Context: Although a quarter of US women undergo elective hysterectomy before menopause, controlled trials that evaluate the benefits and harms are lacking. Objective: To compare the effect of hysterectomy vs expanded medical treatment on health-related quality of life. Design, Setting, and Participants: A multicenter, randomized controlled trial (August 1997-December 2000) of 63 premenopausal women, aged 30 to 50 years, with abnormal uterine bleeding for a median of 4 years who were dissatisfied with medical treatments, including medroxyprogesterone acetate. The participants, who were patients at gynecology clinics and affiliated practices of 4 US academic medical centers, were followed up for 2 years. Interventions: Participants were randomly assigned to undergo hysterectomy or expanded medical treatment with estrogen and/or progesterone and/or a prostaglandin synthetase inhibitor. The hysterectomy route and medical regimen were determined by the participating gynecologist. Main Outcome Measures: The primary outcome was mental health measured by the Mental Component Summary (MCS) of the 36-item Short-Form Health Survey (SF-36). Secondary outcomes included physical health measured by the Physical Component Summary (PCS), symptom resolution and satisfaction, body image, and sexual functioning, as well as other aspects of mental health and general health perceptions. Results: At 6 months, women in the hysterectomy group had greater improvement in MCS scores than women in the medicine group (8 vs 2, P=.04). They also had greater improvement in symptom resolution (75 vs 29, P<.001), symptom satisfaction (44 vs 7, P<.001), interference with sex (41 vs 22, P=.003), sexual desire (21 vs 3, P=.01), health distress (33 vs 13, P=.009), sleep problems (13 vs 1, P=.03), overall health (12 vs 2, P=.006), and satisfaction with health (31 vs 14, P=.01). By the end of the study, 17 (53%) of the women in the medicine group had requested and received hysterectomy, and these women reported improvements in quality-of-life outcomes during the 2 years that were similar to those reported by women randomized to the hysterectomy group. Women who continued medical treatment also reported some improvements (P<.001 for within-group change in many outcomes), with the result that most differences between randomized groups at the end of the study were no longer statistically significant in the intention-to-treat analysis. Conclusions: Among women with abnormal uterine bleeding and dissatisfaction with medroxyprogesterone, hysterectomy was superior to expanded medical treatment for improving health-related quality-of-life after 6 months. With longer follow-up, half the women randomized to medicine elected to undergo hysterectomy, with similar and lasting quality-of-life improvements; those who continued medical treatment also reported some improvements.