• Department of Epidemiology, Johns Hopkins School of Hygiene and Public Health, Baltimore, MD 21205, USA.

OBJECTIVES: Previous reports have documented a geographic variation in the use of radical prostatectomy. We examined whether this phenomenon can be explained by factors other than geography alone.

METHODS: This study was based on the data from nine geographic regions of the Surveillance, Epidemiology, and End Results (SEER) program for the years 1983 through 1994. Patients with localized or regional prostate cancer were included in the analysis. Logistic regression analysis was used to investigate the influence of geographic and demographic factors on the use of radical prostatectomy. The squared multiple correlation coefficient R(2) was used to measure the proportion of variation in the selection of radical prostatectomy explained by each factor of interest.

RESULTS: As previously reported, the use of radical prostatectomy was significantly associated with geographic location; the degree of geographic variation varied as a function of age and was most dramatic in the youngest (younger than 45 years) and the oldest (75 years or older) groups. Overall, however, geography explained less than 2% of the total variation in the use of radical prostatectomy. Age was the most important factor that influenced the use of radical prostatectomy.

CONCLUSIONS: Geography explains only a small proportion of the variation in the use of radical prostatectomy. In fact, of the factors examined, only age appeared to meaningfully explain the variation in the use of radical prostatectomy. Overall, our ability to explain the variation in the use of radical prostatectomy remains meager, and new factors must be identified if we are to better understand how patients and physicians make clinical decisions.