Date of Award


Document Type


Degree Name

Medical Doctor (MD)

First Advisor

Cary Gross

Second Advisor

Kenneth Roberts


Urologist-owned prostate cancer centers that self-refer patients for IMRT present a significant risk of over-utilization, threatening to further contribute to the U.S. cost crisis. None of the available pro-competitive arguments for integrated centers justify an ownership model in which urologist income correlates with referrals for radiation therapy. Facilities are currently able to escape regulatory control by locating in a state without certificate of need laws and taking advantage of the in-office ancillary exception to the Stark Law. The American Medical Associations (AMAs) policy of mandatory disclosure of conflicts of interest is unlikely to curb over-utilization, and may even exacerbate the situation since there is evidence to suggest that such a policy increases the extent to which physicians offer biased advice. The American Society for Therapeutic Radiology and Oncology (ASTRO) has proposed an amendment to the Stark Law that would exempt radiation oncology from the in-office ancillary exception. Such reform, however, may hinder legitimate business relationships and integration of services for cancer patients, and would not prevent urologist-owned prostate cancer centers from making use of the physicians services exception to the Stark Law. Other forms of payment and delivery reform, particularly those that move away from fee-for-service fee schedules, may present more optimal solutions. On a theoretical level, risk-adjusted, comprehensive, long-term capitation contracts coupled with real-time outcome data perfectly align physicians incentives with the long-term health of patients. Such contracts would motivate providers to administer cost-effective, preference sensitive care to prostate cancer patients in order to contain costs and achieve the positive outcome data necessary to attract future patients. It will be practically difficult to design a sufficiently precise risk adjustment system that adequately rewards capitated providers for enrolling the sick, and generate outcome data that ushers patients toward top performing providers. It is therefore correct for legislators to experiment with more modest alternatives to fee-for-service reimbursement such as shared savings, pay-for-performance algorithms, bundled payments and partial capitation.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.