Room: Phillips 407
Time: Mon 13:15 PM-14:45 PM
Chair: Jonathan Skinner (Dartmouth College)
Session Description
Research on self-referral has focused on the effect of imaging self-referral on imaging utilization. Repeated findings of higher utilization have led to concerns about unnecessary radiation exposure and excess cost.
However, there are many issues in self-referral beyond this relatively narrow focus. This session attempts to extend the study of self-referral in three different dimensions.
The first is to address purported benefits of self-referral. Supporters of imaging self-referral claim that its additional imaging leads to more accurate diagnosis and, hence, more appropriate treatment. Thus, non-imaging costs should be lower with self referral, and total episode costs (imaging plus non-imaging) may be less. Danny Hughes addresses this claim, using Medicare data. He studies chronic illnesses, because the cost burden of chronic disease is drawing increasing attention. He finds non-imaging costs tend, if anything, to be higher with self-referral, and total annual costs of caring for chronic illnesses clearly are higher on average with self-referral.
Given these findings (and similar, as-yet unpublished, findings for non-chronic illnesses), self-referral for imaging more clearly seems undesirable. This implies that finding market incentives that discourage imaging self-referral could be valuable for the health care system. To help find such incentives, the second paper, by Jonathan Sunshine, investigates what market characteristics affect the prevalence of self-referral in a market area. Such characteristics might be useful as policy levers.
Determinants of self-referral’s prevalence have not previously been investigated.
His study uses the Dartmouth Atlas’s hospital referral regions as its market areas and the percent of physicians in a market who self-refer for imaging (ascertained through analysis of Medicare data) as one measure of self-referral prevalence. Preliminary, correlation-only analyses reveal varied patterns, including a general association of high managed care penetration with low prevalence of self-referral. Multivariate analysis and attempts to deal with possible endogeneity will be undertaken.
Imaging is not the only service for which self-referral appears problematic. Laboratory tests and physical therapy are others. The third paper, by Jean Mitchell, investigates the largely unresearched world of self-referral outside imaging, studying anatomic pathology (AP). Case studies suggest AP self-referral by urologists generates greatly increased utilization, but there has been no systematic analysis. Prof. Mitchell will use logistic regression of Medicare data to quantify the effect of self-referral for AP on the odds of a man with prostate problems having a prostate biopsy. Then, because payment for pathology services depends on the number of biopsy specimens examined, for men who have a prostate biopsy, the study will use Poisson regression to ascertain the effect of self-referral on the number of biopsy specimens a urologist takes.
Danny Hughes, PhD, is a Senior Researcher at the American College of Radiology (ACR) with interests in physician behavior and econometrics.
Jonathan Sunshine, PhD, is Senior Director for Research at the ACR and Professor (Adjunct) at Yale University School of Medicine. His interests include physician behavior, health policy, and the effect of healthcare on health outcomes.
Professor Jean Mitchell, PhD, of Georgetown University is a health economist with interests in self-referral and physician behavior.
Session Organizer: Mythreyi Bhargavan (American College of Radiology)
The 3rd Biennial Conference of the American Society of Health Economists took place at Cornell University.
Software © 2010 iHEA - International Health Economics Association