Version May 2024
INTRODUCTION — While limited alcohol use appears to confer little overall health risk in appropriately selected patients, alcohol intake above recommended guidelines is clearly associated with increased mortality and other medical morbidities. At the same time, alcohol consumption remains one of the most challenging lifestyle factors to study [1]. Because alcohol intake has been subjected to only two randomized trials one year or longer in length [2,3], our understanding of its health effects remains limited by all of the challenges of observational research.
An overview of the risks and benefits of alcohol consumption will be discussed in this topic. Specific issues related to alcohol use in pregnancy and the cardiovascular benefits and risks of alcohol are discussed in detail separately. (See "Alcohol intake and pregnancy" and "Cardiovascular benefits and risks of moderate alcohol consumption".)
The implications of excessive alcohol use and screening for unhealthy use of alcohol are also discussed in detail separately. (See "Risky drinking and alcohol use disorder: Epidemiology, clinical features, adverse consequences, screening, and assessment" and "Screening for unhealthy use of alcohol and other drugs in primary care".)
DEFINITIONS
Standard drink size — Definitions of a "standard drink" differ, both within and among countries [4,5]. For example, a standard drink is generally considered as follows:
●United States – 14 to 15 grams of alcohol (0.5 to 0.6 fl ounces), equivalent to 12 ounces of beer, 5 ounces of wine, and 1.5 ounces of 80 proof liquor [6]
●Great Britain – 8 grams of alcohol
●Japan – 19.75 grams of alcohol [4]
Standard drinks are depicted for the United States in the figure (figure 1). The Australian Government Department of Health and Aged Care has published standard drink size; these are applicable to France, Hungary, Ireland, New Zealand, Poland, and Spain.
Level of drinking — Definitions of "moderate" and "heavy" drinking also vary. In the United States, the following parameters generally apply (table 1) [7-9] (see "Risky drinking and alcohol use disorder: Epidemiology, clinical features, adverse consequences, screening, and assessment", section on 'Terminology'):
●Moderate drinking or (relatively) low risk for alcohol problems:
•Females: ≤1 drink per day (or ≤3 drinks on any single day) and <7 drinks per week
•Males: ≤2 drinks per day (or ≤4 drinks on any single day) and <14 drinks per week
•People age ≥65: ≤1 drink on any single day and <7 drinks per week [10]
●Heavy drinking – At risk for alcohol problems:
•Females: >3 drinks per day or >7 drinks per week
•Males: >4 drinks per day or >14 drinks per week
●Binge drinking
•Females: ≥4 drinks in one drinking occasion
•Males: ≥5 drinks in one drinking occasion
Is there a safe dose of alcohol? — Established recommendations for safe levels of drinking differ as to amount. The 2020 to 2025 Dietary Guidelines for Americans advise no more than two drinks per day for males and one drink per day for nonpregnant females [11], guidelines that largely match those from the National Institute on Alcohol Abuse and Alcoholism [12]. Some experts challenge the view that there is a safe dose of alcohol, and they suggest avoidance of alcohol completely for specific health outcomes [1,13]. (See 'Mortality' below and 'Cancer' below.)
A safe dose of alcohol is likely to be lower in females than in males. Reasons for this include lower body size, percentage of body weight composed of water, and activity of gastric alcohol dehydrogenase among females, resulting in a heavier effective alcohol exposure for a given level of drinking [14-16].
For some patients, no level of alcohol consumption can reliably be regarded as safe. Examples of contraindications to alcohol use include pregnancy, personal or strong family history of alcohol use disorder, or alcohol-associated hepatic or pancreatic disease. (See "Alcohol intake and pregnancy" and "Management of alcohol-associated steatosis and alcohol-associated cirrhosis", section on 'Abstinence'.)
Alcohol use needs to be limited in patients with other conditions, including active gastritis and esophagitis. (See "Medical management of gastroesophageal reflux disease in adults", section on 'Lifestyle and dietary modification'.)
On the other hand, cardiovascular disease remains the most common cause of death in many countries and light alcohol consumption may decrease cardiovascular risk. Any potential benefit from limited alcohol use needs to be weighed against the risk of alcohol consumption. (See "Cardiovascular benefits and risks of moderate alcohol consumption", section on 'Advice to patients regarding alcohol use'.)
BEVERAGE TYPES — The French paradox, that coronary heart disease mortality is lower in France than would be expected from the high national prevalence of smoking and saturated fat intake, has been attributed to frequent red wine consumption [17]. Beverage type and its relationship to cardiovascular risk is discussed in detail elsewhere. (See "Cardiovascular benefits and risks of moderate alcohol consumption", section on 'Beverage type'.)
Whether beverage type matters for specific diseases other than coronary heart disease is still uncertain. Most evidence regarding alcohol and cancer suggests that beverage type does not substantially alter risk [18].
Resveratrol (3,5,4'-trihydroxystilbene) is a substance produced by plants in response to stress and is found in grape skin and red wine. Resveratrol extends the life of nonmammalian organisms and was demonstrated to improve the metabolic profile and lifespan of mice fed high-fat diets [19]. The resveratrol concentration of red wine is 1.5 to 3 mg per liter. A 70-kg person would need to consume over 1000 liters of red wine per day to achieve the resveratrol concentrations in this study. In addition, a clinical trial of resveratrol among 24 patients with multiple myeloma demonstrated minimal efficacy and an unacceptable safety profile, with five cases of renal failure [20]. Although resveratrol supplements are available, doses are not standardized and side effects are unknown; use of resveratrol supplements is not advised.
ALCOHOL EFFECT ON OVERALL HEALTH
Quality of life — Problem drinking reduces the quality of life for patients and their loved ones. (See "Risky drinking and alcohol use disorder: Epidemiology, clinical features, adverse consequences, screening, and assessment".)
However, few studies have evaluated how moderate alcohol use affects quality of life. Two European cross-sectional studies found that self-reported health was best among subjects who consumed one to two drinks per day (compared with either higher or lower use) [21,22]. A similar relationship between amount of alcohol consumed and use of sick leave was also observed [23]. As noted above, abstainers may be prone to poor subjective health even in their youth [24]. A study of middle-aged nurses who repeatedly reported on their quality of life found small bidirectional positive associations between moderate alcohol intake and physical health-related quality of life, with slightly higher future quality of life among moderate drinkers compared with abstainers [25]. (See 'Limitations of the evidence' below.)
Disability — Study findings have come to different conclusions about the contribution of alcohol use to the risk of becoming disabled. An observational study using data from successive National Health and Nutrition Examination Surveys of residential-dwelling adults aged 50 years and older found a favorable effect of some level of alcohol consumption among adults who reported good or better health [26]. Consuming <15 drinks per week AND <5 drinks per day for males or <4 per day for females was associated with 25 percent lower risk for disability, compared with abstention. However, in a global study that included evaluation of the effect of alcohol on disability, alcohol use (combined across all levels of consumption) was a leading risk factor for disability [27].
Mortality — The relationship between alcohol use and mortality differs according to the amount consumed. Excessive alcohol consumption is linked to excess mortality. (See "Risky drinking and alcohol use disorder: Epidemiology, clinical features, adverse consequences, screening, and assessment", section on 'Mortality'.)
Results of studies about the effect of moderate drinking on mortality rates are mixed. Several studies have found that moderate drinkers generally have lower mortality rates compared with abstainers or heavy drinkers with a risk of death that is generally 10 to 30 percent lower for moderate drinkers compared with abstainers [28-31]. This finding has been observed in prospective studies from around the world, including the United States [28,29,32], Great Britain [33], Italy [34], Trinidad [35], Japan [36], and China [37], among many others. For reasons that are not entirely clear, the mortality curve for alcohol consumption shifts according to the drinking habits of the country studied [38].
However, not all studies have found a mortality benefit for moderate alcohol consumption [27,39,40]. As examples:
●In a pooled analysis including 83 prospective studies, the lowest mortality risk was found in those who consumed <100 g/alcohol per week (approximately eight drinks, using the per-drink estimate from that study) [41]. Above this threshold, mortality increased roughly linearly with increased alcohol consumption. When accounting for drinking pattern, there was a lower total mortality risk among individuals who spread their intake over three or more days per week, with the lowest risk among consumers of 51 to 75 grams of alcohol per week.
●In a meta-analysis of 107 cohort studies involving more than 4.8 million participants, there was no reduction in all-cause mortality among those with low or moderate daily alcohol intake (less than 25 grams of ethanol per day) compared with lifetime nondrinkers [42]. There was, however, an increased risk at higher consumption levels (25 or more grams per day for females and 45 or more grams per day for males).
Studies in older adults have also found mixed results. In a study of over 24,000 United States adults >50 years [39], no level of alcohol consumption was associated with reduced all-cause mortality [43]. However, in another study in United States adults >55 years, those who had one drink per day had a lower mortality compared with abstainers (odds ratio 0.72, 95% CI 0.57-0.91) [44].
Among studies that found a benefit of alcohol on mortality risk, the dose associated with the lowest mortality was around 6 grams per day for males (approximately a drink every other day) and around 2 to 6 grams per day for females [28,32,45].
ALCOHOL EFFECT ON SPECIFIC CONDITIONS
Cardiovascular disease — The effects of alcohol consumption on cardiac disease, hypertension, peripheral vascular disease, and stroke are discussed separately. (See "Cardiovascular benefits and risks of moderate alcohol consumption".)
The association of alcohol use and atrial fibrillation is discussed separately. (See "Epidemiology, risk factors, and prevention of atrial fibrillation", section on 'Alcohol'.)
Cancer — Alcohol use has been associated with increased risk for several types of cancer, with heavy use being associated with the highest risk [31,46-49]. For example, in a 2015 meta-analysis of 572 observational studies, the relative risk (RR) of cancer with heavy alcohol use compared with no or occasional alcohol use was greatest for upper aerodigestive cancers (oral and pharyngeal [RR 5.13, 95% CI 4.31-6.10], esophageal [4.95, 95% CI 3.86-6.34], and laryngeal [2.65, 95% CI 2.19-3.19]), with a lesser elevation in risk for several other cancer types, including screen detectable cancers (breast and colorectal). Further, these associations generally had a dose-risk relationship [50]. Heavy drinkers also had a higher risk of certain other cancers of the digestive system and lungs. An earlier meta-analysis had suggested that the pooled relative risk for cancer mortality also progressively increased with quantity of alcohol use among males; the same relationship was not observed in females [51]. A population-based cohort study in the United States found higher risks of incident cancer and cancer-related mortality with greater alcohol consumption, although a combined endpoint of cancer or all-cause mortality exhibited a J-shaped relationship with alcohol [52].
Even among light and moderate drinkers, some studies have indicated a trend toward an increased cancer risk, regardless of smoking history [53]. In particular, among females who had never smoked, just one alcoholic drink per day has been associated with an increased risk of alcohol-related cancer (RR 1.13, 95% CI 1.06-1.20), mainly driven by an association with an increased risk for breast cancer, but no significantly increased risk was observed among males.
The American Society of Clinical Oncology advises that alcohol drinking is an established risk factor for several malignancies and is a potentially modifiable risk factor for cancer [54]. Furthermore, 2020 guidelines from the American Cancer Society advise complete avoidance of alcohol specifically for cancer prevention due to an increased cancer risk evident even with consumption of less than one drink per day [13].
Breast cancer — There is consistent evidence that breast cancer risk is higher for individuals consuming both low (<1 drink per day) to high (≥3 drinks per day) levels of alcohol compared with abstainers [50,53,55-58]. There appears to be a significant dose response relationship beginning with intakes as low as three to six drinks per week. A history of heavy episodic ("binge") drinking, especially among moderate alcohol users, seems to confer risk among females who have a sister with a history of breast cancer [58].
A 2013 meta-analysis of 110 epidemiologic studies showed a small but significant association between female breast cancer and light alcohol intake (RR 1.05, 95% CI 1.02-1.08) [59]. A large cohort study examining the relationship between alcohol and breast cancer also found a small increased risk of breast cancer with three to six drinks per week compared with abstainers (RR 1.15, 95% CI 1.06-1.24) [57]. There was a 10 percent increase in risk with each 10 grams per day of alcohol intake. Breast cancer risk was linearly correlated with cumulative lifetime alcohol intake and was most strongly associated with drinking patterns in both early and later adult life. In addition, in an analysis of breast cancer incidence among 38,500 females participating in the Women's Health Study, daily alcohol intake again was shown to modestly increase risk [60]. The relative risk for each 10 gram increase in daily alcohol intake was 1.11 (95% CI 1.03-1.20) for estrogen receptor (ER)- and progesterone receptor (PR)-positive breast cancer; the difference in risk was not statistically significant for ER- and PR-negative breast cancer. There has also been an association with alcohol consumption and luminal A and human epidermal growth factor receptor 2 (HER2)-amplified breast cancer [61].
The proportion of breast cancer cases attributed to alcohol intake varies widely. In the United States, the population attributable risk is estimated to be 2 percent [62]. In Italy, where alcohol intake is higher, alcohol has been estimated to account for 11 percent of breast cancer cases [63]. In the European Prospective Investigation into Cancer and Nutrition (EPIC) study of over 250,000 females in eight European countries, the attributable risk for breast cancer was 5 percent [64]. In large part, the increased risk occurred in individuals who drank alcohol in excess of the recommended upper limit.
The effect of alcohol consumption on breast cancer risk may be modified by hormone therapy or intake of folate:
●Combination of alcohol and hormone therapy – An additive risk for the combination of postmenopausal hormone therapy and alcohol intake was reported in the Nurses' Health Study [65]. Compared with females who neither took hormones nor drank alcohol, the risk of breast cancer was significantly increased approximately 1.3-fold in females who either drank alcohol or took postmenopausal hormone supplements but was twofold higher in those who drank more than one alcoholic drink per day and used hormones for five or more years. (See "Menopausal hormone therapy and the risk of breast cancer".)
●Association with folic acid intake – Folic acid intake may attenuate the effect of alcohol consumption on breast cancer. Observational data suggest that females who consume alcohol should also take a daily multivitamin fortified with folic acid [66]. It is uncertain if supplemental folate intake is necessary in the United States, where grain is fortified with folic acid.
In the Nurses' Health Study, consumption of one or more drinks per day was associated with a nonsignificant 5 percent increase in risk of breast cancer among females who consumed at least 300 micrograms of folic acid per day [67]. By contrast, alcohol use was associated with a 32 percent increased risk of breast cancer in those with lower folate intake. A follow-up nested case-control study looked at plasma levels of folate and found that among females who consumed one or more drinks per day, the RR for breast cancer was 0.11 (95% CI 0.02-0.59) among those in the highest compared with lowest quintiles of plasma folate [68]. Another follow-up study showed an association between alcohol consumption and breast cancer in individuals aged 27 to 44 years with a positive family history of breast cancer and folic acid intake of <400 micrograms per day (multivariate hazard ratio [HR] 1.82, 95% CI 1.06-3.12) [69].
Similarly, the Iowa Women's Health Study found that folate intake can modify the higher risk of breast cancer in females who drink alcohol [70]. In contrast to some of the studies described above, in this study the higher risk of breast cancer among regular drinkers (and thus the protective effect of folate) was limited to ER-negative tumors [71]. The reason for this disparate result is not clear.
A number of biologic mechanisms may explain the association between breast cancer and alcohol intake, including increased circulating estrogens and androgens, enhancement of mammary gland susceptibility to carcinogenesis, increased mammary carcinogen deoxyribonucleic acid (DNA) damage, and greater potential for invasiveness of breast cancer cells [55,72].
Alcohol consumption has been associated with both ER- and PR-positive breast cancer. A cohort study of 51,847 postmenopausal Swedish females demonstrated an association between increased alcohol intake and ER-positive breast cancer, regardless of PR status (RR 1.35, 95% CI 1.02-1.80) [73]. However, a subsequent study using data from the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial of over 54,000 females aged 55 to 74 found that, compared with abstainers, those who consumed ≥7 drinks per week had an increased risk of ER-positive/PR-positive breast cancer (HR 1.63, 95% CI 1.30-2.05) but not ER-positive/PR-negative or ER-negative/PR-negative breast cancer [74]. Additionally, the Women's Health Initiative found a positive association of alcohol intake with ER-positive breast cancer but an inverse association with the 300 cases of triple-negative breast cancer (RR 0.57, 95% CI 0.34-0.95) [75].
Gastrointestinal cancer — Several types of gastrointestinal cancer are linked to alcohol consumption, even at low levels of intake, and studies suggest that both the amount and frequency of consumption are associated with increased risk.
●In a Korean population-based cohort study including almost 12 million adults, among those who consumed alcohol, increased frequency of alcohol consumption was associated with an increased incidence of several gastrointestinal malignancies, including esophageal, gastric, colorectal, biliary, and pancreatic cancers [76].
●In a study including 226,000 United States males, the combined mortality rate for cancers of the mouth, larynx, pharynx, esophagus, and liver was 40 percent higher in less-than-daily drinkers compared with abstainers and rose progressively with heavier consumption [28].
●In the large EPIC study, the alcohol-related attributable risk for males and females, respectively, was 44 and 25 percent for upper gastrointestinal cancer, 33 and 18 percent for hepatocellular cancer, and 17 and 4 percent for colorectal cancer [64]. As with breast cancer, the increased risk in large part was found in individuals who drank more than the recommended upper limit.
The following gastrointestinal cancers are associated with alcohol consumption:
●Esophageal – The risk of squamous cell esophageal cancer is related to alcohol use [77,78]. However, alcohol does not appear to be associated with increased risk for esophageal adenocarcinoma [78,79]. This is discussed in detail elsewhere. (See "Epidemiology and pathobiology of esophageal cancer", section on 'Smoking and alcohol' and "Epidemiology and pathobiology of esophageal cancer", section on 'Alcohol'.)
●Colorectal – The incidence of colorectal cancer may be related to alcohol use, but the association appears to be weak, particularly for limited alcohol use, and study results have varied [28,56,80-84]. Most of the studies report an association between moderate or heavy alcohol use and colorectal cancer [50,85]. (See "Epidemiology and risk factors for colorectal cancer", section on 'Alcohol'.)
●Pancreatic – Studies about the impact of alcohol use on pancreatic cancer incidence have shown varying results; the risk, if present, seems related to heavy drinking. One study of 34,000 females found a trend toward increased risk of pancreatic cancer with heavier alcohol consumption (p = 0.11) [86], but many other reports have found no such relationship [87-90]. Two case-control studies found that only heavy drinking (over five drinks per day) was associated with pancreatic cancer [91,92].
●Hepatocellular – Hepatocellular carcinoma has been linked to alcohol use. Much of this association may be mediated by alcohol-related cirrhosis since cirrhosis of any cause is a major cause of this malignancy. Alcohol, possibly even at low levels of consumption, also exacerbates the increased risk of hepatocellular carcinoma incurred by carriers of hepatitis B or hepatitis C [93,94]. (See "Hepatitis C and alcohol" and "Epidemiology and risk factors for hepatocellular carcinoma", section on 'Lifestyle factors'.)
In subgroup analysis, several case-control studies have found little risk of hepatocellular carcinoma among mild drinkers when distinguished from heavy drinkers [95-97]. A large prospective study of middle-aged females found no increased risk for liver cancer among consumers of three to six drinks per week [56]. Although there was a 20 percent higher risk among those drinking 7 to 14 drinks per week, this was not statistically significant. However, risk was significantly increased by 70 percent among females consuming more than 14 drinks per week.
Head and neck cancer — Alcohol consumption is a risk factor for head and neck cancer. Patients who smoke and drink are at greater risk than would be expected from either factor alone [98]. This is discussed in detail elsewhere. (See "Epidemiology and risk factors for head and neck cancer", section on 'Tobacco products' and "Epidemiology and risk factors for head and neck cancer", section on 'Alcohol'.)
Lung cancer — Data are unclear as to whether alcohol intake increases risk for lung cancer. Since there is a strong association between smoking and drinking throughout the world, studies may be confounded if not adequately controlled for tobacco use. One study of 4265 participants in the original Framingham Study cohort and 4793 participants in the Framingham offspring cohort compared risks for lung cancer with levels of drinking [99]. Alcohol intake in these cohorts was mostly light to moderate. No relationship between alcohol and lung cancer was found after adjustment for cigarette use. A 2002 meta-analysis also found no evidence of increased cancer risk except for individuals with very high alcohol intakes (>2000 grams per month or five to six drinks per day) in whom there was still the strong possibility of confounding by cigarette smoking [100]. However, a study using data from over 490,000 individuals from the National Institute of Health/American Association of Retired Persons Diet and Health found that after adjusting for tobacco smoking, compared with no drinking, alcohol consumption of <1 drink per day was associated with a decreased risk for lung cancer (HR 0.89, 95% CI 0.82-0.96), while drinking ≥7 drinks/day may be associated with a modest increased risk (HR 1.11, 95% CI 1-1.24) [101]. The study also suggested that the relationship between alcohol consumption and lung cancer may differ by histologic subtype.
Liver disease — Alcohol-related liver disease is described in detail separately. (See "Pathogenesis of alcohol-associated liver disease" and "Alcoholic hepatitis: Clinical manifestations and diagnosis", section on 'History'.)
Many epidemiologic studies have found that liver disease is unusual among moderate drinkers [35-37]. By contrast, a study of 490,000 adults reported relative rates of death from cirrhosis and alcohol use disorder of 1.2 to 1.5 for those who consumed one drink per day and 2.1 to 2.6 for those who consumed two to three drinks per day [28]. (See "Management of alcohol-associated steatosis and alcohol-associated cirrhosis".)
Some of the risk associated with moderate drinking may arise from its interaction with other hepatotoxic agents, particularly hepatitis C. Hepatitis C and alcohol are discussed in detail elsewhere. (See "Hepatitis C and alcohol".)
Cholelithiasis — Moderate alcohol intake lowers the risk of gallstones [102,103]. In the Nurses' Health Study, females who consumed at least two to three drinks per week were 40 percent less likely to develop symptomatic gallstones than abstainers [102]. Other studies have extended this observation to males [104-107]. The mechanism of this protective benefit may be an alcohol-induced reduction in the biliary cholesterol saturation index [108].
Alcohol consumption that is heavy enough to cause liver disease may reverse this benefit. This is suggested by studies that have found a high prevalence of gallstones in patients with alcohol-related cirrhosis [109].
Pancreatitis — Heavy alcohol use predisposes to both acute and chronic pancreatitis. (See "Etiology of acute pancreatitis" and "Etiology and pathogenesis of chronic pancreatitis in adults".)
Mild alcohol consumption seems to carry an increased RR of chronic pancreatitis relative to abstainers, especially in patients with underlying hypertriglyceridemia, although the absolute risk appears to be low [110,111] and not all studies have found elevated risk in moderate drinkers [112]. Whether mild alcohol consumption exacerbates the risk of pancreatitis in patients with other risk factors for chronic pancreatitis, such as cystic fibrosis transmembrane conductance regulator (CFTR) gene mutations, is unknown.
Renal dysfunction — Moderate alcohol consumption does not appear to be substantially associated with higher or lower risk of renal dysfunction. In the Physicians' Health Study involving 11,023 initially healthy males, there was a trend toward decreased risk of renal dysfunction with greater consumption of alcohol [113]. Other studies have not found such an association [114,115].
Hematologic complications — Hematologic complications of alcohol use are described separately. (See "Hematologic complications of alcohol use".)
Gout — The association between alcohol use and the risk of gout and gout flares are described in detail separately. (See "Nonpharmacologic strategies for the prevention and treatment of gout".)
Osteoporosis — Heavy alcohol use predisposes to hip fracture by causing both osteoporosis and falls [116,117], although the data on moderate use are mixed. Moderate alcohol use is associated with increased bone mineral density in many studies [118-122], perhaps related to higher endogenous estrogen levels [123]. On the other hand, studies of actual fractures variously show more fractures [124-126], fewer fractures [127-130], or no association [131]. A 2008 meta-analysis of observational studies confirmed an apparently nonlinear association of alcohol consumption with osteoporotic fracture but a linear association with bone mineral density [132]. The disparity in the relationships of bone mineral density and fracture risk with alcohol use could relate to an increased risk of trauma or falls that offsets the benefit related to osteoporosis itself. However, a study of alcohol intake and falls among older adults found an increased fall risk only among those who consumed 14 or more drinks per week [133]. (See "Screening for osteoporosis in postmenopausal women and men", section on 'Risk factor assessment' and "Evaluation and treatment of premenopausal osteoporosis", section on 'Screening'.)
Diabetes mellitus — The risk of diabetes mellitus is decreased in people with moderate alcohol consumption [134,135]. A 2005 meta-analysis of 15 cohort studies also found a decreased risk for diabetes among light to moderate, but not heavy, alcohol users. Compared with abstainers, the relative risk of diabetes for those who consumed 6 to 12, 12 to 24, 24 to 48, and ≥48 grams of alcohol per day were 0.70 (95% CI 0.61-0.79), 0.69 (95% CI 0.58-0.81), 0.72 (95% CI 0.62-0.84), and 1.04 (95% CI 0.84-1.29) [135].
A randomized crossover trial in 51 postmenopausal females found that, compared with no alcohol consumption, 30 grams of alcohol per day (two drinks) for eight weeks lowered serum insulin levels while leaving glucose levels unchanged and thus improved insulin sensitivity [136]. Triglyceride levels also decreased during exposure to alcohol. This study provides a plausible mechanism for the lower risk of diabetes associated with moderate drinking.
Three randomized trials in diabetic individuals confirmed the antihyperglycemic effects of moderate drinking. One study in Israel found that one glass of wine per day (red or white) reduced fasting blood sugar by almost 20 mg/dL [137]. A similar trial among 115 post-myocardial infarction diabetic patients found that drinking one glass of red wine per dat for one year was associated with lower fasting insulin (suggesting better insulin sensitivity), lower levels of oxidative stress and inflammatory markers, and a better ejection fraction [2]. One potential mechanism for these effects is the impact of alcohol on postprandial hyperglycemia. In a feeding study among healthy young adults, alcohol given prior to or with a meal reduced the degree of subsequent glycemia [138]. In the longest study of its kind, 224 Israeli subjects with diabetes were randomized to sparkling water, white wine, or red wine daily for two years. Compared with those who drank water, subjects who drank white wine had reduced fasting glucose and insulin resistance; red wine demonstrated nonsignificant effects in the same direction [3]. A meta-analysis of 14 studies also found that in nondiabetic patients, compared with controls, alcohol consumption was associated with reduced hemoglobin A1c and fasting insulin concentrations [139].
Another potential link between moderate drinking and diabetes is the effect of alcohol on adiponectin, an adipocyte hormone which directly improves insulin sensitivity in animal models [140]. Both clinical trials and population studies show that moderate drinking is associated with higher adiponectin levels [141,142].
An important qualifier to the association of alcohol consumption with diabetes is drinking pattern. One cohort study of 312,388 current drinkers found that consuming alcohol with meals was associated with a 12 percent lower risk of diabetes (HR 0.88, 95% CI 0.83-0.93) than was consuming alcohol outside of meals, even adjusting for amount of alcohol consumed. In addition, a dose-dependent lower risk of diabetes with greater alcohol consumption was only observed among participants who consumed alcohol consistently with meals but not in others [143].
Mild to moderate alcohol consumption also may decrease the risk of death due to coronary heart disease in older diabetics [134]. In one study of 983 diabetics with a mean age of 69 years, the RR of coronary heart disease mortality compared with never-drinkers was 0.54 for those who drank less than 2 grams per day (less than one drink per week), and 0.21 for those who drank 14 or more grams per day (approximately one drink or more per day) [144].
On the other hand, alcohol may worsen diabetic neuropathy [145] and, anecdotally, may induce severe hypoglycemia [146]. In a small study (n = 16) of type 1 diabetics, a modest amount of alcohol with the evening meal increased the risk for delayed hypoglycemia [147].
The effect of alcohol on cardiovascular risk in patients with diabetes is discussed in detail elsewhere. (See "Cardiovascular benefits and risks of moderate alcohol consumption", section on 'In patients with diabetes mellitus'.)
Accidents and trauma — Alcohol is implicated in the morbidity and mortality from trauma [148]. Since most of the trauma risk associated with alcohol is acute, exposure to alcohol is generally measured in blood alcohol concentration (BAC) rather than drinks per day or week. In most of the United States, the legal BAC limit for driving is 0.08 percent. This corresponds to approximately four drinks for a 200 pound man but only 2.5 drinks for a 150 pound woman.
However, the risk of involvement in a collision while driving doubles at a BAC of only 0.05 percent [149]. Furthermore, simulated driving ability is impaired with BACs as low as 0.02 percent (figure 2) [150]. Alcohol intake is also associated with a greater severity of injury in motor vehicle accidents [151]. The risk of driving accidents is greatest in the first two years of exposure to alcohol [152].
Whether consuming one or two drinks leads to an increased risk of injury is unclear, but several lines of evidence suggest that it does. As noted above, BACs of 0.02 to 0.05 percent impair the ability to operate motor vehicles; this level of blood alcohol can occur with even a single drink. An analysis of 1150 respondents from the 1990 National Alcohol Survey suggests that the risk of injury increases with even one drink per day [153]. In summary, no safe level of alcohol use exists for the use of potentially dangerous equipment.
The risk of injuries during operation of boats [154], bicycles [155,156], and snowmobiles [157] is increased with alcohol use. Falls, drowning, burns, hypothermia, and occupational injuries are more prevalent in drinkers, particularly heavy drinkers [158]. The United States Federal Aviation Administration has established a BAC offense level of 0.04 percent for pilots in response to data demonstrating impairment in simulated flying ability at this level and above [159].
Violence — Alcohol use increases the risk of violence. Alcohol is involved in one-half to two-thirds of all homicides, at least one-half of serious assaults, and more than one-quarter of all rapes [160]. (See "Ethanol intoxication in adults", section on 'Introduction' and "Intimate partner violence: Epidemiology and health consequences", section on 'Risk factors' and "Peer violence and violence prevention", section on 'Alcohol/drug use'.)
Suicide — Alcohol abuse is associated with an increased risk of suicide [161]. However, moderate drinking does not appear to increase suicide risk [162]. As an example, in a prospective study of over 128,000 enrollees in a Californian-managed care plan, intake of six or more drinks per day was associated with a sixfold higher risk of suicide, but more moderate drinking was not linked to risk [163]. In a prospective study of 49,000 young Swedish males, death from suicide or probable suicide was linearly related to the level of alcohol consumption [164]. A prospective study of 47,000 middle-aged and older males found that risk increased with the quantity of alcohol consumed per occasion and not the frequency of intake [165]. Even in individuals with alcohol use disorder, suicide in most cases is thought to require superimposed episodes of depression rather than the substance use disorder alone [166].
Acute alcohol use is often a factor in suicide attempts. In a case-crossover study of 102 patients seen in seven emergency departments for suicide attempt, the RR for suicide attempt in the six hours following acute ingestion, compared with each patient's usual alcohol consumption, was 9.6 (95% CI 5.7-16.3) [167].
Suicide and alcohol use are described separately. (See "Suicidal ideation and behavior in adults", section on 'Psychiatric disorders'.)
Dementia — Alcohol-related dementia, alcohol-related cerebellar degeneration, Wernicke-Korsakoff syndrome and other chronic neurologic manifestations of excessive alcohol use are described separately. (See "Overview of the chronic neurologic complications of alcohol".)
Given inconsistencies in existing data, as well as limitations inherent in observational studies, the relationship between alcohol use and neurodegenerative dementia remains uncertain. Several studies have demonstrated either a lower risk of dementia or a lesser decline in cognitive function over time with mild to moderate alcohol consumption compared with abstention [168-173]. One study found this relationship for older females but not males [174]. Another study found that this relationship was not sustained when the analysis controlled for premorbid intelligence and physical health [175]. In two studies, wine was associated with a decreased risk of dementia, while other beverages were associated with unchanged or slightly increased risk [176,177]. In one meta-analysis of prospective studies, the lowest risk of dementia (RR 0.9) was associated with consumption of approximately a half-drink (6 grams) per day while excessive drinking (at least 38 grams per day) was associated with increased dementia risk [178]. A study that selected 527 subjects from over 10,000 followed for over 30 years found identical rates of cognitive decline in abstainers and moderate drinkers but higher rates in heavy drinkers [179,180].
Carriers of the apoE genotype were shown to be at increased risk for dementia with increased alcohol consumption in one study [181] but not in another [172].
Pregnancy — Pregnancy, alcohol, and fetal alcohol spectrum disorder are discussed in detail elsewhere. (See "Alcohol intake and pregnancy" and "Fetal alcohol spectrum disorder: Clinical features and diagnosis".)
LIMITATIONS OF THE EVIDENCE — No long-term randomized trials of alcohol administration exist. Our knowledge of the effects of moderate alcohol consumption in humans is derived primarily from two sources: short-term trials analyzing the effect of alcohol on physiologic measures and observational studies comparing moderate drinkers with abstainers. Both of these sources have limitations.
While short-term trials of alcohol intake are more easily controlled, they focus on intermediate measures (such as hemostatic parameters, lipid levels, or cognitive function) rather than disease outcome [182-184]. They also do not inform about the balance between alcohol's long-term risks and benefits, since alcohol use extends over decades. Furthermore, these trials tend to be performed using healthy, young, White, male subjects, and the results cannot necessarily be extrapolated to other populations.
The main limitation of long-term observational studies that compare abstainers with alcohol users is that alcohol use is not distributed randomly among individuals. Simply adjusting for known, measurable differences between abstainers and drinkers may be inadequate to make the two groups comparable. People who abstain are different than those who drink in more ways than how much alcohol they consume. Studies that compare abstainers with other drinkers may be confounded by the following factors:
●Cessation of drinking with illness – Whether they realize it or not, people who are ill may stop drinking. Thus, the group of healthy abstainers may be diluted with former drinkers at high risk for illness and death. The British Regional Heart Study found that a large proportion of abstainers were former drinkers with high rates of high blood pressure, chronic bronchitis, and coronary heart disease [185]. Also, patients with a wide range of chronic diseases may avoid alcohol to prevent medication interactions [186].
●Inclusion of those with a previous alcohol use disorder – People with alcohol use disorder who no longer consume alcohol rarely return to social drinking [187]. They are more likely to completely abstain from alcohol. People may not admit to having a problem with alcohol as the reason behind their current abstention, since it carries a strong social stigma.
●Abstention as an indicator of underlying emotional or physical problems – The decision to avoid alcohol may itself reflect important facts about one's physical and emotional health. Observational data suggest that abstainers differ from drinkers well before differences in specific diseases emerge.
In a prospective study of males from adolescence into older age, 33 percent of lifelong abstainers had poor physical health in childhood compared with 14 percent of moderate drinkers [188]. In addition, only 16 percent of abstainers came from warm childhood homes versus 35 percent of moderate drinkers. Similarly, an analysis of 9605 individuals at age 33 demonstrated that abstainers had double the risk of psychological distress or limiting illness than light or moderate drinkers, regardless of gender [24].
●Moderate alcohol use may indicate resistance to alcohol use disorder – People who are consistently able to drink in moderation are, by definition, not abusers of alcohol. They have undergone a "diagnostic challenge" for alcohol use disorder and passed. The same cannot be said for all lifelong abstainers. If abstainers were to initiate alcohol consumption, some would develop problem drinking. It may be difficult to generalize the results of observational studies comparing moderate drinkers and abstainers. This problem does not apply to studies comparing infrequent with moderate drinkers.
●Inaccuracies of reported alcohol use – Typically, these studies classify subjects based on self-reported alcohol intake, with potential inaccuracy in the reporting of alcohol exposure. Additionally, subjects may change their drinking patterns over the study period [189]. Although alcohol consumption is still much better reported than many other dietary factors [190], inaccuracy limits our ability to identify optimal or safe levels of drinking.
●Confounding by other lifestyle factors – Multiple lifestyle factors (eg, educational status, tobacco use, exercise, diet) are associated with patterns of alcohol use and thus confound the results when alcohol intake is presumed to be linked to a particular outcome. Most studies attempt to adjust for known confounders, but it is difficult to identify or correct for all confounding factors.
A review of data from over 250,000 respondents to the 2003 Behavioral Risk Factor Surveillance System (BRFSS) compared cardiovascular disease-associated risk factors in moderate drinkers (two drinks per day for males and one drink or fewer per day for females) with nondrinkers [191]. Nondrinkers had significantly higher rates of multiple known risk factors for cardiovascular disease, including no leisure-time physical activity, low overall physical activity, diabetes, hypertension, body mass index >30, and fair or poor overall health status. However, many of these differences were attenuated or even reversed when adjusting for basic demographic differences between abstainers and drinkers [192].
In the United States, moderate alcohol use is more common among those with higher socioeconomic status. Because of the profound effect of socioeconomic status on health, its potential for confounding the observed effect of alcohol is considerable. In one study, current drinkers were 15 to 25 percent more likely to have attended school beyond grade 12 than were lifelong nondrinkers [28]. In a 1998 meta-analysis of 10 population studies, lifelong abstainers were of lower socioeconomic status than light drinkers [193].
●Ascertainment bias – It is possible that some of the findings related to the association of drinking and cancers is due to ascertainment bias. For example, the propensity of moderate alcohol drinkers to seek medical care (eg, screening tests for cancer) more often than nondrinkers would increase the likelihood that drinkers will be found to have a screen-detectable cancer (eg, breast or colon cancer).
Support for this concern comes from the 2002 to 2012 BRFSS cross-sectional survey of >2,000,000 Americans, in which moderate alcohol drinkers reported a greater likelihood of being screened with mammography or colonoscopy than did abstainers [194]. By contrast, in an observational study that used a mandated screening protocol, breast cancer incidence was not higher among alcohol users [195]. Further, in this study moderate alcohol use (>0 to 1 drink per day) was associated with a lower risk of colon cancer (hazard ratio [HR] 0.35, 95% CI 0.17-0.73) compared with abstention from alcohol.
●Patterns of drinking – Epidemiologic studies commonly ascertain alcohol consumption in drinks per week or similar averaged cumulative measures. Other investigators separate drinking frequency from quantity consumed per drinking day, which is important given recommended limits to drinking. Strongly preferred are tools that assess at least three dimensions of alcohol intake (frequency, quantity per usual drinking day, and heavy episodic drinking) such as the Alcohol Use Disorders Identification Test-Concise (AUDIT-C) [196]. Without this level of detail, infrequent heavy consumption and regular, limited consumption can be easily confused.
●Other factors – Individuals may abstain from alcohol because they dislike the taste. Others may abstain for personal, religious, or cultural reasons.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Alcohol consumption".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Beyond the Basics topic (see "Patient education: Risks and benefits of alcohol (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Is there as safe "dose" of alcohol? – In patients without comorbidities, federal guidelines in the United States advise no more than two alcoholic drinks per day (figure 1) for males and one per day for nonpregnant females (table 1). Contraindications to alcohol use include pregnancy, personal or strong family history of alcohol use disorder, or alcohol-associated hepatic or pancreatic disease. In addition, other conditions may require that patients limit or avoid alcohol consumption (eg, active gastritis or esophagitis). (See 'Is there a safe dose of alcohol?' above.)
•Alcohol intake in excess of these limits is associated with increased mortality and other medical morbidities (see 'Mortality' above and 'Alcohol effect on specific conditions' above). Medical comorbidities associated with alcohol use include:
-Esophageal, head and neck, hepatocellular cancers (see 'Gastrointestinal cancer' above and 'Head and neck cancer' above)
-Breast cancer (even minimal alcohol use may increase risk; supplemental folate may attenuate the risk) (see 'Breast cancer' above)
-Cirrhosis (see 'Liver disease' above)
-Pancreatitis (see 'Pancreatitis' above)
-Osteoporosis (see 'Osteoporosis' above)
-Accidents and trauma (see 'Accidents and trauma' above)
-Violence (see 'Violence' above)
-Suicide (see 'Suicide' above)
-Hematologic complications (see "Hematologic complications of alcohol use")
-Gout (see "Nonpharmacologic strategies for the prevention and treatment of gout")
•In some studies, moderate alcohol consumption (within guideline limits) (table 1) has been associated with decreased mortality and may have cardiovascular benefits. Differences in study methods and interpretations make clear conclusions difficult. (See 'Mortality' above and 'Cardiovascular disease' above.)
●Limitations of the evidence – There have been no long-term randomized trials of alcohol consumption on clinical outcomes. It remains possible that many of the health benefits and risks of alcohol consumption represent associations unrelated to the intake of alcohol itself. (See 'Limitations of the evidence' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
