Home » Posts tagged 'Ordered Regression Model'

Tag Archives: Ordered Regression Model

Possibility of Cost Offset in Expanding Health Insurance Coverage: Using Medical Expenditure Panel Survey 2008

By Catherine Moon

The Patient Protection and Affordable Care Act aims to substantially reduce the number of the
uninsured over time and asserts that the financial burden of extending insurance coverage to the
previously uninsured will be offset by the benefit of the attendant improvement in their health.
Motivated by this policy, I explore whether health-insurance status and type affect one’s likelihood of
improving or maintaining health using the Medical Expenditure Panel Survey data. I build a set of
ordered regression models for health-status transitions under the first-order Markov assumption and
estimate it using maximum likelihood estimation. I perform a series of likelihood ratio tests for pooling to determine whether the latent propensity index is the same between adjacent initial health-status groups. Empirical results imply that expanding health care to the unwillingly uninsured due to severe
economic constraints and extending the scope of public insurance to that of private insurance will lead to improvement or maintenance of health for the relatively healthy population, implying the possibility of cost off-set in the expansion of coverage and the extension of scope.

View Thesis

Advisor: Frank Sloan, Michelle Connolly | JEL Codes: C12, C25, I12, I13, I18 | Tagged: Health Insurance, Health Transition, Ordered Regression Model, Patient Protection and Affordable Care Act (PPACA), Self-Assessed Health Status, Test for Pooling Adjacent Ordinal Categories

Questions?

Undergraduate Program Assistant
Matthew Eggleston
dus_asst@econ.duke.edu

Director of the Honors Program
Michelle P. Connolly
michelle.connolly@duke.edu