Variation in local treatment of prostate cancer by hospital within New Jersey

• Published in: Journal of clinical oncology Vol. 24; no. 18_suppl; p. 14551
• Main Authors: Blom, T. R., Somer, R. A., Shih, W., Sarno, M., McNally, D., Todd, M., Yao, S., Lu-Yao, G., Dipaola, R., Stein, M. N.
• Format: Journal Article
• Language: English
• Published: 20.06.2006
• ISSN: 0732-183X; 1527-7755
• DOI: 10.1200/jco.2006.24.18_suppl.14551

Abstract only 14551 Background: Although prostate cancer diagnosis and treatment vary through the US, few studies have assessed variations in local therapy between multiple hospitals located within a very narrow geographic region. Methods: To determine the variation of diagnosis and treatment of prostate cancer between hospitals in a local region, data was derived and analyzed from 1,301 patients from fifteen Cancer Institute of New Jersey Oncology Group network hospitals. Tumor registry data from patients with a histological confirmed diagnosis of prostate cancer accessioned over a one-year period between January 1, 2003 and December 31, 2003 was analyzed by a multivariate logistic regression model. Results: Median age of the cohort was 66 and mean number of cases diagnosed at these hospitals for this period was 87 (26–161). When the cohort was assessed overall, 192 cases assessed were African American, 1049 Caucasian, 51 Hispanic, and 32 Asian. Median Gleason score was 6. A total of 463 patients underwent Prostatectomy, and 538 had radiation therapy. Multivariate analysis demonstrated that age and hospital were associated with use of surgery versus no surgery. Race and Gleason score were not associated with use of surgery in this group of hospitals. Multivariate analysis also demonstrated that age and hospital were associated with utilizing radiation therapy versus not utilizing radiation therapy. Again, race and Gleason score were not associated with utilization of radiation therapy in these NJ hospitals. Conclusions: Overall, the most important factors predicting for surgery, and radiation therapy, were age and hospital, which were independent of race and Gleason score. These data support our ongoing assessment of hospital and geographic characteristics that may be responsible for these associations. No significant financial relationships to disclose.