|Home | About | Journals | Submit | Contact Us | Français|
To describe how alcohol use disorders (AUDs) affect women, focusing on gender-specific implications for primary care physicians (PCPs).
An overview of literature from 1966 to 2000 identified by a medline, PsychINFO and HealthSTAR/Ovid Healthstar database search using key words “women,”“alcohol” and “alcoholism.”
Although the prevalence of AUDs is greater in men than in women, women with AUDs are more likely to seek help, but less likely to be identified by their physicians. Psychiatric comorbidities (especially depression and eating disorders) are more common in women with AUDs than in men with AUDs. A past history of sexual and/or physical abuse places a woman at increased risk for AUDs. Women have a greater sensitivity to alcohol, have an accelerated progression from alcohol toxicity, and have increased mortality at lower levels of consumption compared to men. Women and men who are light-to-moderate drinkers have lower coronary artery disease mortality than do abstainers or heavy drinkers. Risk of breast cancer is increased in women who drink ≥1 drinks daily. Common barriers to treatment include: fear of abandonment by partner; fear of loss of children; and financial dependency. Brief interventions have been shown to be effective in reduction of alcohol consumption in women with at-risk drinking. It is unclear if women-only treatment programs improve outcomes.
PCPs should be alert to gender-specific differences for women with AUDs.
Alcohol is the drug of choice among American women. Six million women meet the diagnostic criteria for current abuse or dependence. Until recently, most research, prevention and treatment efforts have been extrapolated from men to women without evidence of whether important and significant gender differences may exist. Within primary care settings, a large proportion of both male and female patients, (approximately 10% to 20%, depending on the severity of the disorder studied), suffer from alcohol use disorders (AUDs).1–3
Women are more likely than men to seek help for alcohol problems in health care settings, yet are less likely to be identified and diagnosed.4–6 For many of these women the visit to their primary care provider may be an important opportunity for screening and potential entry into treatment. Research has shown that screening and brief interventions for AUDs in women by primary care providers are effective in reducing future alcohol consumption.7,8 Given the high prevalence of female patients with AUDs and the fact that currently fewer than half of such cases are properly identified,5,9 primary care physicians (PCPs) need to take a more active role in identifying such patients.
The objective of this paper is to describe how AUDs affect women, the known gender-specific differences in AUDs, and implications for PCPs. Specifically, we will review known differences between men and women suffering from AUDs by epidemiology, genetics, biology, presentation, definition and screening, social context, course and medical consequences, psychiatric comorbidity, victimization and violence, and treatment. Our review of this enormous literature is necessarily selective and we pay greatest attention to those areas of clinical significance to PCPs.
An overview of literature from 1966 to 2000 identified by a medline, PsychINFO and HealthSTAR/Ovid Healthstar database search using key words “women,”“alcohol,” and “alcoholism.” Gender-specific data from cohort studies of general population or large clinical samples are primarily reviewed.
More than 4 times as many American women die of substance abuse–related disease than of breast cancer.10 Increased overall mortality occurs in women who drink more than 2 drinks daily.11 In 1993, the National Institute on Alcohol Abuse and Alcoholism estimated that nearly 6 million American adult women (or about 6% of the adult female population) suffered from AUDs within a 12-month period.12 By comparison, approximately 14% of men were reported to have serious alcohol problems.
The National Comorbidity Study investigated the prevalence of psychiatric disorders, including substance use disorders in a population-based representative sample of individuals 15 to 54 years old in the United States.13 Consistent with other large population-based studies,14,15 substance use disorders were more prevalent in men than in women; the gender differential for alcohol dependence was larger than for drug dependence. Depending on the method of data collection and population studied, the ratio of AUDs in men to women ranges from about 2:1 to 3:1. Recent data,16–20 however, suggest that the gender gap between both adolescent girls and boys and older women and men may be narrowing.
The prevalence of AUDs among female patients in medical settings is substantially higher than are prevalence rates within the general population of women. Rates of lifetime alcohol abuse or dependence among women in primary care settings have ranged from 23% to 25%.21,22 Within outpatient gynecology practices, prevalence rates of current alcohol abuse or dependence have ranged from 12% to 20%.23–27 Inpatient surveys have revealed prevalence estimates of alcohol-related problems in women of 8% to 12%, depending on the service.9
The evidence for a genetic influence on addiction in women is somewhat mixed. Early studies comparing identical and fraternal twins involved only male subjects.28,29 Later research from Denmark showed a probable genetic influence for men but was inconclusive for women.30 Some of these studies indicated that environmental factors may be more important in both the development of and protection from alcohol problems in women.31–33
The Stockholm adoption study, which involved 1,775 adults adopted by nonrelatives early in life, has yielded much information on heredity patterns of alcohol problems in both men and women.34–36 The authors of this study were able to distinguish 2 patterns of heredity relevant to women. The more common type of inheritance, called type 1, was seen in both sexes. This type is characterized by adult onset and less severe alcohol abuse. Alcohol abuse by either biological father or mother, or both, increased risk for alcohol problems by a factor of 3 in these adoptees. In addition, researchers identified a less common, male-only pattern, type 2. This type was characterized by severe, early-onset alcohol abuse associated with criminality in both biological father and son, and revealed stronger hereditary influence than type 1. Neither the biological mothers nor the daughters of these fathers had an increased incidence of alcohol problems, although the daughters had a high incidence of multiple physical illnesses.
Kendler et al.37 performed a twin study in women that revealed a concordance for alcoholism consistently higher in monozygotic compared with dizygotic twin pairs and a heritability of 50% to 60%. In summary, it seems reasonable to conclude that a genetic influence is probable in some, but not all women with AUDs and that there may very well be different patterns of heritability in women compared to men.37,38
Women's physiology—lower average body weight, less total body water, lower concentration of gastric alcohol dehydrogenase and variations in hormones—all contribute to greater sensitivity to alcohol, increased blood alcohol levels, and greater alcohol-related morbidity given equivalent drinking careers compared to men.38,39 Because of the above stated physiologic variations, women become intoxicated after drinking smaller quantities of alcohol than men.39 Hormonal fluctuations with menstrual cycle have been correlated with changes in blood alcohol levels.40
It has also been suggested that differing organization and modulation of neurotransmitters, (particularly neurosteroids), in the male and female brain may influence the response to alcohol.34
In recognition of these differences, the National Institute on Alcohol Abuse and Alcoholism (NIAAA) has defined at-risk drinking for women as more than 7 drinks (either 12 oz of beer, 5 oz of wine, or 1.5 oz of liquor) per week or 3 drinks per occasion, as opposed to 14 drinks per week or 4 drinks per occasion for men (Table 1).41
Clear gender differences exist with respect to how men and women with alcohol problems present for clinical care. While men are much more likely to enter alcohol treatment via social programs (e.g., such as drunk-driving rehabilitation, public-intoxicant intervention programs, and employee-assistance programs) or peer-oriented settings (e.g., Alcoholics Anonymous), women more frequently visit a physician or mental health care provider prior to treatment entry.4,42 Because women with alcohol problems are more likely to present initially within the health care setting, the ratio of men to women with alcohol problems within the outpatient setting is much narrower than in the general population (1.5:1 compared to about 2–3:1, respectively).
Women with AUDs are more likely than are men with AUDs to present to their PCP repeatedly with nonspecific health complaints, nervousness, anxiety, or insomnia.43 Surveys of both patients and health care providers have shown that although women are often seeking care as a “cry for help,” health care providers do not screen adequately for problem alcohol use in women.6 Unless specifically questioned about alcohol use, these women may be misdiagnosed with another primary psychiatric condition.
Physicians prescribe psychoactive drugs more readily to women than to men with the same complaints.44 Thus, women presenting with alcohol problems are not only less likely to be screened for AUDs but also more likely to be prescribed additional sedative drugs, potentially leading to secondary addictions.45
Multiple research studies exist to support the health benefits of screening and intervention for alcohol problems.8,46Table 1 reviews categories and definitions of patterns of alcohol use. The diagnostic differentiation of use, abuse, and dependence has been operationally refined and repeatedly shown to be reliable and valid.47,48 Although historically physicians were trained to focus on alcohol abuse and dependence, recent recommendations by the NIAAA encourage PCPs to screen for lower levels of alcohol consumption that place patients at risk of abuse and dependence and future adverse health consequences.41,49
The United States Preventive Services Task Force has endorsed the use of self-report screening tests for AUDs in the primary care setting.50Table 2 presents descriptions of the most commonly used alcohol screening questionnaires in primary care (the CAGE questionnaire, the AUDIT, and the TWEAK questionnaire). Many studies evaluating efficacy of screening instruments vary by level of severity of AUD addressed and lack of inclusion of women and minorities, making them difficult to compare.
A recent systematic review by Fiellin et al.7 evaluated the accuracy of screening methods for alcohol problems in primary care. On the basis of the studies included in this review, the authors concluded that the literature supports screening for less-severe alcohol problems such as at-risk (see Table 1), hazardous (at risk for adverse consequences from alcohol), and harmful (alcohol is causing physical or psychological harm, but criteria are not met for dependence) drinking by means of the AUDIT and screening for lifetime and current abuse or dependence disorders by means of the CAGE. As these authors identified, one of the limitations of their review included either the lack of inclusion of women or failure to report gender distribution in the primary study.
Bradley et al.51 performed the first systematic review and description of performance of alcohol screening questionnaires in female patients. This analysis reviewed the literature for studies that included data for women comparing brief alcohol screening questionnaires with valid criterion standards for heavy drinking (≥2 drinks per day) or alcohol abuse or dependence in U.S. general clinical populations. The screening questionnaires and cut points that were found to have the best operating characteristics in women are summarized in Table 3.
While the CAGE was relatively insensitive in predominantly white female populations (the lower range of sensitivities), the TWEAK and AUDIT performed adequately in African-American or white women, using lower cut points than usual. Although it is not clear from this data (given the limited number of studies and subgroup analyses) which screening test is optimal in women, for all instruments it is recommended that the cut points for a positive response in women should be lowered to the following: TWEAK: ≥2 points; AUDIT: ≥4 points; CAGE: ≥1 point. In addition, if a positive screen is obtained, further questioning about quantity and frequency of drinking should be pursued.41
Perhaps more important than the role of biology and heredity in AUDs in women may be social context. Women may be less likely to divulge hazardous levels of alcohol use to their physicians.52 Although there is clearly a negative stigma attached to AUDs for both men and women, societal and cultural norms and expectations may make this admission more problematic for women.53–56 Much of the stigma attached to women when they drink is related to failure to fulfill societal expectations about femininity and role of caretaker and mother.57 This stigmatization may lead to increased isolation and decreased opportunities for identification and treatment.
There is some evidence that environmental influences have a greater influence on predisposing and perpetuating AUDs in women than in men. Alcoholic women are more likely to have had nonrelated alcoholic role models in their lives and are more likely to have alcoholic partners than are alcoholic men.53 Alcoholic women are more likely to be left by their partners, especially at the time of entry into treatment, than are alcoholic men.53 Fear of the loss of a partner and potential loss of custody of children are important social barriers to women seeking help for alcohol problems.
Women with alcohol problems have different drinking patterns and societal consequences as a result of drinking compared to men. Women first drink to intoxication at a later age and more often drink alone at home; men are more likely to drink away from home and more often drive after drinking.58 Women are much more likely to attribute their drinking to a traumatic event or stressor and often view their drinking as self-medication.53
Women experience different and increased psychosocial problems at lower levels of alcohol consumption than men.59 While men are more likely to have their drinking affect their jobs and career paths, women are more likely to initially experience disruptions in relationships and family life.60 Women with alcohol problems have less-frequent involvement in the criminal justice system as a result of their drinking, yet as the primary caregivers of children, have the additional worry of prosecution and loss of child custody if their problem is identified.
While some studies have revealed that working outside the home has provided increased opportunity to drink and has narrowed the gender gap, others have shown a protective effect.60 There has also been speculation that women currently employed in historically male-dominated professions (e.g., medicine, business) are at increased risk for alcohol problems.61
A woman is somewhat more at risk for an AUD if she is childless, separated, or divorced and less at risk if she is married.61,62 Women who are not married but living with a partner are 50% more likely to drink heavily than married women. Women in their 30s to 60s are most at risk when they are divorced or widowed, not employed, and have no children living at home. Similar to their male counterparts, white women are more likely to report problem drinking than are Hispanic and African-American women, respectively. Interestingly, the more acculturated to U.S. culture a minority woman is, the more likely she is to drink heavily.
Biologic differences in metabolism may lead to a phenomenon known as “telescoping.”63 Telescoping is defined as an accelerated medical, physiologic, and psychologic progression of alcoholism in women compared to men given the same duration and intensity of drinking careers. This results in a faster progression between major landmarks associated with alcoholism (e.g., regular drinking, loss of control, loss of job and relationships). In addition, women are more likely to have alcoholic liver disease after shorter and less-intense drinking careers than men,11,64–66 and a greater proportion of alcoholic women die from cirrhosis than do alcoholic men.67 If the concept of telescoping holds true in women suffering from alcohol problems, there is a much narrower window for intervention before disease progression in women compared to men, making screening potentially more beneficial.
Mortality rates are higher among women than men who drink heavily.68–70 A meta-analysis of cohort studies evaluating the relationship between alcohol consumption and mortality found that women have higher all-cause mortality at lower levels of alcohol consumption than men, (2 to 3 drinks/day compared to 4 or more drinks per day).71 Overall increased mortality in women has been demonstrated with intake of 2.5 drinks daily compared to women nondrinkers.11 Similar to men, however, data from the Nurses Health Study11 revealed strong evidence for a U-shaped relation between alcohol intake and mortality. In comparison to abstinence and heavier drinking, women with light-to-moderate alcohol consumption had a significantly reduced risk of death.
Multiple large prospective studies11,72–76 assessing alcohol consumption and mortality have revealed that women, like men, experience decreased risk of coronary artery disease events (fatal and nonfatal) with light-to-moderate (usually <2 drinks daily) alcohol consumption, largely confined to women at greatest risk for coronary artery disease. Among younger women and those without risk factors for coronary heart disease, however, protection has not been observed for light-to-moderate drinking, and heavier drinking has been shown to substantially increase mortality, largely due to cirrhosis and breast cancer.68 Mukamul et al., identified that self-reported moderate alcohol consumption in the year prior to myocardial infarction was associated with reduced mortality following infarction after adjusting for multiple confounders, including gender.77
Despite the fact that the mean lifetime dose of alcohol in women alcoholics is less than that of male alcoholics, there are data to suggest that toxic effects of alcohol on the heart (cardiomyopathy and myopathy) are as prevalent in women as men.78 A recent study79 including a large cohort of elderly persons (mean age, 73.7 years; 58.8% female) revealed moderate alcohol consumption associated with decreasing risk for congestive heart failure after controlling for a number of factors, including gender.
Given that current data are obtained from observational studies and known potential harmful effects of alcohol, it is unreasonable at this time to endorse any level of drinking for protection of coronary artery disease.
Heavy alcohol consumption has also been shown to place women at increased risk of death from breast cancer.11 While the link of alcohol consumption with breast cancer includes some conflicting data, several very large studies,80–84 including a recent meta-analysis,85 have revealed a range of 10% to 30% increased risk in women reporting consuming at least 1 drink daily compared to nondrinkers.11,83,86,87
Alcohol abuse is often considered a risk factor for osteoporosis, based on the frequent finding of low bone mass, decreased bone formation, and increased fracture incidence in alcoholics. Present data, however, suggest that there may be significant gender differences in terms of the effect of alcohol on bone. Overall, the evidence generally supports a detrimental effect of chronic alcohol abuse on the skeleton of a subpopulation of men (especially elderly men) and a neutral or generally beneficial effect for moderate alcohol consumption, especially in postmenopausal women.88–94 It has been postulated that this putative beneficial effect in postmenopausal women may be due to a reduction in the increase in bone remodeling that in part mediates age-related bone loss. In contrast, however, studies assessing alcohol and osteoporosis in premenopausal women have shown increased risk for osteoporosis.95,96
Data revealing both increased risk and protection against ovarian cancer from alcohol consumption have also been reported.86,97–99 A recent study100 reviewed the incidence of ovarian cancer among a cohort of alcoholic women in Sweden and found an overall decreased incidence of about 14% among women with a diagnosis of alcoholism. This finding was especially strong among women <60 years of age. The suggested mechanism may be related to reductions in gonadotropin levels in alcoholic women.
Women who drink heavily may have more painful, heavy, and irregular menses compared to women who drink less.101 Heavy alcohol consumption has also been associated with increased rates of infertility, although causality has not been established.102–104 Increased rates of spontaneous abortions have been shown in women who drink more than 2 drinks daily.105
Women may also suffer greater acute and chronic cognitive impairment from lower levels of alcohol consumption than do men.106–109 Ammendola et al.110 evaluated a sample of male and female alcoholics for evidence of peripheral neuropathy. After adjusting for total lifetime dose of ethanol and other risk factors for neuropathy, the results revealed a significantly higher sensitivity of women from the toxic effects of ethanol on peripheral nerve fibers compared to men.
Within the general population, it has been clearly established that affective disorders are much more prevalent in women compared to men. Similarly, women with AUDs suffer from a significantly higher prevalence of affective disorders compared to men with AUDs.111–115 In addition, it has been demonstrated that the onset of psychiatric disorders precedes the onset of substance abuse more often in women than in men.115–117 Women with AUDs report more frequent suicide attempts than men with AUDs, as well as 4 times the frequency of suicide attempts compared to women without AUDs.118
Although treatment for coexistent psychiatric comorbidity concomitantly with treatment for alcohol abuse has become the standard of care, many women with AUDs remain undiagnosed and untreated for a major affective disorder. Since the prevalence of these disorders is higher in women than in men, undertreatment of these comorbidities remains a significant obstacle to success in sobriety for women.
There is substantial evidence that alcoholism and eating disorders co-occur at high rates.119,120 Cross-sectional studies of women with eating disorders, including anorexia nervosa and bulimia, have documented prevalence of alcohol and other substance abuse at rates much higher than those reported in the general female population.121,122 Conversely, women with substance abuse disorders report higher rates of eating disorders.123 A common underlying pathology hypothesis is supported by the finding that patients with eating disorders are more likely to have family histories of alcohol and other drug abuse.124,125
Alcohol leads to increased vulnerability and potential for violence, especially toward women alcoholics. Unfortunately, women with alcohol disorders are frequently the partners of alcohol abusers and become victims of aggression. Women at greatest risk for injury from domestic violence include those whose male partners abuse alcohol or drugs.126 Multiple studies have established both a higher incidence of childhood physical and/or sexual abuse as well as increased vulnerability to future abuse and violence in women with AUDs.54,56,127–130 Breslau et al.131 reviewed sequelae of post-traumatic stress disorder (PTSD) in women and found that women with PTSD suffered increased risk for major depression and AUDs.
Barriers for women seeking help or treatment for AUDs are distinct from those for men and may include: lack of income or underinsurance; fear of abandonment by their husband or partner after help seeking; lack of child care during treatment; exclusion of pregnant women from treatment programs; lack of transportation secondary to overall lower levels of socioeconomic status; and fear of loss of custody of children.4,55,56,132–134 Barriers for men seeking treatment have been shown to be more closely related to loss of career and financial instability.
In the primary care setting, Fleming et al.8 reported the first direct evidence for efficacy of a brief intervention being associated with sustained reductions in alcohol consumption by women of childbearing age. An intervention consisting of two 15-minute physician-delivered counseling visits that included advice, education, and contracting using a scripted workbook resulted in a significant treatment effect, reducing both 7-day alcohol use and binge drinking episodes over the 48-month follow-up period. Interestingly, gender was observed to be a strong independent predictor, with women demonstrating a nearly 2-fold greater relative reduction in alcohol consumption than men who had received the intervention.
Wilk et al.46 performed a meta-analysis of 12 randomized controlled trials assessing the efficacy of brief intervention in heavy drinkers in a variety of settings (outpatient, residential, inpatient). This analysis revealed that heavy drinkers who received a brief intervention were twice as likely to moderate their drinking 6 to 12 months after the intervention compared to heavy drinkers who received no intervention. When specific subcategories of heavy drinkers were analyzed, there was a greater likelihood of alcohol moderation after brief intervention for women compared to men. Previous data have also shown that women have higher rates of abstinence when treated in a medical setting compared to a peer-oriented setting (e.g., Alcoholics Anonymous).62
Until recently, most substance abuse treatment programs were male-oriented, focusing primarily on re-entry into the workplace and less on strengthening or renewing emotional relationships and improving self-esteem, factors that have been shown to be more important to women in recovery.135 Current data are conflicting on whether better outcomes actually result from specialized female-only programs compared to traditional programs. One preliminary study revealed improved 2-year outcomes for women in a women-only treatment program compared to a traditional program.136 A second study, however, did not support this finding.137 It is unclear if these contradictory findings about the efficacy of women-only treatment programs are due to comparing women with varying levels of addiction severity and co-morbidities, small numbers, or inadequate follow-up.
Women in recovery from alcohol treatment report a high frequency of persistent life problems and may require additional surveillance and therapeutic attention to prevent post-treatment relapse. Snow and Anderson138 evaluated factors influencing relapse and recovery among alcohol- and drug-addicted women and found that depression was a major trigger for relapse, as were co-occurring addictions and current violent partnerships.
Cochran et al.139 examined alcohol use patterns among adults interviewed in the 1996 National Household Survey on Drug Abuse. Alcohol use patterns did not differ between heterosexual and homosexual men. Homosexually active women, however, reported using alcohol more frequently and in greater amounts and experienced greater alcohol-related morbidity than did exclusively heterosexually active women.
Older women represent the largest single group of health care users in this country.140 Twelve percent of older women regularly drink in excess of recommended guidelines and are considered at-risk drinkers. Older women are also the largest users of sedatives and narcotic analgesics.141 The fact that many of these patients suffer from multiple comorbidities, may take multiple prescription medications, and are at risk for falls and fractures,142 underscores that these women have specific risks and vulnerabilities to alcohol use and should be aggressively screened for AUDs.
Clearly, women who are thinking about pregnancy or who are pregnant are at significant risk of complications from AUDs. Preconception counseling about risks of alcohol is essential to all women of childbearing age because of the risk associated with drinking before pregnancy is recognized. Although a full discussion of the toxic effects of alcohol on pregnancy is beyond the scope of this article, effects of alcohol in pregnancy include: fetal alcohol syndrome, persistent neurobehavioral deficits, and low-birth-weight infants. In the most recent National Survey of Family Growth, 34% of U.S. women drank alcohol during pregnancy.143 The prevalence of fetal alcohol syndrome is 1.95 per 1,000 live births and is much higher (4.7 per 1,000 births) among daily drinkers.144
Barriers for pregnant women with AUDs seeking treatment include: lack of treatment programs willing to accept pregnant women; lack of availability of child care for other children; and fear of criminalization and loss of child custody.
In summary, clear gender-specific differences exist between women and men suffering from AUDs. These differences are summarized in Table 4. Because women with AUDs commonly seek help in health care settings, the PCP's office is an ideal setting for screening, brief interventions, or referrals to treatment for women suffering from alcohol problems.
PCPs should be aware of lower recommended alcohol consumption levels for women compared to men as well as increased sensitivity to alcohol at lower levels. Identification of an affective disorder, eating disorder, or history of childhood or adult abuse should trigger PCPs to be more aggressively surveillant for AUDs in female patients. Because women with AUDs are more likely than men to have partners with AUDs, attention should be given to partners' roles in the etiology and maintenance of women's alcohol problems. Women, like men, who are light-to-moderate drinkers may experience some protection from cardiovascular disease; however, this needs to be balanced by increased risk of breast cancer, especially in younger women and those at lower risk of coronary artery disease.
Important areas for further research include: genetic studies focusing on distinguishing patterns of AUD inheritance in men and women; research specifically aimed at identifying why women have increased vulnerability to alcohol in nonreproductive organ systems; and the potential for gender differences in the performance of screening instruments for AUDs.
Given the numerous important differences that have been identified between women and men with AUDs, it seems prudent that treatment programs integrate known gender differences into their treatment models. Clearly, more systematic research evaluating outcomes of gender-specific treatment programs is needed. This research should include diverse ethnic, older, and nonpregnant women. Hopefully, increased awareness of and sensitivity to known differences as well as further research may result in improved quality of care and outcomes for female patients with AUDs.