Poor health is associated with episodic heavy alcohol use: evidence from a National Survey

Public Health (2005) 119, 509–517

Poor health is associated with episodic heavy alcohol use: evidence from a National Survey I.S. Okosuna,*, J.P. Sealeb, J.B. Danielc, M.P. Eriksena a

Institute of Public Health, P.O. Box 3995, Georgia State University, Atlanta, GA, USA Department of Family Medicine, Mercer University School of Medicine, Macon, GA, USA c Department of Community Medicine, Mercer University School of Medicine, Macon, GA, USA b

Received 18 December 2003; received in revised form 19 July 2004; accepted 16 August 2004 Available online 19 January 2005

KEYWORDS Episodic heavy drinking; Drinking pattern; Binge drinking

Summary Objectives. The objective of this study was to examine the relationship between self-rated health and episodic heavy drinking in a representative sample of American adults. We also sought to determine ethnic and gender differences in the association between self-rated health and episodic heavy drinking. Methods. Data (nZ4649) from the Third US National Health and Nutrition Examination Survey were utilized for this investigation. Episodic heavy drinking was defined as the consumption of five or more and four or more alcoholic beverages on one occasion for men and women, respectively. Poor health was defined as answering fair or poor to the question: ‘Would you say your health in general is excellent, very good, good, fair or poor?’ Odds ratio from the logistic linear regression analysis was used to estimate the risk for poor health that was associated with episodic heavy drinking. Statistical adjustments were made for age, hypertension, diabetes, current smoking, body mass index and race/ethnicity. Results. Overall, episodic heavy drinking was associated with increased odds of poor self-rated health in men and women. In men, episodic heavy drinking was independently associated with 1.28 (95% CI: 1.07–1.82) increased odds of poor health. The corresponding value in women was 1.86 (95% CI: 1.05–2.28). In men, being Black was associated with wtwo-fold (ORZ1.96; 95% CI: 1.33, 2.89), and being Hispanic was associated with wfour-fold (ORZ3.59; 95% CI: 2.50, 5.14) increased odds of poor self-rated health relative to being White. The corresponding odds ratios in women were 2.97 (95% CI: 1.90, 4.64) and 5.18 (95% CI: 3.23, 8.30). Associations were greater among blacks (adjusted ORZ2.41; 95% CI: 1.81–3.22) and Hispanics (adjusted ORZ4.15; 95% CI: 3.12–5.52) than among whites. Conclusions. Poor health is associated with episodic heavy alcohol consumption. Public health strategies to curb alcohol abuse may improve self-reported health status in these at-risk populations. Q 2004 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: C1 404 651 4249; fax: C1 404 651 1559. E-mail address: [email protected] (I.S. Okosun). 0033-3506/$ - see front matter Q 2004 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.puhe.2004.08.016

510

Introduction Alcohol abuse is an adverse health behavior that is associated with increased risk for many chronic diseases.1–3 Alcohol abuse is a marker for sociobehavioral disintegration, and is significantly associated with increased mortality in different populations.4–7 In the USA, alcohol abuse is associated with approximately 100,000 deaths annually, and it is the third leading preventable cause of death.7 Alcohol abuse is associated with development of cancers of the liver and esophagus.8,9 Recent data on the prevalence of alcohol use among US adults demonstrated that alcohol abuse is more prevalent among low socio-economic groups.10 A classical form of alcohol abuse that is often associated with acute impairment and a substantial portion of all alcohol-related deaths is episodic heavy drinking (sometimes called binge drinking). Episodic heavy drinking is generally defined as the consumption of five or more alcoholic beverages on a single occasion.11 Adverse health effects of episodic heavy drinking include falls, burns, drowning, hypothermia and motor vehicle injuries.12–14 Other consequences of episodic heavy drinking are suicide, alcohol poisoning, hypertension, diabetes, coronary heart disease, gastritis and sexually transmitted diseases.15 Social and economic burdens of episodic heavy drinking are often described in terms of violence, unintended pregnancy, child neglect, fetal alcohol syndrome and lost productivity.15 In the USA, social burdens of episodic heavy drinking are more prevalent among adolescents.16 While much of the association between episodic heavy drinking is described in the literature in terms of specific diseases, only a few studies exist linking episodic heavy drinking with a broad measure of overall health (self-rated health). The reported association between overall health and alcohol consumption in these studies is inconsistent. Investigating university students from Sweden, Vaez and Laflamme observed an association between poor self-rated health and alcohol use.17 An association between poor health and alcohol use was also observed by Powles et al.18 in a population from Melbourne, Australia. However, in Russia, Carlson19 did not observe any association between poor health and episodic heavy drinking. Self-rated health is a summary measure of different domains of health that includes the psychosocial domain.20,21 Self-rated health reflects a complex process of internalization

I.S. Okosun et al. that takes into account both disease exposure experiences and knowledge of disease causes and consequences.20,21 Self-rated health is an important indicator of health that is associated with a wide range of outcomes, from wellbeing to health service utilization,20,21 and even overall mortality.22,23 Despite gender differences in the perception of overall health and episodic heavy drinking,24–26 there is little data to date concerning the relationship between overall health and episodic heavy drinking. Hence, in this study, we examined the relationship between self-rated health and episodic heavy drinking in a representative sample of American adults. We also sought to determine gender differences in the association between episodic heavy drinking and self-rated health.

Methods Data source Data from the Third US National Health and Nutrition Examination Survey (NHANES III) as provided by the National Center for Health Statistics were used in this investigation. The sampling and measurement procedures have been described in detail previously by other investigators.27–29 Briefly, NHANES III is a multistage probability sampling that was conducted in noninstitutionalized US population groups between 1988 and 1994. Only subjects identified as nonHispanic whites, non-Hispanic blacks and Hispanic Americans were eligible for this investigation. This analysis was also restricted to 40–74-year-old subjects because fasting glucose and oral glucose tolerance tests were performed only on subjects in this age category in NHANES III. This study was further confined to individuals for whom anthropometric variables, including age, weight, height and blood pressure (diastolic and systolic), were measured. Weight was measured in a standing position using a Toledo self-zeroing weight scale (Seritex, Carlstadt, NJ, USA). Height was measured in a standing position with a stadiometer. Three blood pressure measurements were obtained from each subject using a standard mercury sphygmomanometer at a 60 s interval between inflation cuffs. The average of the three readings was utilized for this analysis. Details of the methods that were used to collect and assay for glucose have been described. 28 Briefly, plasma glucose was

Episodic heavy drinking and overall health measured using a modified hexokinase enzymatic method. In NHANES III, subjects were asked to rate their health by answering the question: ‘Would you say your health in general is excellent, very good, good, fair or poor?’ This analysis was restricted to subjects (nZ4649) who reported the number of alcoholic drinks per day on a drinking day, and those who provided an answer to the health self-rating question described above. In this study, 39 and 460 subjects were daily drinkers and non-drinkers, respectively. The daily drinkers could be constant moderate or persistent heavy alcohol users or occasional binge drinkers. A comparison of the self-reported health of daily drinkers with episodic heavy drinkers did not show any statistically significant difference. Hence, they were included in this analysis. We excluded subjects with missing values on alcohol use (nZ2423). The excluded subjects were not different from the population examined in our study in terms of age, total cholesterol and body mass index (BMI) values.

Definition of terms We calculated BMI as measured weight in kilograms divided by height in meters squared (kg/m2). BMI is a measure of overall body mass adjusted for height. ‘Episodic heavy drinking’ was defined as the consumption of five or more alcoholic beverages by men, and the consumption of four or more alcoholic beverages by women, on one occasion.11 ‘Type 2 diabetes mellitus’ was defined using the 1988 American Diabetes Association (ADA) diagnostic criteria30. ‘Hypertension’ was defined as systolic blood pressure (SBP)R140 or diastolic blood pressure (DBS)R90 or current use of antihypertensive medication.31 ‘Poor health’ was defined as answering fair or poor, and ‘good health’ was defined as answering excellent, very good or good to the above selfrated health question.

511 drinking and diabetes were age-adjusted by direct methods using the 1990 US population census data. Differences between subjects with poor health and subjects with good health for continuous and categorical variables were assesses by independent t-test and c-square statistics, respectively. Univariate and multiple logistic regression analyses were performed using self-rated overall health (coded as 1 for subjects with poor health and 0 for those in good health) as the dependent variable, and drinking (coded as 1 for episodic heavy alcohol drinkers and 0 for others) as the independent variable, adjusting for age, gender, hypertension, diabetes, current smoking, BMI and race/ethnicity. To investigate the contribution of episodic heavy drinking to racial/ethnic differences in the risk of poor health, dummy variables were used to compare whites with non-whites fitted in the multiple logistic regression models. In order to determine whether the association between episodic drinking and self-rated health was different in men and women, we fitted an interaction term between episodic heavy drinking and gender in the logistic regression model.

Results Fig. 1 shows the distribution of eligible men (nZ2244) and women (nZ2405) by self-rated health status. There was a statistically significant gender difference with respect to perceived health (P!0.05). The proportion of subjects reporting poor health (defined as fair or poor in the self-rating five-point Likert-type scale) was

Statistical methods Statistical programs available in SAS for Windows32 and SUDAAN33 were utilized for this analysis. To account for unequal probabilities of selection, oversampling and non-response, appropriate sample weights were utilized for the analyses. Prevalence estimates were weighted to account for cluster design and to represent the total civilian non-institutionalized population of the USA. The prevalence of episodic heavy

Figure 1 Distribution of self-rated health status in American men (nZ2244) and women (nZ2405).

512

I.S. Okosun et al.

significantly higher in women (26.3%) compared with men (23.7%). The basic characteristics of subjects reporting poor health and good health are shown in Table 1. There were statistically significant differences between the two groups with respect to nearly all of the variables that were investigated. Subjects with poor self-rated heath were older, consumed more drinks on a drinking day, and had higher BMIs (P!0.01). Subjects with poor self-rated health also had higher prevalences of episodic drinking, current smoking, hypertension and type 2 diabetes compared to subjects with good self-rated health (P!0.01). We plotted mean number of alcohol use on a drinking day by health status for men and women.

Table 1 Descriptive characteristics of eligible subjects with self-rated poor (nZ1164) and good (nZ3485) health status. Variables

Poor health (meanGSD)

Good health (meanGSD)

P-value

Number of drinks on a drinking day Age (years) Weight (kg) BMI (kg/m2) Diastolic blood pressure (mmHg) Systolic blood pressure (mmHg)

3.7G1.2

2.7G1.1

!0.001

57.6G9.9 79.0G18.1 29.0G6.1 77.7G33.6

54.7G10.4 78.1G16.9 27.6G5.4 76.8G30.0

!0.001 0.119 !0.001 0.367

134.4G37.0

128.9G32.8

!0.001

%

%

45.6 54.4

49.2 50.8

Gender Male Female Race/ethnicity White Black Hispanic Episodic heavy drinkers Current smokers Hypertension Type 2 diabetes

0.037

!0.001 31.6 31.5 36.9 23.8

56.2 24.0 19.8 12.0

!0.001

30.1

25.1

0.001

58.1 26.5

40.7 9.6

!0.001 !0.001

Note: Episodic heavy drinking was defined as the consumption of five or more alcoholic beverages by men, and the consumption of four or more alcoholic beverages by women, on a drinking day. Poor health was defined as answering fair or poor, and good health was defined as answering excellent, very good or good on the Likert scale. SD, standard deviation.

Figure 2 Mean number of drinks per drinking day versus self-rated health status in American men (nZ2244) and women (nZ2405).

As shown in Fig. 2, a gradient of increasing number of alcoholic drinks per day on a drinking day with decreasing health status was apparent in men and women (P!0.01 for c-square test for trend). Also, there were significant gender differences with respect to the number of alcoholic drinks per day on a drinking day, with men consuming more drinks at each level of self-rated health status (P!0.01). In order to determine the association between episodic heavy drinking and other studied variables including poor self-rated health, we estimated the risk of poor health using odds ratios from unadjusted (Table 2, Model I) and adjusted (Table 2, Model II) logistic regression analyses. In Model II, we adjusted for age, hypertension, diabetes, current smoking and BMI. In the unadjusted models, episodic heavy drinking, increased age, gender, hypertension, diabetes, current smoking and increased BMI were each associated with increased risk of poor self-rated health. The risk of poor selfrated health associated with episodic heavy drinking decreased by approximately 40% when adjusted for age, gender, hypertension, diabetes, BMI, current smoking and race/ethnicity. In the adjusted and unadjusted models, we determined whether or not race/ethnicity was associated with risk of poor self-rated health. Dummy variables were used to compare blacks and Hispanics with whites. In the unadjusted models, being black and being Hispanic was associated with increased risk for poor self-rated health over two-fold and three-fold, respectively. Adjusting for age, hypertension, diabetes, current smoking and BMI, being black remained positively associated with over a two-fold increased risk of

Episodic heavy drinking and overall health Table 2

513

Association between episodic heavy drinking and poor health in American adults (nZ4649).

Variable

Episodic heavy drinking Age Genderb Hypertension Current smokingc Diabetes Body mass index Race/ethnicityd Black Hispanic Episodic heavy drinking!genderb

Model I a

Model II

Model III

OR

95% CI

OR

95% CI

OR

95% CI

2.30 1.03 1.15 2.11 1.29 3.41 1.04

1.77–2.40 1.02–1.04 1.01–1.32 1.84–2.42 1.11–1.49 2.78–4.05 1.03–1.05

1.37 1.03 1.41 1.58 2.02 2.07 1.02

1.02–1.86 1.01–1.04 1.11–1.78 1.25–2.01 1.58–2.58 1.47–2.93 0.99–1.04

1.36 1.03 1.29 1.57 2.01 2.07 1.01

1.15–1.91 1.01–1.04 1.01–1.67 1.25–2.01 1.57–2.57 1.47–2.93 0.99–1.04

2.34 3.31 –

1.98–2.76 2.81–3.89 –

2.41 4.15 –

1.81–3.22 3.12–5.52

2.41 4.17 1.75

1.81–3.22 3.13–5.55 1.08–3.31

Note: OR, odds ratio from the multiple logistic regression model. Episodic heavy drinking was defined as the consumption of five or more alcoholic beverages by men, and the consumption of four or more alcoholic beverages by women, on a drinking day. Poor health was defined as answering fair or poor, and good health was defined as answering excellent, very good or good on the Likert scale. Model I is univariate, with odds of episodic heavy drinking unadjusted for other variables; Models II and III are multivariate, with odds of episodic heavy drinking adjusted for other variables. a Reference group was non-episodic heavy drinkers and non-drinkers. b Values are for females relative to males. c Reference group was non-current smokers. d Values are for blacks and Hispanics relative to whites.

poor self-rated health relative to whites (ORZ2.41; 95% CI: 1.89, 3.22), while being Hispanic was associated with over a four-fold increased risk of poor self-rated health relative to whites (ORZ4.15; 95% CI: 3.12, 5.52). To investigate the contribution of episodic heavy drinking to racial/ethnic differences in the risk of self-reported poor health (Table 2), we compared Model I, where odds ratios for race/ethnicity were unadjusted for episodic heavy drinking, with Model II, where race/ethnicity were adjusted for episodic heavy drinking and other variables. As shown, the odds ratios for race/ethnicity were attenuated. Relative to whites, adjusting for episodic heavy drinking and other variables attenuated the risk for poor self-rated health by 4.3% in blacks and 26% in Hispanics. We fitted an interaction term for episodic heavy drinking and gender in the logistic regression model (Table 2, Model III). As indicated by the significant interaction term in the regression model, the response of episodic heavy drinking for poor selfrated health in men and women is different. We estimated the risk for poor self-rated health that was associated with episodic heavy drinking in gender-specific univariate logistic regression models (not shown). In men, episodic heavy drinking was associated with two-fold increased odds (ORZ2.02; 95% CI: 1.47, 2.77) of poor self-rated

health. In women, episodic heavy drinking was associated with approximately four-fold increased odds (ORZ3.70; 95% CI: 2.22, 6.16) of poor selfrated health. In gender-specific logistic regression models, adjusting for age, hypertension, diabetes, BMI, current smoking and race/ethnicity (Table 3), episodic heavy drinking was associated with 28 and 86% increased odds of self-rated poor health in men and women, respectively. Similar to the results from the unadjusted model, the highest contribution of episodic heavy drinking on racial/ethnic differences in poor self-rated health was observed in women compared with men in the multivariate models. In men, being black was associated with approximately a two-fold (ORZ1.96; 95% CI: 1.33, 2.88), and being Hispanic was associated with over a three-fold (ORZ3.59; 95% CI: 2.50, 5.14) increase in odds of poor self-rated health relative to whites. The corresponding odds ratios in women were 2.97 (95% CI: 1.90, 4.64) and 5.18 (95% CI: 3.23, 8.30).

Discussion Episodic heavy drinking, which may be defined as the consumption of four or more drinks on one occasion, is increasingly becoming a common pattern of alcohol abuse in the USA. A reduction

514 Table 3 2405).

I.S. Okosun et al. Association between episodic heavy drinking and poor health in American men (nZ2244) and women (nZ

Variable a

Episodic heavy drinking Age Hypertension Current smokingb Diabetes BMI Race/ethnicityc Black Hispanic

Men

(95% CI for OR)

Women

(95% CI for OR)

OR

Lower

Upper

OR

Lower

Upper

1.28 1.02 1.44 1.64 2.46 1.01

1.07 1.01 1.06 1.19 1.66 0.96

1.82 1.04 1.98 2.27 3.74 1.04

1.86 1.02 1.88 2.72 1.40 1.02

1.05 1.00 1.28 1.84 0.75 0.99

2.28 1.04 2.77 4.02 2.60 1.06

1.96 3.59

1.33 2.50

2.89 5.14

2.97 5.18

1.90 3.23

4.64 8.30

Note: OR, odds ratio from the multiple logistic regression model. Episodic heavy drinking was defined as the consumption of five or more alcoholic beverages by men, and the consumption of four or more alcoholic beverages by women, on a drinking day. Poor health was defined as answering fair or poor, and good health was defined as answering excellent, very good or good on the Likert scale. a Reference group was non-episodic heavy drinkers and non-drinkers. b Reference group was non-current smokers. c Values are for blacks and Hispanics relative to whites.

in episodic heavy drinking is one of the leading health initiatives described in US Healthy People 2010.34 Although many studies associating alcohol abuse such as alcohol dependence and binge drinking are often described in relation to wellestablished diseases, to our knowledge, only a few studies have described the association between episodic heavy drinking and a global measure of overall health defined by self-rated health. Self-rated health is an important summary indicator of general wellbeing, which takes into account both disease exposure experiences and consequences.20,21,35 Since health is a normatively defined concept, the ways in which individuals evaluate their own health and symptoms, and interpret biological changes and disease labels, are shaped by sociocultural and sociopsychological factors.20,21,36,37 The importance of a self-rated measure is derived from the fact that it provides an overall picture of health in population studies, and the fact that it predicts mortality is important since it provides confidence that we are not only measuring psychological wellbeing or mood but, as expected, overall health.36,37 To our knowledge, this is the first investigation utilizing NHANES III to determine the association between episodic heavy drinking and self-rated health. NHANES III represents the best available data for the US population, since the sampling scheme was representative and national in scope. The training program and quality control measures instituted in NHANES III add credibility to the data. The main finding of this investigation was that overall heath status is associated with the number

of alcoholic drinks per day on a drinking day in men and women. Adjusting for age, hypertension, diabetes, BMI, current smoking and race/ethnicity, episodic heavy drinking was associated with increased odds of poor self-rated health in men and women. This finding is consistent with a recent study that showed an association between frequent binge drinking and poor health-related quality of life and mental distress, including stress, depression and emotional problems.38 We also found that episodic heavy drinking was much more strongly associated with poor self-rated health among women than among men. The higher risks in women than men is consistent with the fact that women are more vulnerable than men to various alcohol-related complications, including liver disease, cardiovascular disease, psychiatric problems and organic brain damage.39–41 These differences may be attributable in part to gender differences in total body water, resulting in higher blood alcohol concentrations even when amounts of alcohol consumed are adjusted for body weight.42 Due to these differences, lower consumption levels are used for defining heavy drinking and binge drinking in women.43,44 It is notable that in spite of the use of a lower threshold for defining binge drinking by women in this study, the risk of poor self-related health was almost twice as high as that reported for men. We found that episodic heavy drinking was more strongly associated with poor self-rated health among minority subjects, especially Hispanics. Previous studies have found that Hispanic males demonstrate higher rates of binge drinking,46 heavy

Episodic heavy drinking and overall health drinking46 and frequent heavy drinking47 than other ethnic groups and, more importantly, show higher rates of alcohol-related problems and cirrhosis mortality.48–50 Possible explanations for these differences include greater stability or continuation of heavy drinking patterns over time45,46 and genetic differences in enzymes that metabolize alcohol.51 In agreement with the studies by Franks et al. which used data from the 1987 US National Medical Expenditure Survey,52 women rated their health poorer than men in our study. Relative to whites, adjusting for episodic heavy drinking age, hypertension, diabetes, BMI and current smoking attenuated the risk for poor self-rated health by 4.3% in blacks and 26% in Hispanics. Some limitations must be taken into account in the interpretation of results from this study. First, bias due to survey non-response and missing values for some variables cannot be ruled out. However, previous studies of National Health and Nutrition Examination Surveys have shown little bias due to non-response.53 Second, as a cross-sectional study, directionality of the associations cannot be clearly established. In a cross-sectional study, it is not possible to rule out that those with poor health are more likely to drink, perhaps to alleviate symptoms. A cohort study is needed to shed light on the temporality of the association. Third, in this study, daily drinkers were included among episodic heavy drinkers. Arguably, these daily drinkers could be moderate rather than constant heavy drinkers. As such, they may also binge drink. However, in this study, we observed no statistically significant differences between the self-reported health of daily drinkers and episodic heavy drinkers. Fourth, several well-known risk factors for poor health (e.g. lack of physical activity and poor nutrition) were not examined. Finally, the definition of episodic heavy drinking as used in this study did not account for the drinker’s alcohol-drinking history, the time period over which the alcohol was consumed, or whether any food was consumed during the drinking episode. Indeed, studies have shown that a large number of episodic heavy drinkers often do not have blood alcohol concentrations of 0.06% or higher.54,55 Arguably, the use of a set cutoff point may imply that alcohol consumption of less than five drinks is acceptable for men and less than four drinks is acceptable for women. However, these cutoff points are consistent with the values used by other investigators to define binge drinking.56,57 Also, the US National Institute of Alcohol Abuse and Alcoholism employs a cutoff point of four or five

515 alcoholic drinks on one occasion to define risky drinking.58

Conclusion Episodic heavy drinking is an adverse health behavior that has consequences for public health. The result of this study indicating an association between episodic heavy drinking and increased risk for poor self-rated health provides a framework to communicate ill effects of this form of alcohol abuse to the general public. There is a need to educate the public about the risk to overall health that is attributable to episodic heavy drinking.

Acknowledgements We thank the US National Center for Health Statistics for providing data that were used in this study.

References 1. Apte MV, Wilson JS. Alcohol-induced pancreatic injury. Best Pract Res Clin Gastroenterol 2003;17:593–612. 2. Nanchahal K, Ashton WD, Wood DA. Alcohol consumption, metabolic cardiovascular risk factors and hypertension in women. Int J Epidemiol 2000;29:57–64. 3. Rehm J, Room R, Monteiro M, Gmel G, Graham K, Rehn N, et al. Alcohol as a risk factor for global burden of disease. Eur Addict Res 2003;9:157–64. 4. Power C, Estaugh V. The role of family formation and dissolution in shaping drinking behaviour in early adulthood. Br J Addict 1990;85:521–30. 5. Ramstedt M. Alcohol consumption and alcohol-related mortality in Canada, 1950–2000. Can J Public Health 2004; 95:121–6. 6. Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000. JAMA 2004;291: 1238–45. 7. Thun MJ, Peto R, Lopez AD, Monaco JH, Henley SJ, Heath Jr CW, et al. Alcohol consumption and mortality among middle-aged and elderly US adults. N Engl J Med 1997;337: 1705–14. 8. Singh GK, Hoyert DL. Social epidemiology of chronic liver disease and cirrhosis mortality in the United States, 1935– 1997: trends and differentials by ethnicity, socioeconomic status, and alcohol consumption. Hum Biol 2000;72: 801–20. 9. Yokoyama A, Omori T. Alcohol and oropharyngolaryngeal and digestive tract cancer. Nihon Arukoru Yakubutsu Igakkai Zasshi 2001;36:551–66. 10. Curran GM, Stoltenberg SF, Hill EM, Mudd SA, Blow FC, Zucker RA. Gender differences in the relationships among SES, family history of alcohol disorders and alcohol dependence. J Stud Alcohol 1999;60:825–32.

516 11. Wechsler H, Austin SB. Binge drinking: the five/four measure. J Stud Alcohol 1998;59:122–4. 12. Kauhanen J, Kaplan GA, Goldberg DE, Salonen JT. Beer binging and mortality: results from the Kuopio ischaemic heart disease risk factor study, a prospective population based study. BMJ 1997;315:846–51. 13. McDonald AJ, Wang N, Camargo Jr. CA. US emergency department visits for alcohol-related diseases and injuries between 1992 and 2000. Arch Intern Med 2004;164:531–7. 14. National Institute of Alcohol Abuse and Alcoholism. Tenth Special Report to the US Congress on Alcohol and Health. Bethesda, MD: National Institutes of Health; 2000. 15. Shultz JM, Rice DP, Parker DL, Goodman RA, Stroh G, Chalmers N. Quantifying the disease impact of alcohol with ARDI software. Public Health Rep 1991;106:443–50. 16. Rehm J, Room R, Graham K, Monteiro M, Gmel G, Sempos CT. The relationship of average volume of alcohol consumption and patterns of drinking to burden of disease: an overview. Addiction 2003;98:1209–28. 17. Vaez M, Laflamme L. Health behaviors, self-rated health, quality of life: a study among first year Swedish university students. J Am Com Health 2003;51:156–62. 18. Powles W, Macaskill G, Hopper JL, Ktenas D. Differences in drinking patterns associated with migration from Greek island to Melbourne, Australia a study of sibships. J Stud Alcohol 1991;52:224–31. 19. Carlson P. Risk behaviors and self-rated health in Russia. J Epidemiol Community Health 2001;55:806–17. 20. Krause NM, Jay GM. What do global self-rated health items measure? Med Care 1994;32:930–42. 21. Idler EL, Hudson SV, Leventhal H. The meaning of self-rated health: a qualitative and quantitative approach. Res Aging 1999;21:458–76. 22. Kaplan GA, Camacho T. Perceived health and mortality: a nine-year follow-up of the human population laboratory cohort. Am J Epidemiol 1983;117:292–304. 23. Strawbridge WJ, Walhagen MI. Self-rated health and mortality over three decades: results from a time-dependent covariate analysis. Res Aging 1999;21:402–16. 24. Prager E, Walter-Ginzburg A, Blumstein T, Modan B. Gender differences in positive and negative self-assessments of health status in a national epidemiological study of Israeli aged. J Women Aging 1999;11:21–41. 25. Serdula MK, Brewer RD, Gillespie C, Denny CH, Mokdad A. Trends in alcohol use and binge drinking, 1985–1999: results of a multi-state survey. Am J Prev Med 2004;26: 294–8. 26. Nelson DE, Naimi TS, Brewer RD, Bolen J, Wells HE. Metropolitan-area estimates of binge drinking in the United States. Am J Public Health 2004;94:663–71. 27. Ezzati TM, Massey JT, Waksberg J, Chu A, Maurer KR. Sample design: Third National Health and Nutrition Examination Survey. Vital Health Stat 1992;2:1–35. 28. Harris MI, Flegal KM, Cowie CC, Eberhardt MS, Goldstein DE, Little RR, et al. Prevalence of diabetes, impaired fasting glucose, and impaired glucose tolerance in US adults. The Third National Health and Nutrition Examination Survey, 1988–1994. Diabetes Care 1998;21: 518–24. 29. National Center for Health Statistics. National Health and Nutrition Examination Survey III: Data Collection Forms. Hyattsville, MD: National Center for Health Statistics; 1991. 30. American Diabetes Association. Screening for Type 2 Diabetes. Diabetes Care 1998;12(suppl. 1):S20–S2.

I.S. Okosun et al. 31. Joint National Committee on Detection, Evaluation and Treatment of High Blood Pressure. Arch Intern Med 1988; 148:1023–38. 32. SAS Release 8.02. SAS Institute, Cary, NC. 33. Shah BV, Barnwell BG, Bieler GS. SUDAAN User’s Manual. Research Triangle Institute, Research Triangle Park, NC. 34. US Department of Health and Human Services. Healthy People 2010. Washington, DC: US Government Printing Office; 2000. 35. Twaddle AC. The concept of health status. Soc Sci Med 1974; 8:29–38. 36. Fylkesnes K, Forde OH. Determinants and dimensions involved in self-evaluation of health. Soc Sci Med 1992;35: 271–9. 37. Maddox G. Commentary on the self-rated health symposium. Res Aging 1999;21:501–6. 38. Okoro CA, Brewer RD, Naimi TS, Moriarty DG, Giles WH, Mokdad AH. Binge drinking and health-related quality of life: do popular perceptions match reality? Am J Prev Med 2004; 26:230–3. 39. Evans SM, Levin FR. Differential response to alcohol in light and moderate female social drinkers. Behav Pharmacol 2004;15:167–81. 40. Becker U, Deis A, Sorensen TI, Gronbaek M, Borch-Johnsen K, Muller CF, et al. Prediction of risk of liver disease by alcohol intake, sex, and age: a prospective population study. Hepatology 1996;23:1025–9. 41. Amodei N, Elkin BB, Burge SK, Rodriguez-Andrew S, Lane P, Seale JP. Psychiatric problems experienced by primary care patients who misuse alcohol. Int J Addict 1994;29:609–26. 42. Mumenthaler MS, Taylor JL, O’Hara R, Yesavage JA. Gender differences in moderate drinking effects. Alcohol Res Health 1999;23:55–64. 43. Graham K, Wilsnack R, Dawson D, Vogeltanz N. Should alcohol consumption measures be adjusted for gender differences? Addiction 1998;93:1137–47. 44. Wechsler H, Dowdall GW, Davenport A, Rimm EB. A genderspecific measure of binge drinking among college students. Am J Public Health 1995;85:982–5. 45. Bennett ME, Miller JH, Woodall WG. Drinking, binge drinking, and other drug use among southwestern undergraduates: three-year trends. Am J Drug Alcohol Abuse 1999;25:331–50. 46. Caetano R, Kaskutas LA. Changes in drinking patterns among whites, blacks and Hispanics, 1984–1992. J Stud Alcohol 1995;56:558–65. 47. Caetano R, Clark CL. Trends in alcohol consumption patterns among whites, blacks and Hispanics: 1984 and 1995. J Stud Alcohol 1998;59:659–68. 48. Caetano R. Alcohol-related health disparities and treatmentrelated epidemiological findings among whites, blacks, and Hispanics in the United States. Alcohol Clin Exp Res 2003;27: 1337–9. 49. Russo D, Purohit V, Foudin L, Salin M. Workshop on alcohol use and health disparities 2002: a call to arms. Alcohol 2004; 32:37–43. 50. Gomberg ES. Treatment for alcohol-related problems: special populations: research opportunities. Recent Dev Alcohol 2003;16:313–33. 51. Konishi T, Calvillo M, Leng AS, Feng J, Lee T, Lee H, et al. The ADH3*2 and CYP2E1 c2 alleles increase the risk of alcoholism in Mexican American men. Exp Mol Pathol 2003;74:183–9. 52. Franks P, Gold MR, Fiscella K. Sociodemographics, selfrated health, and mortality in the US. Soc Sci Med 2003;56:2505–14.

Episodic heavy drinking and overall health 53. Suchman EA, Phillips BS, Streib GF. An analysis of the validity of health questionnaires. Social Forces 1958;36:223–32. 54. Perkins HW, Linkenbach J, DeJong W. Estimated blood alcohol levels reached by ‘binge’ and ‘nonbinge’ drinkers: a survey of young adults in Montana. Psychol Addict Behav 2001;15:317–20. 55. Lange JE, Voas RB. Defining binge drinking quantities through resulting blood alcohol concentrations. Psychol Addict Behav 2001;15:310–6.

517 56. Naimi TS, Brewer RD, Mokdad A, Denny C, Serdula MK, Marks JS. Binge drinking among US adults. JAMA 2003;289:70–5. 57. Wechsler H, Nelson TF. Binge drinking and the American college student: what’s five drinks? Psychol Addict Behav 2001;15:287–91. 58. National Institute of Alcohol Abuse and Alcoholism. Helping Patients with Alcohol Problems. A Health Practitioner’s Guide. http://www.niaaa.nih.gov/publications/Practitioner/HelpingPatients.htm