Tobacco-related diseases are the most preventable health problems in the United States. Although tobacco use is declining, it is responsible for nearly 20% of the deaths in the U.S., and kills approximately 50% of continuous smokers. Smoking-related diseases include many types of cancer, chronic obstructive pulmonary disease, coronary heart disease, stroke, peripheral vascular disease and peptic ulcer disease . Lung cancer is the leading cause of cancer-related death among women and men in the United States .
Similarly, alcohol misuse may contribute to morbidity and mortality resulting in acid-related disorder, arthritis, asthma, cirrhoses of the liver, diseases of the pancreas, hepatitis C, chronic obstructive pulmonary disease, diabetes, hypertension, injuries and overdoses, depression, anxiety, and major psychosis [3, 4]. If, as many studies indicate, lesbians and gay men have higher rates of smoking and alcohol misuse than heterosexuals, it is likely that these segments of the population will also have higher rates of cancer, respiratory disease and cardiovascular morbidity than similarly aged heterosexuals.
Same-gender sexual orientation and sexual behavior have been identified as risk factors for smoking among women [5–15] in population-based studies [10–13, 16], convenience samples with internal comparison groups [14, 17] and convenience samples with outside comparison groups [5–9, 15]. Only two published studies, Gruskin et al. 2001 and Dibble et al 2004 [16, 18], found risk of current smoking was not significantly different for older lesbians compared to heterosexuals. Similarly, multivariate analyses of population-based studies [19–21] and studies relying on convenience samples [9, 22] have found that gay men and men who have sex with men are more likely to be current smokers than heterosexual men.
Evidence about the relationship of sexual orientation with risk for alcohol problems is more mixed. Previous studies have found lesbians to be more likely to have alcohol binges [10, 23] than heterosexual women and/or women who have sex with men only, to be dependent on alcohol [24–28], and to experience negative consequences related to alcohol [25, 26, 29–33]. In addition, they were more likely to seek help for alcohol-related problems and/or to be in recovery [5, 26, 29, 30]. However, several studies have not found differences by sexual orientation and/or sexual behavior [17, 18, 34, 35], and others suggest that if there is an association, it may be restricted to certain age groups [16, 31].
Studies on alcohol use post-19951 are also mixed for gay men. Some studies have found no difference in alcohol use between gays and heterosexuals [21, 24, 26, 27, 36, 37]. Other studies found gay men more likely to report frequent heavy drinking and less likely to be abstinent from alcohol than heterosexual men [9, 22, 31–33, 38].
Although there have been recent improvements in research about sexual orientation and alcohol and tobacco use, significant limitations remain. Some of the population-based studies included fewer than 100 lesbians and/or bisexual women or gays and/or bisexual men [12, 27, 29, 34, 36, 39], limiting the statistical power. While studies specifically targeting lesbians or gays have had larger samples, they have lacked comparison data obtained using the same methodology and instruments, making comparisons problematic. Also, many studies that used sexual behavior as a proxy for sexual orientation focused on recent sexual behavior, hence, misclassifying or eliminating people who were not sexually active in the given time interval.
The current study has several advantages when compared with previous research. We used data from two random sample surveys of the adult membership of the Kaiser Permanente Medical Care Program in Northern California. A participant will not be included in both databases, even if they are picked twice. Only one person in each household was chosen to participate in the survey. Combined they yielded a sample of 12,188 heterosexual women, 210 lesbians, 9342 heterosexual men and 331 gay men aged 20–64, resulting in 1.6% lesbians and 3.4% gay men. These numbers are similar to those of the current research. The most recent methodological strong study using a probability sample found that there were 1.4% lesbians and 3.2% gay men . In a study conducted by Drabble et al,, only.9% reported being lesbian and 1.7% reported being gay . Diamant et al, found 1.1% of her sample to be lesbian. The members of this prepaid health plan described in this study are sociodemographically representative of the non-Medicaid insured adult population in Northern California. Large numbers of participants enabled sufficient statistical power for meaningful comparisons between groups and to conduct multivariate analyses. Finally, the survey asked respondents to identify their sexual orientation rather than sexual behavior, resulting in a sample that includes people who identify as lesbian or gay but who may not have recently been sexually active.
This study is particularly important because it does not use a typical convenience sample and because it clarifies some of the results for gay men which are mixed.