Bottom Line: Adoption of Medicaid expansion in U.S. states appeared to improve both five-year cause-specific and overall survival in cancer patients.
Journal in Which the Study was Published: Cancer Discovery , a journal of the American Association for Cancer Research (AACR)
Author: Elizabeth Schafer, MPH, associate scientist at the American Cancer Society
Background: In 2014, a provision under the Patient Protection and Affordable Care Act (ACA) went into effect that allowed states to opt into expanding Medicaid eligibility to adults with incomes up to 138% of the federal poverty guidelines . Existing studies have linked Medicaid expansion to increased insurance coverage, access to cancer screening services, and improved two-year survival rates, but its impact on longer-term survival among individuals diagnosed with cancer remains unknown, according to Schafer.
How the Study was Conducted: To answer their question, Schafer and her coauthors used a difference-in-differences (DD) study design that compared data from states during 2007-2008—a period prior to Medicaid expansion—with data from 2014-2015, when many states adopted the expansion. This design allowed the research team to treat Medicaid expansion akin to an experimental intervention with effects that could be measured against a control group of states that had not enacted the expansion.
Using the Cancer in North America Survival dataset, Schafer and her coauthors analyzed data from 1,423,983 cancer cases diagnosed in adults aged 18 to 59 years and compared five-year survival from follow-up data obtained in the years following both the 2007-2008 cohort and the 2014-2015 cohort. The analysis assessed and compared 26 expansion states with 12 states that did not expand Medicaid at that time.
In the context of Schafer's study, results presented as percentage points (ppt) in DD are the subtraction of the difference between expansion states and nonexpansion states in 2007-2008 from the difference between expansion and nonexpansion states in 2014-2015. DD of survival measurements shows whether Medicaid expansion correlates with the trajectory of survival rates relative to states where it was not adopted.
Results: After adjusting for covariates, the researchers observed that, among cancer patients living in rural areas, Medicaid expansion was associated with significant DD improvements in cause-specific and overall survival of 2.55 ppt and 3.03 ppt, respectively. Similarly, patients living in high-poverty areas in states with Medicaid expansion experienced significant DD improvements in cause-specific and overall survival: 1.54 ppt and 1.69 ppt, respectively.
Patients with cancers associated with higher mortality also experienced significant improvements in both cause-specific and overall survival if they lived in Medicaid expansion states.
Overall survival among non-Hispanic Black individuals was also significantly higher in Medicaid expansion states, with a DD of 1.05 ppt. Expansion was associated with significant increases in both cause-specific and overall survival in non-Hispanic Whites as well, with DDs of 0.37 ppt and 0.57 ppt, respectively. In expansion states, an adjusted cause-specific survival improvement was observed for all cancer patients irrespective of individual variables, but the improvement was not statistically significant.
Author's Comments: "Studying the impact of Medicaid expansion on longer-term survival outcomes among individuals diagnosed with cancer is important for guiding evidence-based decision-making by policymakers and advocates for maintaining and expanding Medicaid," Schafer said.
"The evidence supporting Medicaid expansion in improving outcomes for cancer patients is clear," she added. "Research has shown that Medicaid expansion can increase cancer screening prevalence, early-stage diagnosis, short-term survival, and now—according to our own analysis—five-year survival. These findings underscore the importance of protecting and expanding Medicaid in the remaining 10 nonexpansion states to improve outcomes for all individuals."
Study Limitations: Limitations of the study include the possible confounding factor of the ACA's general broadening of health insurance coverage, as more people with health insurance and access to early cancer detection and treatment could skew the data. Additionally, the dataset also excluded individuals with missing cause-of-death and sociodemographic information.
Funding & Disclosures: The study was supported by the Intramural Research Department of the American Cancer Society. Schafer reported no conflicts of interest.
