The Unintended Consequences of Reimbursement Changes and Interplay with Practice Patterns

December 5, 2011

As we continue to focus on health care spending, it is important to look at the tools we’ve tried already and learn from our experience – especially our mistakes.

In health care, it seems that every action taken to reduce spending leads to an equal and opposite reaction elsewhere in the system. This issue has been well illustrated by the Congressional debate on the sustainable growth rate (SGR) formula since 1998 (for more on SGR see our 5/10/2010 blog post[POST TITLED “THE WRONG POLICY: PHYSICIANS…”). While health policy analysts and people in both parties agree on the need for a fix—and have lots of ideas for it—every solution is fraught with major problems and creates winners and losers, and that makes the politics of change extremely difficult.

Another illustration of this type of a problem can be seen in payment changes made to oncology drugs. The issue of payment for oncology drugs delivered in physician offices has been a topic of much debate among health plans over the past 10 years. There was little doubt that the amount paid for these drugs by most payers was highly inflated relative to their cost, and that payments were providing “windfall” profits to oncologists who provided chemotherapy in their offices. In January of 2005, Medicare significantly lowered its payments for physician-administered drugs—a change that many private health plans followed. Specifically, the change lowered the amount paid for these drugs from the average wholesale price (AWP) to the average sales price (ASP) of the drugs, plus a 6 percent margin. The AWP was an artificially set price that was not what physicians were actually paying for the drugs. The ASP rate was intended to more than adequately cover the cost of the drug along with the cost to administer it. The ASP was significantly lower than the AWP—in some cases, quite dramatically so.

So, what happened as a result of this change? The November 2, 2011 New England Journal of Medicine reports on a fascinating study that describes some of the impacts of that change with regard to one disease – lung cancer. In the United States overall, the rates of chemotherapy treatment for lung cancer went up after the payment change, by 10 percent within 60 days of the payment change and about double that within six months of the change. That is, as rates for these drugs were cut, on average, physicians in the United States started providing more treatments to patients.

Clearly, this change is a concerning one from a patient care standpoint: one wonders about the impact on patient outcomes and care from this additional chemotherapy. But, perhaps unfortunately, the change is not entirely surprising—target income theory notes that as prices are cut, providers will seek to offset those cuts by providing more or different services (an experience quite in evidence with the SGR).

What was most surprising in the study reported on in the NEJM, however, is that this change was not uniform across the country—it varied quite significantly geographically. For example, use rates for chemotherapy went up quite significantly in New Hampshire, Minnesota, and the District of Columbia (and in many other states), but down in Idaho, South Dakota, and North Dakota. Indeed, there is no discernible pattern among the states that increased use considerably, modestly, or not at all. And, this change was certainly not driven by a change in patient characteristics between the states.

If we look just at the experience of the SGR, we can see that it is essential to think about the unintended consequences of reimbursement changes. But, looking more deeply at the example of these chemotherapy drugs helps us understand that the picture is even more complex than we might have thought. What do we learn when we add in this complexity? If nothing else, this study is a reaffirmation that local practice patterns are essential: geography is indeed destiny and all health care is local.

Looking at cost trends in broad averages isn’t enough. If we are going to have any sustained success at changing this cost picture, it will take aligned incentives that are not focused on price alone and that take human behavior into account. Anything less and we will be destined to repeat the mistakes we have made so brilliantly made in the past.