RESUMO
OBJECTIVE: To compare the Medicare managed care (MC) and fee-for-service (FFS) sectors on stage at diagnosis and treatment patterns for prostate, female breast, and colorectal cancers, and to examine patterns across MC plans. DATA: Surveillance, Epidemiology, and End Results-Medicare linked data. METHODS: Among cases diagnosed at ages 65-79 between 1998 and 2002, we selected all MC enrollees (n = 42,467) and beneficiaries in FFS (n = 82,998) who resided in the same counties. MC and FFS samples were compared using logistic regression, adjusting for demographic, geographic, and clinical covariates. RESULTS: The percentage of late stage cases was similar in MC and FFS for prostate and colorectal cancers; there were slightly fewer late stage breast cancer cases in MC after adjustment (7.3% vs. 8.5%, P < 0.001). Within MC, radical prostatectomy was performed less frequently for clinically localized prostate cancer (18.3% vs. 22.4%, P < 0.0001), and 12 or more lymph nodes were examined less often for resected colon cancer cases (40.9% vs. 43.0%, P < 0.05). Treatment patterns for early stage breast cancer were similar in MC and FFS. Analyses of treatment patterns at the individual plan level revealed significant variation among plans, as well as within the FFS sector, for all 3 types of cancer. CONCLUSIONS: On average, there are few significant differences in cancer diagnosis and treatment between MC and FFS. Such comparisons, however, mask the wide variability among MC plans, as well as FFS providers. Observed variation in patterns of care may be related to patient selection, but can potentially lead to outcome differences. These findings support the need for quality measures to evaluate plan practices and performance.