The Impact of Massachusetts Health Reform on Cancer Diagnosis and Treatment

Breast and colorectal cancers are among the most prevalent forms of cancer in the United States, together accounting for over 380,000 new cancer cases in the US in 2017. Breast cancer (BCA) is the second leading cause of cancer death among US women, and colorectal cancer (CRC) is the second leading cause of cancer death overall. There are substantial disparities in BCA and CRC diagnosis, treatment, and outcomes in the US by insurance and socioeconomic status. Those who are uninsured or underinsured are significantly more likely to be diagnosed at late stages, less likely to receive recommended treatment, and consequently suffer poorer health outcomes. Insurance expansions such as those enacted by the Affordable Care Act (ACA) could improve diagnosis, treatment, and outcomes among the newly insured. Pre-ACA state-level coverage expansions can provide timely evidence of how expansions affect cancer diagnosis and treatment, and where gaps in the system remain.

The 2006 Massachusetts health reform substantially increased coverage in the state. Key provisions of the reform closely parallel those of the ACA, for which Massachusetts served as a model. Few studies have examined impacts of the reform on cancer diagnosis, treatment, or outcomes. This study uses the large expansion of health insurance coverage in Massachusetts as a natural experiment to investigate the effect of insurance coverage on BCA and CRC cancer diagnosis and treatment. We use a quasi-experimental framework to assess the effects of insurance expansions in Massachusetts on BCA and CRC diagnosis and treatment. Data come from the Massachusetts Cancer Registry and the National Cancer Institute’s Surveillance Epidemiology and End Results (SEER) cancer registries from comparison states. We exploit variation across states and in the expected impact of reform across zip codes within Massachusetts to estimate the relationship between health reform implementation and diagnosis and treatment outcomes in Massachusetts.

Preliminary results provide evidence of a shift to earlier stage at cancer diagnosis in Massachusetts after the reform relative to control states, particularly for CRC. Our estimates suggest about a 2 to 3 percentage point decrease in the likelihood that CRC cases are diagnosed at regional or distant stages, representing about a 5% to 8% decrease relative to Massachusetts diagnosis patterns in the pre-period. Evidence is weaker for BCA diagnoses, though models that compare trends in Massachusetts to Georgia, Kentucky, and Michigan (three SEER states with no Medicaid substantial Medicaid expansions during our study period) suggest a shift to earlier stage diagnosis that is similar to that seen for CRC. Within Massachusetts, we find an increase in time to treatment for breast cancer among patients living in high and low-income areas. We observe no change in time to treatment for CRC patients. Additional analysis will explore whether increased time to treatment for BCA patients is due to changes in type of treatment delivered or capacity constraints. Current results suggest that insurance expansions may improve early diagnosis of cancer, which could lead to improved health outcomes and reduced costs, though we see little evidence of changes in treatment for those with cancer.