Evidence Based Reimbursement

It didn’t take long. The day after I wrote that Evidenced Based Medicine would be used to ration medical care the June 8 issue of the Archives of Internal Medicine appeared on my desk. Under the heading of Health Care Reform is “A 300-Year-Old Solution to the Health Care Crisis”. In it Diamond, et al propose that payment for medical services be linked “directly to the expectation of benefit (so-called evidence-based reimbursement) rather than to subjective standards for outcomes and performance.”

They use conservative and aggressive diagnosis and treatment of coronary artery disease as an example. They correctly point out that long term results are the same with either approach though the cost is obviously much much less with conservative management. Next they suggest that “the Centers of Medicare and Medicaid Service (CMMS) set reimbursement for evidence-based care at a higher level than for non-evidence-based care.” (These authors seem obsessed by hyphens.) In other words if conservative care is the equal of aggressive treatment, the latter should be be paid less.

They advise that such an approach to payment be limited to the five most costly medical conditions. But, of course, once we hit the trail of evidence based reimbursement there will be no stopping. What’s wrong with evidence-based payment? At first glance it seems reasonable. The problems start with who will be assembling and analyzing the evidence. Much of what doctors do is based on shaky evidence that shifts regularly. Very often there isn’t sufficient evidence to arrive at an evidence based conclusion. And the evidence frequently is in the eye of the observer.

Once started CMMS will be setting payment that is “Evidence-Based” on just about everything. Every group you can think of will lobby to have its service included on the paid list. There will of course be losers and winners. If the decision is made that starting dialysis on people older than 70 is not cost efficient and you’re in that group – too bad. The justification used to deny care to that group (substitute any other you wish) will not be that it costs too much but that the best medical evidence suggests that the benefit obtained from such treatment is not great enough for it to displace treatment of some other totally unrelated disease.

Consider this startling statement by Diamond, et al: “If evidence based reimbursement policies…were adopted, dramatic changes in health care utilization could be realized virtually overnight (as happened in the 1980s with the advent of Medicare’s diagnosis-related groups that was implemented to control spiraling hospital costs).” The use of DRGs doesn’t seem to have had much effect on spiraling Medicare costs. It’s because Medicare’s cost are verging on ruinous that we’re discussing this problem. Why do they cite something that hasn’t worked to support an argument that’s shaky to start with?

What the authors of this piece seem unable to confront is that there is no centrally applied top down solution to our imploding healthcare system that is likely to save money while caring for more people. I suspect that they would be horrified at the suggestion that what they urge will immediately morph into a device for rationing medical care, but that’s what evidence-based care will become.