Population-based health surveys seldom assess sexual orientation, which results in the absence of a reliable measure of smoking among lesbians, gays, and bisexuals (LGB), a population perceived to have higher risks of tobacco-related diseases. This is the first study to compare the cigarette smoking rate of LGB with that of heterosexual individuals using a population-based sample with both male and female adults, and to identify which sub segments of LGB population are particularly burdened by tobacco use. California Health Interview Survey (CHIS), a population-based telephone survey was used to assess smoking prevalence and its correlates among respondents. Of 44,606 respondents, 343 self-identified as lesbian; 593 self-identified as gay; and 793 identified themselves as bisexual (511 female and 282 male). Statistical analysis was performed using SAS and SUDAAN. Lesbians' smoking rate (25.3%), was about 70% higher than that of heterosexual women (14.9%) Gay men had a smoking prevalence of 33.2%, comparing to heterosexual men (21.3%). After controlling for demographic variables, logistic regression analysis showed that lesbians and bisexual women were significantly more likely to smoke compared with heterosexual women (OR = 1.95 and OR = 2.08, respectively). Gay men were also significantly more likely to smoke than heterosexual men (OR = 2.13; 95% CI = 1.66-2.73). Being 35-44-years-old, non-Hispanic White, and having low-education attainment and low-household income were common demographic predictors of cigarette smoking among LGB. Our study provides the strongest evidence to date that lesbians, bisexual females, and gay men had significantly higher cigarette smoking prevalence rates than their heterosexual counterparts.