The impact of nonclinical factors on practice variations: the case of hysterectomies

OBJECTIVE: This study investigates the role of nonclinical factors (physician characteristics) in explaining variations in hysterectomy practice patterns. DATA SOURCES AND STUDY SETTING: Patient discharge data are obtained from the Arizona state discharge database for the years 1989-1991. Physician data are obtained from the Arizona State Medical Association. The analyses are based on 36,104 cases performed by 339 physicians in 43 hospitals. STUDY DESIGN: This article measures the impact of physician factors on the decision to perform a hysterectomy, controlling for a host of patient and hospital characteristics. Physician factors include background characteristics and training, medical experience, and physician's practice style. Physician effects are evaluated in terms of their overall contribution to the explanatory power of regression models, as well as in terms of specific hypotheses to be tested. DATA COLLECTION: The sources of data were linked to produce one record per patient. PRINCIPAL FINDINGS: As a set, physician factors account for a statistically significant increase in the explanatory power of the model after addition of patient and hospital effects. Parameter estimates provide further support for the hypothesized effects of physicians' background, experience, and practice characteristics. CONCLUSIONS: Overall, the results confirm that nonclinical (physician) factors play a statistically significant role in the hysterectomy decision. Substantively, however, these factors play a smaller, secondary role compared to that of clinical and patient factors in explaining practice variations in hysterectomies. The results suggest that efforts to reduce unnecessary hysterectomies should be directed at identifying the appropriate clinical indications for hysterectomy and disseminating this information to physicians and patients. This may require such intervention strategies as continuing clinical education, promulgation of explicit practice guidelines, peer review, public education, and greater understanding and inclusion of patient preference in the decision process.