Amid all the political battlefronts in the effort to reform our multi-trillion-dollar health care system, some of the most potentially worthwhile initiatives have received little notice—and the notice they have received has threatened to undo them. Each of the health care bills under consideration as we went to press creates a government-supported institute to oversee research comparing the effectiveness of existing medical treatments and practices. The American Recovery and Reinvestment Act of 2009 also allotted $1.1 billion to comparative effectiveness research (CER), whose results are expected to begin appearing within a year or so.
To guide the spending of that money, the National Institute of Medicine made a priority list of situations for which data about outcomes are badly needed—for instance, comparing the effectiveness of various medical and behavioral interventions to prevent the elderly from falling (the complications of which are a leading cause of death), comparing assorted drugs and surgeries alone or in combination in the treatment of specific cancers, comparing the effectiveness of different implants and devices for treating hearing loss, and so forth.* In most cases, the recommendations explicitly state the goal is to compare the effectiveness of treatments and practices in specific patient populations. In other words, not to seek one-size-fits-all answers.
Yet many people have gotten the wrong impression that CER is little more than an excuse to ration care. In early December, Senator Lisa Murkowski of Alaska attempted to insert an amendment to the health care bill forbidding insurers from denying coverage of medical tests or treatments based on comparative effectiveness research findings. She was responding to the brief but loud controversy over a study last fall that questioned the value of routine yearly mammograms for women younger than 50. A study panel appointed by the U.S. Preventive Services Task Force reviewed available evidence and concluded that a blanket recommendation for women to have annual mammograms starting at age 40 is unwarranted. By the group’s calculation, the mass screenings incurred a high likelihood of invasive follow-up testing and anxiety while finding a relatively small number of cancers that would have been lethal if they were caught later. To save 10,000 lives through early detection of tumors, one analysis found, 19 million women in their 40s would have to be screened over 10 years.
Advocates of cutting health spending did not help matters by noting that those figures could add up to $20 million per life saved. The cost-benefit discussion left the impression that the panel had judged the value of those 10,000 hypothetical women’s lives and decided that the price of saving them was too high. But the report could and should have been interpreted differently. It assessed risks, not costs, versus benefits. And it did not say that mass screening is ineffective at catching deadly cancers, merely grossly inefficient, which is as much a commentary on the inadequacies of current screening technologies as on the ineffectiveness of blanket prescriptions.
To make informed decisions, any individual and his or her doctor need evidence, so comparative effectiveness research should, in principle, make more personalized medicine possible. The goal of CER is not to identify the most effective test or therapy for the great majority and impose it on everyone. Nor is it to ration health care—if CER finds that the more expensive treatment is also the most effective, so be it.
Senator Murkowski’s amendment was ultimately defeated, but Senator Barbara Mikulski of Maryland succeeded in adding language to the bill requiring insurers to cover mammograms that doctors deem prudent, and Senator David Vitter of Louisiana added, by unanimous consent, a directive telling the government to disregard the latest task force recommendations. This fear and misunderstanding of comparative effectiveness research are unfortunate. Used properly, the CER studies should give us better medicine, not take it away. That should make support and protection for comparative effectiveness research one of the easiest pieces of the health care puzzle to resolve.
*Erratum (11/23/10): The editors referred to the National Institute of Medicine. The correct name is the Institute of Medicine of the National Academies.