Physician Self-Referral: A Vexing and Costly Challenge

In 2014 the GAO released an analysis of trends in physician self-referrals for four services (advanced imaging, anatomic pathology, radiation therapy, and physical therapy), and the effects of these increases on Medicare Part B spending. According to the audit, in 2010 alone self-referrals of these services resulted in “increased Medicare Part B expenditures by approximately $300 million.”

Schaeffer Center Senior Fellow Bob Kocher went beyond the findings of the GAO audit to discuss the implications of physician self-referral and offer policy recommendations in a JAMA Viewpoint article he co-authored with Eli Y. Adashi.

The term physician self-referral describes the scenario of physicians referring patients to service facilities that are specific to their (or an immediate family member’s) financial interest.  For example, a doctor may refer a patient for an MRI procedure at a facility that the doctor has a financial stake in.

In and of itself, the practice of self-referral may have a patient-centered logic; proponents argue patients receive better continuity of services and care when they are able to access all their needed services under one roof.   But, as Adashi and Kocher point out, the GAO report and numerous other studies show evidence of an exaggerated increase in service use and costs, some of which have been found to be unnecessary or harmful for patients.

Stark Provisions I and II (enacted in 1989 and 1993 respectively) are the prominent statues in place to regulate self-referral, specifically prohibiting a physician to refer designated health services that are Medicare or Medicaid payable to an entity that he or she has a financial relationship with.  However, 35 exceptions to the regulations have largely rendered the Stark provisions ineffective and self-referrals have increases substantially.  Of the four different services the GAO report audited, three showed substantial increases in self-referrals compared to non-self-referred services from 2004-2010.  The trends for IMRT (intensity-modulated radiation therapy for prostate cancer) were particularly disconcerting according to Adashi and Kocher, “The utilization of self-referred IMRT services (driven by urology groups) increased by as much as 356% at a time when the utilization of the non-self-referred variety decreased by 5%.”

Reprioritizing the current fee-for-service environment with fee-for-value alternatives is a worthwhile systems change that might curb the financial incentive inherent in the volume-based model.  However, it must be recognized that legislative action will have to happen for substantial change to occur though. As Adashi and Kocher put it: “The GAO reports must be seen as nothing less than a call for action…Failure to do so would constitute a costly opportunity missed.”