Incentives in health care

Posted by Marc Hodak on July 30, 2009 under Patterns without intention | Be the First to Comment

Here is one of the best articles I have ever read about health care.  It probably could have only been written by a doctor.

The article basically asks why the average cost per patient could be twice as high in one town versus another town in the same state with similar demographics and culture, as well as similar medical outcomes. The author looked at all the usual suspects:  treatments, technologies, and torts.  The culprit turns out to be all three–overutilization of expensive treatments.  This is not necessarily driven by fear of torts, but that doesn’t help.  The more expensive places that prescribe more tests and treatments don’t tend to have better patient outcomes.  In fact, they tend to be worse because every treatment has risks as well as benefits, and it’s quite possible for unwarranted treatments to have a net cost in public health.

When he asked why certain places were more expensive, it unsurprisingly came down to the prevailing incentives of different models of health care practiced in different towns.  In the high-cost model, physicians focused on revenue maximization by looking at the patient as a revenue source.  In these environments, which evolved over time to become the culture of medicine as it’s practiced in that area, doctors tended to over-prescribe tests and treatments where judgment allowed (which covers a lot of illnesses), often referring patients to facilities in which the doctors had a financial interest.  The lower cost, higher quality models were less individualistic.  The doctors worked as a team, easily shared information, and got paid salaries in (generally) non-profit organizations.  They were content to not maximize their personal profits as long as they were comfortably paid and allowed to act as professionals.

The author was agnostic about most of the things bandied about in today’s health care debate.  It doesn’t matter if the government or private insurers are paying for treatments.  The doctors are (properly) in control.  If they are intent on gaming the system, they can game it regardless of who is paying.  Doctors in systems built around putting their patient’s interests first can work quite well with any payers, although they will actually save more money by being given the discretion to spend what they think is appropriate.  The idea of having the patients bear more of their own costs was pooh-poohed as nonsense.  As one doctor asked:

“I’ll do three vessels for thirty thousand, but if you take four I’ll throw in an extra night in the I.C.U.”—that sort of thing?

The bottom line is that the current level of waste in the high cost portion of our health care system is unsustainable.  Worse yet, communities are migrating from the more effective, efficient, collaborative system toward the more individualistic, wasteful, and profitable model.  The key (and this may be my conclusion more than the author’s) is to figure out a way to reward the more collaborative, higher quality, lower cost, model so that it is actually the more profitable as well.

I think the insights provided by the stories and research in this article were incredibly instructive.  I would supplement them with a few observations:

– It seemed to me that the Medicare system is easier to game.  It is also more susceptible to profligacy due to the political influence wielded by the profit-seeking (and highly profitable) model.  It seems even the most individualistic doctors are perfectly capable of banding together into large cooperatives for the purpose of lobbying.

– The private insurers could also be gamed, but they were at least forced to make trade-offs between pushing down costs and having to accede to the judgment of the doctors.  But the doctors are (again, properly) in the drivers seat, so their attitude about “patients first” versus “me first” dominates the cost discussion.

– The author made a passing reference to tort reform, with a quote from the doctors that it was remarkably effective at reducing litigation, but leaving out any further discussion about its impact on defensive medicine generally, which is the easiest place for profit-driven physicians to use their discretion to over-test and over-treat.

– Unsurprisingly, the author omitted any mention of mechanisms that might preferentially reward collaborative, patient-first medicine.  (It’s easy to omit what is not visible.)  He makes the case that community-based solutions are the only kinds likely to work, but punts as to what could drive that.  Well, a greater public awareness of costs and benefits could drive that, combined with some sensitivity to the costs.

– I understand that medical outcomes are devilishly difficult to meaningfully measure.  I really do.  It’s impossible to precisely design metrics that tell a non-practitioner how good Hospital A is versus Hospital B.  But right now, people have nothing.  There is no data on  relative outcomes in different disease or treatment categories, or patient/family satisfaction, or average cost per patient.  This isn’t like E-bay ratings, I know, but it’s not impossible.

– The notion of haggling with a doctor as a rebuttal for the benefits of having more “skin in the game” is a straw man.  Nobody is suggesting bargaining for their care.  I’m suggesting shopping for a plan, and it would definitely help if people had skin in the game in terms of choosing one.  New York actually provides ratings of health insurers on various dimensions, including customer satisfaction.  An insurer or hospital plan with a reputation for stiffing their patients would suffer appropriately.  Similarly, a plan that was too expensive would be forced to at least consider more efficient models of health care.

– Yeah, yeah, adverse selection makes it harder for plans to control their costs, but I’m sure one could design mechanisms around that if they were allowed to evolve. Life insurers generally get younger folks to buy earlier by offering lower rates that provide a permanent, low base from which all future increases are made.  A consortium of health care providers can come together to offer something similar.

“Hey, you, 22 year old kid in perfect health.  Sign up with the 5-Star Alliance, and we will guarantee you rates that go up only one percent above inflation for as long as you’re with any one of us.  If you wait, the base line rate will go up much faster than that.  And don’t you dare wait until you start getting sick to sign on.”

This is an easily visualized world.  It makes perfect sense in the context of other kinds of insurance that people already carry, like property and life.  Getting there, however, will take a few things we don’t currently have.

First, the insurance would have to belong to the individual, not the employer.  It would have to be portable.  Second, the consortium would have to be able to offer national coverage.  This would mean substantially eliminating the ability of states to individually regulate insurers.  In other words, the Federal government would have to do exactly what the commerce clause intended for them to do–tear down barriers to interstate commerce.  Finally, we would have to develop a cultural expectation for people to always have health insurance, right after they have a home, but before they get cable.  We could (and I think should) have a system where people who are too poor or too foolish get insurance until they become uninsurable can get access to a basic, second-rate, government-provided plan.  But we would need to get rid of the notion that anyone who forgoes bearing the cost of insurance has a right to the best health care available, or even average health care.  Nothing encourages adverse selection more than steps to promulgate this “right.”  In New York, where insurers must offer the same rate to anyone regardless of pre-existing conditions, it would be foolish for a young person to pay the high costs of health insurance before they got ill.  So, few of them do.  And it raises the costs for the rest of us, and makes it that much less affordable for them (until they get sick).  It’s a lousy mechanism.

People will argue that asymmetrical information and market imperfections will doom any private system, but they will always lose that argument against an expert creator of mechanisms.  I have never yet been stumped.  No mechanism is perfect, of course, but there is no reason to let the perfect be the enemy of better-than-we’ve-got-today.