Effect of advertising for a state-based health insurance marketplace on information-seeking behavior Public Deposited

Downloadable Content

Download PDF
Last Modified
  • February 22, 2019
  • Shafer, Paul
    • ORCID: http://orcid.org/0000-0003-0654-5821
    • Affiliation: Gillings School of Global Public Health, Department of Health Policy and Management
  • Baum, Laura
    • Other Affiliation: Wesleyan Media Project, Wesleyan University
  • Fowler, Erika Franklin
    • Other Affiliation: Wesleyan Media Project, Wesleyan University
  • Gollust, Sarah
    • Other Affiliation: Division of Health Policy and Management, School of Public Health, University of Minnesota
  • Introduction • Kentucky was an early success stories under the Affordable Care Act through its statebased marketplace, known as kynect • The state sponsored an award-winning multimedia campaign to create awareness and educate its residents about the opportunity to gain coverage • In 2015, the new administration failed to renew a contract that resulted in the cancellation of television advertising for the final four weeks of the 2016 open enrollment period • We leverage this change to identify whether a dose-response relationship exists between state-sponsored television advertising and information seeking about health insurance Methods • Advertising data from 10 media markets across Kentucky was obtained for October 2013 through January 2016 from Kantar Media/CMAG through a partnership with the Wesleyan Media Project • State-level information-seeking was derived from reports made by the Office of the Kentucky Health Benefit Exchange to the Centers for Medicare and Medicaid Services obtained via public records request • Policy variable – state-level population-weighted average count of kynect ads per week • Outcomes – 1) calls to the kynect call center, 2) page views (number of individual pages viewed), 3) visits (including repeats from the same IP address), 4) unique visitors (excluding repeats) for the kynect web site • Linear regression models used to describe how changes in kynect advertising volume affected these outcomes, adjusting for other advertising (healthcare.gov, insurers, insurance agencies, nonprofits, and other state governments), open enrollment periods, and other relevant time periods (e.g., week of Thanksgiving, week of Christmas, etc.) Results • Advertising for kynect fell from an average of 58.8 and 52.3 ads per week during the 2014 and 2015 open enrollment periods, respectively, to 19.4 during the first nine weeks of the 2016 open enrollment period and none during the last four weeks • Each additional kynect ad per week during open enrollment was associated with – 7,973 additional page views for kynect website – 390 additional visits for kynect website – 388 additional unique visitors for kynect website • Advertising was not significantly associated with calls to the kynect call center Conclusions • State-sponsored television advertising was a key driver of information-seeking behavior in Kentucky during the 2014-2016 open enrollment periods • Our findings add to previous evidence indicating that government-sponsored media campaigns are associated with increased enrollment for health insurance exchanges • Reductions in public outreach efforts could have negative effects on the individual market if consumers are uncertain about the continued availability of coverage and so-called “healthy procrastinators” fail to enroll as a result, contributing to lower enrollment and potentially a worse risk pool for insurers Implications • Misconceptions about eligibility and deadlines for enrollment in health insurance coverage are barriers to having all eligible individuals covered • Given the uncertain future of the ACA and the individual non-group market for health insurance in the United States, our findings indicate that state-sponsored media campaigns can be an effective strategy for encouraging information-seeking behavior
Date of publication
Resource type
Rights statement
  • In Copyright
Conference name
  • University of North Carolina at Chapel Hill. Gillings School of Global Public Health. Spotlight on Student Research (2017: Chapel Hill, NC)

This work has no parents.