Short-term effects of the California Earned Income Tax Credit on colorectal cancer risk factors and prevention

• In Event: Policies to Incentivize Engagement in Healthy Behaviors

Abstract

Colorectal cancer (CRC) is the second leading cause of cancer mortality in the United States and is characterized by large socioeconomic disparities in risk factors, access to preventive screening, and survival. We study whether the California Earned Income Tax Credit (CalEITC), which provides relatively higher payments for very low-income families in the “phase-in” section of the federal EITC benefit schedule, is associated with changes in clinical indicators of colorectal cancer risk, utilization of preventive healthcare, and CRC diagnosis among primary care patients in California. In our prior work, we find that the passage of the CalEITC in 2015 was associated with significant improvements in CRC-related risk factors including reductions in cost as a barrier to healthcare and fewer poor mental and physical health days (Rong, Lowenstein, and Rehkopf 2024). Given our prior findings, we hypothesize that the CalEITC will be associated with short-term increases in the utilization of primary care, preventive screening for CRC, and reductions in key modifiable risk factors. 

We draw on detailed, patient-level electronic health record (EHR) data from the American Family Cohort (AFC) from 2014 to 2019. The AFC is a large and geographically diverse electronic health record database from primary care clinics across the US. Unlike most primary care research datasets limited to single payers, states, or hospital systems, the AFC is diverse with respect to health insurance status, race and ethnicity, and urban/rural status. Our sample contains 435,349 primary care patients from 1,708 unique California zip codes. We proxy for exposure to CalEITC using annual, zip code-level EITC claims data from the California Franchise Tax Board and construct area-level measures such as the population share claiming any CalEITC benefits and the inflation-adjusted credit amount per claimant. 

We estimate a series of difference-in-differences and event study models leveraging this area-level variation and exploit temporal variation due to the seasonal disbursement of EITC benefits. Prior research demonstrates that EITC beneficiaries are most likely to receive credits between February and April, allowing us to compare short-term changes in clinical outcomes while differencing-out seasonality in treatment utilization between high- and low-treatment zip codes. Preliminary results show a positive association between the zip code level share of CalEITC claimants and screening rates from February to April among adults ages 45-64, with the association becoming weaker from May to January. Our difference-in-difference models will also test outcomes that are more likely to be affected by short-term income supplementation among households with children, including modifiable risk factors such as tobacco use, physical activity, and diet quality. To strengthen identification, we will also estimate triple-difference models in which we proxy for individual-level EITC eligibility using health insurance information in the EHR database. 

References:

Rong, M., Lowenstein, C., & Rehkopf, D. H. (2024). Effects of the 2015 California EITC Expansion on Colorectal Cancer Risk Factors. Available at SSRN 4931575.

Authors

• Analisa Jia, Stanford University
  Presenting Author

• Christopher Lowenstein, Stanford University
  Non-Presenting Co-Author

• Maxwell Rong, Stanford University
  Non-Presenting Co-Author

• Britni Wilcher, Stanford University
  Non-Presenting Co-Author

• David Harold Rehkopf, Stanford Center on Poverty & Inequality
  Non-Presenting Co-Author