The aim of this case-control study was to identify etiologic factors predictive for the development of severe pelvic organ prolapse. Three hundred and sixty-eight controls from a database describing pelvic organ support in the general population were identified as having known good pelvic organ support. Eighty-seven cases were identified from a urogynecology clinic with severe pelvic organ prolapse. The risk of severe prolapse was modeled using stepwise multiple logistic regression analysis. Additional analyses using chi2 and two-sample t-tests were conducted to determine differences in means for individual variables. Variables examined included age, gravidity, parity, number of vaginal deliveries, weight of largest infant delivered vaginally, menopause status, race, body mass index prior to pelvic surgery, and medical illnesses. The following four variables were selected in the regression analysis as predicting severe prolapse: age, weight of largest vaginal delivery, hysterectomy and previous prolapse surgery. Other variables that demonstrated statistically significant differences between groups by chi2 and two-sample t-tests were gravidity, parity, number of vaginal deliveries, menopausal status, race, history of incontinence surgery and the presence of hypertension. Variables that did not demonstrate any significant differences were body mass index, the presence of chronic obstructive pulmonous disease and diabetes mellitus. Advancing age, increasing weight of infants delivered vaginally, a history of hysterectomy and a history of previous prolapse surgery were found to be the strongest etiologic predictors of severe pelvic organ prolapse in our population.