Understanding The Value In Medicare's Physician Payment Data Dump

Apr 14, 2014, 9:24 AM, Posted by

A 35-year battle is over and the taxpayers have won: We have the right to know how much physicians receive in Medicare dollars in exchange for providing our care. But now that the Centers for Medicare and Medicare government has released data on $77 billion in Medicare Part B payments to providers during 2012, what do we really know—or have—that we didn’t have previously? Information alone isn’t knowledge or, for that matter, insight.

For consumers, the slew of raw data ultimately may be useful if it can be packaged into applications that help them compare the way physicians practice—as the Office of the National Coordinator for Health Information Technology now proposes in a newly announced challenge. Private payers, such as insurers, may also find the Medicare data useful, as they can the information to better understand the practice patterns of providers they include in their networks.

But the greatest value of these data is in strengthening the case for ongoing changes in the Medicare program. It’s instructive that the data released by the Centers for Medicare and Medicaid Services (CMS) pertain only to payments issued by the fee-for-service part of Medicare. Data from the large and growing Medicare Advantage program is omitted, since those dollars flow to private managed health care plans first, and only later to physicians. So what we have is an almost pointillist picture of how Medicare dollars are spent at the largely unmanaged, fee-for-service level, mainly to physicians (the data also include payments to other “suppliers,” such as laboratories, imaging centers and ambulances). And as we’ve known for a long time, much of the picture isn’t pretty.

Fraud and abuse. The most obvious—if expected—finding from the data dump is that at least a small number of physicians are draining big dollars from the system, either in outright fraud or systematic abuse. The Inspector General of the Department of Health and Human Services had already identified 104 clinicians for “improper payment reviews” among a group of 300 who furnished more than $3 million of Part B Services in 2009; some of those physicians have already been prosecuted. HHS and other federal agencies are now using predictive tools to watch for suspicious billing patterns in the Medicare program. Although the newly released data by themselves don’t generate much new insight into fraud and abuse, they should help make the case for stepping up efforts to ferret out more illicit activity in the future.

Perverse incentives. The newly released data also flesh out an old story: that many features of fee-for-service Medicare reflect perverse incentives that influence care. For example, many drugs and biologics administered by physicians in office or hospital outpatient settings are paid for under Part B, and thus are included in these data as payments to physician practices. Because of the way doctors are reimbursed for supplying these drugs—at the drugs’ so-called “average sales price” plus 6 percent—physicians may be encouraged to buy and use more expensive drugs within the same class.

A case in point: ophthalmology practices that use the very similar drugs Avastin and Lucentis, both made by Genentech, to treat age-related macular degeneration. A federally funded clinical trial showed that the two drugs produce equivalent results, but annual treatment costs are $650 for Avastin (which is used “off-label,” since the FDA hasn’t approved it for this use) compared with $50,000 with Lucentis.

Needless to say, 6 percent of a higher priced drug is a larger payout than 6 percent of a lower-priced one. Although ophthalmologists who use Lucentis typically insist that the higher reimbursement isn’t a factor, it’s hard to explain otherwise why there is so much variation in the drugs’ use. This dynamic is why government officials would like to test more “bundled payment” arrangements, in which providers would be paid an overall fee for a care episode, and would thus be incentivized to provide only the most cost-effective care.

Necessary but not sufficient: Payment data without quality data is only half the story, or even less: if a particularly skilled physician treats lots of Medicare patients and gets especially good outcomes, we shouldn’t begrudge her high earnings from the program. But we don’t have enough data on quality in Medicare to link it to the payment data. That will change as Medicare implements the new Physician Quality Reporting System in Medicare over the next several years, and doctors are penalized financially if they don’t report on various quality measures. As a nation, we need this bigger picture if we’re really going to be able to assess whether we’re getting our money’s worth for the trillions of dollars we spend through Medicare.

Slowly but (let’s hope) surely, Medicare is moving toward a system that pays for the value of the care delivered instead of the sheer volume. This week’s data dump reminds us that we should remove all remaining speed bumps and get to new payment and delivery systems as fast as we can.