"Trying to Regulate Imaging Self-Referral is Like Playing Whack-A-Mole"

Dr. Bruce Hillman authored the article "Trying to Regulate Imaging Self-Referral is Like Playing Whack-A-Mole" in the August issue of the AJR. The piece actually reviews and comments on another article, "The Prevalence Of Physician Self-Referral Arrangements After Stark II: Evidence From Advanced Diagnostic Imaging" from the April 17, 2007 edition of Health Affairs, by Dr. Jean M. Mitchell, a professor of public policy at Georgetown University.

Dr. Mitchell undertakes an analysis of imaging self-referral by sifting through data from a "large insurer in California". Her conclusion:

Nearly 33 percent of providers who submitted bills for magnetic resonance imaging (MRI) scans, 22 percent of those who submitted bills for computed tomography (CT) scans, and 17 percent of those who submitted bills for positron-emission tomography (PET) scans were classified as “self-referral.” Among them, 61 percent of those who billed for MRI and 64 percent of those who billed for CT did not own the imaging equipment. Rather, they were involved in lease or payment-per-scan referral arrangements that might violate federal and state laws.

Dr. Mitchell conducts a scholarly analysis of the data to come to her conclusions, and I won't try to reproduce it here. What is quite interesting, however, is the commentary to her article published on the HealthAffairs website. None of the letters were complementary, and one from a neurologist who self-refers was outright contrary:

I believe this article takes a simplistic, one-sided look at self-referral. Our private practice group has had MRI services since 1987, and we feel that we are saving health care costs. Two primary reasons: 1. Third-party insurers pay us much less than the local hospital system; 2. In-office MRI expedites diagnosis and reduces hospitalization rates. Abuses of the system should be curbed but not in a way that monopolizes services and prevent unique practice arrangements that foster excellent patient care.

Yeah, that old patient convenience and dead-granny argument. But saving the insurance companies money? Let's use an arbitrary figure and assume the outpatient charge is HALF of what a scan might cost in the hospital. (This I believe grossly overstates the discrepancy, but go with it.) So, if our self-referring friends order from 2-8 times what they would if they had to send the patients to the groady old stinky hospital, they aren't saving anyone anything at all.

More amusing is a comment from none other than Tim Trysla, "Counsel, Access to Medical Imaging Coalition." That would be the head of AMIC, folks, our ACR-sponsored lobby to help the self-referrers maintain their profits. (OK, I'll be fair and note that AMIC was supposed to help radiologists with outpatient imaging centers, too, but when you lie down with dogs, you get up with dog-hair in your mouth as well as fleas.) Here is Mr. Trysla's eLetter:

We have reviewed the article in Health Affairs on equipment leasing and are hopeful that the author will release the underlying study data so that her peers can replicate her analysis. Since its inception, the Access to Medical Imaging Coalition (AMIC) has worked with policymakers with the goal of ensuring appropriate utilization of medical imaging services. Unfortunately, the Health Affairs article does not offer any data on the appropriateness of the imaging procedures it analyzes -- which is the key to crafting responsible and sustainable policies on access to medical imaging services.

Without having access to the author's data, it is impossible to know if the conclusions she has drawn are reflective of the facts. However, if the arrangements described in the Health Affairs article are in violation of federal or state law, those violations of the law should be prosecuted.

AMIC looks forward to continuing to work with the Congress to enact a reasonable Medicare imaging policy that preserves and strengthens beneficiary access to the right imaging procedure at the right time.

Gee, if he applied any more spin to the situation, he would have the worst case of vertigo in the room. Note the usual political tactic of trying to discredit Dr. Mitchell's data. I'm not sure what data Tim feels is missing here. Oh, yes, the appropriateness data. Well, you know, that is really the crux of the entire self-referral debate, isn't it? Every patient that makes it through the door of a specialist, say a neurologist, has some symptom related to that specialty, such as the neurological system in this example. Therefore, a head CT could be considered appropriate for every last one of them. So, why does the data (from this and other sources) show an increase in the number of scans ordered when the doc can profit from it? Are those in a non-self-referring situation neglecting their patients, or is the system being abused? I guess it is appropriate for a bank-robber to ply his trade at banks, since that is where the money is. Every scan can be justified on an individual basis, and I'm sure Tim knows that. But when you have a massive shift in ordering behavior, something is very wrong.

Dr. Hillman furthers the argument, and even notes that his own research prompted a "change in the opinion of the American Medical Association Council on Ethics and Judicial Affairs (AMA CEJA) on physician conflict of interest and self-referral arrangements":

Physicians are free to enter lawful contractual relationships, including the acquisition of ownership interests in health facilities, products, or equipment. However, when physicians refer patients to facilities in which they have an ownership interest, a potential conflict of interest exists. In general, physicians should not refer patients to a health care facility which is outside their office practice and at which they do not directly provide care or service when they have an investment interest in that facility. The requirement that the physician directly provide the care or services should be interpreted as commonly understood. The physician needs to have personal involvement with the provision of care on site.

Personal involvement by a self-referring clinician with an on-site scanner? Yeah, right. I've never seen that happen, and neither has anyone else in this venue. You would think this closes the book, but of course it does not. Stark II was supposed to keep this stuff at bay, but it didn't. Hillman notes:

The in-office exception to the Stark II regulations and state laws, on which the arrangements described by Mitchell are based, conveys the right of physicians to maintain imaging capabilities—expected at the time of the bill’s passage to be largely plain X-ray and sonography—in their office practices. While hard to rationalize even then, given the research results, the exception recognized the political reality of how difficult it would be to pass the legislation if the law did not contain this exemption.

It would have made no sense to remove X-ray and U/S from clinicians' offices, because in most cases, those clinicians are actually reading and using those studies themselves, in other words, they are directly involved. Just like the AMA suggests. But because scanners have become cheaper and easier to place in an office setting,

Equipment manufacturers have taken advantage of these trends to market high-tech imaging devices to physicians on the basis of their projected financial return.

In fact, the acquisition of high-tech imaging capabilities has become the favored approach of nonradiologists’ practices to replace lost revenue from declining reimbursement for their traditional services [6]. In pretty much all regions of the United States, physicians are becoming more entrepreneurial, even to the point of ceasing to provide poorly reimbursed traditional services in favor of higher-paying services such as imaging. Many are outsourcing interpretations to radiologists—at lower rates than they are receiving from insurers—and making a profit on the professional fees as well. Elevated technical fees for imaging are promoting this activity and facilitating the kinds of lease-by-the hour and “pay-per-click” arrangements described by Mitchell, which are in direct conflict with the AMA opinion detailed above and, as Mitchell notes, quite possibly with existing anti-self-referral legislation.

But, but, but....it's for my patients' convenience! Granny will die if she can't get scanned on my in-office money printing press, I mean CT scanner!

. . .such arrangements cannot be rationalized on the basis of quality of care, convenience, access to care, or any of the other explanations commonly offered by the apologists for self-referral. It’s about the money and, if we accept that physicians are susceptible to financial incentives, in conflict with yet another AMA CEJA opinion addressing conflict of interest:

"Under no circumstances may physicians place their own financial interests above the welfare of their patients. The primary objective of the medical profession is to render service to humanity; reward or financial gain is a subordinate consideration. For a physician to unnecessarily hospitalize a patient, prescribe a drug, or conduct diagnostic tests for the physician’s financial benefit is unethical. If a conflict develops between the physician’s financial interest and the physician’s responsibilities to the patient, the conflict must be resolved to the patient’s benefit [7].

Sadly, Hillman ends on a disheartening note:

Given Mitchell’s [1] demonstration of the susceptibility to self-referring physicians to placing their own financial interests above patients’ health interests, it’s hard to be sanguine about the prospects for further regulation improving on the situation. As a sympathetic health economist once said to me, “Finding ways around regulation is the American national past-time” (Albert Williams, PhD, The RAND Corporation, personal communication, 1985). Closing the Stark II in-office exception is a noble regulatory goal, however, we must recognize that there are limitations to what regulatory actions can accomplish. History tells us that even if a Stark III were to pass Congress, there almost surely will be adverse and unintended consequences.

Hence, the Whack-A-Mole analogy. The ugly little critter just keeps popping up somewhere else. In other words, if we staunch the financial hemorrhaging from imaging self-referral, those who are willing to suspend their morals enough to participate will find some other way to pillage the system. No doubt that is true. However, to carry the analogy a bit further, this mole needs to be whacked with a really big hammer, and really hard. There needs to be prosecution, and punishment, with fines, revocation of licenses, and even jail-time. But of course that won't happen, because the practice is so wide-spread, and there is so much money involved. We are seeing more litigation of sham leasing arrangements, as noted in this blurb from the AMA, but there is a long way to go. A very long way.

The worst part of this whole fiasco is that the patients have become pawns in a big financial game. They trust their clinicians, which is as it should be, and they hang on every word they are told. They truly believe it when they are told that they need a scan, and that it is best if they have it right here in the office, which is after all so very convenient. I have heard patients request to have their scans done at the hospital, and be told that their doctor "would be mad" if they did so. This is a complete abuse of the power a physician has over his patient, and if my own position wasn't so precarious in all this, I would have brought the gentleman in question before the medical board.

Because of the wording of Stark II, many clinicians act as if it is their God-given right to buy a surplus East German CT and run it until the rotor melts, and that radiologists should be ever-so-grateful to read the blurry scans produced in this manner. What we are fighting is the mentality of entitlement, and that is something that seems to be drilled into four-year-olds and physicians in this country with equal vigor. I haven't a clue as to the solution. Except perhaps a bigger whacking hammer.