CLINICAL, PRACTICE SETTING, HEALTH-RELATED QUALITY OF LIFE, AND ATTITUDINAL PREDICTORS OF HYSTERECTOMY USE: RESULTS OF AN 8-YEAR LONGITUDINAL STUDY

Purpose:   Hysterectomy remains one of the most common major surgical procedures performed in this country, and wide practice variations are associated with its use.  We conducted an 8-year longitudinal study of 1420 sociodemographically diverse pre-menopausal women who were seeking care for non-cancerous uterine conditions at an academic medical center, a county hospital, a closed panel HMO, or a community-based practice and had not undergone hysterectomy, to identify predictors of use of and satisfaction with hysterectomy and alternatives.
Methods:  Participants were recruited in two waves (1998/1999; 2004/2004).  The first wave completed face-to-face interviews at baseline and every year thereafter for a maximum of 8 years; the second was followed for 4 years.  During these interviews, participants were asked about their use of and satisfaction with medical, surgical, and complementary treatments; their symptoms and symptom impact; and their attitudes toward hysterectomy and alternatives.  Time tradeoff utilities for current health and 6 hypothetical outcomes of hysterectomy, uterus-preserving surgery, and medical treatment were measured at these interviews, and treatment preferences were assessed using probability and symptom duration tradeoffs.  For this analysis, we fit a Cox proportional hazard model with time-varying covariates to identify clinical, sociodemographic, health-related quality-of-life and attitudinal determinants of hysterectomy use.  
Results:  200 women underwent hysterectomy during their follow-up period. Predictors of hysterectomy included presence of fibroids with bleeding (HR=1.53, p=.05) or pressure (HR=1.87, p=.02; referent group: bleeding without fibroids); entering menopause during the study (HR=.47, p=.002); prior treatment with a GnRH agonist (HR=1.97, p=.002), degree of symptom resolution (HR=.66, p<.001, for a 1-point increment on a 4-point scale) and site of care (women receiving care at the HMO were significantly less likely to undergo hysterectomy than women receiving care at the county facility (HR=.54, p<.01).  In addition, increasing impact of pelvic problems on health-related quality of life (p=.035) and on sexual functioning (p=.003) independently predicted hysterectomy use, as did higher scores on a “benefits of not having a uterus” scale (p<.001) and lower scores on a “hysterectomy concerns” scale (p<.001).
Conclusion: Attitudes toward the uterus and concerns about having a hysterectomy, as well as site of care, independently predict hysterectomy use, after controlling for clinical and sociodemographic characteristics, highlighting the importance of shared decision making in this context and the key role of practice setting.