Research Summary
Research Summary
Overview
Description
Phil's work aims to identify the drivers of performance for healthcare organizations and providers, and the mechanisms by which this performance can change over time. In complex healthcare settings, the optimal choice of treatment can be highly ambiguous. As a consequence, the marginal patient may face radically different care depending upon the choice of provider. Moreover, the complexity of treatment in these settings heightens information asymmetries between patient and provider at the same time as the stakes are at their highest. He explores these issues in two settings: the decision-making and surgical performance of kidney transplant teams and primary care practices adopting the patient-centered medical home model.
In his work on performance in kidney transplantation, Phil explores whether the widely-documented volume-outcome relationship is a consequence of better decision-making (such as in diagnosis or evaluating treatment alternatives) or better execution of the chosen treatments. He finds evidence of a puzzle: larger transplant centers have better post-transplant outcomes; but patients at these centers are less likely to receive a better organ and more likely to die or be removed from the transplant waitlist after an offer is declined, indicating lower-quality decision-making. This tension between improved execution and reduced decision quality implies that practice may not make perfect in complex medical decision-making.
His second project is an ongoing collaboration to evaluate the impact of the patient-centered medical home (PCMH) model, a leading model for primary care delivery reform. His work notes that the mixed evidence on the PCMH model may be a consequence of heterogeneity in implementation, and seeks to incorporate this heterogeneity into impact evaluations. Using techniques like hierarchical clustering to categorize medical home practices, he finds that evaluations which treat the PCMH as a single, undifferentiated intervention may conceal differences across PCMH types that explain important differences in patterns of healthcare utilization.