The study used a two-stage design. The first consisted of sending a postal questionnaire to 2,250 women, randomly selected from the electoral rolls of the study's jurisdiction in New Zealand. The postal questionnaire included questions on general health, mental health, demographics, social support, and experience of sexual and physical abuse both as a child and during adult life. The second stage was an interview phase. All 298 women under the age of 65 who had experienced unwanted sexual advances prior to the age of 16 years old were invited for an interview. A group of the same size was randomly selected from the 716 subjects who did not report sexual abuse either as a child or as an adult; this group was also interviewed; the interviews were conducted over 14 months in 1989 and 1990. The interview included a detailed schedule that collected data on variables that have been associated with an increased risk of abuse during childhood. Measures of adult outcomes were also included in the interview. Findings show that a history of any form of abuse was associated with increased rates of psychopathology, sexual difficulties, decreased self-esteem, and interpersonal problems. The similarities between the three forms of abuse in terms of their association with negative adult outcomes was more apparent than any differences, although there was a trend for sexual abuse to be particularly associated with sexual problems, emotional abuse with low self-esteem, and physical abuse with marital breakdown. Abuse of all types was more frequent in those from disturbed and disrupted family backgrounds. The background factors associated with reports of abuse were themselves often associated with the same range of negative adult outcomes as for abuse. Logistic regressions show that some of the apparent associations between abuse and adult problems were accounted for by this matrix of childhood disadvantages from which abuse so often emerged. 4 tables and 41 references