Newtown, CT – March 7, 2014
The tide of change is rising, and it's got the hand-wringing ivory tower dwellers gasping for air – There are thousands of healthcare services researchers busily employed by hundreds of organizations across the country. It's actually astonishing how many of these people are around. Some focus on the effects of care patterns and protocols, or of new medical innovations, and that work should continue at a rapid pace, even accelerate. There are others, however, who study the effects of micro-environmental changes in what some have described as the mother of all adaptive systems. Unsurprisingly, they have been increasingly calling for slowing down the pace of incentives changes, going as far as suggesting the use of randomized trials. For those of us who have worked outside of health care, it sounds loony…because it is. While randomized trials could and probably should be used to test the effect of new consumer incentives, it's a ridiculous notion to suggest that it should be used to test the effect of provider payments. Last year we published a report in which we summarize the existing knowledge on behavioral motivation of professionals, and the upshot is that, for the most part, payment incentives act as a deterrent to high professional standards. As such, designing payment incentives isn't about optimizing behaviors, but rather reducing negative impulses. That's also why the report released this year by RAND is another headline grabber, but a ho-hum in substance. It's not news that past experiments haven't designed very well, and here's another non-news flash: many current and future experiments won't be designed well either. So what?
What this means to you – The hand-wringers are trying to tell us that we're moving too fast, that changes in payments might not be as effective as thought, that we might be wasting better opportunities. And it all sounds good, even though it's also self-serving. After all, what would all these folks do if the market changed and they weren't needed anymore? How many researchers are there in the rest of industry busily studying the effects of how professionals are paid? A dozen, perhaps a hundred, but certainly not thousands. The truth is that these academics survive by publishing, and slowing down the pace of change to better meet their needs would certainly be beneficial to them. But not for those who are paying the price today for the snail's pace at which payment innovation has been deployed. Hundreds of thousands of consumers are harmed by health care every year. Millions can't afford their health care expenses. Tens of millions are not getting enough services, or too many of the wrong ones. And yet we should slow down and do randomized trials…it's not only a bad joke, it's an insult to all but the privileged few who can afford to turn their noses on the hardscrabble farming that is needed to make the system better for all. Yes there will be mistakes and yes some of the programs will fail, but there will also be successes, and those have to be rapidly deployed. What's needed is more learning networks and informal publications of results, not decade long randomized trials. Fortunately, the tide has started to rise, and if it rises rapidly over the heads of the do-nothing crowd, that might be a form of progress as well.