• Chart of the day: Prostate cancer treatment “selfies”

    Today’s chart comes from the recent GAO report on delivery of intensity-modulated radiation therapy (IMRT) for prostate cancer among Medicare patients by physician groups that do and do not self-refer (h/t Karan Chhabra). Self-referring here means that providers administering IMRT had a financial relationship with the entity that referred the patient for the therapy.

    Surprise! Self-referring groups are a growing source of IMRT delivery; non-self-referring groups are not. Remuneration does not imply causation, but the findings are consistent with the idea that physician groups respond to financial incentives.

    GAO self-ref

    Some other facts from the GAO report:

    • Providers are not required to disclose that they self-refer.
    • IMRT is one of the more costly prostate cancer treatment options, but not clearly superior.
    • Other work has documented that physicians play a large role in determination of a patient’s prostate cancer treatment approach.
    • Other work has also suggests financial incentives may influence treatment decisions.
    • Differences between self-referring and non-self-referring providers shown above could not be explained by differences in age, geographic location, or health.


    • I thought that self-referral was illegal…Stark law… or have doctors figured out some scam to get around the law?

      • I am not familiar with many of the specifics but there are plenty of ways around Stark. For one, it only applies to Medicare and Medicaid. Second, it only applies to designated health services (listed at (6)(h) below).


      • At least up until around 2006 when we decided to sell all our ancillaries, if the equipment is within your immediate physical space, compliments your practice, and you don’t divvy up the profits simply based on individual docs referral numbers, then the self-referral can comply with Stark. Also as I recall, around 2010 there appeared a way to get around the profit distribution compliance, but do not remember the details. I do believe that all the above was and still is illegal in Maryland.

    • The GAO’s “fact” that IMRT is not clearly superior is not a fact (as the word “fact” is defined). More of an assertion. Or an opinion to buttress an underlying premise. It is a fact, however, that higher-risk prostate cancers are not properly treated with surgery or seed brachytherapy (see NCCN guidelines). It is also a fact that if you’re getting external beam radiation for prostate cancer, you should get IMRT (again, see NCCN guidelines). It’s also a fact that if I get prostate cancer, I’ll choose IMRT.

    • IMRT was an its infancy in 2001 with very few clinics anywhere having access to it. The first graduated residents fully trained and facile in IMRT only began being released upon the nation in the 2003-2007 timeframe. IMRT became only “somewhat ubiquitous” during these years, as well. The sad fact (pardon… assertion) to me is that if IMRT were cost-equivalent to “old style radiation,” its widespread use would be of no concern, nor would any urologist’s adoration of it.

    • It would be interesting to know the treatment choices of physicians who develop prostate CA.

      An entirely different type of self referral…

    • ” Remuneration does not imply causation, but the findings are consistent with the idea that physician groups respond to financial incentives.”

      It is also very consistent with the idea that physicians who believe in something are more likely to invest that something than those that do not believe in it.