Several case-control studies, mostly of prevalent disease, have suggested that systemic lupus erythematosus (SLE) is positively associated with cigarette smoking and inversely associated with alcohol consumption. We prospectively investigated the associations of smoking and alcohol consumption with incident SLE in the Black Women's Health Study (BWHS). In 1995, 64,500 African-American women provided information on demographic characteristics, reproductive and medical histories, smoking, and alcohol consumption. Followup questionnaires in 1997 and 1999 ascertained incident cases of SLE. Cox proportional hazards regression was used to estimate incidence rate ratios (IRR) and 95% confidence intervals (CI). Sixty-seven women reported a new diagnosis of SLE and use of appropriate medication for that illness. In multivariate analyses, the IRR for current and past smoking were 1.6 (both 95% CI 0.8-3.3). The risk was greater for women who began smoking before age 19 years (IRR 1.9, 95% CI 1.0-3.6). Neither current alcohol consumption (IRR 1.0, 95% CI 0.4-2.4) nor past alcohol consumption (IRR 0.9, 95% CI 0.3-2.7) was associated with SLE. Our results suggest an increased risk of SLE among smokers, but no effect of alcohol consumption on risk. The inverse association of alcohol consumption with SLE found in studies of prevalent disease may have resulted from women with SLE giving up drinking.