The influence of physician densities and patient characteristics on the decision to treat prostate cancer patients with varying clinical benefit.

Abstract

19 Background: Our study investigates whether regional variations in physician densities or patient populations influence the decision to treat patients with prostate cancer. Methods: Data from the NCI Surveillance, Epidemiology, and End Results (SEER) database was combined with the health system factors from the Area Resource File for Health Service Areas (HSA). Patients treated from 2004-2007 with low likelihood of benefit from treatment (Gleason 6, PSA <10.0, stage T1cN0M0, and age >75) or high likelihood of benefit (Gleason 8-10 or PSA >20.0 or stage T3N0M0, aged <60) were analyzed. Health system factors analyzed were; radiation oncologist, urologist, and primary care provider densities, % unemployment, and % with at least a high school education. Patient factors analyzed included race and marital status. A logistic regression model was built to test the association between variables and attempted curative therapy (radical prostatectomy, EBRT, and/or brachytherapy). Results: For the 2630 pts with the lowest likelihood of benefit, 56.1% underwent some form of attempted curative therapy. White pts and those residing in HSAs with better-educated populations were less likely to be treated, whereas married pts (OR 0.69 [SE 0.07, p<.001]) were more likely to receive treatment. For the 3207 pts with the highest likelihood of clinical benefit from treatment, 91.1% underwent curative therapy as first initial treatment. Being married was again associated with a higher odds of curative therapy (OR 1.68 [SE 0.22], p<.001) whereas being black was associated with a lower odds (OR 0.43 [SE 0.16], p=.02). Both radiation oncologist density and urologist density were not found to influence receipt of treatment in either group. Conclusions: Radiation oncologist and urologist densities do not seem to influence the decision to treat prostate cancer at the regional level, whereas marital status is a significant predictor of attempted treatment. Though health system factors may influence the receipt of proper therapy, a patient’s decision-to receive treatment is potentially more influenced by his interpersonal relationships.