AIMS: (1) To study the prevalence of childhood sexual abuse before the age of 18 years (CSA) and life-time sexual abuse (LSA) in a Swedish female, general population, (2) to analyse associations between CSA and life-time alcohol dependence or abuse (ADA), and (3) to identify possible confounding factors. DESIGN AND PARTICIPANTS: The study was conducted in two phases. Phase 1: an alcohol problem screening questionnaire was sent to 3130 women aged 25-65. The answers were scored. Phase 2: based on the questionnaire scores, a randomly selected stratified sample of 479 women was invited for an interview. Of these, 316 women participated in a structured face-to-face interview. SETTING: A sector of Goteborg city with 100,000 inhabitants. MEASUREMENTS: The interviews focused on substance use and on social, psychological and behavioural characteristics, including experiences of sexual abuse. Clinical psychiatric diagnoses were made according to DSM-III-R. Bivariate analyses and multivariate logistic regression analyses were performed. FINDINGS: The prevalence of CSA and LSA was 9.8% and 13.9%, respectively. CSA increased the risk for life-time ADA and anxiety, but not for depression. When potential confounding factors (e.g. early background factors, depression and anxiety) were adjusted for, CSA under 13 years of age still predicted ADA in multivariate analyses, but CSA under 18 years of age did not. CONCLUSIONS: LSA, and especially CSA under 13 years of age, are factors that should be considered in treatment of women with ADA and in psychiatric treatment of women.