BB&T Washington Editor

WASHINGTON — Imaging services have cost the taxpayer a bundle where Medicare is concerned, so much so that Congressional negotiators folded a provision to tamp down on the rise in those costs into the Deficit Reduction Act of 2005.

Still, several expert witnesses at last month's hearing of the Medicare Payment Advisory Commission (MedPAC) indicated that overuse of imaging is a persistent problem, and their recommendations included pre-authorization as a way to put a dent in imaging overuse.

At the start of one session, staff analyst Ariel Winter said, "we recognize [the benefits of] the technological progress in imaging," but said the "rapid growth in services ... and the varying use by providers" raise concerns about inappropriate use.

Winter introduced three experts on the subject of Part B imaging, the first of whom was Lawrence Casalino, PhD, of the University of Chicago. He said radiologists' share of Medicare imaging payments fell by 9% between 2002 and 2006 while cardiologists' share rose by 10%.

Casalino also said that a 242% increase in MR billed by primary care doctors between 2000 and 2005 suggests "some form of leasing or per-click" arrangements, which are those in which a physician leases radiology equipment from an outside source, usually a freestanding radiology clinic, and marks up the technical fees for billing purposes.

He observed that MRIs will usually take place "if it has any chance of doing a benefit and little chance of doing it harm" and if the patient squeaks loudly enough. He also pointed to "a lot of regional variation," leading to suspicions about appropriateness.

Health plans, Casalino said, will acknowledge that "imaging goes down," when they put up procedural hurdles, but said that when physician "feel squeezed for income, it's easy to order more imaging.

"You can order images without limit, but you can't see patients without limit," he commented, adding that some data indicate that "pretty soon, 2% of cancers in the U.S. will be attributable to CT scans."

As for Medicare pre-authorization, Casalino said "I don't necessarily disagree with this," but suggested "not making it a blanket requirement." His recommendation was to require pre-authorization for high-use providers, which would generate "much less political backlash." He also observed that despite the rhetoric of medical societies, some doctors "know some guys in their communities who think [pre-authorization] should be done too."

Laurence Baker, PhD, of Stanford University (Stanford, California), noted first that the number of computerized axial tomography (CT) units in operation in the U.S. rose from slightly less than 5,400 in 1995 to almost 8,400 nine years later. He described a similar rise in installations of MR equipment over the same period, from a bit more than 2,400 units to almost 5,900.

"Is there a relationship between availability and utilization? There certainly is," Baker said, asserting that "adding a scanner is associated with about 800 additional MRI procedures for the Medicare population" a year. The incremental cost to Medicare is about half a million dollars each year for each new scanner.

"There certainly are some benefits to expanding utilization" of CT and MR technology, Baker said, such as carotid artery angiography, which "reduces [the] side-effect risk" associated with other forms of angiographic imaging. All the same, "expanded availability of MRI appears associated with less clearly less beneficial utilization, and back pain is one good example," he said.

Baker said physician ownership results in "dramatic [utilization] increase over time," citing a study he conducted recently based on billing for technical components by neurologists. His numbers indicated that 9.3% of neurologists who had no history of billing for the technical portion of an MRI will recommend an MRI for lower back pain, but that number rises to 14.5% for those with a technical component billing history.

As for policy recommendations, Baker said the options included limiting financial self-interest and pre-authorization, but he declined to come down decisively in favor of any particular approach.

Bruce Steinwald, director of healthcare research at the Government Accountability Office, reviewed data from a June study of cost-containment procedures. He noted the well-known increase in Medicare imaging between 2000 and 2006, from almost $6.9 billion to more than $14 billion, but remarked that only 11% came from an increase in the number of beneficiaries.

Steinwald also noted vast geographic disparities in per-beneficiary imaging costs per year, with the retirement haven of Florida generating more than $400 per beneficiary per year vs. less than $100 in a number of states.

Pre-authorization works, he said. "We did a whole bunch of interviews" of plans that "are active in managing imaging benefits." Those using prior authorization relied heavily on contract benefits management companies. CMS's approach, on the other hand, "is in essence a post-pay claims review, what's known as pay-and-chase.'"

Steinwald said the net effect of pre-authorization and other efforts by the private plans was a reduction in the rate of growth in imaging, due principally to pre-authorization. "The plans in our study reported that prior authorization ... was the practice most important to managing physician use of imaging." He recommended that "CMS consider more front-end approaches, including prior authorization and privileging," but said CMS should "not necessarily [be] limited to those approaches."

Steinwald said his discussions with CMS indicated the agency is skeptical, citing the probability that "proprietary systems won't work well in a public program" due to public disclosure, and "an appeals process that can overturn decisions." He said GAO understood those concerns, "but we don't think post-payment systems" will constrain costs.

MedPAC chairman Glenn Hackbarth said, "Clearly, there's evidence of rapid growth, and ownership is associated with more growth," but asked whether more scans in doctors' offices are a substitute for hospital outpatient activity and whether they prevent more costly treatment.

As for substitution, Steinwald said, "the evidence is that it continues to grow in other settings" despite that some substitution does take place. Baker added that discernible "offsets in outpatient spending are much smaller than increases in" physician-billed scans, and that there is an offset in terms of better outcomes for some disease states. That case is tougher to make for back pain, for instance, because the data indicate no meaningfully different outcomes.

Casalino chimed in, saying that the answer to "this will vary a lot ... by the type of episode." Those with lower back pain "want to know if they have a herniated disc," and even though it won't change the outcome, patients want to know what to expect.

Commission member Bill Scanlon, an independent health policy consultant, asked whether "we should give post-claims review a chance," which might get around the fact that "Medicare ... has to operate in a transparent world."

Steinwald said CMS does not have the resources for such reviews and that the agency's focus seemed more on fraud than appropriateness. "Our position would be do both (front-end and back-end reviews), and if CMS needs more resources, they should have them." Casalino made the case that post-claim reviews will not get to the problem very effectively because trying to get the money back "would create an even bigger [political] problem."

Commission member Robert Reischauer, PhD, of the Urban Institute (Washington), said, "Any time I see a rapid expansion in a service" along with "innovative financial arrangements, bells begin to go off." He asked, "Is there any way out of this other than bundling and capitation to get the mix of services right?"

Baker replied that he did not see "a lot a lot of data on capitated services," but said bundling "could change the game a lot and would lead to efforts to make things more efficient."

SGR a confounder on Medicare forecast

The MedPAC meeting represented a continuation of its work in developing policy recommendations for cost control for Medicare, which it has been doing since 1999. So it comes as no surprise that there was little in the way of new ideas in circulation.

While some ideas have come into sharper focus over that time, Congress's refusal to hold the line on spending for Part B doctor's fees drew several negative comments at the morning meeting.

In the day's first briefing, staff analyst Evan Christman cited a series of well-known facts and figures, such as the fact that cost growth is "a challenge for all payers" and that the Congressional Budget Office pegged the growth of Medicare as being 2.4% faster than GDP growth between 1975 and 2005. He also reminded the commission that Medicare will devour more than 10% of GDP by 2070, given a standard set of assumptions about demography and economic growth.

The caveat was that "these projections assume the SGR mechanism ... is not overridden," Christman said in reference to the sustainable growth rate mechanism (SGR) that Congress authorized in 2003 to constrain the cost of physician services under Medicare Part B. Congress has repeatedly backed off those growth caps, including the most recent reauthorization of the Medicare budget.

In reference to the percentage of the U.S. economy that goes to healthcare which currently stands at 16% to 17%, depending on the measurement Christman observed that "some have suggested that a nation as wealthy as the U.S. should spend 30% of its income on healthcare," but that view is far from universal and is generally seen as a level that will cripple the economy.

As a matter of Medicare perspective, Christman pointed out that "in 1970, Medicare was three fourths of 1% of GDP." Factors in the growth include technology, which Christman described as including drugs, devices and biologics. "Most analysts believe that 50% or more of the growth [in Medicare spending] is due to technology," he noted, reminding members "there is not an adequate evidence base" to compare clinical effectiveness of various treatments. However, increasing life-spans are another boost, along with the encroaching tidal wave of retirees.

During discussions, panelist Karen Borman, MD, of the University of Central Florida College of Medicine (Orlando, Florida), remarked that "one segment we left out ... is understanding the expectations of patients," which she said feeds demand for discretionary services, including cosmetic procedures. "What you expect and what the government provides as a baseline may be two different things," Borman noted, adding that such data "would be important to add to the conversation."

In reference to an earlier question about the impact of the Medicare drug benefit (Part D) on overall Medicare spending, Tom Dean of Horizon Healthcare (Wesson Springs, South Dakota) said he is aware of "a projection that pharmaceutical costs are going to grow rapidly due to biologics," which he characterized as "a whole new approach to care that will be extremely expensive."

Dean also made the argument that "it's a mistake to leave out the SGR" when calculating future spending, insisting further that "it can't be ignored forever."

Arnold Milstein, MD, of the Pacific Business Group on Health (San Francisco), suggested that MedPAC try to establish specific numbers on the value of various reforms. "Even a rough estimate would be a valuable addition" to the discussion, he said. He also said that any reforms should be described as either delivering a one-time reduction in the baseline of spending or alternately as something that would improve both care and costs over time.

One of the recurrent themes of the panel discussion is that MedPAC's reports to Congress should break out definitions of the problem and proposed solutions into separate chapters.

In reference to a chapter defining the problem, MedPAC Chairman Glenn Hackbarth, an independent consultant, said "the context chapter has an audience" even on Capitol Hill, "but maybe what we need to do is separate into another statement" the proposed solutions. He also recommended a stronger sense of urgency in defining the problem, condensing that section into a smaller, punchier document.

Hackbarth also made reference to a need for a change in the mindset of beneficiaries that acknowledges that "new and better is not always best," adding that "until we have some change in the payment system," that system will perpetuate perverse incentives.

John Bertko, who retired last year as the chief actuary for Humana (Louisville, Kentucky), seconded Hackbarth's call for a stronger sense of urgency, and said that while the 45% threshold for the Medicare Trustees report may seem arbitrary to some, the collapse of the Part A trust fund, which is projected to take place in 2019, "is not arbitrary."

Advisors lean toward full disclosure

The relationship between doctors and makers of drugs and devices has undergone scrutiny in many quarters, including in the meetings of MedPAC, which has staked out a position that all financially meaningful relationships should be reported.

On the other hand, commission members quickly shot down a proposal that any transaction with a value of $25 constitutes a meaningful relationship, with one member describing such a threshold as the price of a trip to the local hamburger joint.

MedPAC staff analyst Ariel Winter led off one session with the remark that "we expect today's session to lead to draft recommendations" to be forwarded to Congress. Winter described relations between drug/device makers and doctors as "pervasive," adding that drug companies spent $7 billion on physician detailing in 2005 and gave free drug samples worth $18 billion the same year.

There are "both benefits and risks" to such relations, Winter said, including that drug and device makers get much-needed expertise, but these relations "may undermine physician objectivity."

He also noted that there is "no mechanism to enforce or report on [the Office of Inspector General] guidelines" that deal with such relations, although five states and the District of Columbia require drug makers to report such relations. Only the Massachusetts law covers device makers as well as drug companies.

One option MedPAC could consider is a national database on such relations, which "could discourage appropriate relationships," Winter said. Some concerns on this would be compliance costs for makers and administrative costs for government, but he also acknowledged that any such laws "might discourage beneficial arrangements" and "would not [entirely] eliminate conflicts of interest."

On the question of how comprehensive such a data system should be, he said one possible low-end threshold for reporting would be for gifts and other perks valued at $25. Large and small makers of drugs, devices and supplies would be included, as would any entities making payments to companies that host continuing medical education sessions, but Winter said monies paid to patient advocacy groups might also be included in reporting requirements.

As for whether a company should be able to withhold data it deems proprietary, he said one policy approach might be to delay reporting "related to development of new products," which could be tied to the date of registration of related clinical trials at the National Institutes of Health web site. Another date might be the date of an FDA approval of the product, but the absolute latest date should be "no later than a set number of years after a payment is made."

As to whether federal law should pre-empt state law on reports of financial ties between doctors and drug and device makers, Winter said one argument "in favor is it would reduce compliance costs for makers." A proposal to allow states to collect data in parallel to federal data collection, even if those states are disinclined to act on those data, would impose the same elevated compliance costs as would be imposed by state-based reporting.

Publication of the data offers a new set of questions to be answered. An Internet-based database that clearly defines payment categories seems an obvious solution, Winter said, but he pointed out a few implementation issues, such as which agency at the Department of Health and Human Services would handle the task. Among the obvious candidates are FDA and CMS, but he reminded the commissioners that "both agencies have funding and resource constraints," and data collection and sorting costs are unknown.

Scanlon said, "I worry about a threshold that's too low" in reference to the $25 threshold, which he hypothesized would turn out a volume of data that might obscure more important information. As for pre-emption, he said one option would be to make it an administrative move rather than a legal one, possibly alluding to a willingness on the parts of the five states and DC to forgo reporting requirements, even if they left the relevant laws on the books.

Panelist Richard Butler of Rush University Medical Center (Chicago), said, "It gives me a headache to think of reporting everything under the sun" in reference to the $25 threshold, to which Reischauer assented. Reischauer suggested a threshold of at least $100, adding that he would "make sure it's indexed [to inflation] because these things have a way of becoming irrelevant." He also said the threshold of $25 argued that doctors can be influenced "for the price of a hamburger."

As for pre-emption, Reischauer said "my guess is the states will get out of this business if there is an adequate federal law."

Dislcosure sought for doctor owners

Winter gave a thumbnail on the potential for conflicts of interest embodied in physician ownership in ambulatory surgical centers (ASCs) and specialty hospitals, noting that the number of specialty hospitals owned at least in part by doctors rose threefold between 2002 and this year. The number of ASCs with a similar ownership was said to have risen by 60% between 2000 and 2006.

The obvious concern is "physician ownership can affect the volume of services," Winter said, but the fact that such arrangements are not made public makes it "difficult for payers and researchers to obtain information on financial relationships."

At present, CMS requires hospitals to report any ownership stake of 5% or more, but the law in question does not provide for public disclosure. Medicare beneficiaries are entitled to know of any ownership stake their doctor has in a hospital regardless of percentage, but such data is not available to CMS or to the general public under the 5% set-point. The rules for ASCs are essentially identical.

Winter said that one option is to force all hospitals and ASCs to report at all ownership thresholds for both CMS and for web publication, and since 5% is already reported, "the additional reporting burden ... should be minimal."

Commissioner Ronald Castellanos of Southwest Florida Urological Associates (Ft. Myers, Florida) said, "I'm fully in favor of public disclosure," adding that he is "all for transparency. I don't see how it can be any other way." This sentiment was echoed by George Miller of Community Mercy Health Partners (Springfield, Ohio), who said he also saw mandatory disclosure as necessary "across the board."

He alluded to the motive for forming specialty hospitals and quipped, "you never see physicians coming together to open an emergency room."

As for the burden of collecting relevant information, Nancy Kane of the Harvard School of Public Health (Boston) remarked that given the multiple levels of financial participation in the modern American corporate organization, "its going to be really hard to figure out who owns what," and asked what kind of audit capacities will be needed and at what cost. Decoding such organizational structures "adds an enormous cost," so the penalty for running afoul of regulations will have to be substantial to provide a deterrent if the audit capacity is not up to the job.

Offering an alternate vector for information, Michael Chernew, PhD, of Harvard Medical School (also Boston), said the issue might be "which activities should be disclosed by whom," making the case that perhaps doctors ought to do the reporting because the effect of such arrangements is really derived from how large a percentage of the doctor's income is derived from that interest.