Impact of Medicare Advantage Enrollment on Utilization of Intensity-Modulated Radiation Therapy and Cost of Care for Cancer Treatment

Abstract

Intensity-modulated radiation therapy (IMRT) is an important driver of rising costs in oncology care, but the level of evidence supporting its routine use varies across disease sites, including breast, lung, and prostate. While Medicare Advantage (MA) plans have incentives to reduce health care spending, the effect of MA enrollment on utilization of high-cost medical services and quality of care is unclear, especially for cancer patients. We examined differences in IMRT utilization and health care spending between MA and traditional fee-for-service (FFS) Medicare for patients receiving radiation therapy (RT) across these disease sites. We performed a retrospective cohort study using 2009-2012 health insurance claims data from the Health Care Cost Institute (HCCI) and the Centers for Medicare and Medicaid Services (CMS). We identified 2,571 MA and 6,062 FFS Medicare beneficiaries from the HCCI and CMS databases, respectively, who were newly diagnosed with non-metastatic breast, lung, or prostate cancer in 2011 and were treated with external beam RT within 12 months of diagnosis. We measured utilization rate of IMRT, unit price of RT delivery, and RT-specific spending among patients within each cancer-type cohort. Outcomes were adjusted for differences in patient characteristics (age, sex, and Charlson co-morbidity index) and geography between MA and FFS beneficiaries. Statistical significance of the differences between groups was assessed using two-sample t tests. In both MA and FFS Medicare, the adjusted rate of IMRT utilization was highest among prostate patients and lowest among breast patients. Across all three disease sites, IMRT utilization was lower in MA than FFS Medicare, but the difference was statistically significant only for the prostate cohort (18.0 vs. 20.2%, p=0.149 for breast; 32.8 vs. 36.0%, p=0.385 for lung; 94.0 vs. 98.5%, p<0.001 for prostate). Average unit price of IMRT delivery was not significantly different in MA versus FFS Medicare. However, per patient RT-related costs were lower among MA patients in the breast and prostate cancer cohorts ($12,630 vs. $13,214 and $25,990 vs. $27,685, respectively; p=0.005 and p<0.001), but not significantly different among lung cancer patients ($15,653 vs. $15,024; p=0.301). IMRT utilization rates in both MA and FFS Medicare were consistent with the relative level of evidence supporting its use in each disease site. MA enrollment was associated with modestly lower IMRT utilization for all patients, with no evidence of a differential impact based on cancer type. The unit price of IMRT delivery was similar in MA versus FFS Medicare. We found no evidence that MA plans adopted utilization management tools that more effectively targeted high-cost therapies based on clinical evidence than in traditional Medicare.