substanzen bei Erwachsenen in Deutschland (Epidemiologischer Suchtsurvey 2012) [Internet]. GESIS Data Archive; 2014 [zitiert 8. Dezember 2021]. https://doi.org/10.4232/1.12042

Crum RM, Chan YF, Chen LS, Storr CL, Anthony JC. Incidence rates for alcohol dependence among adults: prospective data from the Baltimore Epidemiologic Catchment Area Follow-Up Survey, 1981–1996. J Stud Alcohol. 2005;66(6):795–805.

Grant BF, Goldstein RB, Chou SP, Huang B, Stinson FS, Dawson DA, et al. Sociodemographic and psychopathologic predictors of first incidence of DSM-IV substance use, mood and anxiety disorders: results from the Wave 2 National Epidemiologic Survey on Alcohol and Related Conditions. Mol Psychiatry. 2009;14(11):1051–66.

Kessler RC, Amminger GP, Aguilar-Gaxiola S, Alonso J, Lee S, Ustün TB. Age of onset of mental disorders: a review of recent literature. Curr Opin Psychiatry. 2007;20(4):359–64.

Stalder T, Kirschbaum C. Analysis of cortisol in hair–state of the art and future directions. Brain Behav Immun. 2012;26(7):1019–29.

Corp S. Stata statistical software: release 17. College Station: StataCorp LLC.; 2021.

Collins RL, Kashdan TB, Koutsky JR, Morsheimer ET, Vetter CJ. A self-administered Timeline Followback to measure variations in underage drinkers’ alcohol intake and binge drinking. Addict Behav. 2008;33(1):196–200.

Wittchen HU, Zaudig M, Fydrich T. Strukturiertes Klinisches Interview f€ ur DSM-IV (SKID-I und SKID-II). Achse Psych Sto Rungen Achse II Perso Nlichkeitssto Rungen Go Ttingen Hogrefe. 1997.

Gröschl M. Current status of salivary hormone analysis. Clin Chem. 2008;54(11):1759–69.

Whiteside SP, Lynam DR, Miller JD, Reynolds SK. Validation of the UPPS impulsive behaviour scale: a four-factor model of impulsivity. Eur J Pers. 2005;19(7):559–74.

Bernstein DP, Fink L, Handelsman L, Foote J. Childhood trauma questionnaire. Assess Fam Violence Handb Res Pract. 1998.

Rohleder N, Nater UM. Determinants of salivary alpha-amylase in humans and methodological considerations. Psychoneuroendocrinology. 2009;34(4):469–85.

Wolff M, Krönke KM, Venz J, Kräplin A, Bühringer G, Smolka MN, et al. Action versus state orientation moderates the impact of executive functioning on real-life self-control. J Exp Psychol Gen. 2016;145(12):1635–53.

Pooseh S, Bernhardt N, Guevara A, Huys QJM, Smolka MN. Value-based decision-making battery: a Bayesian adaptive approach to assess impulsive and risky behavior. Behav Res Methods. 2018;50(1):236–49.

Kräplin A, Höfler M, Pooseh S, Wolff M, Krönke KM, Goschke T, et al. Impulsive decision-making predicts the course of substance-related and addictive disorders. Psychopharmacology. 2020;237(9):2709–24.

Cohen S, Kamarck T, Mermelstein R. Perceived stress scale. Meas Stress Guide Health Soc Sci. 1994;10:1–2.

Bond FW, Hayes SC, Baer RA, Carpenter KM, Guenole N, Orcutt HK, et al. Preliminary psychometric properties of the Acceptance and Action Questionnaire-II: a revised measure of psychological inflexibility and experiential avoidance. Behav Ther. 2011;42(4):676–88.

Kuntsche E, Kuntsche S. Development and validation of the drinking motive questionnaire revised short form (DMQ–R SF). J Clin Child Adolesc Psychol. 2009;38(6):899–908.

Kräplin A, Dshemuchadse M, Behrendt S, Scherbaum S, Goschke T, Bühringer G. Dysfunctional decision-making in pathological gambling: pattern specificity and the role of impulsivity. Psychiatry Res. 2014;215(3):675–82. https://doi.org/10.1016/j.psychres.2013.12.041 .

Raabe A, Grüsser SM, Wessa M, Podschus J, Flor H. The assessment of craving: psychometric properties, factor structure and a revised version of the Alcohol Craving Questionnaire (ACQ). Addict Abingdon Engl. 2005;100(2):227–34.

Gratz KL, Roemer L. Multidimensional assessment of emotion regulation and dysregulation: development, factor structure, and initial validation of the difficulties in emotion regulation scale. J Psychopathol Behav Assess. 2003;14:41–54.

Spielberger CD. Test anxiety inventory. Corsini Encycl Psychol. 2010;1–1.

Steyer R, Schwenkmezger P, Notz P, Eid M. Testtheoretische Analysen des Mehrdimensionalen Befindlichkeitsfragebogen (MDBF). [Theoretical analysis of a multidimensional mood questionnaire (MDBF).]. Diagnostica. 1994;40(4):320–8.

Derryberry D, Reed MA. Anxiety-related attentional biases and their regulation by attentional control. J Abnorm Psychol. 2002;111(2):225.

Gratz KL, Roemer L. Multidimensional assessment of emotion regulation and dysregulation: Development, factor structure, and initial validation of the difficulties in emotion regulation scale. J Psychopathol Behav Assess. 2004;26(1):41–54.

Ehring T, Zetsche U, Weidacker K, Wahl K, Schönfeld S, Ehlers A. The Perseverative Thinking Questionnaire (PTQ): validation of a content-independent measure of repetitive negative thinking. J Behav Ther Exp Psychiatry. 2011;42(2):225–32.

Smyth C. The Pittsburgh sleep quality index (PSQI). Bd. 25, Journal of gerontological nursing. SLACK Incorporated Thorofare, NJ; 1999. S. 10–10.

Gerlach AL, Andor T, Patzelt J. Die bedeutung von unsicherheitsintoleranz für die generalisierte angststörung modellüberlegungen und entwicklung einer deutschen version der unsicherheitsintoleranz-skala. Z Für Klin Psychol Psychother. 2008;37(3):190–9.

Carver CS. You want to measure coping but your protocol’too long: consider the brief cope. Int J Behav Med. 1997;4(1):92–100.

Beck AT, Steer RA, Brown GK. Beck depression inventory. New York: Harcourt Brace Jovanovich; 1987.

Tangney JP, Baumeister RF, Boone AL. High self-control predicts good adjustment, less pathology, better grades, and interpersonal success. J Pers. 2004;72(2):271–324.

Connor KM, Davidson JR. Development of a new resilience scale: the Connor-Davidson resilience scale (CD-RISC). Depress Anxiety. 2003;18(2):76–82.

Jones A, Button E, Rose AK, Robinson E, Christiansen P, Di Lemma L, et al. The ad-libitum alcohol ‘taste test’: secondary analyses of potential confounds and construct validity. Psychopharmacology. 2016;233(5):917–24.

Steudte-Schmiedgen S, Stalder T, Schönfeld S, Wittchen HU, Trautmann S, Alexander N, et al. Hair cortisol concentrations and cortisol stress reactivity predict PTSD symptom increase after trauma exposure during military deployment. Psychoneuroendocrinology. 2015;59:123–33.

Trautmann S, Muehlhan M, Kirschbaum C, Wittchen HU, Höfler M, Stalder T, et al. Biological stress indicators as risk markers for increased alcohol use following traumatic experiences. Addict Biol. 2018;23(1):281–90.

Hellhammer DH, Wüst S, Kudielka BM. Salivary cortisol as a biomarker in stress research. Psychoneuroendocrinology. 2009;34(2):163–71.

Kirschbaum C, Wüst S, Hellhammer D. Consistent sex differences in cortisol responses to psychological stress. Psychosom Med. 1992;54(6):648–57.

Kudielka BM, Gierens A, Hellhammer DH, Wüst S, Schlotz W. Salivary cortisol in ambulatory assessment—some dos, some don’ts, and some open questions. Psychosom Med. 2012;74(4):418–31.

Field M, Quigley M. Mild stress increases attentional bias in social drinkers who drink to cope: a replication and extension. Exp Clin Psychopharmacol. 2009;17(5):312–9.

Field M, Cox WM. Attentional bias in addictive behaviors: a review of its development, causes, and consequences. Drug Alcohol Depend. 2008;97(1–2):1–20.

Ataya AF, Adams S, Mullings E, Cooper RM, Attwood AS, Munafò MR. Internal reliability of measures of substance-related cognitive bias. Drug Alcohol Depend. 2012;121(1–2):148–51.

Field M, Christiansen P. Commentary on; “Internal reliability of measures of substance-related cognitive bias.” Drug Alcohol Depend. 2012;124(3):189–90.

Zvielli A, Bernstein A, Koster EH. Temporal dynamics of attentional bias. Clin Psychol Sci. 2015;3(5):772–88.

Bogacz R, Wagenmakers EJ, Forstmann BU, Nieuwenhuis S. The neural basis of the speed-accuracy tradeoff. Trends Neurosci. 2010;33(1):10–6.

Liesefeld HR, Janczyk M. Combining speed and accuracy to control for speed-accuracy trade-offs(?). Behav Res Methods. 2019;51(1):40–60.

Brainard DH, Vision S. The psychophysics toolbox. Spat Vis. 1997;10(4):433–6.

Kirschbaum C, Pirke KM, Hellhammer DH. The ’Trier Social Stress Test’—a tool for investigating psychobiological stress responses in a laboratory setting. Neuropsychobiology. 1993;28(1–2):76–81.

Labuschagne I, Grace C, Rendell P, Terrett G, Heinrichs M. An introductory guide to conducting the Trier Social Stress Test. Neurosci Biobehav Rev. 2019;107:686–95.

Dickerson SS, Kemeny ME. Acute stressors and cortisol responses: a theoretical integration and synthesis of laboratory research. Psychol Bull. 2004;130(3):355–91.

Het S, Rohleder N, Schoofs D, Kirschbaum C, Wolf OT. Neuroendocrine and psychometric evaluation of a placebo version of the ‘Trier Social Stress Test.’ Psychoneuroendocrinology. 2009;34(7):1075–86.

Colby SM, Rohsenow DJ, Monti PM, Gwaltney CJ, Gulliver SB, Abrams DB, et al. Effects of tobacco deprivation on alcohol cue reactivity and drinking among young adults. Addict Behav. 2004;29(5):879–92.

Jones A, Rose AK, Cole J, Field M. Effects of alcohol cues on craving and ad libitum alcohol consumption in social drinkers: the role of disinhibition. J Exp Psychopathol. 2013;4(3):239–49.

Van Dyke N, Fillmore MT. Operant responding for alcohol following alcohol cue exposure in social drinkers. Addict Behav. 2015;47:11–6.

Christiansen P, Cole JC, Field M. Ego depletion increases ad-lib alcohol consumption: investigating cognitive mediators and moderators. Exp Clin Psychopharmacol. 2012;20(2):118–28.

Jones A, Guerrieri R, Fernie G, Cole J, Goudie A, Field M. The effects of priming restrained versus disinhibited behaviour on alcohol-seeking in social drinkers. Drug Alcohol Depend. 2011;113(1):55–61.

Quigley BM, Collins RL. The modeling of alcohol consumption: a meta-analytic review. J Stud Alcohol. 1999;60(1):90–8.

Bowley C, Faricy C, Hegarty B, Johnstone S, Smith J, Kelly P, et al. The effects of inhibitory control training on alcohol consumption, implicit alcohol-related cognitions and brain electrical activity. Int J Psychophysiol Off J Int Organ Psychophysiol. 2013;89(3):342–8.

Field M, Eastwood B. Experimental manipulation of attentional bias increases the motivation to drink alcohol. Psychopharmacology. 2005;183(3):350–7.

Jones A, Field M. The effects of cue-specific inhibition training on alcohol consumption in heavy social drinkers. Exp Clin Psychopharmacol. 2013;21(1):8–16.

Miller R, Stalder T, Jarczok M, Almeida DM, Badrick E, Bartels M, et al. The CIRCORT database: reference ranges and seasonal changes in diurnal salivary cortisol derived from a meta-dataset comprised of 15 field studies. Psychoneuroendocrinology. 2016;73:16–23.

Pruessner JC, Kirschbaum C, Meinlschmid G, Hellhammer DH. Two formulas for computation of the area under the curve represent measures of total hormone concentration versus time-dependent change. Psychoneuroendocrinology. 2003;28(7):916–31.

Miller R, Wojtyniak JG, Weckesser LJ, Alexander NC, Engert V, Lehr T. How to disentangle psychobiological stress reactivity and recovery: a comparison of model-based and non-compartmental analyses of cortisol concentrations. Psychoneuroendocrinology. 2018;90:194–210.

Miller R, Plessow F, Kirschbaum C, Stalder T. Classification criteria for distinguishing cortisol responders from nonresponders to psychosocial stress: evaluation of salivary cortisol pulse detection in panel designs. Psychosom Med. 2013;75(9):832–40.

Robins JM, Greenland S. Identifiability and exchangeability for direct and indirect effects. Epidemiol Camb Mass. 1992;3(2):143–55.

Royall RM. Model robust confidence intervals using maximum likelihood estimators. Int Stat Rev Int Stat. 1986;221–6.

Trova AC, Paparrigopoulos T, Liappas I, Ginieri-Coccossis M. Prevention of alcohol dependence. Psychiatr Psychiatr. 2015;26(2):131–40.

Bickel WK, Jarmolowicz DP, Mueller ET, Koffarnus MN, Gatchalian KM. Excessive discounting of delayed reinforcers as a trans-disease process contributing to addiction and other disease-related vulnerabilities: emerging evidence. Pharmacol Ther. 2012;134(3):287–97.

Houben K, Nederkoorn C, Wiers RW, Jansen A. Resisting temptation: decreasing alcohol-related affect and drinking behavior by training response inhibition. Drug Alcohol Depend. 2011;116(1–3):132–6.

López-Caneda E, Rodríguez Holguín S, Cadaveira F, Corral M, Doallo S. Impact of alcohol use on inhibitory control (and vice versa) during adolescence and young adulthood: a review. Alcohol Alcohol Oxf Oxfs. 2014;49(2):173–81.

Schoenmakers T, Wiers RW, Jones BT, Bruce G, Jansen ATM. Attentional re-training decreases attentional bias in heavy drinkers without generalization. Addict Abingdon Engl. 2007;102(3):399–405.

Szeto EH, Schoenmakers TM, van de Mheen D, Snelleman M, Waters AJ. Associations between dispositional mindfulness, craving, and drinking in alcohol-dependent patients: an ecological momentary assessment study. Psychol Addict Behav J Soc Psychol Addict Behav. 2019;33(5):431–41.

Mayhugh RE, Rejeski WJ, Petrie MR, Laurienti PJ, Gauvin L. Differing patterns of stress and craving across the day in moderate-heavy alcohol consumers during their typical drinking routine and an imposed period of alcohol abstinence. PLoS ONE. 2018;13(4):e0195063.

Jones A, Tiplady B, Houben K, Nederkoorn C, Field M. Do daily fluctuations in inhibitory control predict alcohol consumption? An ecological momentary assessment study. Psychopharmacology. 2018;235(5):1487–96.

Duif M, Thewissen V, Wouters S, Lechner L, Jacobs N. Associations between affect and alcohol consumption in adults: an ecological momentary assessment study. Am J Drug Alcohol Abuse. 2020;46(1):88–97.

Download references

Acknowledgements

We acknowledge Helen Lenhardt and Lisa Hofmann for their assistance in data collection.

The funder has no role in study design, data collection and analysis, decision to publish, or preparation of any kind of manuscript.

Open Access funding enabled and organized by Projekt DEAL. The study is funded by the Deutsche Forschungsgemeinschaft (DFG), Grant No. TR 1489/1-1.

Author information

Authors and affiliations.

Department of Psychology, Faculty of Human Science, Medical School Hamburg, Hamburg, Germany

Charlotte Wittgens, Markus Muehlhan & Sebastian Trautmann

ICPP Institute for Clinical Psychology and Psychotherapy, Medical School Hamburg, Hamburg, Germany

Charlotte Wittgens & Sebastian Trautmann

ICAN Institute for Cognitive and Affective Neuroscience, Medical School Hamburg, Hamburg, Germany

Markus Muehlhan

Work Group Addictive Behaviors, Risk Analysis and Risk Management, Faculty of Psychology, Technische University Dresden, Dresden, Germany

Anja Kräplin

Department of Psychiatry and Psychotherapy, Campus Charité Mitte, Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin, Germany

You can also search for this author in PubMed   Google Scholar

Contributions

All authors have read and provided substantial contributions to the final version of the study protocol. ST is the central and principal investigator of the project and is responsible for drafting the initial proposal with all subsequent revisions. CW is the principal investigator for the study and for drafting the final protocol for publication. MM contributed substantially to drafting the initial proposal and to the revision of the initial proposal as well as to the revision and final approval of the manuscript. AK contributed substantially to the revision of the initial proposal and to the revision and final approval of the manuscript. MW provided the paradigms for the proposal and contributed to the revision and final approval of the manuscript. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Charlotte Wittgens or Sebastian Trautmann .

Ethics declarations

Ethics approval and consent to participate.

The ethics approval for the study was given by ethics committee of Technische Universität Dresden (EK 522122016) and the ethics committee of Medical School Hamburg (MSH) (MSH-2020/114). The participants give their written informed consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Wittgens, C., Muehlhan, M., Kräplin, A. et al. Underlying mechanisms in the relationship between stress and alcohol consumption in regular and risky drinkers (MESA): methods and design of a randomized laboratory study. BMC Psychol 10 , 233 (2022). https://doi.org/10.1186/s40359-022-00942-1

Download citation

Received : 08 June 2022

Accepted : 05 October 2022

Published : 15 October 2022

DOI : https://doi.org/10.1186/s40359-022-00942-1

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Risky alcohol consumption
• Acute stress
• Ad-libitum taste-test

BMC Psychology

ISSN: 2050-7283

• General enquiries: [email protected]

An official website of the United States government

Here's how you know

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Volume 35 Issue 2 December 1, 2013

The Burden of Alcohol Use: Excessive Alcohol Consumption and Related Consequences Among College Students

Part of the Topic Series: Alcohol’s Evolving Impact on Individuals, Families, and Society

Aaron White, Ph.D., and Ralph Hingson, Sc.D.

Aaron White, Ph.D., is program director, College and Underage Drinking Prevention Research; and

Ralph Hingson, Sc.D., is director, Division of Epidemiology and Prevention Research, both at the National Institute on Alcohol Abuse and Alcoholism, Bethesda, Maryland.

Research shows that multiple factors influence college drinking, from an individual’s genetic susceptibility to the positive and negative effects of alcohol, alcohol use during high school, campus norms related to drinking, expectations regarding the benefits and detrimental effects of drinking, penalties for underage drinking, parental attitudes about drinking while at college, whether one is member of a Greek organization or involved in athletics, and conditions within the larger community that determine how accessible and affordable alcohol is. Consequences of college drinking include missed classes and lower grades, injuries, sexual assaults, overdoses, memory blackouts, changes in brain function, lingering cognitive deficits, and death. This article examines recent findings about the causes and consequences of excessive drinking among college students relative to their non-college peers and many of the strategies used to collect and analyze relevant data, as well as the inherent hurdles and limitations of such strategies.

Since 1976, when the National Institute on Alcohol Abuse and Alcoholism (NIAAA) issued its first report on alcohol misuse by college students, research advances have transformed our understanding of excessive drinking on college campuses and the negative outcomes that follow from it. For instance, we now know that a broad array of factors influence whether a particular college student will choose to drink, the types of consequences they suffer from drinking, and how they respond to those consequences. We have learned that predisposing factors include an individual’s genetic susceptibility to the positive and negative effects of alcohol, alcohol use during high school, campus norms related to drinking, expectations regarding the benefits and detrimental effects of drinking, penalties for underage drinking, parental attitudes about drinking while at college, whether one is member of a Greek organization or involved in athletics, and conditions within the larger community that determine how accessible and affordable alcohol is. Consequences include missed classes and lower grades, injuries, sexual assaults, overdoses, memory blackouts, changes in brain function, lingering cognitive deficits, and death.

This article reviews recent research findings about alcohol consumption by today’s college students and the outcomes that follow. It examines what we know about the causes and consequences of excessive drinking among college students relative to their non-college peers and many of the strategies used to collect and analyze relevant data, as well as the inherent hurdles and limitations of such strategies.

Excessive Drinking At College

Currently, only two active national survey studies are able to characterize the drinking habits of college students in the United States. The National Survey on Drug Use and Health (NSDUH), an annual survey sponsored by the Substance Abuse and Mental Health Services Administration (SAMHSA), involves face-to-face interviews with approximately 67,500 persons ages 12 and older each year regarding use of alcohol and other drugs. Monitoring the Future (MTF) is an annual, paper-and-pencil national survey of alcohol and other drug use with a sample comprising nearly 50,000 students in 8th, l0th, and 12th grades drawn from roughly 420 public and private schools. Approximately 2,400 graduating seniors are resurveyed in subsequent years, allowing for the monitoring of trends in college drinking.

In addition, two prior surveys yielded data on college drinking that remain valuable and relevant. The National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), sponsored by NIAAA, collected data on alcohol and other drug use from a sample of roughly 46,500 citizens 18 and older using face-to-face computer-assisted interviews. Two waves of data (2001 and 2004) were collected from the same sample, and data from an independent sample are scheduled to be collected in 2013. The Harvard College Alcohol Study (CAS), although no longer active, was a landmark paper-and-pencil survey that provided national data (years 1993, 1997, 1999, and 2001) from roughly 15,000 students on more than 100 college campuses each year (Wechsler and Nelson 2008). Data from both NESARC and Harvard CAS remain useful for examining associations between patterns of drinking at college and the frequency and prevalence of alcohol-related consequences for both drinkers and nondrinkers.

Data from NSDUH and MTF suggest that roughly 65 percent of college students drink alcohol in a given month (see figure 1 for data from MTF), and Harvard CAS all suggest that a large percentage of college students who drink do so to excess. Excessive, or “binge,” drinking is defined in NSDUH, MTF, and NESARC as consuming five or more drinks in an evening, although the instruments vary in the specified time frames given (i.e., once or more in the past month for NSDUH, past 2 weeks for MTF, and multiple time periods for NESARC) (Johnston et al. 2001 a ; SAMHSA 2011). The Harvard CAS was the first national study of college students to utilize a gender-specific definition of binge drinking (i.e., four or more drinks in an evening for females or five or more for males in the past 2 weeks) to equate the risk of alcohol-related harms (Wechsler et al. 1995). The Centers for Disease Control and Prevention (CDC) utilizes the same four or more/five or more gender-specific measures but specifies a 30-day time period (Chen et al. 2011). NIAAA uses the four or more/five or more gender-specific measure but specifies a time frame of 2 hours for consumption, as this would generate blood alcohol levels of roughly 0.08 percent, the legal limit for driving, for drinkers of average weight (NIAAA 2004).

 According to NSDUH, the percentage of 18- to 22-year- old college students who reported drinking five or more drinks on an occasion in the previous 30 days remained relatively stable from 2002 (44 percent) to 2010 (44 percent) (SAMHSA 2011). Among 18- to 22-year-olds not enrolled in college, the percentage who engaged in binge drinking decreased significantly from 2002 (39 percent) to 2010 (36 percent) (see figure 2).

Looking at a longer time period, data from MTF suggest that there have been significant declines in the percentage of college students consuming five or more drinks in the previous 2 weeks, from 44 percent in 1980 to 36 percent in 2011 (Johnston et al. 2012) (see figure 3). This time frame includes the passage of the National Minimum Drinking Age Act of 1984, which effectively increased the drinking age from 18 to 21 in the United States.

 Across the four waves of data collection in the Harvard CAS (1993, 1997, 1999, and 2001), rates of binge drinking remained relatively stable (44, 43, 45, and 44 percent, respectively) (Wechsler et al. 2002) (see figure 4). However, the number of non–binge drinkers decreased, whereas the number of frequent binge drinkers (three or more binge-drinking episodes in a 2-week period) increased. Wechsler and colleagues (2002) reported that binge drinkers consumed 91 percent of all the alcohol consumed by college students during the study period. Frequent binge drinkers, a group comprising only 1 in 5 college students, accounted for 68 percent of all alcohol consumed (Wechsler and Nelson 2008).

Individual and Environmental Contributors to Excessive Drinking

Survey data indicate that males outpace females with regard to binge drinking. According to MTF, in 2011, 43 percent of male and 32 percent of female college students crossed the binge threshold in a given 2-week period. Further, 40 percent of students—more males (44 percent) than females (37 percent)—reported getting drunk in a given month. Research suggests that gender differences in alcohol use by college students have narrowed considerably over the years. In their landmark 1953 report on college drinking, Yale researchers Straus and Bacon indicated that, based on survey data from more than 15,000 students on 27 college campuses, 80 percent of males and 49 percent of females reported having been drunk at some point. Nearly 60 years later, in 2011, data from MTF indicated that 68 percent of males and 68 percent females reported having been drunk. These new, higher levels of drinking among females seem to be ingrained in the youth drinking culture. Whereas binge-drinking rates declined significantly among high-school seniors over the last decade, the effect was driven by a decline among males only. Binge-drinking rates among females remained relatively stable (Johnston et al. 2012) (see figure 5).

Beyond gender, survey studies of college drinking reveal a range of characteristics of both individual students and campus environments that influence the likelihood of binge drinking. Data from the Harvard CAS and other studies reveal that males, Caucasians, members of Greek organizations, students on campuses with lower percentages of minority and older students, athletes, students coping with psychological distress, those on campuses near a high density of alcohol outlets, students with access to cheap drink specials, a willingness to endure the consequences of alcohol misuse, and drinking at off-campus parties and bars all contribute to excessive drinking (Mallett et al. 2013; Wechsler and Kuo 2003; Yusko et al. 2008). Further, students living off campus and/or in Greek housing, those who drink to try to fit it, students with inflated beliefs about the proportion of other students who binge drink, and those with positive expectations about the results of drinking are more likely to drink excessively (Scott-Sheldon et al. 2012; Wechsler and Nelson 2008). Importantly, excessive drinking prior to college relative to other college-bound students is predictive of both excessive drinking at college and experiencing alcohol-related consequences (Varvil-Weld et al. 2013; White et al. 2002).

Strengths and Weaknesses of Binge-Drinking Measures

Several studies indicate that crossing commonly used binge-drinking thresholds increases a college student’s risk of experiencing negative alcohol-related consequences. For instance, data from the Harvard CAS indicate that students who binge one or two times during a 2-week period are roughly three times as likely as non–binge drinkers to get behind in school work, do something regretful while drinking, experience a memory blackout, have unplanned sex, fail to use birth control during sex, damage property, get in trouble with police, drive after drinking, or get injured (Wechsler et al. 2000). The more often a student binges, the greater the risk of negative outcomes. Further, the more binge drinking that occurs on a campus, the more likely non–binge drinkers and abstainers are to experience secondhand consequences of alcohol use, such as having studying or sleep disrupted, being a victim of sexual assault, and having property damaged (Wechsler and Nelson 2008).

Because of the increased risk of consequences to self and others that occurs when a person drinks at or beyond the binge threshold, a great deal of emphasis is placed on tracking the percentage of college students that cross binge thresholds. Although this has proven extremely valuable, as Wechsler and Nelson (2001, p. 289) state, “Alcohol use is a complex behavior. No single measure will capture all the relevant aspects of alcohol use.” One limitation of using a single threshold is that it removes data regarding just how heavily students actually drink (Alexander and Bowen 2004; Read et al. 2008) and assigns the same level of risk to all students who cross the thresholds regardless of how far beyond the threshold they go. This is an important consideration as recent studies suggest that plenty of college students who cross the binge threshold when they drink go far beyond it.

In a study of 10,424 first-semester college freshmen, more than one-half of all males and one-third of all females categorized as binge drinkers drank at levels two or more times the binge threshold (8 or more drinks for women and 10 or more drinks for men) at least once in the 2 weeks before the survey. Indeed, one in four binge-drinking males consumed 15 or more drinks at a time during that period (White et al. 2006). Naimi and colleagues (2010) reported that 18- 24-year-olds in the United States drink an average of 9.5 drinks per binge episode, nearly twice the standard binge threshold. Data from MTF also reveal that both college students and their non-college peers often drink at levels that exceed the binge threshold. On average, between 2005 and 2011, 7 percent of college females surveyed and 24 percent of college males consumed 10 or more drinks at least once in a 2-week period, compared with 7 percent of females and 18 percent of males not in college. Further, 2 percent of all college females surveyed and 10 percent of college males consumed 15 or more drinks in a 2-week period. Rates among non-college peers were similar, at 2 percent among females and 9 percent among males (Johnston et al. 2012). For a 140-pound female, consuming 15 drinks over a 6-hour period would produce an estimated blood alcohol level above 0.4 percent, a level known to have claimed, directly, several lives on college campuses in recent years. For a 160-pound male, drinking in this way would lead to a blood alcohol level above 0.3 percent, a potentially lethal level associated with memory blackouts and injury deaths.

Data from the Harvard CAS suggested that students who binge drink frequently (three or more times in a 2-week period) are at particularly high risk of negative alcohol-related outcomes. Compared with students who binge drink one or two times in a 2-week period, those who binge three or more times are twice as likely to experience alcohol-induced memory losses (27 percent vs. 54 percent, respectively), not use protection during sex (10 percent vs. 20 percent, respectively), engage in unplanned sex (22 percent vs. 42 percent, respectively), and get hurt or injured (11 percent vs. 27 percent, respectively), and are equally likely to need medical treatment for an overdose (1 percent vs. 1 percent). Whereas binge frequency is associated with an increased risk of negative outcomes, additional research indicates that there is a relationship between how often a student binges and the peak number of drinks he or she consumes. White and colleagues (2006) reported that 19 percent of frequent binge drinkers consume three or more times the binge threshold (12 or more drinks for females and 15 or more for males) at least once in a 2-week period compared with only 5 percent of infrequent binge drinkers. As a result of the association between frequency of binge drinking and peak levels of consumption, it is difficult to determine if the increase in risk that comes with frequent bingeing is a result of the number of binge episodes, per se, or the number of drinks consumed in an episode.

Importantly, although evidence suggests that many students drink at levels far beyond the binge threshold, additional research suggests that the majority of alcohol-related harms on college campuses result from drinking at levels near the standard four/five-drink measure. This is related to the well-known prevention paradox in which the majority of health problems, such as alcohol-related consequences, tend to occur among those considered to be at lower risk (Rose 1985).

For a particular individual, the odds of experiencing alcohol-related harms increase as the level of consumption increases (Wechsler and Nelson 2001). However, at the population level, far fewer people drink in this manner. As a result, more total consequences occur among those who drink at relatively lower risk levels. For instance, based on data from roughly 9,000 college-student drinkers across 14 college campuses in California, Gruenewald and colleagues (2010) estimated that more than one-half of all alcohol-related consequences resulted from drinking occasions in which four or fewer drinks were consumed. Similarly, using national data from nearly 50,000 students surveyed across the four waves of the Harvard CAS, Weitzman and Nelson (2004) observed that roughly one-quarter to one-third of alcohol-related consequences, including getting injured, vandalizing property, having unprotected sex, and falling behind in school, occurred among students who usually consume three or four drinks per occasion. Such findings raise the possibility that a reduction in high peak levels of consumption might not necessarily result in large overall reductions in alcohol-related consequences on a campus. However, a reduction in high peak levels of drinking would certainly help save the lives of students who drink at these high levels.

In summary, while binge-drinking thresholds are useful for sorting students into categories based on levels of risk, a single threshold cannot adequately characterize the drinking habits of college students or the risks associated with alcohol use on college campuses (Read et al. 2008). It is not uncommon for college students to far exceed standard binge thresholds. Presently, only MTF tracks and reports the incidence of drinking beyond the binge threshold on college campuses. Such data are important as they allow for better tracking of changes in the drinking habits of students. For instance, it is possible that the number of students who drink at extreme levels could increase, whereas the overall percentage of students who binge drink declines or remains stable. Such a phenomenon might help explain why some consequences of excessive alcohol use, like overdoses requiring hospitalization, seem to be on the rise despite relatively stable levels of binge drinking on college campuses across several decades. Finally, although sorting students into binge drinking categories fails to capture high peak levels of consumption among students, a large proportion of harms actually occurs at or near the standard four or more/five or more threshold.

Do Students Know How to Define Standard Servings?

Despite concerns about the accuracy of self-report data for assessing levels of alcohol use among college students and the general population, such surveys remain the most common tool for assessing alcohol use. One major concern is whether students and other young adults are aware of what constitutes a single serving of alcohol. Research shows that college students and the general public tend to define and pour single servings of alcohol that are significantly larger than standard drinks, suggesting they might underestimate their true levels of consumption on surveys (Devos-Comby and Lange 2008; Kerr and Stockwell 2012). For instance, White and colleagues (2003, 2005) asked students to pour single servings of different types of alcohol beverages into cups of various sizes. Overall, students poured drinks that were too large. When asked to simply define standard drinks in terms of fluid ounces, students tended to overstate the number of ounces in a standard drink. The average number of ounces of liquor in student-defined mixed drinks was 4.5 ounces rather than the 1.5 ounces in actual standard drinks (White et al. 2005). When students were provided with feedback regarding discrepancies between their definitions of single servings and the actual sizes of standard drinks, they tended to revise their self-reported levels of consumption upward, leading to a significant increase in the number of students categorized as binge drinkers (White et al. 2005). Such findings suggest that students underreport their levels of consumption on surveys, raising the possibility that more students drink excessively than survey data indicate.

Although a lack of knowledge regarding standard serving sizes could lead students to underestimate, and thus under­report, how much they drink, field research suggests that the discrepancy between self-reported and actual levels of consumption might be smaller than expected from lab studies. For instance, Northcote and Livingston (2011) conducted a study in which they monitored the number of drinks consumed by research participants in bars and then asked them to report their consumption a few days later. Reports by study participants were consistent with the observations made by researchers for participants who had consumed less than eight total drinks. Only those who consumed eight drinks or more tended to underestimate their consumption. When comparing estimated blood alcohol concentrations (BAC) based on self-report to actual BAC readings in college students returning to campus from bars, actual BAC levels tended to be lower, rather than higher, than levels calculated using self-reported consumption (Kraus et al. 2005). Similarly, when actual BAC levels are compared with estimated BAC levels in bar patrons, estimates are spread evenly between accurate, underestimates, and overestimates (Clapp et al. 2009).

In short, although self-reported drinking data might not be perfect, and college students lack awareness of how standard drink sizes are defined, research does not suggest that the discrepancies between self-reported and actual drinking levels are large enough to question the general findings of college drinking surveys.

Paper-and-Pencil, Face-To-Face, and Electronic Surveys: Does It Make a Difference?

National surveys of college drinking often utilize paper-and-pencil questionnaires (e.g., MTF and Harvard CAS) or face-to-face computer-assisted personal interviews (e.g., NSDUH and NESARC). It now is possible to collect survey data electronically via the Internet and also using handheld devices, such as smartphones and personal digital assistants. This raises questions about the comparability between traditional survey methods and electronic data collection.

Several studies comparing traditional (e.g., paper and pencil) and electronic means of data collection suggest that the approaches yield generally similar results from survey participants (Boyer et al. 2002; Jones and Pitt 1999; LaBrie et al. 2006; Lygidakis et al. 2010). For instance, in a comparison of Web-based and paper-and-pencil survey approaches, Knapp and Kirk (2003) found no differences in outcomes, suggesting that Web-based surveys do not diminish the accuracy or honesty of responses. Similarly, LaBrie and colleagues (2006) observed similar outcomes of self-reported alcohol consumption in a paper-and-pencil survey and an electronic survey. However, other studies suggest that students actually feel more comfortable answering personal questions truthfully when completing questionnaires electronically (Turner et al. 1998), which can lead to higher levels of self-reported substance use and other risky behaviors. Both Lygidakis and colleagues (2010) and Wang and colleagues (2005) indicate that adolescents completing electronic surveys reported higher levels of alcohol and other drug use compared with those completing paper-and-pencil versions.

Response rate is an important consideration, with higher response rates increasing the representativeness of the sample and limiting the likelihood that response biases will influence the outcomes. Two national paper-and-pencil surveys mentioned above, MTF and Harvard CAS, report response rates for college students of approximately 59 percent. For MTF, this response rate represents a retention rate, as the participants were followed up after high school. Response rates for the in-person computer-assisted personal interviews, NSDUH and NESARC, which assess college student drinking but are not limited to college students, are roughly 77 percent and 81 percent, respectively. Currently, there is no basis for assessing response rates for national Web-based assessments of college drinking. However, smaller studies suggest that response rates might be comparable, if not higher, than other approaches. McCabe and colleagues (2002) reported that, among 7,000 undergraduate students, one-half of whom were surveyed about alcohol and other drug use via the Internet and half surveyed via paper-and-pencil surveys delivered through the mail, the response rates were 63 percent for the Web survey and 40 percent for the paper-and-pencil survey. Further, response rates for Web-based surveys can be improved by sending reminders via e-mail (van Gelder et al. 2010).

In summary, in recent years an increasing number of researchers have utilized electronic survey methods to collect college-drinking data. At present, evidence suggests that these methods can yield results quite similar to those obtained from traditional survey methods and that response rates might actually be higher.

Alcohol-Related Consequences Among College Students

Drinking to intoxication leads to widespread impairments in cognitive abilities, including decisionmaking and impulse control, and impairments in motor skills, such as balance and hand-eye coordination, thereby increasing the risk of injuries and various other harms. Indeed, research suggests that students who report “getting drunk” even just once in a typical week have a higher likelihood of being injured, experiencing falls that require medical treatment, causing injury in traffic crashes, being taken advantage of sexually, and injuring others in various ways (O’Brien et al. 2006). Students who drink with the objective of getting drunk are far more likely to experience a range of consequences, from hangovers to blackouts, than other students who drink (Boekeloo et al. 2011).

National estimates suggest that thousands of college students are injured, killed, or suffer other significant consequences each year as a result of drinking. However, researchers have questioned the manner in which such national estimates are calculated. In many cases, the lack of college identifiers in datasets means that the actual amount of annual alcohol-attributable harm that occurs among college students is unknown. Although the Harvard CAS collected data regarding the consequences of drinking, its final year of administration was 2001. Currently, assessing the damage done, on a national level, by college drinking requires estimating rates of consequences using a variety of data sources. Such assessments are complicated by the fact that outcomes considered to be negative consequences by researchers (e.g., blackouts and hangovers) are not always perceived as negative by students (Mallett et al. 2013). Further, college students often drink off campus, such as during spring breaks and summer vacations, meaning that many alcohol-related consequences experienced by college students are not necessarily associated with college itself. As such, our understanding of alcohol-related consequences among college students remains somewhat cloudy.

In one set of estimates, Hingson and colleagues (2002, 2005, 2009) utilized census data and national datasets regarding traffic crashes and other injury deaths to estimate the prevalence of various alcohol-related harms among all young people aged 18–24. Next, they attributed an amount of harm to college students equal to the proportion of all 18- to 24-year-olds who were enrolled full time in 4-year colleges (33 percent in 2005, the most recent year analyzed) (Hingson et al. 2009). Because college students drink more heavily than their non-college peers, it is possible this approach underestimated the magnitude of alcohol-related consequences on college campuses. Hingson and colleagues (2002, 2005, 2009) also used the percentage of college students who reported various alcohol-related behaviors (e.g., being assaulted by another drinking college student) in national surveys to derive national estimates of the total numbers of college students who experienced these consequences.

Based on the above strategies along with other sources of data, researchers have estimated the following rates and prevalence of alcohol-related harms involving college students:

• Death:  It is possible that more than 1,800 college students between the ages of 18 and 24 die each year from alcohol- related unintentional injuries, including motor-vehicle crashes (Hingson et al. 2009).
• Injury:  An estimated 599,000 students between the ages of 18 and 24 are unintentionally injured each year under the influence of alcohol (Hingson et al. 2009).
• Physical Assault:  Approximately 646,000 students between the ages of 18 and 24 are assaulted each year by another student who has been drinking (Hingson et al. 2009).
• Sexual Assault:  Perhaps greater than 97,000 students between the ages of 18 and 24 are victims of alcohol-related sexual assault or date rape each year (Hingson et al. 2009).
• Unsafe Sex:  An estimated 400,000 students between the ages of 18 and 24 had unprotected sex and nearly 110,000 students between the ages of 18 and 24 report having been too intoxicated to know if they consented to having sex (Hingson et al. 2002).
• Health Problems:  More than 150,000 students develop an alcohol-related health problem each year (Hingson et al. 2002).
• Suicide Attempts:  Between 1.2 and 1.5 percent of college students indicate that they tried to commit suicide within the past year as a result of drinking or drug use (Presley et al. 1998).
• Drunk Driving:  Roughly 2.7 million college students between the ages of 18 and 24 drive under the influence of alcohol each year (Hingson et al. 2009).
• Memory Loss:  National estimates suggest that 10 percent of non–binge drinkers, 27 percent of occasional binge drinkers, and 54 percent of frequent binge drinkers reported at least one incident in the past year of blacking out, defined as having forgotten where they were or what they did while drinking (Wechsler et al. 2000; White 2003).
• Property Damage:  More than 25 percent of administrators from schools with relatively low drinking levels and more than 50 percent from schools with high drinking levels say their campuses have a “moderate” or “major” problem with alcohol-related property damage (Wechsler et al. 1995).
• Police Involvement:  Approximately 5 percent of 4-year college students are involved with the police or campus security as a result of their drinking (Wechsler et al. 2002) and an estimated 110,000 students between the ages of 18 and 24 are arrested for an alcohol-related violation such as public drunkenness or driving under the influence (Hingson et al. 2002). A more recent national study reported that 8.5 percent of students were arrested or had other trouble with the police because of drinking (Presley and Pimentel 2006).
• Alcohol Abuse and Dependence:  Roughly 20 percent of college students meet the criteria for an alcohol use disorder in a given year (8 percent alcohol abuse, 13 percent alcohol dependence). Rates among age mates not in college are comparable (17 percent any alcohol use disorder, 7 percent alcohol abuse, 10 percent alcohol dependence) (Blanco et al. 2008).

With regard to assessing the number of college students who die from alcohol each year, in addition to the lack of college identifiers in datasets, another barrier is the fact that levels of alcohol often are not measured in nontraffic fatalities. As such, attributable fractions, based on analyses of existing reports in which alcohol levels were measured postmortem, are used to estimate the number of deaths by various means that likely involved alcohol. The CDC often uses attributable fractions calculated by Smith and colleagues (1999) based upon a review of 331 medical-examiner studies. An updated approach is needed. The combination of including college identifiers in medical records and measuring alcohol levels in all deaths would allow for accurate assessments of the role of alcohol in the deaths of college students and their non-college peers.

Academic Performance

About 25 percent of college students report academic consequences of their drinking, including missing class, falling behind in class, doing poorly on exams or papers, and receiving lower grades overall (Engs et al. 1996; Presley et al. 1996 a , b; Wechsler et al. 2002). Although some published research studies have not found a statistically significant association between binge drinking and academic performance (Gill 2002; Howland et al. 2010; Paschall and Freisthler 2003; Williams 2003; Wood et al. 1997), studies linking binge drinking to poorer academic performance outnumber the former studies two to one. Presley and Pimentel (2006) reported that in a national survey of college students, those who engaged in binge drinking and drank at least three times per week were 5.9 times more likely than those who drank but never binged to perform poorly on a test or project as a result of drinking (40.2 vs. 6.8 percent), 5.4 times more likely to have missed a class (64.4 vs. 11.9 percent), and 4.2 times more likely to have had memory loss (64.2 vs. 15.3 percent) (Thombs et al. 2009). Singleton and colleagues (2007, 2009), in separate prospective studies, found negative associations between heavy alcohol use and grade point average. Jennison (2004), based on a national prospective study, reported binge drinkers in college were more likely to drop out of college, work in less prestigious jobs, and experience alcohol dependence 10 years later. Wechsler and colleagues (2000) and Powell and colleagues (2004), based on the Harvard CAS, found frequent binge drinkers were six times more likely than non–binge drinkers to miss class and five times more likely to fall behind in school. White and colleagues (2002) observed that the number of blackouts, a consequence of heavy drinking, was negatively associated with grade point average (GPA). It is important to note that although data regarding GPA often are collected via self-report, the negative association between alcohol consumption and GPA holds even when official records are obtained (Singleton 2007). Collectively, the existing research suggests that heavy drinking is associated with poorer academic success in college.

Alcohol Blackouts

Excessive drinking can lead to a form of memory impairment known as a blackout. Blackouts are periods of amnesia during which a person actively engages in behaviors (e.g., walking, talking) but the brain is unable to create memories for the events. Blackouts are different from passing out, which means either falling asleep or becoming unconscious from excessive drinking. During blackouts, people are capable of participating in events ranging from the mundane, such as eating food, to the emotionally charged, such as fights and even sexual intercourse, with little or no recall (Goodwin 1995). Like milder alcohol–induced short-term memory impairments caused by one or two drinks, blackouts primarily are anterograde, meaning they involve problems with the formation and storage of new memories rather than problems recalling memories formed prior to intoxication. Further, short-term memory often is left partially intact. As such, during a blackout, an intoxicated person is able to discuss events that happened prior to the onset of the blackout and to hold new information in short-term storage long enough to have detailed conversations. They will not, however, be able to transfer new information into long-term storage, leaving holes in their memory. Because of the nature of blackouts, it can be difficult or impossible to know when a drinker in the midst of one (Goodwin 1995).

There are two general types of blackouts based on the severity of the memory impairments. Fragmentary blackouts, sometimes referred to as gray outs or brown outs, are a form of amnesia in which memory for events is spotty but not completely absent. This form is the most common. En bloc blackouts, on the other hand, represent complete amnesia for events (Goodwin 1995).

Blackouts surprisingly are common among college students who drink alcohol. White and colleagues (2002) reported that one-half (51 percent) of roughly 800 college students who had ever consumed alcohol at any point in their lives reported experiencing at least one alcohol-induced blackout, defined as awakening in the morning not able to recall things one did or places one went while under the influence. The average number of total blackouts in those who experienced them was six. Of those who had consumed alcohol during the 2 weeks before the survey was administered, 9 percent reported blacking out. Based on data from 4,539 inbound college students during the summer between high-school graduation and the start of the freshmen year, 12 percent of males and females who drank in the previous 2 weeks experienced a blackout during that time (White and Swartzwelder 2009). Data from the Harvard CAS indicate that blackouts were experienced in a 30-day period by 25 percent of students in 1993 and 27 percent of students in 1997, 1999, and 2001 (Wechsler et al. 2002). A small study by White and colleagues (2004), in which 50 students with histories of blackouts were interviewed, suggests that fragmentary blackouts are far more common than en bloc blackouts. Roughly 80 percent of students described their last blackout as fragmentary.

Blackouts tend to occur following consumption of relatively large doses of alcohol and are more likely if one drinks quickly and on an empty stomach, both of which cause a rapid rise and high peak in BAC (Goodwin 1995; Perry et al. 2006). For this reason, pregaming, or prepartying, which typically involves fast-paced drinking prior to going out to an event, increases the risk of blacking out. Labrie and colleagues (2011) reported that 25 percent of 2,546 students who engaged in prepartying experienced at least one blackout in the previous month. Playing drinking games and drinking shots were risk factors. Further, skipping meals to restrict calories on drinking days is associated with an increased risk of blackouts and other consequences (Giles et al. 2009).

Because blackouts typically follow high peak levels of drinking, it is not surprising that they are predictive of other alcohol-related consequences. Mundt and colleagues (2012) examined past-year blackouts in a sample of more than 900 students in a randomized trial of a screening and brief intervention for problem alcohol use and found that blackouts predicted alcohol-related injuries over a subsequent 2-year period. Compared with students who had no history of blackouts, those who reported one to two blackouts at baseline were 1.5 times more likely to experience an alcohol-related injury, whereas those with six or more blackouts were 2.5 times more likely. In a follow-up report based on the same sample, Mundt and Zakletskaia (2012) estimated that among study participants, one in eight emergency-department (ED) visits for alcohol-related injuries involved a blackout. On a campus of 40,000 students, this would translate into roughly $500,000 in annual costs related to blackout-associated ED visits.

In the study of 50 students with blackout histories by White and colleagues (2004), estimated peak BACs during the night of the last blackout generally were similar for males (0.30 percent) and females (0.35 percent), although it is unlikely that self-reported alcohol consumption during nights in which blackouts occur is highly accurate. A study of amnesia in people arrested for either public intoxication, driving under the influence, or underage drinking found that the probability of a fragmentary or en bloc blackout was 50/50 at a BAC of 0.22 percent and the probability of an en bloc blackout, specifically, was 50/50 at a BAC of 0.31 percent, based on breath alcohol readings (Perry et al. 2006). In their study of blackouts in college students, Hartzler and Fromme (2003 a ) noted a steep increase in the likelihood of blackouts above a BAC of 0.25 percent, calculated from self-reported consumption. Thus, from existing research, it seems that the odds of blacking out increase as BAC levels climb and that blackouts become quite common at BAC levels approaching or exceeding 0.30 percent. As such, the high prevalence of blackouts in college students points to the magnitude of excessive consumption that occurs in the college environment. It should be noted, however, that BAC levels calculated based on self-reported consumption are unlikely to be accurate given the presence of partial or complete amnesia during the drinking occasion.

It seems that some people are more sensitive to the effects of alcohol on memory than others and are therefore at increased risk of experiencing blackouts. Wetherill and Fromme (2011) examined the effects of alcohol on contextual memory in college students with and without a history of blackouts. Performance on a task was similar while the groups were sober, but students with a history of blackouts performed more poorly when intoxicated than those without a history of blackouts. Similarly, Hartzler and Fromme (2003 b ) reported that when mildly intoxicated, study participants with a history of blackouts performed more poorly on a narrative recall task than those without a history of blackouts. When performing a memory task while sober, brain activity measured with functional magnetic resonance imaging is similar in people with a history of blackouts and those without such a history (Wetherill et al. 2012). However, when intoxicated, those with a history of blackouts exhibit lower levels of activity in several regions of the frontal lobes compared with subjects without a history of blackouts.

Thus, studies suggest that there are differences in the effects of alcohol on memory and brain function between those who experience blackouts and those who do not. Research by Nelson and colleagues (2004), using data from monozygotic twins, suggests that there could be a significant genetic component to these differences. Controlling for frequency of intoxication, the researchers found that if one twin experienced blackouts, the other was more likely than chance to experience them as well. Further, Asian-American students with the aldehyde dehydrogenase ALDH2*2 allele 1  are less likely to experience blackouts than those without it, even after adjusting for maximum number of drinks consumed in a day (Lucsak et al. 2006).

1  The ALDH2*2 allele results in decreased action by the enzyme acetaldehyde dehydrogenase, which is responsible for the breakdown of acetaldehyde. The accumulation of acetaldehyde after drinking alcohol leads to symptoms of acetaldehyde poisoning, such as facial flushing and increased heart and respiration rates.

Several challenges hinder the assessment of blackouts and the events that transpire during them. Blackouts represent periods of amnesia. As such, it is difficult to imagine that self-reported drinking levels are highly accurate for nights when blackouts occur. Further, in order for a person to know what transpired during a blackout, and sometimes to be aware that a blackout occurred at all, they need to be told by other individuals. Often, the information provided by these other individuals is unreliable as they were intoxicated themselves (Nash and Takarangi 2011). Thus, it is quite likely that self-reported rates and frequencies of blackouts, drinking levels during nights in which blackouts occur, and the rates of various types of consequences that occur during them, are underestimated.

Alcohol Overdoses

When consumed in large quantities during a single occasion, such as a binge episode, alcohol can cause death directly by suppressing brain stem nuclei that control vital reflexes, like breathing and gagging to clear the airway (Miller and Gold 1991). Even a single session of binge drinking causes inflammation and transient damage to the heart (Zagrosek et al. 2010). The acute toxic effects of alcohol in the body can manifest in symptoms of alcohol poisoning, which include vomiting, slow and irregular breathing, hypothermia, mental confusion, stupor, and death (NIAAA 2007 b ; Oster-Aaland et al. 2009). Using data from the Global Burden of Disease Study, the World Health Organization (WHO) estimated that, in 2002, alcohol poisoning caused 65,700 deaths worldwide, with 2,700 poisoning deaths occurring in the United States (WHO 2009). New stories about alcohol overdoses among college students and their non-college peers have become increasingly common, a fact that is perhaps not surprising given the tendency toward excessive drinking in this age-group.

To investigate the prevalence of hospitalizations for alcohol overdoses—which stem from excessive intoxication or poisoning—among college-aged young people in the United States, White and colleagues (2011) examined rates of inpatient hospitalizations for 18- to 24-year-olds between 1999 and 2008 using data from the Nationwide Inpatient Sample, which contains hospital discharge records from roughly 20 percent of all hospitals in the country. Hospitalizations for alcohol overdoses without any other drugs involved increased 25 percent among 18- to 24-year-olds from 1999 to 2008, highlighting the risks involved in heavy drinking. In total, nearly 30,000 young people in this age-group, more males (19,847) than females (9,525) were hospitalized for alcohol overdoses with no other drugs involved in 2008. Hospitalizations for overdoses involving other drugs but not alcohol increased 55 percent over the same time period, while those involving alcohol and drugs in combination rose 76 percent. In total, there were 59,000 hospitalizations in 2008 among 18- to 24-year-olds for alcohol overdoses only or in combination with other drugs. Given that 33 percent of people in this age-group were full-time college students at 4-year colleges in 2008, a conservative estimate would suggest approximately 20,000 hospitalizations for alcohol overdoses alone or in combination with other drugs involved college students, although the exact number is not known.

Data from the Drug Abuse Warning Network (DAWN) indicate that ED visits for alcohol-related events increased in a similar fashion as those observed for inpatient hospitalizations. Among those ages 18 to 20, ED visits for alcohol-related events with no other drugs increased 19 percent, from 67,382 cases in 2005 to 82,786 cases in 2009. Visits related to combined use of alcohol and other drugs increased 27 percent, from 27,784 cases in 2005 to 38,067 cases in 2009. In 2009, 12 percent of ED visits related to alcohol involved use of alcohol in combination with other drugs (SAMHSA 2011).

Alcohol interacts with a wide variety of illicit and prescription drugs, including opioids and related narcotic analgesics, sedatives, and tranquilizers (NIAAA 2007 a ; Tanaka 2002). Importantly, BAC required for fatal overdoses are lower when alcohol is combined with prescription drugs. An analysis of 1,006 fatal poisonings attributed to alcohol alone or in combination with other drugs revealed that the median postmortem BACs in those who overdosed on alcohol alone was 0.33 percent, compared with 0.13 percent to 0.17 percent among those who overdosed on a combination of alcohol and prescription drugs (Koski et al. 2003, 2005). The combined use of alcohol and other drugs peaks in the 18- to 24-year-old age range (McCabe et al. 2006), suggesting that college-aged young adults are at particularly high risk of suffering consequences from alcohol-and-other-drug combinations.

The above findings reflect the fact that heavy consumption of alcohol quickly can become a medical emergency. One does not need to get behind the wheel of a car after drinking or jump off a balcony into a swimming pool on a dare to risk serious harm. Simply drinking too much alcohol is enough to require hospitalization and potentially cause death. Further, combining alcohol with other drugs can increase the risk of requiring medical intervention substantially. Thus, efforts to minimize the consequences of alcohol-related harms on college campuses should not lose sight of the fact that alcohol often is combined with other drugs and, when this is the case, the risks can be greater than when alcohol or drugs are used alone.

Measuring the true scope of medical treatment for alcohol overdoses among college students is difficult for several reasons. First, in datasets such as the Nationwide Emergency Department Sample (NEDS) and the Nationwide Inpatient Sample (NIS), no college identifiers are included to indicate whether a young person treated for an alcohol overdose is enrolled in college. Many schools do not track or report the number of students treated for an alcohol overdose, and many students drink excessively when away from campus. Further, schools that implement Good Samaritan or Amnesty policies, which allow students to get help for overly intoxicated peers without fear of sanctions, could create the false impression that overdoses are on the rise. For instance, after Cornell University implemented an amnesty policy, they witnessed an increase in calls to residence assistants and 911 for help dealing with an intoxicated friend (Lewis and Marchell 2006). Given the dangerous nature of alcohol overdoses, with or without other drugs involved, it is important to improve the tracking of these events at colleges and in the larger community.

Sexual Assault

Sexual assault is a pervasive problem on college campuses, and alcohol plays a central role in it. A study of roughly 5,500 college females on two campuses revealed that nearly 20 percent experienced some form of sexual assault while at college (Krebs et al. 2009). Data from the Harvard CAS suggested that 5 percent of women surveyed were raped while at college (Mohler-Kuo et al. 2004). In a national sample of students who completed the Core Alcohol and Drug Survey in 2005, 82 percent of students who experienced unwanted sexual intercourse were intoxicated at the time. Similarly, nearly three-quarters (72 percent) of respondents to the Harvard CAS study who reported being raped were intoxicated at the time. In many cases, rape victims are incapacitated by alcohol. In one study, 3.4 percent of rape victims reported being so intoxicated they were unable to consent (Mohler-Kuo et al. 2004). In a study of 1,238 college students on three campuses over a 3-year period, 6 percent of students reported being raped while incapacitated by alcohol (Kaysen et al. 2006).

Research suggests that the involvement of alcohol increases the risk of being victimized in several ways, such as by impairing perceptions that one is in danger and by reducing the ability to respond effectively to sexual aggression (Abbey 2002; McCauley et al. 2010; Testa and Livingston 2009). Further, alcohol might increase the chances that a male will commit a sexual assault by leading them to misinterpret a female’s friendly gestures or flirtation as interest in sex and by increasing sexual aggression (Abbey 2002). When asked to read a story about a potential date rape involving intoxicated college students, both male and female subjects who are intoxicated were more likely to view the female as sexually aroused and the male as acting appropriately (Abbey et al. 2003).

It is widely held that sexual assaults, with and without alcohol involvement, are underreported on college campuses. Title IX of the Education Amendments Act of 1972, a Federal civil rights law, requires universities to address sexual harassment and sexual violence. However, universities vary with regard to how they handle such cases, and a student’s perception of safety and protection can influence the likelihood of reporting a sexual assault. Indeed, many universities have indicated changes in rates of reports of assaults consistent with changes in campus policies regarding how such cases are handled. As such, although it is clear that alcohol often is involved in sexual assaults on college campuses, questions about the frequency and nature of such assaults remain.

Spring Break and 21st Birthday Celebrations— Event-Specific Drinking Occasions

More college students drink, and drink more heavily, during specific celebratory events, such as spring break and 21st birthday celebrations, than during a typical week. Spring break is a roughly weeklong recess from school that takes place in the spring at colleges throughout the United States. While some students continue to work, travel home, or simply relax, others use the opportunity to travel to beaches and other party destinations. During spring break, approximately 42 percent of students get drunk on at least 1 day, 11 percent drink to the point of blacking out or passing out, 32 percent report hangovers, and 2 percent get into trouble with the police (Litt et al. 2013). Students with a history of binge drinking and those intending to get drunk tend to drink the heaviest (Patrick et al. 2013), suggesting that prevention efforts aimed at altering students’ intentions to get drunk while on spring break might lead to a reduction in peak drinking and the consequences that follow (Mallett et al. 2013). Interestingly, students who typically are light drinkers are more likely than those who typically are binge drinkers to experience consequences from excessive drinking during spring break (Lee et al. 2009).

In addition to spring break, 21st birthday celebrations are another event-specific opportunity for students to drink excessively. An estimated 4 out of 5 college students drink alcohol to celebrate their 21st birthdays (Rutledge et al., 2008) and many students drink more than they plan. Of 150 male and female college students surveyed about their intentions to drink during their upcoming 21st birthday celebrations, 68 percent consumed more than they anticipated while only 21 percent drank less and 11 percent were accurate. On average, males intended to consume 8.5 drinks but consumed 12.5, while females expected to drink 7 but had 9 (Brister et al., 2010). As with spring-break drinking, students with a history of binge drinking and those who intended to drink heavily on their 21st birthday consumed the most (Brister et al., 2011). In one study, roughly 12 percent of students reported consuming 21 or more drinks while celebrating, and one-third of females (35 percent) and nearly half of males (49 percent) reached estimated BACs above 0.25 percent (Rutledge et al., 2008). Such high levels of consumption substantially increase the odds of sexual assaults, fights, injuries, and death (Mallett et al., 2013). Research indicates that brief interventions conducted in the week leading up to the 21st birthday celebration can reduce levels of consumption and associated consequences, suggesting that the risks of experiencing alcohol related consequences stemming from 21st birthday celebrations could be partially mitigated through specifically timed prevention efforts (Neighbors et al. 2009, 2012).

We have learned a considerable amount about the drinking habits of college students and the consequences that follow since NIAAA first reported on the matter in 1976. Surprisingly, drinking levels have remained relatively stable on and around college campuses over the last 30 years, with roughly two out of five male and female students engaging in excessive, or binge, drinking. Excessive drinking results in a wide range of consequences, including injuries, assaults, car crashes, memory blackouts, lower grades, sexual assaults, overdoses and death. Further, secondhand effects from excessive drinking place non–binge-drinking students at higher risk of injury, sexual assaults, and having their studying disrupted.

Estimates of the rates of alcohol use and related consequences are imperfect. Lack of knowledge of standard drink sizes and the effects of alcohol on memory formation all complicate the collection of accurate data from traditional self-report surveys. Underreporting of sexual assaults leads to difficulty in estimating the true extent of the problem. Lack of college identifiers in mortality records and the fact that alcohol levels are tested too infrequently in non–traffic-related deaths leaves uncertainty regarding the actual number of college students who die each year from alcohol-related causes. Similarly, college identifiers are not present in most crime reports and hospital reports.

Although it is beyond the scope of this review to examine efforts to prevent excessive drinking on college campuses, it should be noted that important strides have been made in this area (Carey et al. 2012). In addition, data from MTF suggest that levels of binge drinking are decreasing among 12th graders, particularly males. Hopefully, as our understanding of the nature of the problem continues to improve with better measurement strategies, improvements in prevention approaches combined with declines in precollege drinking will lead to reductions in both the levels of alcohol consumption by college students and the negative consequences that result.

Disclosures

The authors declare that they have no competing financial interests.

Abbey, A. Alcohol-related sexual assault: A common problem among college students.  Journal of Studies on Alcohol  (Suppl 14):118–128, 2002.  PMID: 12022717

Abbey, A.; Buck, P.O.; Zawacki, T.; and Saenz, C. Alcohol’s effects on perceptions of a potential date rape.  Journal of Studies on Alcohol  64(5):669–677, 2003.  PMID: 14572189

Alexander, E.N., and Bowen, A.M. Excessive drinking in college: Behavioral outcome, not binge, as a basis for prevention.  Addictive Behaviors  29(6):1199–1205, 2004.  PMID: 15236823

Blanco, C.; Okuda, M.; Wright, C. et al. Mental health of college students and their non-college-attending peers: Results from the National Epidemiologic Study on Alcohol and Related Conditions.  Archives of General Psychiatry  65(12):1429–1437, 2008.  PMID: 19047530

Boekeloo, B.O.; Novik, M.G.; and Bush, E. Drinking to get drunk among incoming freshmen college students.  American Journal of Health Education  42(2):88–95, 2011.  PMID: 23440674

Boyer, K.K.; Olson, J.R.; Calatone, R.J.; and Jackson, E.C. Print versus electronic surveys: A comparison of two data collection methodologies.  Journal of Operations Management  20(4):357–373, 2002.

Brister, H.A.; Wetherill, R.R.; and Fromme, K. Anticipated versus actual alcohol consumption during 21st birthday celebrations.  Journal of Studies on Alcohol and Drugs  71(2):180–183, 2010.  PMID: 20230714

Brown, S.A.; Tapert, S.F.; Granholm, E.; and Delis, D.C. Neurocognitive functioning of adolescents: Effects of protracted alcohol use.  Alcoholism: Clinical and Experimental Research  24(2):164–171, 2000.  PMID: 10698367

Butler, L.H., and Correia, C.J. Brief alcohol intervention with college student drinkers: Face-to face versus computerized feedback.  Psychology of Addictive Behaviors  23(1):163–167, 2009.  PMID: 19290702 .

Campbell, C.A.; Hahn, R.A.; Elder, R.; et al. The effectiveness of limiting alcohol outlet density as a means of reducing excessive alcohol consumption and alcohol-related harms.  American Journal of Preventive Medicine  37(6):556–569, 2009.  PMID: 19944925

Carey, K.B.; Scott-Sheldon, L.A.; Carey, M.P.; and DeMartini, K.S. Individual-level interventions to reduce college student drinking: A meta-analytic review.  Addictive Behaviors  32(11):2469–2494, 2007.  PMID: 17590277

Carey, K.B.; Scott-Sheldon, L.A.; Elliott, J.C.; et al. Face-to-face versus computer-delivered alcohol interventions for college drinkers: A meta-analytic review, 1998 to 2010.  Clinical Psychology Review  32(8):690–703, 2012.  PMID: 23022767

Carpenter, C., and Dobkin, C. The minimum legal drinking age and public health.  Journal of Economic Perspectives  25(2):133–156, 2011.  PMID: 21595328

Chavez, P.R.; Nelson, D.E.; Naimi, T.S.; and Brewer, R.D. Impact of a new gender-specific definition for binge drinking on prevalence estimates for women.  American Journal of Preventive Medicine  40(4):468–471, 2011.  PMID: 21406282

Clapp, J.D.; Johnson, M.; Voas, R.B.; et al. Reducing DUI among US college students: Results of an environmental prevention trial.  Addiction  100(3):327–334, 2005.  PMID: 15733246

Clapp, J.D.; Min, J.W.; Trim, R.S.; et al. Predictors of error in estimates of blood alcohol concentration: A replication.  Journal of Studies on Alcohol and Drugs  70(5):683–688, 2009.  PMID: 19737492

Crego, A.; Rodriguez-Holguin, S.; Parada, M., et al. Reduced anterior prefrontal cortex activation in young binge drinkers during a visual working memory task.  Drug and Alcohol Dependence  109(1–3):45–56, 2010.  PMID: 20079980

DeJong, W.; Schneider, S.K.; Towvim, L.G.; et al. A multisite randomized trial of social norms marketing campaigns to reduce college student drinking.  Journal of Studies on Alcohol  67(6):868–879, 2006.  PMID: 17061004

DeJong, W.; Schneider, S.K.; Towvim, L.G.; et al. A multisite randomized trial of social norms marketing campaigns to reduce college student drinking: A replication failure.  Substance Abuse  30(2):127–140, 2009.  PMID: 19347752

Devos-Comby, L., and Lange, JjE. “My drink is larger than yours”? A literature review of self-defined drink sizes and standard drinks.  Current Drug Abuse Reviews  1(2):162–176, 2008.  PMID: 19630715

Elder, R.W.; Lawrence, B.; Ferguson, A.; et al. The effectiveness of tax policy interventions for reducing excessive alcohol consumption and related harms.  American Journal of Preventive Medicine  38(2):217–229, 2010.  PMID: 20117579

Elliott, J.C.; Carey, K.B.; and Bolles, J.R. Computer-based interventions for college drinking: A qualitative review.  Addictive Behaviors  33(8):994–1005, 2008.  PMID: 18538484

Engs, R.C.; Diebold, B.A.; and Hanson, D.J. The drinking patterns and problems of a national sample of college students, 1994.  Journal of Alcohol and Drug Education  41:13–33, 1996.

Fell, J.C.; Fisher, D.A.; Voas, R.B.; et al. The impact of underage drinking laws on alcohol- related fatal crashes of young drivers.  Alcoholism: Clinical and Experimental Research  33(7):1208–1219, 2009.  PMID: 19389192

Fleming, M.F.; Balousek, S.L.; Grossberg, P.M.; et al. Brief physician advice for heavy drinking college students: A randomized controlled trial in college health clinics.  Journal of Studies on Alcohol and Drugs  71(1):23–31, 2010.  PMID: 20105410

Giles, S.M.; Champion, H.; Sutfin, E.L.; et al. Calorie restriction on drinking days: An examination of drinking consequences among college students.  Journal of American College Health  57(6):603–609, 2009.  PMID: 19433398

Gill, J.S. Reported levels of alcohol consumption and binge drinking within the UK undergraduate student population over the last 25 years.  Alcohol and Alcoholism  37(2):109–120, 2002.  PMID: 11912065

Goodwin, D.W. Alcohol amnesia.  Addiction  90(3):315–317, 1995.  PMID: 7735016

Hanson, K.L.; Medina, K.L.; Padula, C.B.; et al. Impact of adolescent alcohol and drug use on neuropsychological functioning in young adulthood: 10-year outcomes.  Journal of Child & Adolescent Substance Abuse  20(2):135–154, 2011.  PMID: 21532924

Hartzler, B., and Fromme, K. Fragmentary and en bloc blackouts: Similarity and distinction among episodes of alcohol-induced memory loss.  Journal of Studies on Alcohol  64(4):547–550, 2003 a .  PMID: 12921196

Hartzler, B., and Fromme, K. Fragmentary blackouts: Their etiology and effect on alcohol expectancies.  Alcoholism: Clinical and Experimental Research  27(4):628–637, 2003 b .  PMID: 12711925

Hingson, R.W.; Heeren, T.; Edwards, E.M.; and Saitz, R. Young adults at risk for excess alcohol consumption are often not asked or counseled about drinking alcohol.  Journal of General Internal Medicine  27(2):179–184, 2012.  PMID: 21935753

Hingson, R.; Heeren, T.; and Winter, M. Lower legal blood-alcohol limits for young drivers.  Public Health Reports  109(6):738–744, 1994.  PMID: 7800781

Hingson, R.; Heeren, T.; Winter, M.R.; and Wechsler, H. Early age of first drunkenness as a factor in college students’ unplanned and unprotected sex attributable to drinking.  Pediatrics  111(1):34–41, 2003 a .  PMID: 12509551

Hingson, R.; Heeren, T.; Winter, M.; and Wechsler, H. Magnitude of alcohol-related mortality and morbidity among U.S. college students ages 18-24: Changes from 1998 to 2001.  Annual Review of Public Health  26: 259–279, 2005 a .  PMID: 15760289

Hingson, R.; Heeren, T.; Zakocs, R., et al. Age of first intoxication, heavy drinking, driving after drinking and risk of unintentional injury among U.S. college students.  Journal of Studies on Alcohol  64(1):23–31, 2003 b .  PMID: 12608480

Hingson, R.; Heeren, T.; Zakocs, R.; et al. Magnitude of alcohol-related mortality and morbidity among U.S. college students ages 18-24.  Journal of Studies on Alcohol  63(2): 136–144, 2002.  PMID: 12033690

Hingson, R.; McGovern, T.; Howland, J.; et al. Reducing alcohol-impaired driving in Massachusetts: The Saving Lives Program.  American Journal of Public Health  86(6):791–797, 1996.  PMID: 8659651

Hingson, R., and White, A. Magnitude and prevention of college alcohol and drug misuse: U.S. college students ages 18-24. In: Kay, J., and Schwartz, V., Eds.  Mental Health Care in the College Community.  London: John Wiley & Sons, 2010, pp. 289–324.

Hingson, R.W.; Zakocs, R.C.; Heeren, T.; et al. Effects on alcohol related fatal crashes of a community based initiative to increase substance abuse treatment and reduce alcohol availability.  Injury Prevention  11(2):84–90, 2005 b .  PMID: 15805436

Hingson, R.; Zha, W.; and Weitzman, E.R. Magnitude of and trends in alcohol-related mortality and morbidity among U.S. college students ages 18-24, 1998-2005. Journal of Studies on Alcohol and Drugs (Suppl. 16):12–20, 2009.  PMID: 19538908

Holder, H.D.; Gruenewald, P.J.; Ponicki, W.R.; et al. Effect of community-based interventions on high-risk drinking and alcohol-related injuries.  JAMA: Journal of the American Medical Association  284(18):2341–2347, 2000.  PMID: 11066184

Howland, J.; Rohsenow, D.J.; Greece, J.A.; et al. The effects of binge drinking on college students’ next-day academic test-taking performance and mood state.  Addiction  105(4):655–665, 2010.  PMID: 20403018

Ichiyama, M.A.; Fairlie, A.M.; Wood, M.D.; et al. A randomized trial of a parent-based intervention on drinking behavior among incoming college freshmen.  Journal of Studies on Alcohol and Drugs  (Suppl. 16):67–76, 2009.  PMID: 19538914

Jennison, K.M. The short-term effects and unintended long-term consequences of binge drinking in college: A 10-year follow-up study.  American Journal of Drug and Alcohol Abuse  30(3):659–684, 2004.  PMID: 15540499

Johnston, L.D.; O’Malley, P.M.; and Bachman, J.G.  Monitoring the Future National Survey Results on Drug Use, 1975–2000. Volume I: Secondary School Students.  NIH Publication No. 01–4924. Bethesda, MD: National Institute on Drug Abuse, 2001 a .

Johnston L.D.; O’Malley, P.M.; and Bachman, J.G.  Monitoring the Future National Survey Results on Drug Use,1975–2000. Volume II: College Students and Adults Ages 19–40.  NIH Publication No. 01–4925. Bethesda, MD: National Institute on Drug Abuse, 2001 b .

Johnston, L.D.; O’Malley, P.M.; Bachman, J.G.; and Schulenberg, J.E.  Monitoring the Future National Survey Results on Drug Use, 1975–2006. Volume II: College Students and Adults Ages 19–45.  NIH Publication No. 07–6206. Bethesda, MD: National Institute on Drug Abuse, 2007.

Johnston, L.D.; O’Malley, P.M.; Bachman, J.G.; and Schulenberg, J.E.  Monitoring the Future National Survey Results on Drug Use, 1975–2011. Volume II: College Students and Adults Ages 19–50.  Ann Arbor: Institute for Social Research, the University of Michigan, 2012.

Jones, R., and Lacey, J.  Alcohol and Highway Safety 2001: A Review of the State of Knowledge . Report No. DOT HS–809–383. Washington, D.C.: National Highway Traffic Safety Administration, 2001.

Jones, R., and Pitt, N. Health surveys in the workplace: Comparison of postal, email and World Wide Web methods.  Occupational Medicine (London)  49(8):556–558, 1999.  PMID: 10658310

Kaysen, D.; Neighbors, C.; Martell, J.; et al. Incapacitated rape and alcohol use: A prospective analysis.  Addictive Behaviors  31(10):1820–1832, 2006.  PMID: 16446044

Kerr, W.C.; and Stockwell, T. Understanding standard drinks and drinking guidelines.  Drug and Alcohol Review  31(2):200–205, 2012.  PMID: 22050262

Knight, J.R.; Wechsler, H.; Kuo, M.; et al. Alcohol abuse and dependence among U.S. college students.  Journal of Studies on Alcohol  63(3):263–270, 2002.  PMID: 12086126

Koski, A.; Ojanperä, I.; and Vuori, E. Interaction of alcohol and drugs in fatal poisonings.  Human & Experimental Toxicology  22(5):281–287, 2003.  PMID: 12774892

Koski, A,; Vuroi, E.; and Ojanperä, I. Relation of postmortem blood alcohol and drug concentrations in fatal poisonings involving amitriptyline, propoxyphene and promazine.  Human & Experimental Toxicology  24(8):389–396, 2005.  PMID: 16138729

Kraus, C.L.; Salazar, N.C.; Mitchell, J.R.; et al. Inconsistencies between actual and estimated blood alcohol concentrations in a field study of college students: Do students really know how much they drink?  Alcoholism: Clinical and Experimental Research  29(9):1672–1676, 2005.  PMID: 16205367

Krebs, C.P.; Lindquist, C.H.; Warner, T.D.; et al. College women’s experiences with physically forced, alcohol- or drug-enabled, and drug-facilitated sexual assault before and since entering college.  Journal of American College Health  57(6):639–647, 2009.  PMID: 19433402

LaBrie, J.; Earleywine, M.; Lamb, T.; and Shelesky, K. Comparing electronic-keypad responses to paper-and-pencil questionnaires in group assessments of alcohol consumption and related attitudes.  Addictive Behaviors  31(12):2334-2338, 2006.  PMID: 16626878

LaBrie, J.W.; Huchting, K.; Tawalbeh, S.; et al. A randomized motivational enhancement prevention group reduces drinking and alcohol consequences in first-year college women.  Psychology of Addictive Behaviors  22(1):149–155, 2008.  PMID: 18298242

LaBrie, J.W.; Hummer, J.; Kenney, S.; et al. Identifying factors that increase the likelihood for alcohol-induced blackouts in the prepartying context.  Substance Use & Misuse  46(8):992–1002, 2011.  PMID: 21222521

Larimer, M.E., and Cronce, J.M. Identification, prevention and treatment: A review of individual-focused strategies to reduce problematic alcohol consumption by college students.  Journal of Studies on Alcohol  (Suppl. 14):148–163, 2002.  PMID: 12022721

Larimer, M.E., and Cronce, J.M. Identification, prevention, and treatment revisited: Individual-focused college drinking prevention strategies 1999-2006.  Addictive Behaviors  32(11):2439–2468, 2007.  PMID: 17604915

Lee, C.M.; Lewis, M.A.; and Neighbors, C. Preliminary examination of spring break alcohol use and related consequences.  Psychology of Addictive Behaviors  23(4):689–694, 2009.  PMID: 20025375

Liang, L., and Huang, J. Go out or stay in? The effects of zero tolerance laws on alcohol use and drinking and driving patterns among college students.  Health Economics  17(11):1261–1275, 2008.  PMID: 18219708

Litt, D.M.; Lewis, M.A.; Patrick, M.E.; et al. Spring break versus spring broken: Predictive utility of spring break alcohol intentions and willingness at varying levels of extremity.  Prevention Science , 2013 Feb 13. [Epub ahead of print].  PMID: 23404667

Luczak, S.E.; Shea, S.H.; Hsueh, A.C.; et al. ALDH2*2 is associated with a decreased likelihood of alcohol-induced blackouts in Asian American college students.  Journal of Studies on Alcohol  67(3):349–353, 2006.  PMID: 16608143

Lygidakis, C.; Rigon, S.; Cambiaso, S.; et al. A web-based versus paper questionnaire on alcohol and tobacco in adolescents.  Telemedicine Journal and E-Health  16(9):925– 930, 2010.  PMID: 20958200

Mallett, K.A.; Varvil-Weld, L.; Borsari, B.; et al. An update of research examining college student alcohol-related consequences: New perspectives and implications for interventions.  Alcoholism: Clinical and Experimental Research  37(5):709–716, 2013.  PMID: 23241024

McCabe, S.E.; Cranford, J.A; and Boyd, C.J. The relationship between past-year drinking behaviors and nonmedical use of prescription drugs: Prevalence of co-occurrence in a national sample.  Drug and Alcohol Dependence  84(3):281–288, 2006.  PMID: 16621337

McCartt, A.T.; Hellinga, L.A.; and Wells, J.K. Effects of a college community campaign on drinking and driving with a strong enforcement component.  Traffic Injury Prevention  10(2):141–147, 2009.  PMID: 19333826

McCauley, J.L.; Calhoun, K.S.; and Gidycz, C.A. Binge drinking and rape: A prospective examination of college women with a history of previous sexual victimization.  Journal of Interpersonal Violence  25(9):1655–1668, 2010.  PMID: 20068115

Medina, K.L.; McQueeny, T.; Nagel, B.J.; et al. Prefrontal cortex volumes in adolescents with alcohol use disorders: Unique gender effects.  Alcoholism: Clinical and Experimental Research  32(3):386–394, 2008.  PMID: 18302722

Meilman, P.W.; Leichliter, J.S.; and Presley, C.A. Greeks and athletes: Who drinks more?  Journal of American College Health  47(4):187–190, 1999.  PMID: 9919850

Meilman, P.W.; Presley, C.A.; and Cashin, J.R. The sober life at the historically black colleges.  Journal of Blacks in Higher Education  9:98–100, 1995.

Meilman, P.W.; Presley, C.A.; and Lyerla, R. Black college students and binge drinking.  Journal of Blacks in Higher Education  8:70–71, 1994.

Miller, N.S.; and Gold, M.S. Alcohol. New York: Plenum Publishing Company, 1991.

Miron, J.A., AND Tetelbaum, E. Does the minimum legal drinking age save lives?  Economic Inquiry  47(2):317–336, 2009.

Mohler-Kuo, M.; Dowdall, G.B.; Koss, M.P.; and Wechsler, H. Correlates of rape while intoxicated in a national sample of college women.  Journal of Studies on Alcohol  65(1):37–45, 2004.  PMID: 15000502

Moreira, M.T.; Smith, L.A.; and Foxcroft, D. Social norms interventions to reduce alcohol misuse in university or college students.  Cochrane Database of Systematic Reviews  (Online) (3):CD006748, 2009.  PMID: 19588402

Mundt, M.P.; and Zakletskaia, L.I. Prevention for college students who suffer alcohol-induced blackouts could deter high-cost emergency department visits.  Health Affairs (Millwood)  31(4):863–870, 2012.

Mundt, M.P.; Zakletskaia, L.I.; Brown, D.D.; and Fleming, .MF. Alcohol-induced memory blackouts as an indicator of injury risk among college drinkers.  Injury Prevention  18(1):44–49, 2012.  PMID: 21708813

Naimi, T.S.; Nelson, D.E.; and Brewer, R.D. The intensity of binge alcohol consumption among U.S. adults.  American Journal of Preventive Medicine  38(2):201–207, 2010.  PMID: 20117577

Nash, R.A., and Takarangi, M.K. Reconstructing alcohol-induced memory blackouts.  Memory  19(6):566–573, 2011.  PMID: 21919584

National Institute on Alcohol Abuse and Alcoholism (NIAAA).  Alcohol Policy Information System, 2010.  Available at  http://www.alcoholpolicy.niaaa.nih.gov/ . Accessed February 16, 2010.

NIAAA.  NIAAA Newsletter, Winter 2004,  Number 3. Bethesda, MD: National Institutes of Health, U.S. Department of Health and Human Services (DHHS), 2004. Available at:  http://pubs.niaaa.nih.gov/publications/Newsletter/winter2004/Newsletter_Number3.pdf

NIAAA.  Harmful Interactions: Mixing Alcohol with Medicines.  Bethesda, MD: National Institutes of Health, U.S. Department of Health and Human Services (DHHS), revised 2007 a  (NIH Publication No. 07-5329). Available at:  http://pubs.niaaa.nih.gov/publications/Medicine/medicine.htm

NIAAA.  Parents – Spring Break Is Another Important Time to Discuss College Drinking.  Bethesda, MD: National Institutes of Health, DHHS, 2007 b  (NIH Publication No. 05-5642).

NIAAA. Task Force of the National Advisory Council on Alcohol Abuse and Alcoholism.  A Call to Action: Changing the Culture of Drinking at U.S. Colleges.  Bethesda, MD: National Institutes of Health, U.S. Department of Health and Human Services (DHHS), 2002 (NIH Publication No. 02–5010).

NIAAA.  What Colleges Need to Know Now: An Update on College Drinking Research.  Bethesda, MD: National Institutes of Health, U.S. Department of Health and Human Services (DHHS), 2007 c  (NIH Publication No. 07–5010).

National Research Council Institute of Medicine of the National Academies.  Reducing Underage Drinking: A Collective Responsibility.  Washington, DC: The National Academies Press, 2004.

Neighbors, C.; Lee, C.M.; Lewis, M.A.; et al. Internet-based personalized feedback to reduce 21st-birthday drinking: A randomized controlled trial of an event-specific prevention intervention.  Journal of Consulting and Clinical Psychology  77(1):51–63, 2009.  PMID: 19170453

Neighbors, C.; Lee, C.M.; Atkins, D.C.; et al. A randomized controlled trial of event-specific prevention strategies for reducing problematic drinking associated with 21st birthday celebrations.  Journal of Consulting and Clinical Psychology  80(5):850–862, 2012.  PMID: 22823855

Nelson, E.C.; Heath, A.C.; Bucholz, K.K.; et al. Genetic epidemiology of alcohol-induced blackouts.  Archives of General Psychiatry  61(3):257–263, 2004.  PMID: 14993113

Norberg, K.E.; Bierut, L.J.; and Grucza, R.A. Long-term effects of minimum drinking age laws on past-year alcohol and drug use disorders.  Alcoholism: Clinical and Experimental Research  33(12):2180–2190, 2009.  PMID: 19775322

Northcote, J., and Livingston, M. Accuracy of self-reported drinking: Observational verification of ‘last occasion’ drink estimates of young adults.  Alcohol and Alcoholism  46(6):709–713, 2011.  PMID: 21949190

O’Brien, M.C.; McCoy, T.P.; Champion, H.; et al. Single question about drunkenness to detect college students at risk for injury.  Academic Emergency Medicine  13(6):629– 636, 2006.  PMID: 16614453

O’Hare, T.M. Drinking in college: Consumption patterns, problems, sex differences and legal drinking age.  Journal of Studies on Alcohol  51(6):536–541, 1990.  PMID: 2270062

Oster-Aaland, L.; Lewis, M.A.; Neighbors, C.; et al. Alcohol poisoning among college students turning 21: Do they recognize the symptoms and how do they help?  Journal of Studies on Alcohol and Drugs  (Suppl 16):122–130, 2009.  PMID: 19538920

Parada, M.; Corral, M.; Caamaño-Isorna, F.; et al. Binge drinking and declarative memory in university students.  Alcoholism: Clinical and Experimental Research  35(8):1475– 1484, 2011.  PMID: 21575014

Parada, M.; Corral, M.; Mota, N. et al. Executive functioning and alcohol binge drinking in university students.  Addictive Behaviors  37(2):167–172, 2012.  PMID: 21996093

Paschall, M.J.; Antin, T.; Ringwalt, C.L.; and Saltz, R.F. Effects of AlcoholEdu for College on alcohol-related problems among freshmen: A randomized multicampus trial.  Journal of Studies on Alcohol and Drugs  72(4):642–650, 2011 a .  PMID: 21683046

Paschall, M.J.; Antin, T.; Ringwalt, C.L.; and Saltz, R.F. Evaluation of an internet-based alcohol misuse prevention course for college freshmen: Findings of a randomized multi-campus trial.  American Journal of Preventive Medicine  41(3):300–308, 2011 b .  PMID: 21855745

Paschall, M.J., and Freisthler, B. Does heavy drinking affect academic performance in college? Findings from a prospective study of high achievers.  Journal of Studies on Alcohol  64(4):515–519, 2003.  PMID: 12921193

Patrick, M.E.; Lewis, M.A.; Lee, C.M.; and Maggs, J.L. Semester and event-specific motives for alcohol use during Spring Break: Associated protective strategies and negative consequences.  Addictive Behaviors  38(4):1980–1987, 2013.  PMID: 23384451

Perry, P.J.; Argo, T.R.; Barnett, M.J.; et al. The association of alcohol-induced blackouts and grayouts to blood alcohol concentrations.  Journal of Forensic Sciences  51(4):896– 899, 2006.  PMID: 16882236

Powell, L.A.; Williams, J.; and Wechsler, H. Study habits and the level of alcohol use among college students.  Education Economics  12(2):135–149, 2004.

Presley, C.A.; Leichliter, J.S.; and Meilman, P.W.  Alcohol and Drugs on American College Campuses: A Report to College Presidents. Third in a Series: 1995, 1996, and 1997 , Carbondale, IL: Core Institute, Southern Illinois University, 1998.

Presley, C.A.; Meilman, P.W.; and Cashin, J.R.  Alcohol and Drugs on American College Campuses: Use, Consequences, and Perceptions of the Campus Environment. Volume IV: 1992-94.  Carbondale, IL: The Core Institute, 1996 a .

Presley, C.A.; Meilman, P.W.; Cashin, J.R.; and Lyera, R.  Alcohol and Drugs on American College Campuses: Use, Consequences, and Perceptions of the Campus Environment. Volume III: 1991-93.  Carbondale, IL: The Core Institute, 1996 b .

Presley, C.A.; Meilman, P.W., and Leichliter, J.S. College factors that influence drinking.  Journal of Studies on Alcohol  (Suppl 14):82–90, 2002.  PMID: 12022732

Presley, C.A., and Pimentel, E.R. The introduction of the heavy and frequent drinker: A proposed classification to increase accuracy of alcohol assessments in postsecondary educational settings.  Journal of Studies on Alcohol  67(2):324–331, 2006.  PMID: 16562416

Preusser, D.; Ulmer, R.; and Preisser, C.  Obstacles to Enforcement of Youthful (Under 21) Impaired Driving.  Washington, DC: National Highway Traffic Safety Administration, 1992 (DOT HS 807–878).

Read, J.P.; Beattie, M.; Chamberlain, R.; and Merrill, J.E. Beyond the “Binge” threshold: Heavy drinking patterns and their association with alcohol involvement indices in college students.  Addictive Behaviors  33(2):225–234, 2008.  PMID: 17997047

Reboussin, B.A.; Song, E.Y.; and Wolfson, M. The impact of alcohol outlet density on the geographic clustering of underage drinking behaviors within census tracts.  Alcoholism: Clinical and Experimental Research  35(8):1541–1549, 2011.  PMID: 21463343

Rutledge, P.C.; Park, A.; and Sher, K.J. 21st birthday drinking: Extremely extreme.  Journal of Consulting and Clinical Psychology  76(3):511–516, 2008.  PMID: 18540744

Saltz, R.F.; Welker, L.R.; Paschall, M.J.; et al. Evaluating a comprehensive campus-community prevention intervention to reduce alcohol-related problems in a college population.  Journal of Studies on Alcohol and Drugs  (Suppl. 16):21–27, 2009.  PMID: 19538909

Schaus, J.F.; Sole, M.L.; McCoy, T.P.; et al. Alcohol screening and brief intervention in a college student health center: A randomized controlled trial.  Journal of Studies on Alcohol and Drugs  (Suppl. 16):131–141, 2009.  PMID: 19538921

Schulenberg, J.; Maggs, J.L.; Long, S.W.; et al. The problem of college drinking: Insights from a developmental perspective.  Alcoholism: Clinical and Experimental Research  25(3):473–477, 2001.  PMID: 11290861

Schweinsburg, A.D.; McQueeny, T.; Nagel, B. J.; et al. A preliminary study of functional magnetic resonance imaging response during verbal encoding among adolescent binge drinkers.  Alcohol  44(1):111–117, 2010.  PMID: 20113879

Scribner, R.A.; Mason, K.E.; Simonsen, N.R.; et al. An ecological analysis of alcohol outlet density and campus-reported violence at 32 U.S. colleges.  Journal of Studies on Alcohol and Drugs  71(2):184–191, 2010.  PMID: 20230715

Scribner, R.; Mason, K.; Theall, K.; et al. The contextual role of alcohol outlet density in college drinking.  Journal of Studies on Alcohol and Drugs  69(1):112–120, 2008.  PMID: 18080071

Scribner, R.A.; Theall, K.P.; Mason, K.; et al. Alcohol prevention on college campuses: The moderating effect of the alcohol environment on the effectiveness of social norms marketing campaigns.  Journal of Studies on Alcohol and Drugs  72(2):232-239, 2011.  PMID: 21388596

Shults, R.A.; Elder, R.W.; Sleet, D.A.; et al. Reviews of evidence regarding interventions to reduce alcohol-impaired driving.  American Journal of Preventive Medicine  21(Suppl. 4):66-88, 2001.  PMID: 11691562

Singleton, R.A. Collegiate alcohol consumption and academic performance.  Journal of Studies on Alcohol and Drugs  68(4):548–555, 2007.  PMID: 17568960

Singleton, R.A. Jr., and Wolfson, A.R. Alcohol consumption, sleep, and academic performance among college students.  Journal of Studies on Alcohol and Drugs  70(3):355– 363, 2009.  PMID: 19371486

Smith, G.S.; Branas, C.C.; and Miller, T.R. Fatal nontraffic injuries involving alcohol: A metaanalysis.  Annals of Emergency Medicine  33(6):659–668, 1999.  PMID: 10339681

Squeglia, L.M.; Spadoni, A.D.; Infante, M.A.; et al. Initiating moderate to heavy alcohol use predicts changes in neuropsychological functioning for adolescent girls and boys.  Psychology of Addictive Behaviors  23(4):715–722, 2009.  PMID: 20025379

Squeglia, L.M.; Sorg, S.F.; Schweinsburg, A.D.; et al. Binge drinking differentially affects adolescent male and female brain morphometry.  Psychopharmacology (Berlin)  220(3):529–539, 2012.  PMID: 21952669

Straus, R., and Bacon, S. D. (1953).  Drinking in college.  New Haven, CT: Yale University Press.

Substance Abuse and Mental Health Services Administration (SAMHSA).  The DAWN Report: Trends in Emergency Department Visits Involving Underage Alcohol Use: 2005 to 2009.  Rockville, MD, Substance Abuse and Mental Health Services Administration, 2011.

Substance Abuse and Mental Health Services Administration (SAMHSA).  Results from the 2001 National Household Survey on Drug Abuse: Volume 1: Summary of National Findings.  Rockville, MD: Substance Abuse and Mental Health Services Administration, 2002 (DHHS Publication No. SMA 02–3758, 2002).

Substance Abuse and Mental Health Services Administration (SAMHSA).  Results from the 2005 National Survey on Drug Use and Health: National Findings.  Rockville, MD: Substance Abuse and Mental Health Services Administration, 2006 (DHHS Publication No. SMA 06–4194).

Substance Abuse and Mental Health Services Administration (SAMHSA).  Results from the 2007 National Survey on Drug Use and Health: National Findings.  Rockville, MD: Substance Abuse and Mental Health Services Administration, 2008 (NSDUH Series H-34, DHHS Publication No. SMA 08–4343).

Substance Abuse and Mental Health Services Administration (SAMHSA).  Summary of Findings of the 1999 National Household Survey on Drug Abuse.  Rockville, MD: Substance Abuse and Mental Health Services Administration, 2000 (DHHS Publication No. SMA 00–3466).

Tanaka, E. Toxicological interactions between alcohol and benzodiazepines.  Journal of Toxicology. Clinical Toxicology  40(1):69–75, 2002.  PMID: 11990206

Testa, M., and Livingston, J.A. Alcohol consumption and women’s vulnerability to sexual victimization: Can reducing women’s drinking prevent rape?  Substance Use & Misuse  44(9-10):1349–1376, 2009.  PMID: 19938922

Thombs, D.L.; Olds, R.S.; Bondy, S.J.; et al. Undergraduate drinking and academic performance: A prospective investigation with objective measures.  Journal of Studies on Alcohol and Drugs  70(5):776–785, 2009.  PMID: 19737503

Timberlake, D.S.; Hopfer,C.J.; Rhee, S.H.; et al. College attendance and its effect on drinking behaviors in a longitudinal study of adolescents.  Alcoholism: Clinical and Experimental Research  31(6):1020–1030, 2007.  PMID: 17403064

Treno, A.J.; Gruenewald P.J.; Lee, J.P.; and Remer, L.G. The Sacramento Neighborhood Alcohol Prevention Project: Outcomes from a community prevention trial.  Journal of Studies on Alcohol and Drugs  68(2):197–207, 2007.  PMID: 17286338

Turner, C.F.; Ku, L.; Rogers, S.M.; et al. Adolescent sexual behavior, drug use, and violence: Increased reporting with computer survey technology.  Science  280(5365):867– 873, 1998.  PMID: 9572724

Turrisi, R.; Larimer, M.E.; Mallett, K.A.; et al. A randomized clinical trial evaluating a combined alcohol intervention for high-risk college students.  Journal of Studies on Alcohol and Drugs  70(4):555–567, 2009.  PMID: 19515296

Upcraft, M.L.  Today’s First-Year Students and Alcohol.  Paper prepared for the Task Force on College Drinking, National Advisory Council on Alcohol Abuse and Alcoholism, Bethesda, MD, 2000.

Varvil-Weld, L.; Mallett, K.A.; Turrisi, R.; et al. Are certain college students prone to experiencing excessive alcohol-related consequences? Predicting membership in a high-risk subgroup using pre-college profiles.  Journal of Studies on Alcohol and Drugs  74(4):542–551, 2013.  PMID: 23739017

Voas, R.B., and Williams, A.F. Age differences of arrested and crash-involved drinking drivers.  Journal of Studies on Alcohol  47(3):244–248, 1986.  PMID: 3724162

Voas, R.B.; Tippetts, A.S.; and Fell, J. The relationship of alcohol safety laws to drinking drivers in fatal crashes.  Accident Analysis and Prevention  32(4):483–492, 2000.  PMID: 10868751

Wagenaar, A.C., and Toomey, T.L. Effects of minimum drinking age laws: Review and analyses of the literature from 1960 to 2000.  Journal of Studies on Alcohol  (Suppl. 14):206–225, 2002.  PMID: 12022726

Wagenaar, A.C.; Murray, D.M.; and Toomey T.L. Communities Mobilizing for Change on Alcohol (CMCA): Effects of a randomized trial on arrests and traffic crashes.  Addiction  95(2):209–217, 2000.  PMID: 10723849

Wagenaar, A.C.; O’Malley, P.M.; and LaFond, C. Lowered legal blood alcohol limits for young drivers: Effects on drinking, driving, and driving-after-drinking behaviors in 30 states.  American Journal of Public Health  91(5):801–804, 2001.  PMID: 11344892

Wagenaar, A.C.; Salois, M.J.; and Komro, K.A. Effects of beverage alcohol price and tax levels on drinking: A meta-analysis of 1003 estimates from 112 studies.  Addiction  104(2):179–190, 2009.  PMID: 19149811

Wagenaar, A.C.; Erickson, D.J.; Harwood, E.M.; and O’Malley, P.M. Effects of state coalitions to reduce underage drinking: A national evaluation.  American Journal of Preventive Medicine  31(4):307–315, 2006.  PMID: 16979455

Wang, Y.C.; Lee, C.M; Lew-Ting, C.Y.; et al. Survey of substance use among high school students in Taipei: Web-based questionnaire versus paper-and-pencil questionnaire.  Journal of Adolescent Health  37(4):289–295, 2005.  PMID: 16182139

Wechsler, H.; Davenport, A.; Dowdall, G.; et al. Health and behavioral consequences of binge drinking in college: A national survey of students at 140 campuses.  JAMA: Journal of the American Medical Association  272(21):1672–1677, 1994.  PMID: 7966895

Wechsler, H.; Davenport, A.E.; Dowdall, G.W.; et al. Binge drinking, tobacco, and illicit drug use and involvement in college athletics. A survey of students at 140 American colleges.  Journal of American College Health  45(5):195–200, 1997.  PMID: 9069676

Wechsler, H.; Dowdall, G.W.; Davenport, A.; and Castillo, S. Correlates of college student binge drinking.  American Journal of Public Health  85(7):921–926, 1995.  PMID: 7604914

Wechsler, H.; Dowdall, G.W.; Davenport, A.; and Rimm, E.B. A gender-specific measure of binge drinking among college students.  American Journal of Public Health  85(7):982– 985, 1995.  PMID: 7604925

Wechsler, H.; Dowdall, G.W.; Maenner, G.; et al. Changes in binge drinking and related problems among American college students between 1993 and 1997. Results of the Harvard School of Public Health College Alcohol Study.  Journal of American College Health  47(2):57–68, 1998.  PMID: 9782661

Wechsler, H., and Kuo, M. Watering down the drinks: The moderating effect of college demographics on alcohol use of high-risk groups.  American Journal of Public Health,  93(11):1929–1933, 2003.  PMID: 14600068

Wechsler, H.; Kuh, G.; and Davenport, A.E. Fraternities, sororities and binge drinking: Results from a national study of American colleges.  NASPA Journal  46(3):395–416, 2009.

Wechsler, H.; Lee, J.E.; Kuo, M., et al. Trends in college binge drinking during a period of increased prevention efforts. Findings from 4 Harvard School of Public Health College Alcohol Study surveys: 1993-2001.  Journal of American College Health  50(5):203– 217, 2002.  PMID: 11990979

Wechsler, H.; Lee, J.E.; Kuo, M.; and Lee, H. College binge drinking in the 1990s: A continuing problem. Results of the Harvard School of Public Health 1999 College Alcohol Survey.  Journal of American College Health  48(5):199–210, 2000.  PMID: 10778020

Wechsler, H., and Nelson, T.F. What we have learned from the Harvard School of Public Health College Alcohol Study: Focusing attention on college student alcohol consumption and the environmental conditions that promote it.  Journal of Studies on Alcohol and Drugs  69(4): 481–490, 2008.  PMID: 18612562

Wechsler, H., and Nelson, T. F. Binge drinking and the American college students: What’s five drinks?  Psychology of Addictive Behaviors,  15(4), 287–291, 2001.  PMID: 11767258

Weitzman, E.R.; Folkman, A; Folkman, M.P.; and Wechsler, H. The relationship of alcohol outlet density to heavy and frequent drinking and drinking-related problems among college students at eight universities.  Health & Place  9(1):1–6, 2003.  PMID: 12609468

Wetherill, R.R., and Fromme, K. Acute alcohol effects on narrative recall and contextual memory: An examination of fragmentary blackouts.  Addictive  Behaviors 36(8):886– 889, 2011.  PMID: 21497445

Wetherill, R.R.; Schnyer, D.M.; and Fromme, K. Acute alcohol effects on contextual memory BOLD response: Differences based on fragmentary blackout history.  Alcoholism: Clinical and Experimental Research  36(6):1108–1115, 2012.  PMID: 22420742

White, A.M. What happened? Alcohol, memory blackouts, and the brain.  Alcohol Research & Health  27(2):186-196, 2003.  PMID: 15303630

White, A.M.; Hingson, R.W.; Pan, I.J.; and Yi, H.Y. Hospitalizations for alcohol and drug overdoses in young adults ages 18-24 in the United States, 1999-2008: Results from the Nationwide Inpatient Sample.  Journal of Studies on Alcohol and Drugs  72(5):774– 786, 2011.  PMID: 21906505

White, A.M.; Jamieson-Drake, D.W.; and Swartzwelder, H.S. Prevalence and correlates of alcohol–induced blackouts among college students: Results of an e–mail survey.  Journal of American College Health  51(3):117–119; 122-131, 2002.  PMID: 12638993

White, A.M.; Kraus, C.L.; Flom, J.D.; et al. College students lack knowledge of standard drink volumes: Implications for definitions of risky drinking based on survey data.  Alcoholism: Clinical and Experimental Research  29(4):631–638, 2005.  PMID: 15834229

White, A.M.; Kraus, C.L.; McCracken, L.A.; and Swartzwelder, H.S. Do college students drink more than they think? Use of a free-pour paradigm to determine how college students define standard drinks.  Alcoholism: Clinical and Experimental Research  27(11):1750–1756, 2003.  PMID: 14634490

White, A.M.; Kraus, C.L.; and Swartzwelder, H. Many college freshmen drink at levels far beyond the binge threshold.  Alcoholism: Clinical and Experimental Research  30(6):1006-1010, 2006.  PMID: 16737459

White, A.M.; Signer, M.L.; Kraus, C.L.; and Swartzwelder, H.S. Experiential aspects of alcohol–induced blackouts among college students.  American Journal of Drug and Alcohol Abuse  30(1):205–224, 2004.  PMID: 15083562

White, A.M., and Swartzwelder, H.S. Inbound college students drink heavily during the summer before their freshman year: Implications for education and prevention efforts.  American Journal of Health Education  40:909–96, 2009.

Williams, J.; Powell, L.M.; and Weschler, H. Does alcohol consumption reduce human capital accumulation? Evidence from the College Alcohol Study.  Applied Economics  35(10):1227–1245, 2003.

Wolfson, M.; Champion, H.; McCoy, T.P.; et al. Impact of a randomized campus/community trial to prevent high-risk drinking among college students.  Alcoholism: Clinical and Experimental Research  36(10):1767–1778, 2012.  PMID: 22823091

Wood, P.K.; Sher, K.J.; Erickson, D.J.; and DeBord, K.A. Predicting academic problems in college from freshman alcohol involvement.  Journal of Studies on Alcohol  58(2):200– 210, 1997.  PMID: 9065898

Wood, M.D.; Sher, K.J.; and McGowan, A.K. Collegiate alcohol involvement and role attainment in early adulthood: Findings from a prospective high-risk study.  Journal of Studies on Alcohol  61(2):278–289, 2000.  PMID: 10757139

World Health Organization (WHO).  Alcohol and Injuries: Emergency Department Studies in an International Perspective.  Geneva: World Health Organization, 2009.

Zagrosek, A.; Messroghli, D.; Schulz, O.; et al. Effect of binge drinking on the heart as assessed by cardiac magnetic resonance imaging.  JAMA: Journal of the American Medical Association  304(12):1328–1330, 2010.  PMID: 20858877

• Get new issue alerts Get alerts

Secondary Logo

Journal logo.

Colleague's E-mail is Invalid

Your message has been successfully sent to your colleague.

Save my selection

Alcohol consumption and awareness of its effects on health among secondary school students in Nigeria

Editor(s): Kufa., Tendesayi

a Department of Home Economics and Hospitality Management Education, Faculty of Vocational and Technical Education, University of Nigeria

b Department of Food Science and Technology, Ebonyi State University Abakaliki, Ebonyi State

c Department of Educational Foundations, Faculty of Education, University of Nigeria, Nsukka, Enugu State

d Department of Home Economics Education, Ebonyi State University Abakaliki, Ebonyi State

e Department of Human Kinetics and Health Education, Faculty of Education, University of Nigeria, Nsukka, Enugu State, Nigeria.

Correspondence: Amaka Bibian Ezeanwu, Department of Home Economics and Hospitality Management Education, Faculty of Vocational and Technical Education, University of Nigeria, Nsukka, P.M.B. 410001, Enugu State, Nigeria (e-mail: [email protected] ).

Abbreviations: % = Percentage, ADCQSSS = Alcoholic Drinks Consumption Questionnaire for Secondary School Students, CI = confidence interval, DNC = does not consume, EA = extremely aware, HC = highly consume, M ± SD = means and standard deviation, MA = moderately aware, MC = moderately consume, NA = not aware, RC = rarely consume, SA = slightly aware, t = t test statistic.

Authorship: NME, CE, UCU, and HAN conceived the study. NME, BNA, HAN, CE, JIO, and BAE designed the study procedure. All the authors were involved in the data collection. CE, BNA, UCU, NME, JIO, and BAE carried out the analysis and interpretation of these data. NME, CE, UCU, HAN, BNA, JIO, and BAE drafted the manuscript. CE, HAN, BNA, NME, JIO, UCU, and BAE critically revised the manuscript for intellectual content. All authors read and approved the final manuscript.

The authors report no conflicts of interest.

This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms. http://creativecommons.org/licenses/by-nc-sa/4.0

Received June 23, 2017

Received in revised form November 4, 2017

Accepted November 8, 2017

Alcohol consumption among secondary school students is a major public health issue worldwide; however, the extent of consumption among secondary school students and their understanding of its effects on human health remain relatively unknown in many Nigerian States. This study aimed to determine the extent of alcohol consumption and of the awareness of its negative effects on human health among secondary school students.

The study used a cross-sectional survey design. Self-report questionnaire developed by the researchers was administered to representative sample (N = 1302) of secondary school students in the study area. The data collected from the respondents were analyzed using means and t test.

The results showed that male secondary school students moderately consumed beer (55.2%) and local cocktails (51.5%), whereas their female counterparts reported rare consumption of these 2 alcoholic drinks (44.8%; 48.5% respectively). The findings also indicated rare consumption of distilled spirits among both male and female students in the investigated area, whereas wine, liquor, local spirits, and palm wine were consumed moderately, regardless of gender. Finally, male and female secondary school students differed significantly in their awareness of the negative effects of alcohol consumption on health.

There is a need to intensify efforts to further curtail the extent of alcohol consumption and increase awareness of the negative effects of alcohol use on human health among secondary school students.

1 Introduction

Alcohol consumption is a serious public health challenge worldwide, including in Nigeria. Although the level of alcohol consumption differs widely around the world, the burden of disease and death remains significant in most regions, with Europe and America having the highest alcohol attributable fractions at 6.5% and 5.6%, respectively. [1] Recent evidence also indicates that alcohol consumption is now the world's third largest risk factor for disease and disability; almost 4% of all deaths globally are attributed to alcohol. [2] However, alcohol is the most commonly used psychoactive drug in both young people and adults in Nigeria. [3–5] Some of the factors contributing to alcohol consumption among Nigerians include the absence of alcohol policies, easy access to alcoholic drinks, and lack of implementation of a minimum drinking age by both the government and the brewers. [6]

According to Bada and Adebiyi, [7] it is not rare for Nigerian secondary school students to consume alcoholic drinks; this consumption could be due to their curiosity as adolescents, an irresistible urge, emotional disturbances such as anxiety, the subculture, and the influence of advertisements. Several previous studies have shown the prevalence of alcohol consumption among the Nigerian population, but they did not explore adolescent students’ understanding of its negative health effects. For instance, Lasebikan and Ola [8] found that the prevalence of lifetime alcohol use was 57.9% and that of current alcohol use was 27.3% among a sample of Nigerian semirural community dwellers. Through in-person interviews with Nigerian adults, previous research by Gureje et al [3] revealed that the lifetime prevalence of alcohol consumption was 56%. A recent study by Alex-Hart et al [9] showed that the prevalence of current alcohol consumption among a sample of Nigerian secondary school students was 30.6% and that 38.1% of current drinkers had also been drunk in the past 30 days, with 17.2% being drunk very frequently.

Alcohol consumption negatively affects human health across the lifespan. Previous studies show that alcohol consumption is associated with a burden of diseases such as cancer, [10] pancreatitis, liver cirrhosis, tuberculosis, pneumonia, diabetes mellitus, alcohol use disorder, malignancies, psychiatric morbidity, and injury. [11] Although 18 years of age is the legal limit for alcohol consumption per policy in many parts of the world, sociocultural influences [12,13] seem to hinder strict adherence to this public health policy in Nigerian society. The objective of the present study was therefore to investigate the level of alcohol consumption and knowledge of its negative effects on health among secondary school students in Nigeria. Specifically, the study sought to determine the responses of secondary school students regarding the extent of their alcohol consumption and the extent to which students are aware of the negative health effects of alcohol consumption.

2.1 Ethical consideration

Ethical committee approval was obtained by the authors for this study (Ethical Approval Number: VTE/ERA/0023). Furthermore, parents of the selected participants signed an informed consent form to indicate their approval. School principals of the selected students provided informed consent and conveyed their approval to the researchers in writing. Participants were informed that they were free to participate or to decline participation in the study.

2.2 Study design

The current study adopted a cross-sectional survey design.

2.3 Study setting

This study was conducted in public secondary schools in Ebonyi and Enugu States, Nigeria.

2.4 Study participants

The participants comprised 1302 senior secondary school students who were purposively selected to participate in the study. The study sample size was determined using G*Power 3.1 software [14,15] based on a statistical power of 0.95. Brown [16] stated that if the observed statistical power is large enough (≥0.80), the sample size can be considered adequate for the study. Figure 1 shows the results of the sample size determination. Table 1 summarizes the characteristics of the participants.

2.5 Assumptions about the sample size calculations

We conducted an a priori analysis to determine the study sample size based on the assumptions that a required sample size can be computed as a function of user-specified values for the required significance level α , the desired statistical power 1– β , and the to-be-detected population effect size. [14,15,17,] According to Uzoagulu, [18] researchers should endeavor to make use of statistical technique to determine sample size, and should be aware that the fewer a sample size is, the greater the possibility of sampling error. Thus, the assumptions underlying the use of a priori analysis for sample size calculations were considered appropriate for the current study.

2.6 Sampling strategy

Before sampling, the researchers and assistants purposively visited 40 secondary schools each in the surveyed States to seek for the school principals’ approval, and to explain to them the purpose of the study, and possibility of being included or excluded from the study later on due to certain criteria. All the school principals visited gave their informed consent, and their schools were therefore qualified for sampling. Through multistage sampling technique, the researchers selected the current study sample. First, the simple random sampling technique (balloting without replacement) was used to select only 31 school secondary schools from each State, making a total of 62 secondary schools surveyed. This technique was used in order to give each of the secondary schools the opportunity of being selected and thus eliminate selection bias. Furthermore, 21 senior students from each of the selected schools in the 2 States were selected to participate in the study through stratified random sampling. The samples were stratified by gender (male, n = 651, 50%; female, n = 651, 50%) and other demographics as summarized in Table 1 . Both the schools and their students were selected on the basis of certain inclusion criteria set by the researchers.

2.7 Inclusion and exclusion criteria

The inclusion criteria included that the school principal must provide informed consent in writing; the respondent (student) must be in senior secondary school class two or three, agree to participate freely, inform their parents/guardians about the study, and provide a letter of consent from them. A participant must also be at least 16 years of age and above. Those who did not meet these criteria were excluded from the current study.

2.8 Measures

The Alcoholic Drinks Consumption Questionnaire for Secondary School Students (ADCQSSS) is a structured questionnaire developed by the researchers based on previous literature. [2,8,19] The ADCQSSS consists of 22 items divided into 2 major sections (A and B). Section A assesses respondents’ personal data (age, gender, religion, socioeconomic background, and educational class level). Section B has 2 parts; part one contains 7 items that evaluate the extent to which students consume alcohol with regard to a variety of alcoholic drinks (i.e., Beer, Distilled Spirits, Wine, Liquor, Local Spirits, Local Cocktails, and Palm Wine), and part two contains 15 items that ask respondents about their awareness of the negative effects of alcoholic drinks on human health. The ADCQHSS has a 5-point rating scale from Do not consume/Not aware (0) to Highly Consume/Extremely aware (4). The ADCQSSS was validated by 2 experts in Home Economics and Hospitality Management Education and 2 other independent experts in Educational Research, Measurement, and Evaluation. Cronbach alpha reliability coefficient of the ADCQSSS was 0.72 for part one, 0.76 for part two, and 0.85 for the entire scale, based on data from the current study sample.

2.9 Data collection

To overcome the challenges of participant attrition and nonretrieval of instruments, which are common to many cross-sectional surveys, the questionnaires were distributed and retrieved from each respondent on the spot with the help of 4 research assistants. Respondents met with the researchers and assistants in school halls to complete the questionnaire during long break periods in school. The respondents were guided appropriately and given sufficient time (15–20 minutes) to avoid incomplete responses. Respondents were encouraged to call the attention of any of the researchers or assistants if they need additional clarification on any item or how to complete the questionnaire. Given these measure, responses and return rates were 100%.

2.10 Data entry, management procedure, and analysis

The data collected from the respondents were analyzed using means, percentage, and t test. Item scores were included as the dependent variables and sex as the independent variables. Using purposively determined mean benchmark values, the item scores for the first part of the ADCQHSS section B (the extent to which students consumed alcoholic drinks) were interpreted as follows: Highly Consume (HC) = 3.50 to 4.00; Moderately Consume (MC) = 3.00 to 3.49; Rarely Consume (RC) = 2.50 to 2.99; and Does Not Consume (DNC) = 1.00 to 2.49. Using similarly purposively set mean benchmark values, the item scores for the second part of the ADCQHSS section B (awareness of the negative effects of alcoholic drinks on health) were interpreted as follows: Extremely Aware (EA) = 3.50 to 4.00; Moderately Aware (MA) = 3.00 to 3.49; Slightly Aware (SA) = 2.50 to 2.99; and Not Aware (NA) = 1.00 to 2.49. The t test was used to examine the differences between male and female students at a 0.05 level of significance. To perform the t tests, item scores were treated as test variables, whereas sex was used as the grouping variable. During coding, the numerical value of 1 was used as the label for male students, whereas the value of 2 was applied for female students. Before performing the t tests, the normality of the distribution of the data was assessed using Shapiro–Wilks normality test. The data were found to be normally distributed ( P = .95). Furthermore, we described the percentage (%) of students scoring below or above given thresholds ( see Tables 2 and 3 ). A database created from Microsoft Excel was used for data management, which involved compiling, organizing, defining, and managing data. Thereafter, the statistical software used for analysis was the Statistical Package for the Social Sciences (SPSS) version 21 (IBM Corp., Chicago, IL). [20] To assure quality, we checked for missing data and violation of assumptions using the IBM SPSS statistical software. There were no missing data.

The results in Table 2 reveal the extent of participants’ consumption of alcoholic drinks by gender. Male secondary school students (55.2%) in this study reported moderate consumption of beer (M ± SD = 3.17 ± .83), while females (44.8%) reported rare consumption of beer (2.57 ± .57). In addition, distilled spirits (male=50.3%; female=49.7%) were rarely consumed by both genders, whereas wine, liquors, local spirits, and palm wine were moderately consumed by both (see Table 2 ). These results imply that secondary school students in the study area consumed different types of alcoholic beverages. In addition, the results in Table 2 reveal a significant difference between male and female students in their extent of consumption of beer [ t = -15.25, P = .000, 95% confidence interval (95% CI) -0.676 to 0.522] and local cocktails ( t = -3.92, P = .000, 95% CI -0.263 to -0.088) at 1300 degrees of freedom, as the corresponding P values were lower than the chosen level of significance (.05). This finding means that male and female students in the investigated area consumed these alcoholic drinks unequally due to gender. Male secondary school students moderately consumed beer and local cocktails, whereas their female counterparts were rare consumers of these 2 alcoholic drinks (see Table 2 ).

Furthermore, the results in Table 2 reveal nonsignificant differences between male and female students in the extent to which they consume distilled spirits ( t = -0.991, P = .322, 95% CI -0.109 to 0.036], wine ( t = -0.268, P = .789, 95% CI -0.089 to 0.068), liquor ( t = -0.846, P = .398, 95% CI -0.138 to 0.055), local spirits ( t = -0.098, P = .922, 95% CI -0.097 to 0.088), and palm wine ( t = -0.066, P = .947, 95% CI -0.094 to 0.088), given that the P values for all t tests ranged from .32 to .95 at 1300 degrees of freedom and were therefore higher than the chosen level of significance (.05). This finding suggests that male and female students in the investigated area consumed these alcoholic drinks similarly regardless of gender (see Table 2 ).

Table 3 summarizes the extent to which students were aware of the negative effects of alcohol consumption on health. The results show that male and female students differed significantly in their awareness of the following negative effects of alcohol consumption on human health: excessive drinking can cause alcoholic hepatitis ( t = 39.39, P = .000, 95% CI 1.347–1.488); heavy alcohol intake increases the risk of many forms of cancers ( t = 12.92, P = .000, 95% CI 0.526–0.715); excessive alcohol intake can result in sleep disturbances ( t = -5.01, P = .000, 95% CI -0.329 to 0.144); alcohol abuse can increase the risk of injuries and accidents ( t = -2.65, P = .008, 95% CI -0.193 to 0.029); alcohol abuse can cause liver disease ( t = 3.38, P = .001, 95% CI 0.059–0.221); alcohol abuse can damage the salivary glands ( t = -4.14, P = .000, 95% CI -0.215–0.077); alcohol abuse can lead to gum disease and tooth decay ( t = 6.23, P = .000, 95% CI 0.201–0.386); people who abuse alcohol suffer from malnutrition ( t = 32.98, P = .000, 95% CI 0.972–1.095); excessive alcohol intake can cause a woman to stop menstruating and become infertile ( t = 6.44, P = .000, 95% CI 0.179–0.337); an immune system weakened by alcohol abuse has difficulty fighting off illness ( t = -7.66, P = .000, 95% CI -0.382 to 0.226); and heavy drinking can cause damage to the heart ( t = -15.46, P = .000, 95% CI -0.748 to 0.579) (see Table 3 ).

Finally, the results in Table 3 further show that male and female students were similarly aware of the following effects of alcohol on human health, in that no significant differences were found: excessive alcohol intake can affect coordination, interfering with balance and the ability to walk ( t = 0.70, P = .484, 95% CI -0.055 to 0.117); heavy alcohol use can result in alcohol dependence ( t = 0.00, P = .998, 95% CI -0.088 to 0.088); alcohol use can make people with depression feel worse ( t = -0.86, P = .388, 95% CI -0.121 to 0.047); and erectile dysfunction is a side effect of alcohol abuse in men ( t = 1.66, P = .097, 95% CI -0.009 to 0.104) (see Table 3 ).

4 Discussion

The current study determined the extent of alcoholic drink consumption and of awareness of its negative effects on human health among secondary school students in Nigeria. First, our findings showed that male secondary school students moderately consume beer and local cocktails, whereas their female counterparts were rare consumers of these 2 alcoholic drinks. These findings support those of Lasebikan and Ola, [8] who found that current alcohol drinking was highly related to male gender. In addition, the current study showed that both male and female students in the investigated area rarely consumed distilled spirits, whereas wine, liquor, local spirits, and palm wine were consumed moderately by the students regardless of their gender. According to the authors, more than two-thirds of the current drinking population are moderate drinkers. [8] Bada and Adebiyi [7] noted that it is not difficult to identify Nigerian secondary school students who consume alcohol. Despite what literature might suggest as reasons for students’ alcohol consumption, Cox et al [21] indicate that negative reasons for alcohol consumption are stronger determinants of drinking problems than are positive reasons among both secondary-level school students. Sociocultural factors may also explain the reason for the differences as well as similarities on the extent of alcoholic drink consumption in Nigerian population. [12,13]

Furthermore, the present study showed that male and female secondary school students differed significantly in their awareness of the following negative effects of alcohol consumption on health: excessive drinking can cause alcoholic hepatitis; heavy alcohol intake increases the risk of many forms of cancers; excessive alcohol intake can result in sleep disturbances; alcohol abuse can increase the risk of injuries and accidents; alcohol abuse can cause liver disease; alcohol abuse can damage the salivary glands; alcohol abuse can lead to gum disease and tooth decay; people who abuse alcohol suffer from malnutrition; excessive alcohol intake can cause a woman to stop menstruating and become infertile; an immune system weakened by alcohol abuse has difficulty fighting off illness; and heavy drinking can cause damage to your heart. Finally, our findings also showed that male and female students had similar levels of awareness of the following effects of alcohol on human health, as no significant differences were found: excessive alcohol intake can affect coordination, interfering with balance and the ability to walk; heavy alcohol use can result in alcohol dependence; alcohol use can make people with depression feel worse; and erectile dysfunction is a side effect of alcohol abuse in men. These outcomes support previous studies, [1,9] which show that alcohol increases the risk of numerous diseases and all injury outcomes. According to the WHO, [2] alcohol consumption is now the world's third largest risk factor for disease and disability, with almost 4% of all global deaths attributed to alcohol.

5 Limitations

Our study has several limitations. First, this study used a cross-sectional survey design, which does not enable conclusions regarding causality. However, a cross-sectional survey was considered necessary because previous studies did not investigate the extent of alcohol consumption or of awareness of its negative effects on human health among secondary school students. In the future, longitudinal studies are needed to determine the causal relationship between alcoholic drink consumption and awareness of its negative effects on students’ health. Second, all participants were recruited from secondary schools in 2 States in Nigeria. This approach may have limited the ability to generalize these findings to other populations. Third, our study used self-reported assessments. The instrument showed good validity; however, future research may need to use observational assessments and interview.

6 Implications

If no further research or action is implemented to determine the extent of alcohol consumption and awareness of its negative effects on human health among secondary school students in Nigeria, in other parts of the country in particular, then alcohol consumption among adolescent students may be associated with increased school-based violence, student neglect and abuse, and absenteeism in school, among other social issues. In addition, policy interventions and other actions to reduce the patterns of alcohol use among the student population may not be realistic. Therefore, further research is needed to examine the patterns and prevalence of alcohol consumption and of awareness of its negative effects on human health among secondary school students across the globe.

7 Conclusion

The present study suggests that male secondary school students moderately consume beer and local cocktails, whereas their female counterparts are slight consumers of these 2 alcoholic drinks. Furthermore, both male and female students in the investigated area slightly consumed distilled spirits, whereas wine, liquor, local spirits, and palm wine were consumed moderately by students regardless of gender. Finally, male and female secondary school students significantly differed in their awareness of some of the negative effects of alcohol consumption on health. Overall, secondary school students are not very aware of some of the negative effects of alcohol on human health. Accordingly, health education teachers, school health counselors, and school administrators should combine their professional experiences to promote health education interventions and health counseling programs aimed at reducing students’ engagement in alcohol consumption. Schools should organize seminars for students to provide education on the health-related issues surrounding alcohol consumption. Addiction counselors should also organize awareness campaigns to orient Nigerian secondary school students to the damages caused by alcohol consumption. Parents should properly monitor and counsel their adolescent students on matters relating to alcohol consumption and its effect on their health.

Acknowledgments

We would like to thank the editor and anonymous reviewers for their constructive remarks regarding this work. We also are thankful to AuthorAID and American Journal Experts (AJEs) for their editing support. We are very much grateful to the research assistants and all the schooling adolescents who made this study a success.

alcohol; consumption; health; Nigeria; secondary school; students

• + Favorites
• View in Gallery

The Truth About the Health Benefits of Alcohol

New research challenges long-held beliefs about the perks of drinking..

Posted August 10, 2024 | Reviewed by Margaret Foley

• What Is Alcoholism?
• Find counselling to overcome addiction
• Scientists and clinicians have long debated about the health benefits conferred by alcohol use.
• A worldwide study sheds light on the relationship between drinking and chronic diseases.
• The findings challenge the notion of health benefits of drinking for those under 40.

Although it has been widely believed for decades that there are health benefits linked with moderate amounts of alcohol consumption, not everyone who drinks experiences health benefits from it. According to the largest study evaluating the relationship between alcohol use and chronic disease, which included over 1 billion people across the world, the safety and potential benefits of drinking alcohol depend largely on your age. Specifically, this groundbreaking study found that there are no health benefits of drinking among those who are under the age of 40 . This younger group is not only susceptible to the health risks and harms related to alcohol use (including motor vehicle accidents, injuries, and suicides related to drinking), but alcohol consumption did not benefit their health or reduce their risk of chronic diseases.

What about those who are over the age of 40?

• People who are over 40 may see health benefits from alcohol use, including reduced risk of cardiovascular disease, stroke, and diabetes.
• Reduced risks of these chronic diseases were observed among people without underlying health conditions.
• These benefits are linked with the use of a limited quantity of alcohol (i.e., no more than one to two standa rd drinks per day).

What should people who drink keep in mind to stay in the “healthy” zone?

• The potential health benefits of drinking small to moderate quantities of alcohol for certain people not only vanish if a person who is drinking moderately transitions to heavy drinking, but they are replaced with various potential health risks.
• Health risks for those who drink heavily can include high blood pressure, liver disease, and the development of addiction .
• The limited quantities of alcohol use that may be linked to health benefits for certain people over 40 are “daily” guidelines and do not apply cumulatively (i.e., “saving up” four days’ worth of daily drinks and having them all on one day is considered heavy drinking—with potential health risks rather than benefits).

If you or someone you know drinks alcohol, it is important to bring intention to the quantity that you consume, with knowledge of what is considered heavy or risky alcohol use.

How much is too much?

• Consuming more than three drinks in one day for women, or more than four for men, is considered heavy drinking.
• Problematic drinking is defined not only by the quantity of alcohol a person consumes, but also by one’s ability to control their use of alcohol.
• When a person is losing control over their drinking, they may frequently drink more than they planned to; make rules for themselves about how much or how often they will drink but have trouble keeping them; experience withdrawal symptoms when the effects of alcohol wear off; have problems in important relationships or in meeting responsibilities because of their drinking; experience cravings; and/or find themselves continuing to drink despite some of these problems.

If you or someone you care about has trouble controlling their alcohol use, consult a professional. I cover this topic in more depth in my book, Addiction: What Everyone Needs to Know , and on my podcast .

GBD 2020 Alcohol Collaborators (2022). Population-level risks of alcohol consumption by amount, geography, age, sex, and year: a systematic analysis for the Global Burden of Disease Study 2020. Lancet (London, England) , 400 (10347), 185–235. https://doi.org/10.1016/S0140-6736(22)00847-9

Suzette Glasner, Ph.D., is a licensed clinical psychologist and an Associate Professor in the Department of Psychiatry and Biobehavioral Sciences at UCLA.

• Find a Therapist
• Find a Treatment Center
• Find a Psychiatrist
• Find a Support Group
• Find Online Therapy
• International
• New Zealand
• South Africa
• Switzerland
• Asperger's
• Bipolar Disorder
• Chronic Pain
• Eating Disorders
• Passive Aggression
• Personality
• Goal Setting
• Positive Psychology
• Stopping Smoking
• Low Sexual Desire
• Relationships
• Child Development
• Self Tests NEW
• Therapy Center
• Diagnosis Dictionary
• Types of Therapy

Sticking up for yourself is no easy task. But there are concrete skills you can use to hone your assertiveness and advocate for yourself.

• Emotional Intelligence
• Gaslighting
• Affective Forecasting
• Neuroscience

A .gov website belongs to an official government organization in the United States.

A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

• Alcohol Use and Your Health
• Preventing Alcohol-Related Harms
• Underage Drinking
• Data on Excessive Alcohol Use
• U.S. Deaths from Excessive Alcohol Use
• Publications
• About Surveys on Alcohol Use
• About Standard Drink Sizes
• CDC Alcohol Program
• Alcohol Outlet Density Measurement Tools
• Resources to Prevent Excessive Alcohol Use
• Online Alcohol Tools and Apps
• Funding to Prevent Excessive Alcohol Use

Related Topics:

• View All Home
• Alcohol-Related Disease Impact (ARDI) Application
• Check Your Drinking. Make a Plan to Drink Less.
• Controle su forma de beber. Haga un plan para beber menos.
• Addressing Excessive Alcohol Use: State Fact Sheets

At a glance

Excessive alcohol use might be more common than you think. These data show how much and how often people binge drink in the United States, and its high costs to our nation.

Alcohol use, by the numbers

Among adults in the United States: 1

• More than half drink alcohol.
• 17% binge drink. This means they have four or more drinks (women) or five or more drinks (men) on an occasion.
• Nearly all of the adults who drink heavily also binge drink.

Binge drinking, by the numbers

Of the four ways that people drink excessively, binge drinking is the most common.

• Over 90% of U.S. adults who drink excessively report binge drinking. 2
• But people who binge drink are at higher risk for serious health effects from alcohol compared to people who do not binge drink.

Below we report on binge drinking statistics across the United States:

• The percentage of people who binge drink.
• The number of drinks they have on an occasion.
• How often people binge drink.

Percentage of adults who binge drink:

Among people who binge drink, this is how many drinks they have on an occasion: 3, how often adults report binge drinking in the past 30 days: 3, frequency of adult binge drinking by state, 2022.

* Number of binge drinking occasions in the past 30 days among adults who reported binge drinking. †Among adults who binge drink, 25% of them report binge drinking this many times or more in the past month. § Half of adults who binge drink report doing so this many times or more in the past month.

The cost of excessive drinking in the United States

Excessive drinking cost the United States about $249 billion in 2010 (the most recent data available). A 4 This includes:

• Lost labor and lower worker performance in the workplace (72%)
• Property damage, crashes, and criminal justice needs (17%)
• Health care costs for injuries (11%)

Added to these costs is the toll on people's health, quality of life, safety, and well-being.

The cost impacts everyone, whether they drink or not.

On average, the direct and indirect costs of excessive drinking add up to about $807 per person in the United States.

Every alcoholic drink consumed creates an extra $2.05 in economic costs to address alcohol-related impacts.

State-level costs

The economic costs of excessive drinking also weigh on individual states across the country. 4

• The median cost to states (including Washington DC) was $3.5 billion in 2010.
• Costs ranged from $488 million in North Dakota to $35 billion in California.
• Costs were mostly the result of binge drinking (77%).

Governments paid for about $2 of every $5 spent to address the impacts of excessive alcohol use. B 4

Keep in mind‎

Data table: costs of excessive alcohol use by state.

• These costs have likely increased because of factors like inflation and increased health care and public safety costs. However, this cost estimate still provides an idea of the minimum costs of excessive drinking.
• This includes local, state, and federal governments.
• Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System. Accessed February 29, 2024. https://www.cdc.gov/brfss/
• Esser MB, Hedden SL, Kanny D, Brewer RD, Gfroerer JC, Naimi TS. Prevalence of alcohol dependence among US adult drinkers, 2009–2011 . Prev Chronic Dis. 2014;11:140329.
• Centers for Disease Control and Prevention. Chronic Disease Indicators. Behavioral Risk Factor Surveillance System, United States, 2022. Accessed March 14, 2024. https://www.cdc.gov/cdi/index.html
• Sacks JJ, Gonzales KR, Bouchery EE, Tomedi LE, Brewer RD. 2010 National and State Costs of Excessive Alcohol Consumption . Am J Prev Med. 2015;49(5):e73–e79.
• Esser MB, Sherk A, Liu Y, Naimi TS. Deaths from excessive alcohol use — United States, 2016-2021. MMWR Morb Mortal Wkly Rep . 2024;73:154–161. doi: http://dx.doi.org/10.15585/mmwr.mm7308a1

Alcohol Use

Excessive alcohol use can harm people who drink and those around them. You and your community can take steps to improve everyone’s health and quality of life.

For Everyone

Public health.

• All topics »
• Fact sheets
• Feature stories
• Publications
• Questions & answers
• Tools and toolkits
• Coronavirus disease (COVID-19) pandemic
• Ukraine emergency
• Environment and health

• Calls for experts
• Initiatives
• European Programme of Work
• Sustainable Development Goals
• The Pan-European Mental Health Coalition
• Empowerment through Digital Health
• The European Immunization Agenda 2030
• Healthier behaviours: incorporating behavioural and cultural insights
• Moving towards UHC
• Protecting against health emergencies
• Promoting health and well-being
• News stories
• Media releases
• Photo stories
• Questions and answers
• Media Contacts

Newsletters

• European Health Information Gateway
• European health report
• Core health indicators
• WHO Immunization Data portal
• Noncommunicable diseases (NCD) dashboard 
• Events 
• Teams »
• Data and digital health
• Policy & Governance f. Health through the Life Course
• Groups and networks »
• Health Evidence Network (HEN)

The European Health Report 2021 »

• Conflict in Israel and the occupied Palestinian territory
• Armenian refugee health response
• Climate crisis: extreme weather
• Türkiye and Syria earthquakes
• About health emergencies
• Health emergencies newsletter 
• Health emergencies list

• Regional Director
• Executive Council
• Technical centres
• Faces of WHO
• Regional Committee for Europe
• Standing Committee
• Partners 
• Groups and networks
• WHO collaborating centres

74th session of the WHO Regional Committee for Europe

No level of alcohol consumption is safe for our health

The risks and harms associated with drinking alcohol have been systematically evaluated over the years and are well documented. The World Health Organization has now published a statement in The Lancet Public Health: when it comes to alcohol consumption, there is no safe amount that does not affect health.

It is the alcohol that causes harm, not the beverage

Alcohol is a toxic, psychoactive, and dependence-producing substance and has been classified as a Group 1 carcinogen by the International Agency for Research on Cancer decades ago – this is the highest risk group, which also includes asbestos, radiation and tobacco. Alcohol causes at least seven types of cancer, including the most common cancer types, such as bowel cancer and female breast cancer. Ethanol (alcohol) causes cancer through biological mechanisms as the compound breaks down in the body, which means that any beverage containing alcohol, regardless of its price and quality, poses a risk of developing cancer.

The risk of developing cancer increases substantially the more alcohol is consumed. However, latest available data indicate that half of all alcohol-attributable cancers in the WHO European Region are caused by “light” and “moderate” alcohol consumption – less than 1.5 litres of wine or less than 3.5 litres of beer or less than 450 millilitres of spirits per week. This drinking pattern is responsible for the majority of alcohol-attributable breast cancers in women, with the highest burden observed in countries of the European Union (EU). In the EU, cancer is the leading cause of death – with a steadily increasing incidence rate – and the majority of all alcohol-attributable deaths are due to different types of cancers.

Risks start from the first drop

To identify a “safe” level of alcohol consumption, valid scientific evidence would need to demonstrate that at and below a certain level, there is no risk of illness or injury associated with alcohol consumption. The new WHO statement clarifies: currently available evidence cannot indicate the existence of a threshold at which the carcinogenic effects of alcohol “switch on” and start to manifest in the human body.

Moreover, there are no studies that would demonstrate that the potential beneficial effects of light and moderate drinking on cardiovascular diseases and type 2 diabetes outweigh the cancer risk associated with these same levels of alcohol consumption for individual consumers.

“We cannot talk about a so-called safe level of alcohol use. It doesn’t matter how much you drink – the risk to the drinker’s health starts from the first drop of any alcoholic beverage. The only thing that we can say for sure is that the more you drink, the more harmful it is – or, in other words, the less you drink, the safer it is,” explains Dr Carina Ferreira-Borges, acting Unit Lead for Noncommunicable Disease Management and Regional Advisor for Alcohol and Illicit Drugs in the WHO Regional Office for Europe. 

Despite this, the question of beneficial effects of alcohol has been a contentious issue in research for years.

“Potential protective effects of alcohol consumption, suggested by some studies, are tightly connected with the comparison groups chosen and the statistical methods used, and may not consider other relevant factors”, clarifies Dr Jürgen Rehm, member of the WHO Regional Director for Europe’s Advisory Council for Noncommunicable Diseases and Senior Scientist at the Institute for Mental Health Policy Research and the Campbell Family Mental Health Research Institute at the Centre for Addiction and Mental Health, Toronto, Canada.

We are missing the bigger picture

Globally, the WHO European Region has the highest alcohol consumption level and the highest proportion of drinkers in the population. Here, over 200 million people in the Region are at risk of developing alcohol-attributable cancer.

Disadvantaged and vulnerable populations have higher rates of alcohol-related death and hospitalization, as harms from a given amount and pattern of drinking are higher for poorer drinkers and their families than for richer drinkers in any given society.

“So, when we talk about possible so-called safer levels of alcohol consumption or about its protective effects, we are ignoring the bigger picture of alcohol harm in our Region and the world. Although it is well established that alcohol can cause cancer, this fact is still not widely known to the public in most countries. We need cancer-related health information messages on labels of alcoholic beverages, following the example of tobacco products; we need empowered and trained health professionals who would feel comfortable to inform their patients about alcohol and cancer risk; and we need overall wide awareness of this topic in countries and communities,” adds Dr Ferreira-Borges.

The Lancet Public Health: Health and cancer risks associated with low levels of alcohol consumption

The Lancet: Alcohol and health

Alcohol and cancer in the WHO European Region: an appeal for better prevention

Turning down the alcohol flow. Background document on the European framework for action on alcohol, 2022-2025

72nd session Regional Committee for Europe: European regional action framework for behavioural and cultural insights for health, 2022-2027

Making the WHO European Region SAFER: developments in alcohol control policies, 2010-2019

Log in using your username and password

• Search More Search for this keyword Advanced search
• Latest content
• For authors
• Browse by collection
• BMJ Journals

You are here

• Volume 12, Issue 2
• Consumption and effects of caffeinated energy drinks in young people: an overview of systematic reviews and secondary analysis of UK data to inform policy
• Article Text
• Article info
• Citation Tools
• Rapid Responses
• Article metrics

• http://orcid.org/0000-0002-9571-3147 Claire Khouja 1 ,
• http://orcid.org/0000-0002-7016-978X Dylan Kneale 2 ,
• Ginny Brunton 3 ,
• Gary Raine 1 ,
• Claire Stansfield 2 ,
• Amanda Sowden 1 ,
• Katy Sutcliffe 2 ,
• James Thomas 2
• 1 Centre for Reviews and Dissemination , University of York , York , UK
• 2 EPPI-Centre, Social Science Research Unit , UCL Institute of Education, University College London , London , UK
• 3 Faculty of Health Sciences , Ontario Tech University , Oshawa , Ontario , Canada
• Correspondence to Claire Khouja; claire.khouja{at}york.ac.uk

Background This overview and analysis of UK datasets was commissioned by the UK government to address concerns about children’s consumption of caffeinated energy drinks and their effects on health and behaviour.

Methods We searched nine databases for systematic reviews, published between 2013 and July 2021, in English, assessing caffeinated energy drink consumption by people under 18 years old (children). Two reviewers rated or checked risk of bias using AMSTAR2, and extracted and synthesised findings. We searched the UK Data Service for country-representative datasets, reporting children’s energy-drink consumption, and conducted bivariate or latent class analyses.

Results For the overview, we included 15 systematic reviews; six reported drinking prevalence and 14 reported associations between drinking and health or behaviour. AMSTAR2 ratings were low or critically low. Worldwide, across reviews, from 13% to 67% of children had consumed energy drinks in the past year. Only two of the 74 studies in the reviews were UK-based. For the dataset analysis, we identified and included five UK cross-sectional datasets, and found that 3% to 32% of children, across UK countries, consumed energy drinks weekly, with no difference by ethnicity. Frequent drinking (5 or more days per week) was associated with low psychological, physical, educational and overall well-being. Evidence from reviews and datasets suggested that boys drank more than girls, and drinking was associated with more headaches, sleep problems, alcohol use, smoking, irritability, and school exclusion. GRADE (Grading of Recommendations, Assessment, Development and Evaluation) assessment suggests that the evidence is weak.

Conclusions Weak evidence suggests that up to a third of children in the UK consume caffeinated energy drinks weekly; and drinking 5 or more days per week is associated with some health and behaviour problems. Most of the evidence is from surveys, making it impossible to distinguish cause from effect. Randomised controlled trials are unlikely to be ethical; longitudinal studies could provide stronger evidence.

PROSPERO registrations CRD42018096292 – no deviations. CRD42018110498 – one deviation - a latent class analysis was conducted.

• nutrition & dietetics
• epidemiology
• public health
• community child health

Data availability statement

Data are available upon reasonable request. All the data in the overview are publicly available, but not necessarily without charge. Those for the dataset analysis are available from the UK Data Service.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ .

https://doi.org/10.1136/bmjopen-2020-047746

Statistics from Altmetric.com

Request permissions.

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

The main strength of this study was the novel use of a secondary data analysis to fill a gap in the evidence that was identified by the overview.

A strength of the overview was its robust methods, and that it only included reviews that used systematic methods.

A limitation of the overview was the strength of evidence of the primary research, most of which was from cross-sectional surveys.

The main limitations of the dataset analysis were that longitudinal data were not available, and the survey data could not be combined due to differences between surveys in their designs and measures reported.

Introduction

Caffeinated energy drinks (CEDs) are drinks containing caffeine, among other ingredients, that are marketed as boosting energy, reducing tiredness, and improving concentration. They include brands such as Red Bull, Monster Energy, and Rockstar. There is widespread concern about their consumption and effects in children and adolescents (under 18 years old). 1–4 Some professional organisations have suggested banning sales to children. 2 In the UK, warnings, aimed at children and pregnant women, are required on the packaging for drinks that contain over 150 mg/L of caffeine. 5 An average 250 mL energy drink contains a similar amount of caffeine to a 60 mL espresso, and the European Food Safety Authority proposes a safe level of 3 mg of caffeine per kg of body weight per day for children and adolescents. 6 Many drinks also contain other potentially active ingredients, such as guarana and taurine, and more sugar than other soft drinks, although there are sugar-free options. 7–9 Children may be more at risk of ill effects than adults. 10 11 Effects could be physical (eg, headaches), psychological (eg, anxiety) or behavioural (eg, school attendance or alcohol consumption). 12 Available systematic reviews report a wide range of findings, including positive effects on sports performance.

In 2018, the UK government ran a consultation on implementing a ban on sales to children, 13–15 and in March 2019 they published a policy paper. 16 The research reported here was commissioned by the Department of Health and Social Care (DHSC), England, in 2018, to identify and assess the evidence on the use of CEDs by children. As the deadline was short, and as initial searches identified several systematic reviews, a systematic review of systematic reviews (referred to as overview, from this point onwards) was conducted. As only two UK studies were identified within the reviews included in the overview, UK datasets were sought, and a secondary analysis of relevant data was carried out to supplement the international literature and ensure relevance to UK policy. Full reports are available. 17 18

The research questions (RQ) were:

RQ1. What is the nature and extent of CED consumption among people aged 17 years or under in the UK?

RQ2. What impact do CEDs have on young people’s physical and mental health, and behaviour?

This paper summarises the overview and dataset analysis. 17 18 For the overview, a literature search was conducted during May 2018 and updated on 2 July 2021. EPPI-Reviewer software 19 was used to manage the data. The gaps, identified by the overview and a search for primary studies, guided the search, conducted during August 2018, for UK datasets and their subsequent analysis. STATA v13 20 was used to analyse the datasets. Ethical approval was granted by UCL’s Ethics Committee. Protocols were registered on PROSPERO (CRD42018096292 and CRD42018110498).

Search strategies

For the overview, we searched nine databases, focusing on research in health, psychology, science or social science, or general research. We completed forward citation searching in Google Scholar for 13 included reviews. The databases searched and the MEDLINE search strategy are in the online supplemental file (section 1). The search terms were based on three concepts: caffeine, energy drink, and systematic review. The searches were limited to the publication year of 2013 onwards, to identify the most recent systematic reviews. For the dataset analysis, search terms were based on caffeine and energy drink. We searched the UK Data Service 21 (accessing over 6000 UK nation population datasets), with no restrictions.

Supplemental material

Inclusion criteria.

For the overview:

Systematic review published since 2013

Extractable data on children under 18 years of age

Available in English

Patterns of CED use or associations with physical, mental, social or behavioural effects.

Four reviewers (GB, CK, GR and CS) screened references based on their titles and abstracts, and then screened potential includes on their full texts. The four reviewers double-screened batches of 10 references until their decisions to include or exclude each paper were the same on at least nine of the 10 (90%), then they screened individually. Disagreements and indecisions were resolved by another of the four reviewers, where necessary.

For the dataset analysis:

Downloadable datasets, representative of the UK or a constituent country

Information on the levels and patterns of CED consumption

Data on children under 18 years of age (adults could provide the data on their behalf)

Reporting primary (frequency, amount, or occurrence of drinking/not drinking (comparator)) or secondary (sugar consumption, cardiovascular health, mental health, neurological conditions, educational outcomes, substance misuse, sports performance or sleep characteristics) measures.

After a pilot batch, for which two reviewers (GB and DK) assessed datasets independently and discussed their decisions to include or exclude, the remaining datasets were screened, independently.

Data extraction

From the systematic review reports that met the overview inclusion criteria, we extracted details on/for: systematic review methods; included studies; CED consumption; associations with physical, mental, social or behavioural effects; and risk of bias assessment. One reviewer (GB, CK, GR or CS) extracted these data, which were checked by another reviewer. For the dataset analysis, one reviewer (GB or DK) extracted dataset characteristics (sample size, etc); details on participants (age, gender, etc) and consumption (how it was measured, etc); well-being and health outcomes, including potential confounders; and information on missing data and for risk of bias assessment.

The data extracted from the systematic reviews were synthesised in a narrative format due to variation between reviews. Prevalence was synthesised by the measure used, where possible. Associations were synthesised by whether they were physical, mental, behavioural, or social/educational, and summary tables were produced. One reviewer (GB, CK, GR or CS) synthesised the data and another checked each synthesis.

Each dataset was analysed for prevalence and frequency of CED consumption, and any variations by children’s characteristics. Most of the cross-sectional analyses were bivariate (exploring interactions between two features), with binary and multinomial logistic regression used to control for confounders. A latent class analysis (LCA) was conducted, 22 for one dataset. The latent profiles were based on children’s health experiences, such as headaches, anxiety, or dizziness. The observed variables (11 indicators of child well-being) and latent variables (five classes of well-being) were identified from the data. Class membership was used as the dependent variable in multinomial logistic regressions. Descriptive associations were explored in bivariate analyses of the 11 indicators, separately. The results from individual datasets were synthesised in a narrative because meta-analysis was not deemed to be appropriate. Missing data were not imputed, as it was not possible to determine if they were missing at random. One reviewer (DK) analysed the data.

Risk of bias

AMSTAR2 23 was used to assess the risk of bias in the included systematic reviews, because some reviews included randomised controlled trials (RCTs) as well as non-RCTs. AMSTAR2 has questions on the protocol, inclusion criteria, search, selection, data extraction, risk of bias assessment, reporting, synthesis (RCTs and non-RCTs), and conflicts of interest; a question on relevance was added. The strength of the evidence was assessed using GRADE (Grading of Recommendations, Assessment, Development and Evaluation) criteria, 24 which can be used to determine whether the evidence is strong or weak, based on any risk of bias, including in study design and size, consistency of the results, relevance to the population, and potential publication bias. Overlap, where the same primary studies appear in more than one review, was assessed. 25 Overlap can lead to double counting of the results of a study, giving these more influence than those of other studies. 26 Two reviewers (CK and GR) assessed risk of bias; random samples were checked by a third reviewer (GB). Datasets were not formally assessed, but all datasets met the quality assurance criteria of the UK Data Service. 27 Data on exposure (quantity, frequency and type of drink), sample frame (characteristics of participants), and level of participation (response rate) were extracted, by one reviewer (DK), to determine their parameters. 17 In line with National Institutes of Health guidance, 28 no overall risk of bias score was produced for each dataset because overall scores can be misleading where the risk of bias on each criterion has a different impact on the reliability of the conclusions.

Patient and public involvement

We did not include young people in the research process.

The overview searches identified 1102 references, after deduplication (see figure 1 ); 126 were screened on full texts. We included 15 reviews; six reported information on prevalence, 12 29–33 and 14 reported associations. 12 29–32 34–42 The reasons for exclusion, based on assessment of the full text, are reported in the online supplemental file (section 2). Most were excluded because they did not use systematic review methods or did not report information on children.

• Download figure
• Open in new tab
• Download powerpoint

Flow diagram for the overview. CED, caffeinated energy drinks; T&A, title and abstract.

Three reviews focused on CEDs in children. 12 30 41 One 35 focused on children, with a section on CEDs alongside other drinks. The other 11 reported information on children alongside data for adults; one 29 with CEDs alongside other drinks, and two 31 32 focusing on alcohol mixed with CEDs. For summary and full characteristics, see the online supplemental file (section 3) and the full report. 18

For the dataset analysis, as there was no facility to export results, it was not possible to record the flow of datasets through screening. Five datasets met the inclusion criteria; analyses were not possible for one dataset 43 (see table 1 ). For full descriptions, see the full report. 17

• View inline

Description of the five datasets included in the secondary data analysis

There was a high risk of bias in all but three of the reviews—Visram et al , 12 and Bull et al 37 Yasuma et al 41 (details in the online supplemental file , section 4)—meaning that some relevant evidence may have been missed. Overlap between studies in the reviews was slight (corrected covered area 3.2%; see the online supplemental file , section 5). The reviews did not include any analyses of the UK datasets that we analysed. Within the reviews, there were four small randomised controlled trials, while most studies were surveys with a high risk of bias; the application of GRADE criteria, which are used to assess the overall strength of the evidence found, suggests that the evidence is weak. Exposure, sample frame and level of participation for the datasets are reported in appendix 1 of the full report. 17

UK studies in the overview

Of the 74 studies identified by the reviews that are summarised in the overview, two were UK surveys. One 44–46 was a longitudinal (two time-points) cross-sectional survey of 11- to 17-year-olds in the south-west of England. The other 47 was a survey of 13- to 18-year-olds across 22 European countries, one of which was the UK (2.6% of respondents).

Below and in tables 2–4 , the overview results are summarised by research question, followed by highlights of the dataset analysis within each topic. The full results of the overview 18 and dataset analysis 17 are available online.

Characteristics and main findings of reviews reporting prevalence of consumption

Prevalence of CED consumption across datasets by school year (approximately weekly consumption with weighted percentages and unweighted sample sizes - see notes below)

Characteristics and main findings of the reviews reporting associations with consumption

RQ1. Nature and extent of CED consumption

The overview included six reviews with data on prevalence of children’s CED consumption, and these are summarised in table 2 .

Across reviews, prevalence varied by study location, population age range, and definition of drinking (ever drunk, in the past year, regularly, with alcohol, etc) from 13% to 67% of children having a CED in the past year. 30 32 One meta-analysis 29 of four studies in the Gulf states suggested that about two thirds of children consumed CEDs (not further defined; 65.3%, 95% CI 41.6 to 102.3 (as reported in the paper)). Across reviews, weekly or monthly drinking ranged from 13% to 54% 48 of children. In one study, across Europe, UK children had the highest proportion of caffeine intake from CEDs, at 11%, 47 but this might reflect a lower intake from coffee or tea. Across reviews, 10% 49 to 46% 50 of children had tried CEDs with alcohol.

In the UK dataset analysis, self-reported prevalence was relatively consistent across UK countries (see table 3 ), although there were differences in the questions asked. About a quarter of children aged 13 to 14 years consumed one drink or more per week (Smoking and Drinking Survey of Young People (SDSYP) data). 51 Prevalence ranged from 3% to 32% of children—slightly lower than found in the overview.

Characteristics of drinkers

In the overview, more boys reported drinking CEDs than girls. 12 29–32 Prevalence by age was inconsistent: for example, within the reviews, one study 48 found that girls started drinking CEDs when they were younger; while one 52 suggested that drinking prevalence peaked at 14 to 15 years; and another 53 suggested that more older boys drank CEDs than younger boys, but more younger girls drank them than older girls. Prevalence by ethnicity was also inconsistent. Children with minority ethnicity drank more than white children, 12 32 but white children drank more than black or Hispanic children, when drinks were mixed with alcohol. 12 In the UK, drinking was associated with being male, older and lower socioeconomic status. 45

In the dataset analysis, the SDSYP reported the most detailed information on sociodemographic characteristics. As in most of the overview evidence, prevalence increased with age, so that between a quarter and a third of children aged 15 to 16 years reported consuming one or more CED per week. More boys (29.3%) than girls (18.1%), and more children living in the North of England than in the South (for example, 33.1% in the North-East vs 16.5% in the South-East), consumed at least one can a week. More children who were eligible for free school meals (29.5%), than those who were not eligible (22.6%), drank CEDs weekly. These differences were robust to the impact of potential confounders (see the online supplemental file , section 6). Unlike the evidence from the overview, which suggested differences in consumption by ethnicity, the proportion of weekly CED consumers was within 3 percentage points of the average across all ethnic groups.

Motives and context

Three reviews reported on motives or context for consumption. 12 29 32 The context was parties and socialising with friends or family 12 32 35 or exams. 29 Children’s motives included taste (particularly with alcohol), for energy, curiosity, friends drinking them, and parental approval or disapproval. Across the reviews, single studies suggested that more girls than boys drank CEDs to suppress appetite, 54 while more boys than girls drank them for performance in sport. 55 And about half of children knew that the drinks contained caffeine, 56 while those who knew that the content might be harmful drank less. 57

Motives and context were not measured in the UK datasets.

RQ2. Associations with drinking CEDs

Fourteen reviews reported associations and are summarised in table 4 . Most reviews included cross-sectional evidence (surveys) or individual case studies. Three reviews 12 40 42 reported prospective trials (four small RCTs in total), which assessed physical performance, cardiovascular response, or the effects of sleep education; one review reported prospective cohort studies.

As most of the evidence was from surveys, measured at a single time-point, cause cannot be distinguished from effect.

Physical health associations

Associations between drinking CEDs and physical symptoms were reported in all but one 40 of the 14 reviews. CEDs improved sports performance. 58 59 There was consistent evidence of associations with headaches, stomach aches and low appetite, 12 35 42 and with sleep problems. 12 30 35 42 Within the reviews, a trial of boys randomised to receive different doses of CED reported dose-dependent increases in diastolic blood pressure and decreases in heart rate. 60 Across reviews, 34 36–39 nine cases of adverse events were reported; eight children had cardiovascular events, and one had renal failure, following a single drink, moderate drinking, or excessive drinking (in a day or for weeks).

Analysis of the Health Behaviour in School Children (HBSC) 2013/14 data found that children drinking CEDs once a week or more, compared with those drinking less often, were statistically significantly more likely to report physical symptoms occurring more than once a week, such as headaches (22.2% vs 16.8%), sleep problems (13.6% vs 8.5%) and stomach problems (31.2% vs 23.1%).

Mental health associations

Associations between drinking CEDs and mental health were inconsistent. 12 29 30 32 35 40 42 One review reported that improvements in mental health and hyperactivity were found in children who were randomised to receive an intervention to lower their intake of CEDs. 61 Associations were found with stress, anxiety or depression, 12 30 35 40 42 but two reviews 12 40 also found studies that did not find an association. Some reviews included evidence of associations with self-harm or suicidal behaviour, 30 35 40 42 and with irritation and anger. 12 30 35 40 42

Secondary analyses of the HBSC 2013/14 data found that children who consumed CEDs at least once a week were statistically significantly more likely, than those who did not, to report low mood (20.3% vs 14.9%) and irritability (30.8% vs 18.0%) on a weekly basis.

Behavioural associations

Some evidence of associations between drinking CEDs and behaviour was reported. 12 30–32 35 42 Drinking CEDs was associated with alcohol, smoking and substance misuse at a single time point, 12 30 35 and at follow-up. 41 CED consumption at baseline predicted alcohol consumption at follow-up. 12 Consumption was associated with increased hyperactivity and inattention, and with sensation seeking. 12 30 35 Injuries were associated with drinking CEDs with alcohol 12 31 and without alcohol. 12 30

Analysis of the SDSYP data found that higher proportions of children who consumed one or more cans per week had tried alcohol (59.1%) and smoking (39.7%), compared with non-CED consumers (alcohol 28.9%, smoking 10.4%).

Social or educational associations

Consistent associations between drinking CEDs and social or educational outcomes were reported. 12 32 Within reviews, one UK study 45 found an association between drinking CEDs once a week or more and poor school attendance. CEDs mixed with alcohol were associated with lower grades and more absence from school. 32

Analysis of the SDSYP data found that almost half of children who had been truant or excluded reported drinking a can of CED on a weekly basis (49.5%), compared with less than a fifth of those who had not been truant or excluded (18.5%).

Well-being profiles

Using the HBSC 2013/14 dataset, we identified 11 indicators of well-being: weekly experience of irritability, sleep difficulties, nervousness, dizziness, headaches, stomach aches, and low mood; as well as low life satisfaction, feeling pressured by schoolwork some or a lot of the time, dislike of school, and low self-rated academic achievement. From these, using LCA, we identified five profiles: low psychological well-being (18.2% of children), high overall well-being (48.6%), low educational well-being (6.7% of children), low physical well-being (13.0%), and low overall well-being (13.5%). See the online supplemental file (section 6) for details.

After controlling for age, gender, rurality, smoking status, alcohol status and Family Affluence Scale (a measure of socioeconomic status; for more information see Hartley et al 62 ), the relative risk of having a low well-being profile, compared with a high well-being profile, was substantially higher for children who consumed CEDs at least 5 days a week (frequent), compared with those who rarely or never did. Relative to a high well-being profile, frequent consumers had a higher risk of low psychological well-being (RR 2.11, 95% CI 1.56 to 2.85) and low physical well-being (RR 2.52, 95% CI 1.76 to 3.61), and were over four times more likely to have low educational well-being (RR 4.81, 95% CI 3.59 to 6.44) and low overall well-being (RR 4.15, 95% CI 2.85 to 6.00). These data suggest that CED consumption is a marker of low well-being, but the analyses also showed that consumption was one of a cluster of factors (eg, smoking and drinking alcohol) in children with low well-being.

Summary of the evidence

Prevalence varied according to the measures used and the ages of children. In the overview, CED consumption prevalence was up to 67% of children in the past year and, in the dataset analyses, up to 32% of children were consuming a CED at least 1 day a week, meaning that up to a third of UK children are regularly consuming caffeine. Evidence from the overview and the dataset analyses consistently suggests that boys drink more than girls, and that drinking tends to increase with age. Some evidence from the overview suggested higher prevalence in children from ethnic minority backgrounds, but no such association was detected in the UK data analysis. This could be due to factors such as area of residence or social class affecting well-being in children from ethnic minorities, where well-being is driving the differences in prevalence of CED consumption, rather than minority background. Reviews included in the overview found that most drinking of CEDs occurred at parties, around exams, with friends, or with family, and motives included taste, energy, curiosity, appetite suppression, and sports performance, which was reported to be improved. There was some evidence that knowledge of content was low, and that children who knew that the content might be harmful drank less, suggesting that education could reduce drinking.

Evidence from the overview suggests worse sleep, and raised blood pressure, with CED consumption, compared with reduced or no consumption. Both the overview and the dataset analysis found that children who consumed CEDs reported headaches, stomach aches and sleep issues more frequently than those who did not; although most studies were cross-sectional, some in the overview were longitudinal, showing changes over time. 18 The overview identified consistent evidence of associations with self-harm, suicide behaviour, alcohol use*, smoking*, substance misuse*, hyperactivity, irritation*, anger, and school performance, attendance, and exclusion (*also found in the UK dataset analysis). This was consistent with findings reported in non-systematic reviews. 10 63 64

The UK dataset analysis suggested that children who consumed CEDs 5 or more days a week had lower psychological, physical, educational and overall well-being than non-drinkers. It remains unclear whether drinking CEDs contributes to low well-being, or low well-being leads to CED consumption, or both. Alternatively, there may be a common cause, such as social inequality.

Strengths and limitations

The overview was limited by the amount of information reported in the included systematic reviews, and by their method limitations; all had a high risk of bias. They mainly included cross-sectional surveys or case reports, which means that cause or effect cannot be determined where an association is found. However, some prospective studies, including four small RCTs, were included in the reviews and where there were common measures, the evidence from these RCTs and from most of the cross-sectional studies within the reviews was consistent. This suggests that the associations found could be reliable. A strength of our work is that the UK evidence in the overview (two studies within the reviews) was supplemented by the analysis of UK data, which was mostly consistent with the non-UK evidence. These data support the idea that there is a link between drinking CEDs and poorer health and behaviour in children, although the cause is unclear. Overlap between reviews in the overview was slight (unsurprisingly, given the different foci of the reviews). There was no overlap between the reviews and the dataset analysis, meaning that the latter added new information. The wide range of tools used to measure prevalence made it difficult to summarise the overview evidence, and meta-analysis of the individual participant UK data was not possible, meaning that the conclusions are based on weaker evidence from single sources.

Recommendations for research

Standardisation is needed in the measurement of the prevalence of drinking—defining the dosage (in drinks and/or caffeine), timing (daily, weekly, etc) and population (age, ethnicity, etc). There was little evidence on children under 12 years old, and both the overview and dataset analysis found little evidence from the UK. Longitudinal data, from the UK datasets, should be collected to understand better the impact of consumption. RCTs may not be ethical, even where benefits are predicted, such as where children who consume CEDs are randomised to interventions to reduce or stop their drinking to see if this improves their well-being.

Based on a comprehensive overview of available systematic reviews, we conclude that up to half of children, worldwide, drink CEDs weekly or monthly, and based on the dataset analysis, up to a third of UK children do so. There is weak but consistent evidence, from reviews and UK datasets, that poorer health and well-being is found in children who drink CEDs. In the absence of RCTs, which are unlikely to be ethical, longitudinal studies could provide stronger evidence.

Ethics statements

Patient consent for publication.

Not applicable.

Ethics approval

This study does not involve human participants.

Acknowledgments

Thank you to Irene Kwan for assisting with data extraction for the review.

• Committee on Nutrition and the Council on Sports Medicine and Fitness
• UK Government.
• Haskell CF ,
• Kennedy DO ,
• Wesnes KA , et al
• Keast RSJ ,
• Swinburn BA ,
• Sayompark D , et al
• Curran CP ,
• Marczinski CA
• Henderson R , et al
• Cheetham M ,
• Riby DM , et al
• UK Government
• Department of Health and Social Care (DHSC)
• Department of Health and Social Care
• Brunton G ,
• Sowden A , et al
• Raine G , et al
• Brunton J ,
• ↵ Uk data service . Available: https://ukdataservice.ac.uk/ [Accessed 3 Apr 2020 ].
• Huang L , et al
• Reeves BC ,
• Wells G , et al
• Brennan S ,
• McKenzie J ,
• Middleton P
• Antoine S-L ,
• Mathes T , et al
• McKenzie JE ,
• UK Data Service
• National Institutes of Health
• El Kashef A ,
• AlGhaferi H
• Stockwell T
• Verster JC ,
• Johnson SJ , et al
• Babayan Z , et al
• Bleich SN ,
• Vercammen KA
• Burnett K , et al
• Goldfarb M ,
• Tellier C ,
• Thanassoulis G
• Cervellin G ,
• Sanchis-Gomar F
• Richards G ,
• Imamura K ,
• Watanabe K , et al
• Nadeem IM ,
• Shanmugaraj A ,
• Sakha S , et al
• University of London, I.o.E., Centre for Longitudinal Studies,
• EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA)
• Al-Hazzaa H ,
• Waly MI , et al
• Jasionowski A
• Nobile CGA , et al
• NHS Digital
• Gambon DL ,
• Boutkabout C , et al
• Hammond D ,
• McCrory C , et al
• Bryant Ludden A ,
• Musaiger AO ,
• Gallimberti L ,
• Chindamo S , et al
• Abian-Vicen J ,
• Salinero JJ , et al
• Gallo-Salazar C ,
• Abián-Vicén J , et al
• Temple JL ,
• Briatico LN
• Man Yu MW , et al
• Hartley JEK ,
• Zucconi S ,
• Volpato C ,
• Adinolfi F , et al
• Seifert SM ,
• Schaechter JL ,
• Hershorin ER , et al
• Public Health England
• Northern Ireland Statistics and Research Agency

Supplementary materials

Supplementary data.

This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

• Data supplement 1
• Press_Release.pdf

Twitter @katysutcliffe

Contributors GB, CK, GR and CS worked on all stages of the overview. GB, CK, DK and GR worked on the overview update. GB and DK completed all stages of the secondary data analysis. GB, KS, AS and JT supervised the work. All authors discussed the results and contributed to the final manuscript. JT is the guarantor of this work.

Funding This overview and secondary data analysis was funded by the National Institute for Health Research (NIHR) Policy Research Programme (PRP) for the Department of Health and Social Care (DHSC). It was funded through the NIHR PRP contract with the EPPI Centre at UCL (Reviews facility to support national policy development and implementation, PR-R6-0113-11003). Any views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the NIHR or the DHSC.

Competing interests None declared.

Provenance and peer review Not commissioned; externally peer reviewed.

Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Read the full text or download the PDF:

Inference for Local Projections

Òscar Jordà

Atsushi Inoue

Guido M. Kuersteiner

Download PDF (531 KB)

2024-29 | August 19, 2024

Inference for impulse responses estimated with local projections presents interesting challenges and opportunities. Analysts typically want to assess the precision of individual estimates, explore the dynamic evolution of the response over particular regions, and generally determine whether the impulse generates a response that is any different from the null of no effect. Each of these goals requires a different approach to inference. In this article, we provide an overview of results that have appeared in the literature in the past 20 years along with some new procedures that we introduce here.

Suggested citation:

Inoue, Atsushi, Òscar Jordà, and Guido M. Kuersteiner. 2024. “Inference for Local Projections.” Federal Reserve Bank of San Francisco Working Paper 2024-29. https://doi.org/10.24148/wp2024-29

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Alcohol Res Health
• v.34(2); 2011

The Risks Associated With Alcohol Use and Alcoholism

Alcohol consumption, particularly heavier drinking, is an important risk factor for many health problems and, thus, is a major contributor to the global burden of disease. In fact, alcohol is a necessary underlying cause for more than 30 conditions and a contributing factor to many more. The most common disease categories that are entirely or partly caused by alcohol consumption include infectious diseases, cancer, diabetes, neuropsychiatric diseases (including alcohol use disorders), cardiovascular disease, liver and pancreas disease, and unintentional and intentional injury. Knowledge of these disease risks has helped in the development of low-risk drinking guidelines. In addition to these disease risks that affect the drinker, alcohol consumption also can affect the health of others and cause social harm both to the drinker and to others, adding to the overall cost associated with alcohol consumption. These findings underscore the need to develop effective prevention efforts to reduce the pain and suffering, and the associated costs, resulting from excessive alcohol use.

Alcohol consumption has been identified as an important risk factor for illness, disability, and mortality ( Rehm et al. 2009 b ). In fact, in the last comparative risk assessment conducted by the World Health Organization (WHO), the detrimental impact of alcohol consumption on the global burden of disease and injury was surpassed only by unsafe sex and childhood underweight status but exceeded that of many classic risk factors, such as unsafe water and sanitation, hyper-tension, high cholesterol, or tobacco use ( WHO 2009 ). This risk assessment evaluated the net effect of all alcohol consumption—that is, it also took into account the beneficial effects that alcohol consumption (primarily moderate consumption) can have on ischemic diseases 1 and diabetes ( Baliunas et al. 2009 ; Corrao et al. 2000 ; Patra et al. 2010 ; Rehm et al. 2004 ). Although these statistics reflect the consequences of all alcohol consumption, it is clear that most of the burden associated with alcohol use stems from regular heavier drinking, defined, for instance, as drinking more than 40 grams of pure alcohol per day for men and 20 grams of pure alcohol per day for women 2 ( Patra et al. 2009 ; Rehm et al. 2004 ). In addition to the average volume of alcohol consumption, patterns of drinking—especially irregular heavy-drinking occasions, or binge drinking (defined as drinking at least 60 grams of pure alcohol or five standard drinks in one sitting)—markedly contribute to the associated burden of disease and injury ( Gmel et al. 2010 ; Rehm et al. 2004 ). This article first defines which conditions necessarily are caused by alcohol use and for which conditions alcohol use is a contributing factor. It then looks more closely at the most common disease risks associated with excessive alcohol use, before exploring how these risks have influenced guidelines for drinking limits. The article concludes with a discussion of the alcohol-related risk of harm to people other than the drinker.

Disease and Injury Conditions Associated With Alcohol Use

Conditions for which alcohol is a necessary cause.

More than 30 conditions listed in the WHO’s International Classification of Diseases, 10th Edition (ICD–10) ( WHO 2007 ) include the term “alcohol” in their name or definition, indicating that alcohol consumption is a necessary cause underlying these conditions (see table 1 ). The most important disease conditions in this group are alcohol use disorders (AUDs), which include alcohol dependence and harmful use or alcohol abuse. 3 AUDs are less fatal than other chronic disease conditions but are linked to considerable disability ( Samokhvalov et al. 2010 d ). Overall, even though AUDs in themselves do not rank high as a cause of death globally, they are the fourth-most disabling disease category in low- to middle-income countries and the third-most disabling disease category in high-income countries ( WHO 2008 ). Thus, AUDs account for 18.4 million years of life lost to disability (YLDs), or 3.5 percent of all YLDs, in low- and middle-income countries and for 3.9 million YLDs, or 5.7 percent of all YLDs, in high-income countries. However, AUDs do not affect all population subgroups equally; for example, they mainly affect men, globally representing the second-most disabling disease and injury condition for men. In contrast, AUDs are not among the 10 most important causes of disabling disease and injury in women ( WHO 2008 ).

Disease Conditions That by Definition Are Attributable to Alcohol (AAF = 100%)

Note: ICD codes in italics represent subcodes within a main code of classification.

Abbreviations: AAF = alcohol-attributable fraction.

Alcoholic liver disease and alcohol-induced pancreatitis are other alcohol-specific disease categories that are of global importance. However, no global prevalence data on these disease categories exist because they cannot be validly assessed on a global level. Thus, these conditions are too specific to assess using verbal autopsies and other methods normally used in global-burden-of-disease studies ( Lopez et al. 2006 ; pancreatitis can be estimated indirectly Rajaratnam et al. 2010 ). Nevertheless, via the prevalence of alcohol exposure the prevalence of alcohol-attributable and relative risk for the wider, unspecific liver cirrhosis and alcohol-induced disease categories ( Rehm et al. 2010 a ).

Conditions for Which Alcohol Is a Component Cause

Disease and injury conditions for which alcohol consumption is a component cause contribute more to the global burden of disease than do alcohol-specific conditions. Overall, the following are the main disease and injury categories impacted by alcohol consumption (listed in the order of their ICD–10 codes):

• Infectious disease;
• Neuropsychiatric disease;
• Cardiovascular disease;
• Liver and pancreas disease; and
• Unintentional and intentional injury.

For all chronic disease categories for which detailed data are available, those data show that women have a higher risk of these conditions than men who have consumed the same amount of alcohol; however, the differences are small at lower levels of drinking ( Rehm et al. 2010 a ). The following sections will look at these disease categories individually.

Individual Disease and Injury Conditions Associated With Alcohol Use

Infectious diseases.

Although infectious diseases were not included in the WHO’s comparative risk assessments for alcohol conducted in 2000 ( Rehm et al. 2004 ) and 2004 ( Rehm et al. 2009 b ), evidence has been accumulating that alcohol consumption has a detrimental impact on key infectious diseases ( Rehm et al. 2009 a , 2010 a ), such as tuberculosis ( Lönnroth et al. 2008 ; Rehm et al. 2009 c ), infection with the human immunodeficiency virus (HIV) ( Baliunas et al. 2010 ; Shuper et al. 2010 ), and pneumonia ( Samokhvalov et al. 2010 c ). In fact, recent studies (Rehm and Parry 2009 ; Rehm et al. 2009 a ) found that the overall impact of alcohol consumption on infectious diseases is substantial, especially in sub-Saharan Africa.

One of the pathways through which alcohol increases risk for these diseases is via the immune system, which is adversely affected by alcohol consumption, especially heavy drinking ( Rehm et al. 2009 c ; Romeo et al. 2010 ). As a result, although risk for infectious diseases does not differ greatly for people drinking less than 40 grams of pure alcohol per day compared with abstainers, this risk increases substantially for those who drink larger amounts or have been diagnosed with an AUD ( Lönnroth et al. 2008 ; Samokhvalov et al. 2010 c ). In addition, alcohol consumption is associated with poorer outcomes from infectious disease for heavy drinkers by way of social factors. Thus, people with alcohol dependence often are stigmatized and have a higher chance of becoming unemployed and destitute; as a result, they tend to live in more crowded quarters with higher chances for infection and lower chances of recovery ( Lönnroth et al. 2009 ).

The relationship between alcohol consumption and HIV infection and acquired immunodeficiency syndrome (AIDS) is different from that with other infectious diseases. To become infected with HIV, people must exchange body fluids, in most cases either by injecting drugs with a contaminated needle or, more commonly in low-income societies, engaging in unsafe sex. Thus, although significant associations exist between alcohol use, especially heavy drinking, and HIV infection via alcohol’s general effects on the immune system ( Baliunas et al. 2010 ; Kalichman et al. 2007 ; Shuper et al. 2009 , 2010 ), it cannot be excluded that other variables, including personality characteristics, psychiatric disorders, and situational factors may be responsible for both risky drinking and unsafe sex ( Shuper et al. 2010 ). Researchers frequently have pointed out that personality characteristics, such as a propensity for risk-taking, sensation-seeking, and sexual compulsivity, may be involved in the risk of HIV infection. Indeed, a recent consensus meeting determined that there is not yet sufficient evidence to conclude that alcohol has a causal impact on HIV infection ( Parry et al. 2009 ). However, it can be argued that experimental studies in which alcohol consumption led to a greater inclination to engage in unsafe sex indicate that some causal relationship between alcohol and HIV infection exists (e.g., George et al. 2009 ; Norris et al. 2009 ).

Once a person is infected with HIV, alcohol clearly has a detrimental impact on the course of the disease, especially by interfering with effective antiretroviral treatment ( Pandrea et al. 2010 ). A recent meta-analysis found that problem drinking—defined as meeting the National Institute on Alcohol Abuse and Alcoholism (NIAAA)’s criteria for at-risk drinking or having an AUD—was associated with being less than half as likely to adhere to antiretroviral treatment guidelines ( Hendershot et al. 2009 ). Because the level of adherence to the treatment regimen affects treatment success as well as outright survival, alcohol consumption clearly is associated with negative outcomes for people living with HIV and AIDS.

Recently, the Monograph Working Group of the International Agency for Research on Cancer concluded that there was sufficient evidence for the carcinogenicity of alcohol in animals and classified alcoholic beverages as carcinogenic to humans ( Baan et al. 2007 ). In particular, the group confirmed, or newly established, the causal link between alcohol consumption and cancer of the oral cavity, pharynx, larynx, esophagus, liver, colorectum, and female breast. For stomach and lung cancer, carcinogenicity was judged as possible but not established. For all sites where alcohol’s causal role in cancer is established, there is evidence of a dose-response relationship, with relative risk rising linearly with an increasing volume of alcohol consumption ( Corrao et al. 2004 ).

The molecular and biochemical mechanisms by which chronic alcohol consumption leads to the development of cancers of various organs are not fully understood. It has been suggested that these mechanisms differ by target organ and include variations (i.e., polymorphisms) in genes encoding enzymes responsible for ethanol metabolism (e.g., alcohol dehydrogenase, aldehyde dehydrogenase, and cytochrome P450 2E1), increased estrogen concentrations, and changes in folate metabolism and DNA repair ( Boffetta and Hashibe 2006 ; Seitz and Becker 2007 ). In addition, the International Agency for Research on Cancer group concluded that acetaldehyde—which is produced when the body breaks down (i.e., metabolizes) beverage alcohol (i.e., ethanol) but also is ingested as a component of alcoholic beverages— itself is carcinogenic. It likely plays an important role in the development of cancers of the digestive tract, especially those of the upper digestive tract ( Lachenmeier et al. 2009 ; Seitz and Becker 2007 ).

The relationship between alcohol consumption and diabetes is complex. A curvilinear relationship exists between the average volume of alcohol consumption and the inception of diabetes ( Baliunas et al. 2009 )—that is, lower alcohol consumption levels have a protective effect, whereas higher consumption is associated with an increased risk. The greatest protective effect has been found with a consumption of about two standard drinks (28 grams of pure alcohol) per day, and a net detrimental effect has been found starting at about four standard drinks (50 to 60 grams of pure alcohol) per day.

Neuropsychiatric Disorders

With respect to neuropsychiatric disorders, alcohol consumption has by far the greatest impact on risk for alcohol dependence. However, alcohol also has been associated with basically all mental disorders (e.g., Kessler et al. 1997 ), although the causality of these associations is not clear. Thus, mental disorders may be caused by AUDs or alcohol use, AUDs may be caused by other mental disorders, or third variables may be causing both AUDs and other mental disorders. This complex relationship makes it difficult to determine the fraction of mental disorders actually caused by alcohol consumption (see Grant et al. 2009 ).

The relationship between alcohol and epilepsy is much clearer. There is substantial evidence that alcohol consumption can cause unprovoked seizures, and researchers have identified plausible biological pathways that may underlie this relationship ( Samokhvalov et al. 2010 a ). Most of the relevant studies found that a high percentage of heavy alcohol users with epilepsy meet the criteria of alcohol dependence.

Cardiovascular Diseases

The overall effect of alcohol consumption on the global cardiovascular disease burden is detrimental (see table 2 ). Cardiovascular disease is a general category that includes several specific conditions, and alcohol’s impact differs for the different conditions. For example, the effect of alcohol consumption on hypertension is almost entirely detrimental, with a dose-response relationship that shows a linear increase of the relative risk with increasing consumption ( Taylor et al. 2009 ). A similar dose-response relationship exists between alcohol consumption and the incidence of atrial fibrillation 4 ( Samokhvalov et al. 2010 b ). On the other hand, for heart disease caused by reduced blood supply to the heart (i.e., ischemic heart disease), the association with alcohol consumption is represented by a J-shaped curve ( Corrao et al. 2000 ), with regular light drinking showing some protective effects. Irregular heavy drinking occasions, however, can nullify any protective effect. In a recent systematic review and meta-analysis comparing the effects of different drinking patterns in people with an overall consumption of less than 60 grams of pure alcohol per day, Roerecke and Rehm (2010) found that consumption of 60 grams of pure alcohol on one occasion at least once a month eliminated any protective effect of alcohol consumption on mortality. The authors concluded that the cardio-protective effect of moderate alcohol consumption disappears when light to moderate drinking is mixed with irregular heavy-drinking occasions. These epidemiological results are consistent with the findings of biological studies that—based on alcohol’s effects on blood lipids and blood clotting—also predict beneficial effects of regular moderate drinking but detrimental effects of irregular heavy drinking ( Puddey et al. 1999 ; Rehm et al. 2003 ).

Global Burden of Alcohol-Attributable Disease in Disability-Adjusted Life Years (DALYs) (in 1,000s) by Sex and Disease Category for the Year 2004

NOTE: M = men; W = women; T = total.

SOURCE: Rehm et al. 2009 a,b .

The effects of alcohol consumption on ischemic stroke 5 are similar to those on ischemic heart disease, both in terms of the risk curve and in terms of biological pathways ( Patra et al. 2010 ; Rehm et al. 2010 a ). On the other hand, alcohol consumption mainly has detrimental effects on the risk for hemorrhagic stroke, which are mediated at least in part by alcohol’s impact on hypertension.

Overall, the effects of alcohol consumption on cardiovascular disease are detrimental in all societies with large proportions of heavy-drinking occasions, which is true for most societies globally ( Rehm et al. 2003 a ). This conclusion also is supported by ecological analyses or natural experiments. For example, studies in Lithuania ( Chenet et al. 2001 ) found that cardiovascular deaths increased on weekends, when heavy drinking is more common. Also, when overall consumption was reduced in the former Soviet Union (a country with a high proportion of heavy-drinking occasions) between 1984 and 1994, the death rate from cardiovascular disease declined, indicating that alcohol consumption had an overall detrimental effect on this disease category ( Leon et al. 1997 ).

Diseases of the Liver and Pancreas

Alcohol consumption has marked and specific effects on the liver and pancreas, as evidenced by the existence of disease categories such as alcoholic liver disease, alcoholic liver cirrhosis, and alcohol-induced acute or chronic pancreatitis. For these disease categories, the dose-response functions for relative risk are close to exponential ( Irving et al. 2009 ; Rehm et al. 2010 b ), although the risks associated with light to moderate drinking (i.e., up to 24 grams of pure alcohol per day) are not necessarily different from the risks associated with abstention. Thus, the incidence of diseases of the liver and pancreas is associated primarily with heavy drinking.

It is important to note that given the same amount of drinking, the increase in the risk for mortality from these diseases is greater than the increase in risk for morbidity, especially at lower levels of consumption. This finding suggests that continued alcohol consumption, even in low doses, after the onset of liver or pancreas disease, increases the risk of severe consequences.

Unintentional Injuries

The link between alcohol and almost all kinds of unintentional injuries has long been established. It depends on the blood alcohol concentration (BAC) and shows an exponential dose-response relationship ( Taylor et al. 2010 ). Alcohol affects psychomotor abilities, with a threshold dose for negative effects generally found at BACs of approximately 0.04 to 0.05 percent (which typically are achieved after consuming two to three drinks in an hour); accordingly, injury resulting from alcohol’s disruption of psychomotor function could occur in people with BACs at this level ( Eckardt et al. 1998 ). However, the epidemiological literature shows that even at lower BACs, injury risk is increased compared with no alcohol consumption ( Taylor et al. 2010 ).

The acute effects of alcohol consumption on injury risk are mediated by how regularly the individual drinks. People who drink less frequently are more likely to be injured or to injure others at a given BAC compared with regular drinkers, presumably because of less tolerance ( Gmel et al. 2010 ). This correlation was demonstrated with respect to traffic injuries in a reanalysis ( Hurst et al. 1994 ) of a classic study conducted in Grand Rapids, Michigan ( Borkenstein et al. 1974 ). It also is important to realize that even if the absolute risk for injury may be relatively small for each occasion of moderate drinking (defined as drinking up 36 grams pure alcohol in one sitting), the lifetime risks from such drinking occasions sums up to a considerable risk for those who often drink at such a level ( Taylor et al. 2008 ).

Intentional Injuries

Alcohol consumption is linked not only to unintentional but also to intentional injury. Both average volume of alcohol consumption and the level of drinking before the event have been shown to affect suicide risk ( Borges and Loera 2010 ). There also is a clear link between alcohol consumption and aggression, including, but not limited to, homicides ( Rehm et al. 2003 b ). Several causal pathways have been identified that play a role in this link, including biological pathways acting via alcohol’s effect on receptors for the brain signaling molecules (i.e., neurotransmitters) serotonin and γ-aminobutyric acid or via alcohol’s effects on cognitive functioning ( Rehm et al. 2003 b ). Cultural factors that are related to both differences in drinking patterns and beliefs and expectations about the effects of alcohol also influence the relationship between drinking and aggression ( Bushman and Cooper 1990 ; Graham 2003 ; Leonard 2005 ; Room and Rossow 2001 ).

Implications of Alcohol-Related Risks for Drinking Guidelines

Overall, the various risks associated with alcohol use at various levels can be combined to derive low-risk drinking guidelines. Such analyses found that overall, any increase in drinking beyond one standard drink on average per day is associated with an increased net risk for morbidity and mortality in high-income countries ( Rehm et al. 2009 ). Moreover, at any given consumption level this risk increase is larger for women than for men. NIAAA has translated the epidemiological findings into low-risk drinking limits of no more than 14 standard drinks per week for men and 7 standard drinks per week for women ( NIAAA 2010 ). These guidelines also specify that to limit the risk of acute consequences, daily consumption should not exceed four standard drinks for men and three for women ( NIAAA 2010 ).

Overall Global Impact of Alcohol Consumption on Burden of Disease

The most recent systematic overview on the effects of alcohol on global burden of disease was based on data for the year 2004 ( Rehm et al. 2009 a , b ) (see table 2 ). The analyses found that although AUDs (which constitute the major part of the neuropsychiatric disorders listed in the table) clearly are important contributors to global burden of disease, they only account for less than one-third of the overall impact of alcohol consumption. Almost equally important are the acute effects of alcohol consumption on the risk of both unintentional and intentional injury. In addition, alcohol has a sizable effect on the burden of disease associated with infectious diseases, cancer, cardiovascular disease, and liver cirrhosis. However, alcohol consumption also has beneficial effects on the burden of disease, mainly on diabetes and the ischemic disease subcategory of cardiovascular diseases. Yet these effects are by far outweighed by the detrimental consequences of alcohol consumption.

Effects of Alcohol on People Other Than the Drinker

So far, the discussion has centered on alcohol’s effects on health as measured by indicators that primarily are based on the records of hospitals and health systems. Reflecting the information contained in those records, most of the effects considered refer to the health of the drinker. However, this analytic approach omits two large classes of adverse consequences of alcohol: social harm to the drinker and social and health harms to others that result from the drinker’s alcohol consumption. According to the Constitution of the WHO ( WHO 1946 ), health is “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity” (p. 100); this definition therefore takes into account not just physical and mental harms but also social harms, both for the drinker and for others.

A few examples of harm to others are included in the analysis of alcohol’s contribution to the global burden of disease listed in table 2 . These include perinatal conditions attributable to the mother’s drinking during pregnancy and injuries, particularly assault injuries. However, the scope of alcohol-related social harm and of harm to others stretches well beyond these items. Thus, a recent study in Australia ( Laslett et al. 2010 ) identified the following harms to others associated with drinking:

• Harms identified based on records—these included deaths and hospitalizations (e.g., attributed to traffic injuries because of driving under the influence), child abuse or neglect cases involving a caregiver’s drinking, and domestic and other assaults; and
• Harms based on survey reports—these included negative effects on coworkers, household members, other relatives and friends, strangers, and on the community as a whole.

These effects were quite prevalent. Thus, the researchers estimated that within 1 year, more than 350 deaths were attributed to drinking by others, and more than 10 million Australians (or 70 percent of all adults) were negatively affected by a stranger’s drinking ( Laslett et al. 2010 ).

Social Harm

Drinkers also experience a range of social harms because of their own drinking, including family disruption, problems at the workplace (including unemployment), criminal convictions, and financial problems ( Casswell and Thamarangsi 2009 ; Klingemann and Gmel 2001 ). Unfortunately, assessment of these problems is much less standardized than assessment of health problems, and many of these harms are not reported continuously. Social-cost studies provide irregular updates of alcohol-attributable consequences in selected countries (for an overview, see Rehm et al. 2009 b ; Thavorncharoensap et al. 2009 ). These studies regularly find that health care costs comprise only a small portion of the overall costs associated with alcohol use and that most of the alcohol-associated costs are attributable to productivity losses. In total, the costs associated with alcohol use seem to amount to 1 to 3 percent of the gross domestic product in high-income countries; the alcohol-associated costs in South Korea and Thailand, the only two mid-income countries for which similar studies are available, were at about the same level.

Conclusions

As this review has shown, alcohol use is associated with tremendous costs to the drinker, those around him or her, and society as a whole. These costs result from the increased health risks (both physical and mental) associated with alcohol consumption as well as from the social harms caused by alcohol. To reduce alcohol’s impact on the burden of disease as well as on other social, legal, and monetary costs, it therefore is imperative to develop effective interventions that can prevent or delay initiation of drinking among those who do not drink, particularly adolescents, and limit consumption to low-risk drinking levels among those who do consume alcohol. The remaining articles in this journal issue present several such intervention approaches that are being implemented and evaluated in a variety of settings and/or are targeted at different population subgroups. Together with alcohol-related prevention policies, the implementation of specific interventions with proven effectiveness can help reduce the pain and suffering, and the associated costs, resulting from excessive alcohol use.

Acknowledgments

Financial support for this study was provided by NIAAA contract HHSN267200700041C to conduct the study titled “Alcohol- and Drug-Attributable Burden of Disease and Injury in the U.S.”

The views expressed here do not necessarily reflect the views of the funding agency.

F inancial D isclosure

Jürgen Rehm, Ph.D., received a salary and infrastructure support from the Ontario Ministry of Health and Long-Term Care. No potential conflicts of interest relevant to this article were reported.

1 Ischemic diseases are all conditions that are related to the formation of blood clots, which prevent adequate blood flow to certain tissues.

2 In the United States, a standard drink usually is considered to contain 0.6 fluid ounces (or 14 grams) of pure alcohol. This is the amount of ethanol found in approximately 12 ounces of beer, 5 ounces of wine, or 1.5 ounces of distilled spirits. However, many drinks, as actually poured, contain more alcohol. Thus, for example, a glass of wine often contains more than 5 fluid ounces and therefore may correspond to one and a half or even two standard drinks.

3 The condition referred to as “harmful use” in the ICD–10 loosely corresponds to “alcohol abuse,” as defined in the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Diseases, 4th Edition (DSM–IV).

4 Atrial fibrillation is an abnormal heart rhythm involving the two upper chambers (i.e., atria) of the heart.

5 A stroke is the disruption of normal blood flow to a brain region. In the case of an ischemic stroke, this is caused by blockage of a blood vessel that prevents the blood from reaching neighboring brain areas. In the case of a hemorrhagic stroke, rupture of a blood vessel and bleeding into the brain occurs, which prevents normal blood supply to other brain regions.

• Baan R, Straif K, Grosse Y, et al. Carcinogenicity of alcoholic beverages. Lancet Oncology. 2007; 8 (4):292–293. [ PubMed ] [ Google Scholar ]
• Baliunas D, Rehm J, Irving H, Shuper P. Alcohol consumption and risk of incident human immunodeficiency virus infection: A meta-analysis. International Journal of Public Health. 2010; 55 (3):159–166. [ PubMed ] [ Google Scholar ]
• Baliunas DO, Taylor BJ, Irving H, et al. Alcohol as a risk factor for type 2 diabetes: A systematic review and meta-analysis. Diabetes Care. 2009; 32 (11):2123–2132. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Boffetta P, Hashibe M. Alcohol and cancer. Lancet Oncology. 2006; 7 (2):149–156. [ PubMed ] [ Google Scholar ]
• Borges G, Loera CR. Alcohol and drug use in suicidal behaviour. Current Opinion in Psychiatry. 2010; 23 (3):195–204. [ PubMed ] [ Google Scholar ]
• Borkenstein RF, Crowther FR, Shumate RP, et al. The role of the drinking driver in traffic accidents (The Grand Rapids Study) Blutalkohol. 1974; 11 (Suppl. 1):1–131. [ Google Scholar ]
• Bushman BJ, Cooper HM. Effects of alcohol on human aggression: An integrative research review. Psychological Bulletin. 1990; 107 :341–354. [ PubMed ] [ Google Scholar ]
• Casswell S, Thamarangsi T. Reducing the harm from alcohol: Call to action. Lancet. 2009; 373 (9682):2247–2257. [ PubMed ] [ Google Scholar ]
• Chenet L, Britton A, Kalediene R, Petrauskiene J. Daily variations in deaths in Lithuania: The possible contribution of binge drinking. International Journal of Epidemiology. 2001; 30 (4):743–748. [ PubMed ] [ Google Scholar ]
• Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of alcohol consumption and the risk of 15 diseases. Preventive Medicine. 2004; 38 (5):613–619. [ PubMed ] [ Google Scholar ]
• Corrao G, Rubbiati L, Bagnardi V, et al. Alcohol and coronary heart disease: A meta-analysis. Addiction. 2000; 95 (10):1505–1523. [ PubMed ] [ Google Scholar ]
• Eckardt MJ, File SE, Gessa GL, et al. Effects of moderate alcohol consumption on the central nervous system. Alcoholism: Clinical and Experimental Research. 1998; 22 (5):998–1040. [ PubMed ] [ Google Scholar ]
• George WH, Davis KC, Norris J, et al. Indirect effects of acute alcohol intoxication on sexual risk-taking: The roles of subjective and physiological sexual arousal. Archives of Sexual Behavior. 2009; 38 (4):498–513. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Gmel G, Kuntsche E, Rehm J. Risky single occasion drinking: Bingeing is not bingeing. Addiction. 2010 Oct. doi: 10.1111/j.1360-0443.2010.03167.x. Published online: [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Graham K. Social drinking and aggression. In: Mattson M, editor. Neurobiology of Aggression: Understanding and Preventing Violence. Totowa, NJ: Humana Press; 2003. pp. 253–274. [ Google Scholar ]
• Grant BF, Goldstein RB, Chou SP, et al. Sociodemographic and psychopathologic predictors of first incidence of DSM-IV substance use, mood and anxiety disorders: Results from the Wave 2 National Epidemiologic Survey on Alcohol and Related Conditions. Molecular Psychiatry. 2009; 14 (11):1051–1066. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hendershot CS, Stoner SA, Pantalone DW, Simoni JM. Alcohol use and antiretroviral adherence: Review and meta-analysis. Journal of Acquired Immune Deficiency Syndromes. 2009; 52 (2):180–202. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Hurst PM, Harte D, Firth WJ. The Grand Rapids dip revisited. Accident: Analysis and Prevention. 1994; 26 (5):647–654. [ PubMed ] [ Google Scholar ]
• Irving HM, Samokhvalov AV, Rehm J. Alcohol as a risk factor for pancreatitis. A systematic review and meta-analysis. Journal of the Pancreas. 2009; 10 (4):387–392. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Kalichman SC, Simbayi LC, Kaufman M, et al. Alcohol use and sexual risks for HIV/AIDS in sub-Saharan Africa: Systematic review of empirical findings. Prevention Science. 2007; 8 (2):141–151. [ PubMed ] [ Google Scholar ]
• Kessler RC, Crum RM, Warner LA, et al. Lifetime co-occurrence of DSM-III-R alcohol abuse and dependence with other psychiatric disorders in the National Comorbidity Survey. Archives of General Psychiatry. 1997; 54 (4):313–321. [ PubMed ] [ Google Scholar ]
• Klingemann H, Gmel G. Mapping Social Consequences of Alcohol Consumption. Dordrecht, Netherlands: Kluwer Academic Publishers; 2001. [ Google Scholar ]
• Lachenmeier DW, Kanteres F, Rehm J. Carcinogenicity of acetaldehyde in alcoholic beverages: Risk assessment outside ethanol metabolism. Addiction. 2009; 104 (4):533–550. [ PubMed ] [ Google Scholar ]
• Laslett AM, Catalano P, Chikritzhs T, et al. The Range and Magnitude of Alcohol’s Harm to Others. Fitzroy, Victoria, Australia: Turning Point Alcohol & Drug Centre; 2010. [ Google Scholar ]
• Leon DA, Chenet L, Shkolnikov VM, et al. Huge variation in Russian mortality rates 1984– 1994. Artefact, alcohol, or what? Lancet. 1997; 350 (9075):383–388. [ PubMed ] [ Google Scholar ]
• Leonard KE. Alcohol and intimate partner violence: When can we say that heavy drinking is a contributing cause of violence? Addiction. 2005; 100 (4):422–425. [ PubMed ] [ Google Scholar ]
• Lönnroth K, Jaramillo E, Williams BG, et al. Drivers of tuberculosis epidemics: The role of risk factors and social determinants. Social Science & Medicine. 2009; 68 (12):2240–2246. [ PubMed ] [ Google Scholar ]
• Lönnroth K, Williams BG, Stadlin S, et al. Alcohol use as a risk factor for tuberculosis: A systematic review. BMC Public Health. 2008; 8 :289. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Lopez AD, Mathers CD, Ezzati M, et al. Global Burden of Disease and Risk Factors. New York & Washington: The World Bank and Oxford University Press; 2006. [ PubMed ] [ Google Scholar ]
• National Institute on Alcohol Abuse and Alcoholism (NIAAA) Rethinking Drinking: Alcohol and Your Health. Rockville, MD: NIAAA; 2010. Pub. No. 10–3770. [ Google Scholar ]
• Norris J, Stoner SA, Hessler DM, et al. Cognitive mediation of alcohol’ s effects on women’s in-the-moment sexual decision making. Health Psychology. 2009; 28 (1):20–28. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Pandrea I, Happel KI, Amedee AM, et al. Alcohol’s Role in HIV Transmission and Disease Progression. Bethesda, MD: NIAAA; 2010. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Parry C, Rehm J, Poznyak V, Room R. Alcohol and infectious diseases: An overlooked causal linkage? Addiction. 2009; 104 (3):331–332. [ PubMed ] [ Google Scholar ]
• Patra J, Taylor B, Irving H, et al. Alcohol consumption and the risk of morbidity and mortality from different stroke types: A systematic review and meta-analysis. BMC Public Health. 2010; 10 :258. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Patra J, Taylor B, Rehm J. Deaths associated with high-volume drinking of alcohol among adults in Canada in 2002: A need for primary care intervention? Contemporary Drug Problems. 2009; 36 (2):283–301. [ Google Scholar ]
• Puddey IB, Rakic V, Dimmitt SB, Bellin LJ. Influence of pattern of drinking on cardiovascular disease and cardiovascular risk factors: A review. Addiction. 1999; 94 (5):649–663. [ PubMed ] [ Google Scholar ]
• Rajaratnam JK, Marcus JR, Levin-Rector A, et al. Worldwide mortality in men and women aged 15-59 years from 1970 to 2010: A systematic analysis. Lancet. 2010; 375 (9727):1704–1720. [ PubMed ] [ Google Scholar ]
• Rehm J, Parry C. Alcohol consumption and infectious diseases in South Africa. Lancet. 2009; 374 (9707):2053. [ PubMed ] [ Google Scholar ]
• Rehm J, Anderson P, Kanteres F, et al. Alcohol, Social Development and Infectious Disease. Toronto, ON: Centre for Addiction and Mental Health; 2009a. [ Google Scholar ]
• Rehm J, Baliunas D, Borges GL, et al. The relation between different dimensions of alcohol consumption and burden of disease: An overview. Addiction. 2010a; 105 (5):817–843. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rehm J, Mathers C, Popova S, et al. Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders. Lancet. 2009b; 373 (9682):2223–2233. [ PubMed ] [ Google Scholar ]
• Rehm J, Rehn N, Room R, et al. The global distribution of average volume of alcohol consumption and patterns of drinking. European Addiction Research. 2003a; 9 (4):147–156. [ PubMed ] [ Google Scholar ]
• Rehm J, Room R, Graham K, et al. The relationship of average volume of alcohol consumption and patterns of drinking to burden of disease: An overview. Addiction. 2003b; 98 (9):1209–1228. [ PubMed ] [ Google Scholar ]
• Rehm J, Room R, Monteiro M, et al. Comparative Quantification of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva, Switzerland: World Health Organization; 2004. Alcohol use; pp. 959–1109. [ Google Scholar ]
• Rehm J, Samokhvalov AV, Neuman MG, et al. The association between alcohol use, alcohol use disorders and tuberculosis (TB): A systematic review. BMC Public Health. 2009c; 9 :450. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Rehm J, Sempos C, Trevisan M. Alcohol and cardiovascular disease—More than one paradox to consider. Average volume of alcohol consumption, patterns of drinking and risk of coronary heart disease: A review. Journal of Cardiovascular Risk. 2003c; 10 (1):15–20. [ PubMed ] [ Google Scholar ]
• Rehm J, Taylor B, Mohapatra S, et al. Alcohol as a risk factor for liver cirrhosis: A systematic review and meta-analysis. Drug and Alcohol Review. 2010b; 29 (4):437–445. [ PubMed ] [ Google Scholar ]
• Rehm J, Zatonski W, Taylor B, et al. Epidemiology and alcohol policy in Europe. Addiction. 2011 [Epub ahead of print] [ PubMed ] [ Google Scholar ]
• Roerecke M, Rehm J. Irregular heavy drinking occasions and risk of ischemic heart disease: A systematic review and meta-analysis. American Journal of Epidemiology. 2010; 171 (6):633–644. [ PubMed ] [ Google Scholar ]
• Romeo J, Warnberg J, Marcos A. Drinking pattern and socio-cultural aspects on immune response: An overview. Proceedings of the Nutrition Society. 2010; 69 (3):341–346. [ PubMed ] [ Google Scholar ]
• Room R, Rossow I. The share of violence attributable to drinking. Journal of Substance Use. 2001; 6 (4):218–228. [ Google Scholar ]
• Samokhvalov AV, Irving H, Mohapatra S, Rehm J. Alcohol consumption, unprovoked seizures and epilepsy: A systematic review and meta-analysis. Epilepsia. 2010a; 51 (7):1177–1184. [ PubMed ] [ Google Scholar ]
• Samokhvalov AV, Irving HM, Rehm J. Alcohol as a risk factor for atrial fibrillation: A systematic review and meta-analysis. European Journal of Cardiovascular Prevention & Rehabilitation. 2010b; 17 (6):706–712. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Samokhvalov AV, Irving HM, Rehm J. Alcohol consumption as a risk factor for pneumonia: A systematic review and meta-analysis. Epidemiology and Infection. 2010c; 138 (12):1789–1795. [ PubMed ] [ Google Scholar ]
• Samokhvalov AV, Popova S, Room R, et al. Disability associated with alcohol abuse and dependence. Alcoholism: Clinical and Experimental Research. 2010d; 34 (11):1871–1878. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Seitz HK, Becker P. Alcohol metabolism and cancer risk. Alcohol Research & Health. 2007; 30 (1):38–47. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Shuper PA, Joharchi N, Irving H, Rehm J. Alcohol as a correlate of unprotected sexual behavior among people living with HIV/AIDS: Review and meta-analysis. AIDS and Behavior. 2009; 13 (6):1021–1036. [ PubMed ] [ Google Scholar ]
• Shuper PA, Neuman M, Kanteres F, et al. Causal considerations on alcohol and HIV/AIDS: A systematic review. Alcohol and Alcoholism. 2010; 45 (2):159–166. [ PubMed ] [ Google Scholar ]
• Taylor B, Irving HM, Baliunas D, et al. Alcohol and hypertension: Gender differences in dose-response relationships determined through systematic review and meta-analysis. Addiction. 2009; 104 (12):1981–1990. [ PubMed ] [ Google Scholar ]
• Taylor B, Irving HM, Kanteres F, et al. The more you drink, the harder you fall: A systematic review and meta-analysis of how acute alcohol consumption and injury or collision risk increase together. Drug and Alcohol Dependence. 2010; 110 (1–2):108–116. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Taylor B, Rehm J, Room R, et al. Determination of lifetime injury mortality risk in Canada in 2002 by drinking amount per occasion and number of occasions. American Journal of Epidemiology. 2008; 168 (10):1119–1125. [ PubMed ] [ Google Scholar ]
• Thavorncharoensap M, Teerawattananon Y, Yothasamut J, et al. The economic impact of alcohol consumption: A systematic review. Substance Abuse Treatment, Prevention, and Policy. 2009; 4 :20. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• World Health Organization (WHO) Preamble to the Constitution of the World Health Organization as adopted by the International Health Conference; New York. 19–22 June, 1946; signed on 22 July 1946 by the representatives of 61 States (Official Records of the World Health Organization no. 2, p. 100) and entered into force on 7 April 1948. Available at http://www.who.int/about/definition/en/print.html ; accessed 2/18/2011. [ Google Scholar ]
• WHO. International Classification of Diseases and Related Health Problems. 10th Revision. Geneva, Switzerland: WHO; 2007. (version for 2007) [ Google Scholar ]
• WHO. The Global Burden of Disease: 2004 Update. Geneva, Switzerland: WHO; 2008. [ Google Scholar ]
• WHO. Global Health Risks. Mortality and Burden of Disease Attributable to Selected Major Risks. Geneva, Switzerland: WHO; 2009. [ Google Scholar ]