Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)

Graduate Group

Health Care Management & Economics

First Advisor

Daniel Polsky

Second Advisor

Mark Pauly


How do market expansions that increase patient loads among offices affect provider treatment behavior? This question is immediately relevant to recent insurance expansions, where increases in insurance coverage may increase market demand for care. Because this may result in positive income shocks for providers, providers may alter how they treat patients across the entire practice to reallocate existing resources. Hence, a market expansion may not only impact treatment of new patients but also treatment of existing patients.

A simple extension of the McGuire and Pauly (1991) model predicts that an insurance expansion may be an effective policy lever not only to increase dental or medical utilization among populations prone to underutilization, but may also decrease provider-initiated overutilization among existing patients. I then substantiate what decreases in provider-initiated overutilization looks like using the clinical dental literature and reimbursement rates across procedures. The hypotheses generated by this model are then tested empirically using a novel source of dental claims data from Delta Dental of Michigan from 2008-2013. I leverage an exogenous increase in demand for provider services from a dependent coverage insurance expansion in the overall dental market. I use an instrumental variables strategy to isolate increases in patient load that come solely from dependent expansion, and isolate attention to continuously insured patients among dental practices to 1) remove any omitted factors influencing both practice intensity and patient load and 2) detect changes in treatment behavior initiated by providers, holding fixed demand-side factors. I find offices face up to a 8.64\% increase in office loads from the dependent expansion, which leads to substitution away from intensive cavity procedures towards routine procedures. These empirical results are consistent with the predictions of models of demand inducement and provider behavior. I then follow with a more detailed discussion of why demand inducement is an issue in the dental market and why an insurance expansion is only a partial solution. To conclude, I discuss the policy levers available for addressing the misalignment of clinical efficacy and provider reimbursement, using the theory of optimal insurance and optimal coverage of preventive care.