Scared to death? Information avoidance and diagnostic testing

TitleScared to death? Information avoidance and diagnostic testing
Publication TypeThesis
Year of Publication2019
AuthorsZhong, Y
Academic DepartmentEconomics
UniversityUniversity of North Carolina
CityChapel Hill, NC
Thesis TypeDissertation
ISBN Number9781392203422
KeywordsDecision making, Diabetes, Screenings

The use of preventive care in the U.S. is half the recommended level. Previous economic studies suggest that price is not the only significant deterrent. In fact, empirical evidence suggests that some people are health information avoidant: they prefer not knowing information about their health even when diagnostic testing is free and very accurate. This study evaluates the roles of many contributors to an individual's demand for type-2 diabetes screening. To explain the puzzle of information avoidance, I apply insights from the economics theoretical literature to incorporate health anxiety, which represents the stress or disutility associated with the anticipation of bad outcomes, as another psychic cost of taking a test. With data from the Health and Retirement Study, I jointly estimate a set of quasi-structural equations derived from a forward-looking individual's optimization problem regarding diabetes screening. In the model, the individual chooses the number of doctor visits (at which a diabetes screening is a stochastic outcome to reflect provider differences and additional individual behaviors) and lifestyle behaviors. Underlying disease governs her diabetes state and she has imperfect information about her true health. Results suggest that health anxiety, as well as monetary costs, time costs, subjective health and longevity expectations are each important contributors to an individual's diabetes screening behavior. Individuals' lifestyle behaviors also respond to health information associated with screening tests. Multiple policy experiments aiming to improve screening behavior and population health are evaluated using the estimated dynamic model.

Citation Key10114