Munich Personal RePEc Archive

A Structural Misclassifcation Model to Estimate the Impact of Physician Incentives on Healthcare Utilization

Arrieta, Alejandro (2007): A Structural Misclassifcation Model to Estimate the Impact of Physician Incentives on Healthcare Utilization.

[thumbnail of MPRA_paper_6718.pdf]

Download (158kB) | Preview


The issue of over-utilization of medical procedures has generated strong debate in the United States. It is well acknowledged that, in the agency relationship between physicians and patients, the informational advantage gives doctors an incentive to deviate from the appropriate treatment as defined for a patient's health status, thus incurring over- or under- utilization. However, the empirical consequence of this problem has not been adequately considered. In particular, physician agency breaks the correspondence between appropriate treatment and observed treatment, generating a problem whose characteristics and efects on estimation are analogous to a classifcation error. However, the error is non-random.

Empirical literature that does not consider the misclassifcation problem understates the impact of clinical and non-clinical factors on healthcare utilization. This paper proposes a structural misclassification model in which the physician behavior is modeled to characterize the structure of the measurement error. The model captures the interaction between a physician's incentives and a patient's health status, and returns consistent estimators. It also lets us identify the degree of deviation from appropriate treatment (misclassifcation probability) due to physician incentives, and to compute risk-adjusted utilization rates based on clinical factors only. The model is applied to the cesarean section deliveries performed in the state of New Jersey during the 1999-2002 period. Our results show a moderate but growing rate of non-clinically required c-sections of around 3.2%. We conclude that the growth of the c-section rates in New Jersey over these years is explained mainly by non-clinical factors.

Atom RSS 1.0 RSS 2.0

Contact us: mpra@ub.uni-muenchen.de

This repository has been built using EPrints software.

MPRA is a RePEc service hosted by Logo of the University Library LMU Munich.