Self-referral Linked to Rising Medicare Expenditures

Several recent reports indicate that physicians engaging in self-referral for radiation oncology and diagnostic imaging services is playing a role in escalating Medicare costs. A Wall Street Journal article published last week establishes a correlation between a marked rise in the use of intensity-modulated radiation therapy (IMRT) to treat prostate cancer and financial incentives being leveraged by groups of urologists across the country. These groups, the article states, take advantage of loopholes in the Stark laws to team up with radiation oncologists and “capture the lucrative reimbursements IMRT commands from Medicare”. An analysis of Medicare claims undertaken by the Journal pegs IMRT usage as significantly higher in New York, Florida, Pennsylvania, New Jersey and Texas, where 22 of the 37 U.S. self-referral urology groups identified by the newspaper are located. The average IMRT usage for recently diagnosed prostate-cancer patients was 42% in those states in 2008—but only one-third that total in other state. The nation’s largest self-referral group, Long Island-based Integrated Medical Professionals (IMP) PLLC, employs 103 physicians across six counties and owns 11 linear accelerators for delivering radiation. Deepak Kapoor, MD, who heads up the group, conceded in an interview with the Journal earlier this year that approximately one out of six, or 17%, of IMP’s prostate-cancer patients are treated with IMRT. However, an analysis of IMP’s Medicare claims later performed for the newspaper reveal that between its launch in mid-2006 and the end of 2008, Integrated Medical administered IMRT to 601, or 53%, of 1,132 Medicare patients recently diagnosed with prostate cancer. The Journal article points out that IMP received $26 million for the care of those patients. If they referred these patients elsewhere for treatment, the practice would have received just $2.6 million. IMRT is by far the most expensive treatment option for prostate cancer patients at as much as $40,000 per patient; other options include prostatectomy, for which Medicare pays $16,000, and radioactive seed implantation, for which it pays up to $19,000 per patient, and watchful waiting, which involves monitoring the patient with tests at regular intervals. Scrutiny of the urology groups is indeed mounting. The American Society for Radiation Oncology (ASTRO) has been vocal in its criticism of urologists' practice of referring patients for treatment in facilities. ASTRO's members also have a financial stake in the issue, since self-referral urology groups compete directly with them for radiation business. Moreover, this past April, House Ways and Means Committee Chairman Sander Levin (D-Mich.) and two other congressmen requested that the U.S. Government Accountability Office study the impact self-referral groups on Medicare spending. On the Imaging SideMeanwhile, recent studies indicate that the most commonly promoted advantages of self-referral—convenient, same-day imaging for patients and shorter illness durations—are rarely realized. The studies were cited as part of a series of articles in the December 2010 issue of Health Affairs (Vol. 29:12, pp. 2231-2259) and were conducted by researchers from the American College of Radiology and Stanford University. In the first study, ACR Senior Director for Research Jonathan Sunshine and colleague Mythreyi Bhargavan analyzed Medicare data amassed in 2006 and 2007. The data showed that self-referral resulted in same-day imaging for only 15% of advanced procedures, such as CT or MRI. A second study, conducted by ACR Assistant Director for Research Danny Hughes along with several colleagues, found that self-referral increased health care costs but did not decrease a patient's length of illness, with the exception of when doctors self-referred patients for x-rays for common conditions such as sinusitis, chest pain, or respiratory illness. The mean total episode cost was 4% to 10% higher with self-referral than without, depending on the type of imaging studied. Mean imaging cost per episode was found to be 27% to 40% higher with self-referral. As for the Stanford University study, Laurence Baker, MD, chief of health services research at Stanford University, concluded that physicians ordered substantially more scans once they began billing for MRI. Baker analyzed changes in imaging use and in overall spending by patients of orthopedists and neurologists who began billing for MRI scans between 1999 and 2005. MRI procedures used by the former within 30 days of a patient's first visit increased by 38%, from an average of 74 procedures per 1,000 visits before acquiring MRI to 103 afterward. The average 30-day MRI rate for the latter rose by about 20%, from 241 per 1,000 episodes prior to MRI equipment acquisition to 287 after MRI equipment acquisition. Baker also discovered that increased costs were spurred not only by the performance of additional MRI exams, but by additional follow-up procedures.