OBJECTIVE:We evaluated the relative importance of predictors of hysterectomy. STUDY DESIGN: A questionnaire survey of an Australia-wide sample of adult female twins was conducted; self-report data were validated against reports from treating physicians. RESULTS: A total of 3096 women (94%) and 366 physicians (87%) responded. The sensitivity of patient report of hysterectomy was 98.2%. Best predictors of hysterectomy were endometriosis (odds ratio 4.85, 95% confidence interval 3.17-7.43), medical consultation for menorrhagia (odds ratio 3.55, 95% confidence interval 2.47-5.12), joint effects of fibroids with medical consultation for chronic or persisting pelvic pain (odds ratio 3.34, 95% confidence interval 1.42 to 7.87), having smoked >40 cigarettes per day (odds ratio 3.24, 95% confidence interval 1.10-9.55), joint effects of fibroids with consultation for menstrual problems (odds ratio 2.61, 95% confidence interval 1.36-5.01), and tubal ligation (odds ratio 1.77, 95% confidence interval 1.31-2.39). Less-important predictors were age and higher education level (protective). CONCLUSION: Consulting a physician about pelvic pain and menstrual problems, especially heavy bleeding, are recognized steps toward hysterectomy. Of particular interest for future genetic analyses are the high odds of hysterectomy for women with endometriosis, fibroids, or menorrhagia. (Am J Obstet Gynecol 1999;180:945-54.) Key words: Hysterectomy, validation, predictors, risk factors Studies investigating pathways to hysterectomy performed for benign and nonemergency conditions have been rare. Clinical reviews have focused on the impact of new or changing techniques available to surgical practice and on developments in medical and pharmaceutical treatments. A health policy agenda concerning appropriate use of health care resources has influenced many other reviews of rates and indications. The first phase of the current study, with use of a genetic epidemiologic approach, attempted to identify predictors of hysterectomy and their causal role with use of data on Australian twins. 1 To our surprise, we found strong evidence for genetic factors influencing liability to hysterectomy, accounting for 66% of the total variance. Furthermore, these genetic influences were stable across birth cohorts in spite of differences in incidence resulting from aging and other factors such as secular changes in surgical practice and health service financing. 1 We therefore hypothesized that the high heritability of hysterectomy reflects genetic influences on gynecologic factors that are commonly indications for elective hysterectomy. Here we describe a survey implemented to collect data to test this hypothesis. In this study we focus on the relative importance of risk factors for hysterectomy and the phenotypic relationships between them. Genetic analyses making full use of the twin structure of the sample will be the subject of a further study. Given this epidemiologic focus, we also aimed to test the validity of retrospective self-reported data against d...