Review of Evidence on Alcohol and Health (2025)
Chapter: Summary
Summary
The Dietary Guidelines for Americans (DGA), a joint publication of the U.S. Department of Agriculture (USDA) and the U.S. Department of Health and Human Services (HHS), provide guidance to Americans to help them maintain health, achieve nutrient sufficiency, and help prevent diet-related chronic diseases through healthful dietary patterns. Included in the DGA recommendations is guidance for adults who consume beverages containing alcohol. This DGA guidance on alcohol is included because it is a source of energy for those who consume it and consider it part of their diet, and thus should be taken into consideration as a contributor to total caloric intake. Consumption of alcohol has been linked to a range of health outcomes, including those that are potentially detrimental to health. Thus, the DGA recommend that individuals should not start drinking for any reason and that drinking less is better for health than drinking more. For those who do consume alcohol, the DGA recommend drinking in moderation by limiting intake to two drinks or fewer in a day for men and one drink or fewer in a day for women on days alcohol is consumed. Further, alcohol should not be consumed by some individuals, including for example, those under the legal drinking age or those who are pregnant or lactating. The DGA recommendations are informed by systematic reviews conducted by the Dietary Guidelines Advisory Committee (DGAC) with support from the Nutrition Evidence Systematic Review (NESR) group within USDA.
THE COMMITTEE’S TASK
In 2023, Congress asked USDA to contract with the National Academies of Sciences, Engineering, and Medicine (the National Academies) to undertake an independent review of the evidence on the relationship between alcohol consumption and eight health outcomes previously published by USDA and HHS and reviewed by NESR. The review was limited to the eight questions related to alcohol consumption and health outcomes listed in the statement of task (Box S-1). Additionally, the National Academies was asked to prioritize the evidence and determine whether it was sufficient to support a systematic review that could be used to answer each question. In response to this congressional request, the National Academies empaneled a committee of 14 experts in the areas covering the eight areas of health as specified in the statement of task, as well as systematic reviews, previous experience with the DGA, and public health.
APPROACH TO THE TASK
To approach its task, the committee convened two public information-gathering sessions, including a public comment session. Based on the eight questions from the statement of task, the committee developed search strategies to support evidence scans of the published literature from multiple databases. Because there were sparse publications for the three questions related to lactation, the committee decided these questions should not have a systematic review.
The committee determined that the evidence for each of the other five health outcomes (i.e., weight, cancer, cardiovascular disease, neurocognition, all-cause mortality) was sufficient to conduct a de novo systematic review. An important requirement was to have a comparison group that did not combine never drinkers with former drinkers because of the resulting “abstainer bias” that would occur; therefore, results in this report are not directly comparable to past evidence that does include such abstainer bias. These systematic reviews were registered in the PROSPERO international database for systematic reviews and carried out by the Academy of Nutrition and Dietetics Evidence Practice Center at the request of the committee.
In assessing the evidence, the committee interpreted its task as requiring a focus on data related to moderate alcohol consumption. Although individual studies used terminology variations such as light-to-moderate, the committee adopted the term moderate, which it defined as:
Consuming alcoholic beverages up to the limit defined by the Dietary Guidelines for Americans, i.e., two drinks or 28 grams of alcohol in a day for men and one drink or 14 grams of alcohol in a day for women.
Evidence that met this definition of moderate served as the upper threshold of alcohol consumption that the committee considered when developing its findings and conclusions.
To determine the certainty of its conclusions, the committee used a framework based on the following methods from the U.S. Preventive Services Task Force:
- High certainty: Evidence includes consistent results from good-quality studies in relevant populations assessing effects on health outcomes; the conclusion is unlikely to be affected by future studies. (Note that it is unlikely to be rated as high certainty without a randomized controlled trial).
- Moderate certainty: Evidence is sufficient to determine effects on health outcomes but is constrained by issues raised in the quality assessment of the evidence; additional information from future studies could change the conclusion.
- Low certainty: Evidence is insufficient to assess effects on health outcomes; additional information from future studies may allow for assessment.
Low certainty was concluded when the results of eligible studies were inconsistent or when the data were too sparse. When the level of certainty could not be assigned, the committee determined that no conclusion could be drawn. This determination was made when there was a statistically nonsignificant meta-analysis result or there were no eligible studies.
SYSTEMATIC REVIEWS
With a goal of completing this report in time to inform the 2025–2030 DGA, the committee decided to undertake de novo systematic reviews rather than perform updates and reanalysis of past reviews. To determine whether to request a systematic review of studies published since the last DGA, the committee established a process whereby the committee reviewed articles published within the search time frames.
All-Cause Mortality
According to the Centers for Disease Control and Prevention, heart disease, cancer, accidents, and stroke are the leading causes of death in the United States. Previous research studies have demonstrated that modifiable lifestyle factors, including alcohol consumption, are associated with these causes of death. With respect to alcohol consumption, there is strong evidence that heavy drinking has adverse effects on the risk of these leading
causes of death. However, the association of moderate alcohol consumption with all-cause mortality is less clear.
A NESR systematic review on all-cause mortality was conducted for the DGA 2020–2025; therefore, the search dates for this report were January 2019 to September 2023. Of the 27 included studies, 12 had sufficient data to assess the association of moderate alcohol consumption with all-cause mortality, and 8 of those 12 studies contributed to the overall estimate quantified in a meta-analysis. Risk-of-bias assessment showed concerns attributable to confounding and/or exposure assessment.
Findings
Finding 3-1: On the basis of a meta-analysis of eight eligible studies, there was a 16 percent lower risk of all-cause mortality among those who consumed moderate levels of alcohol compared with those who never consumed alcohol (RR = 0.84, 95%CI [0.81, 0.87]).
Finding 3-2: On the basis of a meta-analysis of three eligible studies, a 23 percent lower risk of all-cause mortality was found among females who consumed moderate amounts of alcohol compared with females who never consumed alcohol (RR = 0.77, 95%CI [0.6, 0.97]). An assessment of four studies showed a 16 percent lower risk of all-cause mortality among males who consumed moderate amounts of alcohol compared with males who never consumed alcohol (RR = 0.84, 95%CI [0.81, 0.88]). The committee found no evidence for a difference in the effect size by sex, as reflected in the p-value of 0.56 for the test for heterogeneity between the sexes.
Finding 3-3: On the basis of a meta-analysis of two eligible studies, a 20 percent lower risk of all-cause mortality was found among persons less than 60 years of age who consumed moderate amounts of alcohol compared with persons less than 60 years of age who never consumed alcohol (RR = 0.80, 95%CI [0.74, 0.86]). An assessment of four eligible studies found an 18 percent lower risk of all-cause mortality among persons 60 years of age or older who consumed moderate amounts of alcohol compared with persons 60 years of age or older who never consumed alcohol (RR = 0.82, 95%CI [0.77, 0.87]). The committee found no evidence for a difference in the effect size by age, as reflected in the p-value of 0.61 for the test for heterogeneity between the age groups. This comparison was not graded for certainty of the evidence.
Finding 3-4: On the basis of a meta-analysis of five studies published between 2019 and 2023, the committee found that, among moderate
alcohol consumers, higher versus lower amounts of moderate alcohol consumption were associated with similar risks of all-cause mortality (RR = 0.96, 95%CI [0.87, 1.06]). The committee also found no evidence for a difference in this effect size by sex, as reflected in the p-value of 0.82 for the test for heterogeneity between the sexes.
Conclusion
Conclusion 3-1: Based on data from the eight eligible studies from 2019 to 2023, the committee concludes that compared with never consuming alcohol, moderate alcohol consumption is associated with lower all-cause mortality (moderate certainty).
Weight Changes
Alcohol consumption may directly or indirectly lead to changes in body weight, body composition, and body mass index (BMI) by providing energy as well as affecting metabolism, appetite, and satiety. Moderate alcohol consumption may have differential effects on weight and adiposity relative to biological sex, age, physical activity level, and other individual-level factors. Genetics also contributes to heterogenous pathophysiological responses to alcohol intake.
Databases searched from January 2010 through February 2024 identified seven eligible studies for a systematic review. A meta-analysis was not conducted due to the heterogeneity in populations, exposures, comparators, outcomes, and study designs. Three studies examined associations between different amounts of moderate alcohol consumption and weight, and two examined associations with BMI. Five studies examined moderate alcohol consumption, and the risk of overweight/obesity, four examined waist circumference, and one study examined waist-to-hip ratio and body fat percentage associations. Of the seven eligible studies, risk of bias was primarily caused by the measurement of alcohol consumption and attrition.
Findings
Finding 4-1: Abstainer bias was evident in all seven eligible studies published between 2010 and 2024; therefore, for weight-related outcomes (weight, BMI, risk of overweight/obesity, waist circumference) comparisons between those who consumed moderate alcohol and those who never consumed alcohol could not be made.
Finding 4-2: On the basis of three eligible studies, there was insufficient evidence to evaluate associations between the amount of moderate alcohol
consumption and changes in body weight among men. Among women, the evidence was inconsistent. There were concerns related to sparse evidence, risk of bias (mainly due to confounding), and imprecision in the studies.
Finding 4-3: On the basis of two eligible studies, higher versus lower amounts of moderate alcohol consumption among men were associated with similar changes in BMI. Among women, the evidence was inconsistent. There were concerns related to risk of bias (mainly due to confounding) and imprecision in the studies.
Finding 4-4: On the basis of four eligible studies, higher versus lower amounts of moderate alcohol consumption among men were associated with similar risks of overweight and/or obesity. Among women, the evidence was inconsistent. There were concerns related to risk of bias, mainly due to confounding, and imprecision in the studies.
Finding 4-5: On the basis of three eligible studies, the evidence for changes in waist circumference comparing higher versus lower amounts of moderate alcohol consumption was inconsistent for women and for men. There were concerns related to sparse evidence and risk of bias (mainly due to confounding).
Conclusions
Conclusion 4-1: The committee determined that there was insufficient evidence to draw a conclusion regarding the association between weight-related outcomes and moderate alcohol consumption compared with never consuming alcohol.
Conclusion 4-2: The committee determined that there was insufficient evidence to draw a conclusion regarding the association between amounts of moderate alcohol consumption and changes in weight.
Conclusion 4-3: The committee concludes that higher versus lower amounts of moderate alcohol consumption among men were associated with similar changes in BMI (low certainty). Among women the evidence was inconsistent regarding changes in BMI.
Conclusion 4-4: The committee concludes that among men who moderately consume alcohol, higher versus lower amounts of moderate alcohol consumption were associated with similar risks of overweight and/or obesity (low certainty). Among women the evidence was inconsistent regarding changes in overweight and/or obesity.
Conclusion 4-5: The committee determined that there was insufficient evidence to draw a conclusion regarding the association between amounts of moderate alcohol consumption and changes in waist circumference.
Cancer
Alcohol has been identified as a carcinogen in humans, although the mechanisms of action about the role of carcinogenesis are not completely understood. The committee identified specific cancers for systematic review—i.e., oral cavity, pharynx, esophagus, colorectum, and female breast—as outcomes of interest based on evidence from previously published reviews. The committee’s systematic review focused on cancer incidence and excluded studies that exclusively examined prevalence, cancer recurrence, cancer-related mortality, or survival. As for all the analyses, studies were excluded that did not specify that only never drinkers were included in the comparison group to prevent abstainer bias.
Studies of the relationship between moderate alcohol consumption and each of bladder, endometrial, gastric, pancreas, prostate, lung, and thyroid cancer, as well as several studies that examined combined sites such as the head and neck or biliary tract and renal tract (14 studies in total), were identified in the evidence scan. A systematic review for these cancer sites was not conducted due to the small number of studies per cancer type. The committee evaluated this body of evidence and determined that there was insufficient evidence to establish certainty for an association of moderate alcohol consumption with any of these other sites.
Based on the scope of primary literature identified in the evidence scans, the committee decided to proceed with a systematic review to answer the question regarding alcohol and cancer incidence. This systematic review included studies published between January 2010 and February 2024.
Findings
Finding 5-1: A meta-analysis of four eligible studies found a 10 percent higher risk of breast cancer among persons consuming moderate amounts of alcohol compared with persons never consuming alcohol (RR = 1.10, 95%CI [1.02, 1.19]). There were some concerns related to risk of bias, mainly due to confounding and exposure assessment, in the studies contributing to this comparison.
Finding 5-2: A meta-analysis of seven eligible studies found a 5 percent higher risk of breast cancer for every 10–14 grams (0.7–1.0 U.S.
drinks) increment of higher alcohol consumption per day (RR = 1.05, 95%CI [1.04, 1.06]). On the basis of two eligible studies, consumption of higher compared to lower amounts of moderate alcohol was associated with a higher risk of breast cancer. One study reported a hazard ratio (HR) of 1.05 (95%CI [1.02, 1.09]) for women who consumed higher amounts of moderate alcohol (0.6–<1.1 drinks/day) compared with those who consumed lower amounts of moderate alcohol 0.2–0.5 drinks/day. Another study reported an HR of 1.06 (95%CI [1.01, 1.11]) for breast cancer associated with 0.4–1.1 drinks per day compared to <0.4 drinks per day. There were some concerns related to risk of bias, mainly due to confounding and exposure assessment.
Finding 5-3: On the basis of five eligible studies and a meta-analysis of three of these studies, compared with never drinkers, moderate alcohol consumption was associated with a statistically nonsignificant higher risk of colorectal cancer overall among males and females. There were some concerns with the studies related to risk of bias, mainly due to confounding and exposure assessment.
Finding 5-4: On the basis of two eligible studies, consumption of higher amounts of moderate alcohol was associated with a higher risk of colorectal cancer. One study reported an HR of 1.09 (95%CI [1.02, 1.17]) for colorectal cancer among males who consumed higher amounts of moderate alcohol (0.7–<2.1 drinks/day) compared with males who consumed lower amounts of moderate alcohol (<0.7 drinks/day). Another study reported a HR of 1.05 (95%CI [1.03, 1.07]) for colorectal cancer associated with each 15 grams (1.1 U.S. drinks) increment of higher alcohol consumption per day. There were some concerns related to risk of bias (mainly due to confounding), exposure assessment, and indirectness stemming from estimating linear trends based on alcohol consumption that may have exceeded the moderate range in some individuals in the latter study.
Finding 5-5: There was insufficient evidence to support an association between moderate alcohol consumption and risks of oral cavity, pharyngeal, esophageal, and laryngeal cancers.
Finding 5-6: Upon evaluating the body of evidence, there were several sites where there was emerging evidence that was insufficient to establish certainty for an association of moderate alcohol consumption. These sites included cancer of the head and neck, thyroid, lung, gastric, small intestine, pancreas, biliary tract, renal track, bladder, prostate, and endometrium.
Conclusions
Conclusion 5-1: The committee concludes that compared with never consuming alcohol, consuming a moderate amount of alcohol was associated with a higher risk of breast cancer (moderate certainty).
Conclusion 5-2: The committee concluded that, among moderate alcohol consumers, higher versus lower amounts of moderate alcohol consumption were associated with a higher risk of breast cancer (low certainty).
Conclusion 5-3: The committee determined that no conclusion could be drawn regarding the association between moderate alcohol consumption compared with lifetime nonconsumers and risk of colorectal cancer.
Conclusion 5-4: The committee concluded that among moderate alcohol consumers higher versus lower amounts of moderate alcohol consumption were associated with a higher risk of colorectal cancer (low certainty).
Conclusion 5-5: The committee determined that no conclusion could be drawn regarding an association between moderate alcohol consumption and oral cavity, pharyngeal, esophageal, or laryngeal cancers.
Cardiovascular Disease
Coronary heart disease and stroke, both forms of cardiovascular disease (CVD), are the first and fifth leading causes of death in the United States, respectively. It is well recognized that modifiable lifestyle factors, including alcohol consumption, may influence the risk of myocardial infarction (MI) and stroke. While heavy alcohol consumption has been associated with a higher risk of MI and hemorrhagic stroke, prior observational studies have suggested that moderate alcohol consumption is associated with a lower risk of CVD.
The evidence scan identified 19 systematic reviews. Eight of the reviews considered CVD outcomes broadly, and the remaining 11 focused on specific CVD outcomes. A subset of studies identified in the scan examined the associations of moderate alcohol consumption with particular care to include people who never consumed alcohol as the comparison group. The committee decided to proceed with a systematic review of associations of moderate alcohol consumption, compared with never consuming alcohol, on the risk of nonfatal MI, nonfatal stroke, and CVD death (referred to as major adverse cardiovascular events [MACE-3]) using studies published from January 2010 through February 2024.
Findings
Finding 6-1: A meta-analysis of two eligible studies found that among persons who consumed moderate amounts of alcohol compared with persons who never consumed alcohol, there was a 22 percent lower risk of MI (RR = 0.88, 95%CI [0.68, 0.90]). No studies reported data for males alone. One study reported a 21 percent lower risk of MI among females only; these results were consistent with the estimate for both sexes combined. There were some concerns related to risk of bias in the studies, mainly due to confounding.
Finding 6-2: A meta-analysis of seven eligible studies found an 11 percent lower risk of stroke among persons consuming moderate amounts of alcohol compared with persons never consuming alcohol (RR = 0.89, 95%CI [0.86, 0.93]). These results were driven by ischemic stroke, which showed a 12 percent lower risk (RR = 0.88, 95%CI [0.86, 0.90]). Separate examination of hemorrhagic strokes was infrequent; thus, no estimate of effect for this health outcome could be made. There were some concerns related to risk of bias among the studies, mainly due to confounding and exposure assessment.
Finding 6-3: A meta-analysis of four eligible studies found an 18 percent lower risk of CVD mortality among persons who consumed moderate amounts of alcohol compared with those who never consumed alcohol (RR = 0.82, 95%CI [0.76, 0.89]). The committee further found a 23 percent lower risk in females (RR = 0.77, 95%CI [0.70, 0.85]), and an 18 percent lower risk in males (RR = 0.82, 95%CI [0.71, 0.94]). Very limited data stratified by age were available; however, one study showed that the effect size and direction for moderate alcohol consumption compared with no alcohol consumption was consistent among persons aged less than 60 years (33 percent lower risk of CVD mortality) and among persons aged 60 years or older (19 percent lower risk of CVD mortality). There were some concerns related to risk of bias, mainly due to confounding, in the studies contributing to this comparison.
Conclusions
Conclusion 6-1: The committee concludes that compared with never consuming alcohol, consuming moderate amounts of alcohol is associated with a lower risk of nonfatal MI (low certainty).
Conclusion 6-2: The committee concludes that compared with never consuming alcohol, consuming moderate amounts of alcohol is associated with a lower risk of nonfatal stroke (low certainty).
Conclusion 6-3: The committee concludes that compared with never consuming alcohol, consuming moderate amounts of alcohol is associated with a lower risk of CVD mortality in both females and males (moderate certainty).
Neurocognition
The mainstay of research on the effects of alcohol consumption on neurocognition stems from investigations of people diagnosed with alcohol use disorder (AUD). By contrast, a paucity of research has examined moderate drinking, often defined by exclusion from AUD criteria. The few studies of moderate drinking that have used objective neuropsychological tests report performance advantages in some areas and impairment in others.
Dementia, Alzheimer’s disease, and cognitive decline were examined longitudinally. Dementia and Alzheimer’s disease were assessed separately because dementia is an umbrella diagnosis that may include Alzheimer’s disease, a diagnosis determined by experts using accepted criteria for dementia. Cognitive decline was determined with quantitative measures of episodic memory, cognitive screening, or phonemic or semantic word fluency.
Findings
Finding 7-1: Four eligible studies with data from 2010 to 2024 reported that the risk of developing dementia was higher among those consuming higher amounts of moderate alcohol than lower amounts of moderate alcohol. One study reported that, when compared with long-term moderate consumers, long-term abstinence or decreasing consumption from midlife to older age was associated with higher risk of dementia. Two studies reported that moderate drinkers had a lower risk of developing dementia than never drinkers, and one study found no association between moderate consumption levels of alcohol and the development of dementia.
Finding 7-2: On the basis of six eligible studies with data from 2010 to 2024, the committee found the risk of Alzheimer’s disease or dementia among those who consumed higher amounts of moderate alcohol versus lower amounts was inconsistent.
Finding 7-3: On the basis of nine eligible studies with data from 2010 to 2024, there was insufficient evidence to support an association between moderate versus never drinking or occasional drinking and the risk of cognitive decline. There were concerns with the studies related
to differences in measurement instruments, differences in comparator groups, and imprecise results.
Conclusions
Conclusion 7-1: The committee concludes there was insufficient evidence about the association between the risk of dementia for those with no alcohol consumption compared to those with moderate alcohol consumption or for those who consume higher versus lower amounts of moderate alcohol.
Conclusion 7-2: The committee concludes there was insufficient evidence regarding the association between amounts of moderate alcohol consumption and the risk of developing Alzheimer’s disease.
Conclusion 7-3: The committee determined that there was insufficient evidence to draw an association between moderate alcohol consumption versus never or occasional consumption and the risk of cognitive decline.
REVIEW OF LACTATION
Human milk provides all essential and conditionally essential nutrients in amounts adequate to meet an infant’s needs. It also provides a complex array of biologically active components, maternal cells, and microbes that contribute enzymatic, hormonal, and immunomodulatory functions to the developing infant. Bioactive components associated with alcohol enter milk after maternal consumption; however, their putative effects on lactation, milk composition, and infant outcomes are understudied, and research results have been inconsistent. Nonetheless, use of alcohol during breastfeeding is generally discouraged.
Because there had not been a systematic literature search by a DGAC on breastfeeding and alcohol since 2010, the committee conducted a systematic search to identify all eligible papers published between January 2010 and April 2024. Among studies identified for review, two were identified in the initial evidence scan, four additional publications were identified in a second systematic search, and one using a hand search. A systematic review with a narrative synthesis of the studies was conducted for any level of alcohol consumption (i.e., not limited to moderate) by the committee in lieu of a systematic review given the sparse literature across the three lactation-related questions in the statement of task.
Findings
No studies published since 2010 addressed the question of maternal alcohol consumption during breastfeeding and postpartum weight loss. Thus, the committee was unable to evaluate this association.
Finding 8-1: There was insufficient evidence to determine any association between maternal alcohol consumption at any level during lactation and milk composition or milk production.
Finding 8-2: There was insufficient evidence to determine an association between maternal alcohol consumption at any level during lactation and infant development.
Conclusions
Conclusion 8-1: The committee determined that no conclusion could be drawn regarding any associations between maternal alcohol consumption during lactation and milk composition or milk production.
Conclusion 8-2: The committee determined that no conclusion could be drawn regarding the association between maternal alcohol consumption during lactation and infant development.
RESEARCH GAPS
Throughout the systematic review of current literature and the preparation of this report, the committee identified a consistent set of research gaps that, when addressed, could strengthen the existing evidence on moderate alcohol consumption and health outcomes. Overarching limitations identified in the committee’s review of evidence include abstainer bias; a lack of standard definitions of alcohol consumption levels and a lack of standardized cutoffs for exposure categories; underreporting of alcohol consumption by participants; lack of data stratified by smoking status, age, sex, and genetic ancestry to evaluate possible interactions between alcohol consumption and health outcomes; and limitations of observational studies. The committee urges that all studies addressing the effects of alcohol consumption on human health speak to these limitations and consider including menopausal status as well as postpartum women (both breastfeeding and nonbreastfeeding) and their infants when possible.
