With the debate on health care reform in full force, it's notable that the proposed bills all focus on extending coverage but do not have credible plans for controlling costs. It's a matter of the greatest national urgency to provide universal coverage, but we can't do it without bending back ever-rising cost curves. Reversing the escalation of health care costs is going to need more than legislation, yet it can be done without imposing rationing as critics of reform fear.
The key to understanding how this can happen begins with recognizing the crucial distinction between providing good care to one patient at a time, and managing whole patient populations for good health outcomes.
As an example, let's say you're diabetic and you're seeing your family physician for your annual checkup. Your doctor can tell you how well or poorly you're doing in managing your chronic disease over time. Patient by patient, the quality of health care by family physicians in this country is pretty good, and, of course, the goal of universal coverage is that everyone have access to a primary care physician.
But now, let's look at diabetes management not just for you, but for all the diabetics in your physician's practice, who number in the hundreds. If one asks how well or poorly are they doing as a group, there's no good way to know. The information is in the wrong form, scattered across the hundreds of disparate paper records that 97% of small medical practices still use. Since most of the total cost in our health case system has to do with management of chronic diseases like diabetes and hypertension, this choice of example is central to the cost management issue.
A typical medical practice is like an old-fashioned business which keeps all of its records on paper. It can probably track down any individual transaction if it needs to, but it's basically helpless when it comes to overall measurements of performance. And that's the big problem.
What would make a difference is if the physician had patient data in electronic medical records (EMR's). Beyond the value of EMR's for individual patients, those that permit data aggregation are indispensable to assessing overall performance.
It's a significant peculiarity of the world of routine medical care that such aggregate performance measures are not only unavailable, but regarded as unimportant or even unwanted.
Worse, many existing EMR's are in fact really billing documentation modules used for managing insurance claims and not in the least useful for tracking population health. But with the right kind of EMR, it would be straight-forward to determine measures of the health of a particular population or sub-population.
Why does this matter? It's not as if the legacy health care software vendors are itching to bring these capabilities to market or that their customers who manage medical practices are clamoring for them either.
It matters because once performance can be measured, it opens up the possibility of reforming payment systems to pay for outcomes (outputs) instead of inputs (office visits, lab tests, etc.). If a physician can improve the overall health of diabetics in their panel, they could be paid more. Even if we wanted to base payments on this today, we couldn't, but with the right information technology infrastructure we could.
Physicians today, as human beings, are not exempt from the perverse economic pressures created by fee-for-service regimes to see more patients for shorter appointments and order more tests and procedures. If the incentives were changed to pay to foster better health outcomes, I am convinced physician behavior would change over time.
Of course, putting such a system of payment reform in place would be a huge undertaking. To get to the starting line, though, we need to be able to measure population health, and that's where information technology can play a crucial enabling role. Therein lies the real hope for controlling medical costs.
The culture of the practice of medicine, however, is enormously resistant to this approach. There's a great deal of suspicion and misunderstanding about IT among practicing doctors. One hears things like, "I don't want to be turned into a data entry clerk, and I don't want some machine between me and my patients." It reminds me of what I heard from managers and executives a generation ago about personal computers. "I'm not going to spend my time typing at a keyboard. That's for secretaries and clerks. It's wasteful and inefficient."
Another contributor to the problem is an over-insistence on medicine as an art, which therefore does not lend itself to outcomes-based assessment. While there are plenty of hard cases in which experience, human judgment and intuition are terribly important, there are many opportunities to provide basic health care services in more efficient ways and to use IT to help measure effectiveness. We have to get serious about this if there is any hope of controlling health care costs.
Follow Mitchell Kapor on Twitter: www.twitter.com/mkapor
