Three essays in health economics

Abstract

This thesis presents three essays on policy-related topics in Health Economics. The specific policy topics this thesis explores are health insurance and inequality, spillovers or peer effects of health behaviours, and the impact of human resources in healthcare. The first essay in this thesis, I analyse the redistributive effects of a publicly financed healthcare expansion. Using data from the Oregon Health Insurance Experiment (OHIE) we analyse the redistributive impacts of a publicly financed health insurance expansion. We use a residual inclusion methodology combined with quantile regression to estimate the heterogeneity in the effects of health insurance across the income distribution. We find that there are indeed redistributive impacts, even in the small income window we have access to, which would otherwise be concealed in a linear regression. Those at the lower end of the income distribution tend to have a substantial increase in their disposable income as a result of health insurance coverage, while those at the upper end see no change in income. We additionally estimate that increased employment in at risk households is driving this effect. In next essay I analyse the spillover effects of a diabetes diagnosis. Diabetes is a unique condition, in that a positive change in lifestyle and behaviour, is both the first line treatment and the recommended method of preventing the disease. It is theoretically possible that by jointly partaking in diabetes treatment, partners of people with diabetes would substantially benefit from their partners’ diabetes diagnosis. Using blood data from the Health Survey for England, and a fuzzy regression kink design, we causally estimate the effect of a diabetes diagnosis on health-related behaviours of the individual with diabetes, as well as, their partners. We find that a diagnosis of diabetes results in a significant increase in the probability of exercising and a decrease in the probability of currently being a smoker both for the diabetic individual and their partner. However, we find limited evidence of other lifestyle changes. From a public health perspective, our results are especially important for the evaluation of diabetes related policies, while positive spillovers, particularly within households, should be taken into account in the evaluation process. In the final essay of this thesis, I analyse the impact of primary care physicians on health outcomes. Worldwide there is a growing concern that there are insufficient primary care physicians to meet demand. There is, however, mixed evidence on how effective primary care is in influencing population health outcomes. I estimate the effect of an increasing in primary care physicians using the Programa Mais M´edicos. Although previous studies have used the Programa Mais M´edicos to analyse the impact of a primary care physician supply, I exploit the variation in physicians allocated to each municipality and use only treated municipalities to identify the impact of primary care physicians on hospitalisations and mortality. I estimate the impact of primary care physicians using a generalised synthetic control estimator and find limited evidence of primary care physicians impacting health outcomes. These results question the notion that primary care physicians are a cost-effective means of improving population health.

More information

Identifiers

ORCID for Rhys, Llewellyn Thomas: orcid.org/0000-0003-0124-0391

ORCID for Emmanouil Mentzakis: orcid.org/0000-0003-1761-209X

Catalogue record

Export record