This case-control analysis presents odds ratios for active and passive cigarette smoke exposure and cervical intraepithelial neoplasia of levels II and III (CIN II and CIN III) while controlling for confounders. From 1987 to 1988, 103 biopsy-conformed incident cases of CIN II or III and 268 controls with normal cervical cytology were enrolled. Seventy % of cases were cigarette smokers, while only 30% of controls had ever smoked. The adjusted odds ratio for current cigarette smoking was 3.4 (95% confidence interval, 1.7-7.0). The following confounders were included in logistic regression models: age, race, education, number of sex partners, contraceptive use, sexually transmitted disease history, and Pap smear history. The risk of CIN II/III increased with increasing years of cigarette smoking and with increasing pack-years of exposure. Smoking was associated more strongly with CIN III than CIN II. The effect of passive cigarette smoke exposure was explored separately for smokers and nonsmokers and was found not to be consistently associated with CIN II/III when controlling for confounders.