+ The Impact of Health Insurance on Preventive Care and Health Behaviors: Evidence from the First Two Years of the ACA Medicaid Expansions (with Kosali Simon and John Cawley, Journal of Policy Analysis and Management)

The U.S. population receives suboptimal levels of preventive care and has a high prevalence of risky health behaviors. One goal of the Affordable Care Act (ACA) was to increase preventive care and improve health behaviors by expanding access to health insurance. This paper estimates how the ACA-facilitated state-level expansions of Medicaid in 2014 affected these outcomes. Using data from the Behavioral Risk Factor Surveillance System, and a difference-in-differences model that compares states that did and did not expand Medicaid, we examine the impact of the expansions on preventive care (e.g., dental visits, immunizations, mammograms, cancer screenings), risky health behaviors (e.g., smoking, heavy drinking, lack of exercise, obesity), and self-assessed health. We find that the expansions increased insurance coverage and access to care among the targeted population of low-income childless adults. The expansions also increased use of certain forms of preventive care, but there is no evidence that they increased ex ante moral hazard (i.e., there is no evidence that risky health behaviors increased in response to health insurance coverage). The Medicaid expansions also modestly improved self-assessed health.

Press coverage: Newswise, UPI, Health Medicine Network, Healio, Science Daily, Becker's Hospital Review, Insurance News Net, EurekAlert, myScience, Breitbart, Medical XPress, Health Management Technology, IU Bloomington Newsroom, Indiana Daily Student, Cornell Chronicle, WVPE 88.1, WBOI 89.1, APPAM Public Policy News, Health Affairs Blog


Before 2014, the primary path to Medicaid eligibility for non-elderly childless adults was the Supplemental Security Income program (SSI). The SSI program typically confers immediate Medicaid eligibility to adults with a work-limiting disability, low income, and limited assets. After 2014, in the 32 states and DC that expanded Medicaid, low-income adults became eligible for Medicaid without having to obtain “disabled” status. In the non-expansion states, SSI (which allows for lower earnings limits than the 2014 Medicaid expansion) remained the primary avenue for non-elderly childless adults to obtain Medicaid. We use data from the American Community Survey and the Social Security Administration to estimate the impact of the Medicaid expansion on SSI participation. We find that in states that expanded Medicaid, participation in the SSI program decreased by about 3 percent relative to pre-expansion levels. This suggests Medicaid expansion may reduce SSI participation.


+ Medicaid Expansion under the Affordable Care Act and Insurance Coverage in Rural and Urban Areas (with Kosali Simon and Michael Hendryx, Journal of Rural Health)

Purpose: To analyze the differential rural-urban impacts of the Affordable Care Act Medicaid expansion on low-income childless adults’ health insurance coverage. Methods: Using data from the American Community Survey years 2011-2015, we conducted a difference-in-differences regression analysis to test for changes in the probability of low-income childless adults having insurance in states that expanded Medicaid versus states that did not expand, in rural versus urban areas. Analyses employed survey weights, adjusted for covariates, and included a set of falsification tests as well as sensitivity analyses. Findings: Medicaid expansion under the Affordable Care Act increased the probability of Medicaid coverage for targeted populations in rural and urban areas, with a significantly greater increase in rural areas (P < .05), but some of these gains were offset by reductions in individual purchased insurance among rural populations (P < .01). Falsification tests showed that the insurance increases were specific to low-income childless adults, as expected, and were largely insignificant for other populations. Conclusions: The Medicaid expansion increased the probability of having “any insurance” for the pooled urban and rural low-income populations, and it specifically increased Medicaid coverage more in rural versus urban populations. There was some evidence that the expansion was accompanied by some shifting from individual purchased insurance to Medicaid in rural areas, and there is a need for future work to understand the implications of this shift on expenditures, access to care and utilization.




+ Changes in Insurance Coverage Among Cancer Patients After the Affordable Care Act (with Lindsay Sabik, Kosali Simon, and Benjamin Sommers) (under review)

The Affordable Care Act (ACA) has reduced uninsurance to an historic low. Debate continues about potential changes to the law, which could affect coverage for millions, particularly those with pre-existing conditions. Meanwhile, cancer is the leading cause of death among Americans under age 65. Cancer treatment is often unaffordable for uninsured patients, and some studies suggest expanding insurance could improve cancer diagnosis, treatment, and outcomes. Our objective was to quantify changes in health insurance under the ACA among patients newly diagnosed with cancer.

+ The Medicaid Expansions of 2014 Increased Early Stage Cancer Diagnoses (with John Cawley, Kosali Simon, and Lindsay Sabik) (under review)

Importance: Cancer detection, particularly at an early stage, is a key public health priority in the United States. Objective: To determine whether the 2014 Medicaid expansions facilitated by the Affordable Care Act (ACA) impacted early cancer diagnosis rates for non-elderly adults. Design: Using SEER Cancer Registry data from 2010 through 2014, a difference-in-differences model of cancer diagnosis rates was estimated, both overall and by stage, comparing changes in county-level diagnosis rates in states that expanded Medicaid in 2014 to those that did not expand Medicaid. Setting: Population-based analysis of the SEER Cancer Registry data. Participants: The SEER Cancer Registry collected comprehensive data on all cases of cancer reported in participating areas, covering approximately 28% of the population and representative of national demographics. N=3,055 county-year observations. Exposure(s): Expansions of Medicaid under the Affordable Care Act (ACA) in January 2014. Main Outcome(s) and Measure(s): Outcomes analyzed include county-level cancer diagnosis rate per 100,000 population, early stage cancer diagnosis rate per 100,000 population, late stage cancer diagnosis rate per 100,000 population, and ratio of early to late stage diagnoses. Results: Among the 611 counties in this study, early stage diagnoses increased by 15.0 per 100,000 population (95% CI, 4.2 to 25.9) or 6%. Late stage and overall diagnoses did not change significantly, and the ratio of early to late stage diagnoses increased by 0.17 (95% CI, 0.04 to 0.30). Conclusions and Relevance: In their first year, Medicaid expansions were associated with an increase in early stage cancer diagnosis. This suggests that expanding public health insurance may be an avenue for increasing early cancer detection, which is associated with improved patient outcomes, including reduced mortality.

+ Gains in Health Insurance Coverage Explain Variation in Democratic Vote Share in the 2016 Presidential Election" (with Alex Hollingsworth, Aaron Carroll, John Cawley, and Kosali Simon) (under review)

Results of the 2016 Presidential election departed considerably from polling forecasts. The Affordable Care Act (ACA) was a major issue in the election, and one possible factor explaining geographic variation in the Presidential vote is changes in health insurance coverage over time. We estimate how gains in health insurance coverage at the county level correlate with the change in the Democratic share of the Presidential vote since the previous Presidential election (2016 versus 2012).

+ The Impact of the Tennessee Medicaid Disenrollment on Rural vs. Urban Populations: An ACA Presage? (with Michael Hendryx and Kosali Simon)

Objective: To examine differential impacts of the Tennessee Medicaid disenrollment on insurance coverage, access to care, and self-assessed health for rural vs. urban, White vs. non-White, and male vs. female populations. Methods: We analyzed the 2001-09 Behavioral Risk Factors Surveillance System by geography, race/ethnicity, sex, year (2001-04 and 2006-09), state (Tennessee, Alabama, Arkansas, Georgia, Kentucky, and Virginia), and year-by-state interactions. We estimated difference-in-differences and triple differences regression models comparing outcomes in Tennessee to outcomes in control states, before and after the 2005 Medicaid disenrollment. Results: After the disenrollment, insurance coverage rates dropped more for rural people, White people, and women. Conclusions: While the overall coverage and access to care impacts of the Tennessee Medicaid disenrollment were negative, some groups were hurt more than others. Policy implications: At a time of policy uncertainty, our study provides suggestive evidence that large cutbacks in public insurance programs may particularly hurt rural, White, and female populations.



  • “Insurance Market Consolidation and Health Insurance Premiums” (with Haizhen Lin)
  • “Effects of Expanded Insurance Coverage Options on Insurance Status for Self-Employed Individuals” (with Sandra Decker & Asako Moriya)
  • “The Impact of Public Health Insurance Expansion on Crime”