Purpose:
The commencement of quality-improvement initiatives such as Pay for Performance and the Physician Consortium for Performance Improvement has underscored calls to evaluate the quality of cancer care on a patient level for nationally representative samples.

Methods:
We sampled early-stage prostate cancer cases diagnosed in 2000 through 2001 from the American College of Surgeons National Cancer Data Base and explicitly reviewed medical records from 2,775 men (weighted total=55,160 cases) treated with radical prostatectomy or external-beam radiation therapy. We determined compliance with 29 quality-of-care disease-specific structure and process indicators developed by RAND, stratified by race, geographic region, and hospital type.

Results:
Overall compliance exceeded 70% for structural and pretherapy disease assessment indicators but was lower for documentation of pretreatment functioning (46.4% to 78.4%), surgical pathology (37.1% to 86.3%), radiation technique (62.6% to 88.3%), and follow-up (55%). Geographic variations were observed as higher compliance in the South Atlantic division than the New England division for having at least one board-certified urologist (odds ratio [OR], 9.2; 95% CI, 1.9 to 45.0), at least one board-certified radiation oncologist (OR, 3.3; 95% CI, 1.2 to 9.0), use of Gleason grading (OR, 4.1; 95% CI, 1.2 to 13.8), and administering total radiation dose≥70Gy (OR, 3.1; 95% CI, 1.6 to 6.1). Teaching/research hospitals and Comprehensive Cancer Centers had higher compliance than Community Cancer Centers, whereas racial differences were not observed for any indicator.

Conclusion:
The significant and unwarranted variations observed for these quality indicators by census division and hospital type illustrate the inconsistencies in prostate cancer care and represent potential targets for quality improvement. The lack of racial disparities suggests equity in care once a patient initiates treatment.