Presenter(s)
H. S. Mahmood1, D. DeStephano2, E. Elkin2, K. N. Kim3, D. P. Horowitz4, and S. K. Cheng5; 1Columbia University Medical Center, New York, NY, 2Columbia University, New York, NY, 3Columbia University Vagelos College of Physicians and Surgeons, New York, NY, 4Department of Radiation Oncology, Columbia University Irving Medical Center, New York, NY, 5Herbert Irving Comprehensive Cancer Center, Columbia University Irving Medical Center, New York, NY
Purpose/Objective(s): More than half of the 63 million Americans covered by Medicare are enrolled in Medicare Advantage (MA) plans, while the rest are insured in the traditional the fee-for-service (FFS) model. MA plans continue to increase in popularity, with an growing percentage of patients choosing such plans, including low-income individuals. Historically, patients in Medicare managed care plans were diagnosed at earlier stages of some cancers, but it is unclear whether those differences remain as the program has expanded. Our goal is to assess whether stage at diagnosis for multiple cancer subtypes differs between patients with FFS Medicare and those in MA.
Materials/Methods: Using the SEER Medicare Database, we identified patients diagnosed with breast, colorectal, melanoma, prostate, cervical, and lung cancer in 2007-2021. Patients were classified by SEER summary stage as distant/regional disease vs localized/in situ disease. We compared the percentage of patients with regional/distant disease in FFS and MA. We estimated the adjusted odds ratio (AOR) of being diagnosed at a later stage, controlling for age, race, ethnicity, urban/rural setting, Medicaid status, socioeconomic status, year of diagnosis, and state, using multivariable logistic regression.
Results: The percentage of patients diagnosed with distant or regional stages for breast cancer (n = 517,031) in FFS was 26.3% vs 25.2% in MA, with an AOR of 1.030 (1.016-1.045) p<.0001, colorectal cancer (n = 334,133) was 55.8% vs 56.2%, AOR: 1.046 (1.030-1.063) p<.0001, melanoma (n = 236,189) was 10.9% vs 10.6%, AOR: 1.022 (0.992-1.052) p=0.16, prostate cancer (n= 522,711) was 20.7% vs 21.5%, AOR: 1.047 (1.031-1.062) p<.0001, cervical cancer (n = 10,214) was 70.2% vs 70.9%, AOR: 1.020 (0.927-1.122) p=0.69, lung cancer (n = 557,172) was 72.3% vs 73.3%, AOR: 0.933 (0.920-0.945) p<.0001.
Conclusion: In this large and contemporary analysis of screenable cancers, we found FFS patients to be consistently diagnosed at later stages than MA over the years, although absolute differences were generally small even when statistically significant. Lung cancer was a notable exception. Percentages do not always correspond with AORs because MA enrollment has increased over time, while the proportion of patients diagnosed at later stages is not constant over time. This was notable for prostate and colorectal cancers. These findings may reflect continued differing enrollee selection into MA or greater screening intensity and care coordination in MA plans. Further longitudinal analyses of temporal trends will be important to determine whether these differences hold over time as MA enrollment continues to expand.