Earlier this week we gave you an overview of the new Medicare Payment Advisory Commission (MedPAC). Today we want to go into a bit more detail on MedPAC's ideas about a comparative effectiveness research entity—including governance models and funding streams.

MedPAC believes the entity should sponsor and fund studies that compare the clinical effectives of two or more procedures, drugs, or devices to investigate what works well, for whom, when. It would be independent (non-political) and produce objective information in a transparent fashion. Interestingly, it would make no coverage or payment recommendations—although patients, payers, and physicians are free to reach their own conclusions using the data. It would be funded with both public and private money, have an independent board, and contract most research out to outside groups, including academic research centers, the NIH, and the Agency for Healthcare Research and Quality.

The MedPAC report covers three items: the design of the entity's board, its placement, and its funding. But the most compelling part is a three-page sidebar that highlights the poor clinical decision making that occurs when we don't have the information we need. Five cases are highlighted, starting on page 113 in chapter five. But let me summarize just one so that you can appreciate the need that exists.

Until six years ago, hormone replacement therapy was prescribed broadly to older women with menopausal symptoms. Its wide usage was based on decades of observational (read: non-scientific) evidence that suggested it was associated with greater cardiovascular health. In the previous decade, it jumped quite a bit, from 58 million prescriptions in 1995 to 91 million in 2001. Total sales in 2000 were $1.2 billion (about the GDP of Moldova ).

Then in 2002, a large, NIH-sponsored study called the Women's Health Initiative published findings from the first randomized primary prevention trial (read: scientific) of postmenopausal hormone treatment. The results showed that hormone replacement therapy posed more health risk than benefits, including increased risk for heart disease, breast cancer, stroke, blood clots, and dementia. Eighteen months later, use had dropped by nearly half. In other words, taking nothing was better.

But that's not to say this treatment is useless; for example, a recent study shows that it might prevent age-related eye disease. Again, a major point of comparative effectiveness information is that very few things are always good or always bad. It's for navigating the grey areas that we need research.

The point of all this is to show that this information will result in better health outcomes and lower health care spending growth. As MedPAC has stated, we need solutions to both, and comparative effectiveness is a vital tool to address them.