I have been thinking lately about the state of the field of health services research. Having plied this trade for nearly 30 years, it struck me that many of the unanswered questions that I encountered as a doctoral student remain unanswered. I plan to post occasional blogs in which I pose these questions, discuss the state of the research, and explain why it is critical that we come up with better answers. The first question is really the big kahuna: If you get what you pay for, how much should you spend?
Everyone seems to agree that the U.S. spends too much money on healthcare. This has led many to embrace machete policies: Slash payments to doctors. Slash payments to hospitals. Slash payments to drug companies. Slash the number of specialists. Slash, slash, slash.
There is abundant research that past machete policies directed towards healthcare providers have adversely affect healthcare quality and access. There is also abundant evidence supporting the view that machete policies would curtail medical innovation. Medical providers and drug companies cite this evidence whenever they are threatened with payment cuts, proclaiming that any reductions from current levels would be disastrous for the American public. (This is not to deny that a lot of spending is inefficient; unfortunately, no one has devised the machete that only slashes inefficient spending.) As I will explain in a moment, we are in no position to assess such claims. But note first that if cutting healthcare spending would be disastrous, the implication is astonishing: increasing spending would be wonderful. (I suppose it is possible, by sheer happenstance, that we are spending exactly the right amount of money on doctors, hospitals, and specialists, but the odds of that are about the same as the odds of getting a royal fizbin – astronomical.)
Unfortunately, health services research has taught us just enough to be totally confused about this issue. As I mentioned above, research tells us that if we follow a machete policy, quality and access will fall. But research does not come close to telling us everything we need to know about this tradeoff before we can reach a sensible policy decision. (Congressional Republicans deny the existence of this tradeoff and refuse to discuss it. This makes it easy for them to take out their machetes.) To determine whether we should slash, slash, slash we need to know two more things. First, exactly how much will quality and access suffer per dollar saved? Second, is too much?
Economists have produced interesting answers to the second question. Studying things like the tradeoff between wages and job safety, economists have concluded that we need to save at least several hundred thousand dollars to offset the loss of just one quality adjusted life year. The exact dollar amount remains open to considerable debate. Remarkably, this is the easier number to pin down.
We just don’t know how much quality will be harmed by a machete policy. We have some strong evidence that further reductions in Medicaid payments will cause substantial harm for patients with certain conditions such as heart disease, but those are cutbacks from levels that are already low and might not apply to other medical conditions. We cannot say whether these results will translate for Medicare or the privately insured. There are a few studies of Medicare cutbacks and while the results again suggest that there will be quality and access reductions, the standard errors are large enough so that all we can say is that the reductions will be anywhere from small to large. To make matters worse, these studies do not focus on physicians or distinguish among specialties, and are silent about cuts to home health care and other providers. And for the studies of medical innovation, we are quite sure that if drug profits fall, we will see fewer drugs. A lot fewer? We don’t know. Nor do we know if we will see one less statin drug or if we will never see the drug that cures Alzheimer’s. About all we can say, based on research by Harvard’s David Cutler and his colleagues, is that new medical technology is worthwhile in the aggregate (the health benefits of an entire generation of medical technology exceed the costs). But even the estimable Professor Cutler cannot identify the marginal impact of a machete policy on technological change.
When I started plying my trade, health care spending accounted for 10 percent of the GDP. Most people thought that was too high, but they could offer no logical reason why. It just felt that way. Today, spending stands at 17 percent of a much higher GDP. Nearly everyone thinks this is too much though a logical explanation is still elusive. It just feels like we are spending too much. Yet hardly anyone is proposing that we return to 10 percent though no one can state what the right number is. What will we be saying thirty years from now?
The bottom line when it comes to assessing health care spending, we don’t know much about the bottom line. We publish papers showing that some specific policy had some specific outcome, which is all well and good. At least we know that Congressional Republicans are putting their collective heads in the sand. But beyond that all we can do is make conjectures. The Congressional Budget Office is charged with making these guesses for federal legislation, but they have little to go on. They can project cost savings with a tiny degree of confidence. But no one has asked them if the savings are worth pursuing, which is just as well, because they cannot possible answer that question.
In a nutshell: When it comes to healthcare spending, we do get what we pay for. If only we knew how much we wanted to spend.