Assessing the Impact of Early Medicaid Expansion on Insurance, Stage at Diagnosis, and Survival among Non-Elderly Cancer Patients Receiving Radiation Therapy

Abstract

Early evidence suggests that access to care has increased in non-elderly cancer patients due to insurance uptake associated with Medicaid expansion mandated by the Affordable Care Act (ACA). However, this has not been well studied in states that implemented the ACA early in 2010-2011, especially among radiotherapy (RT) recipients. Here, we assessed how early Medicaid expansion is associated with changes in insurance rates, stage at diagnosis, and survival among RT recipients. Using the Surveillance, Epidemiology, and End Results 18 database, we identified RT recipients aged 18-64 years diagnosed with a first primary malignancy between 2007 and 2014. Cases diagnosed 3 months before and 3 months after the date in which the states expanded were excluded to allow for a wash-out/phase-in period. Additionally, we excluded occult cases and 2014 diagnoses from states that expanded Medicaid in 2014. We compared changes in insurance rates, early (0-II), and late (IV) stage at diagnosis, and survival in cases from states that expanded Medicaid early (EEXP) to states that did not (NEEXP) using difference-in-differences analyses applied to linear probability and Cox proportional hazards models. A p-value < 0.05 was considered statistically significant. A total of 323,972 cases were identified. In EEXP states relative to NEEXP states, a decrease in the rate of uninsured (-1.28 percentage points [PP], 95% CI = -1.6 to -0.96, p < 0.001) was observed but with no change in the rate of Medicaid insurance (-0.01 PP, 95% CI = -0.48 to 0.46, p = 0.97). However, expansion-related increases in Medicaid insurance rates were found in counties with the highest quartile of rates of non-high school graduates (1.35 PP, 95% CI = 0.12 to 2.60, p = 0.032). An increase in early stage diagnoses (1.46 PP, 95% CI = 0.82 to 2.09, p < 0.001) and a decrease in late stage diagnoses (-1.26 PP, 95% CI = -1.8 to -0.72, p < 0.001) were identified in EEXP states relative to NEEXP states. However, no change in survival (HR: 1.01, 95% CI = 0.98 to 1.04, p = 0.47) was observed. For individual cancer sites, expansion-related increases in early stage diagnoses for breast (1.85 PP, 95% CI = 0.8 to 2.89, p < 0.001), colorectal (3.2 PP, 95% CI = 0.05 to 6.36, p = 0.047), and lung (1.64 PP, 95% CI = 0.51 to 2.78, p = 0.005) cancers were observed. Similarly, expansion-related decreases in late stage diagnoses for cervical (-5.19 PP, 95% CI = -8.63 to -1.75, p = 0.003), colorectal (-3.4 PP, 95% CI = -5.93 to -0.87, p = 0.008), and lung (-1.21 PP, 95% CI = -2.31 to -0.1, p = 0.002) cancers were identified. Medicaid expansion is associated with decreases in rates of uninsured, increases in early stage diagnoses, and decreases in late stage diagnoses among RT recipients. Further studies, including evaluating the impact of the 2014 Medicaid expansion on cancer prognosis, are needed to support these findings.