(KQED online)

April 29, 2009

Limiting Treatment to Those Who Need It

by Dana P. Goldman

The United States will produce more than $14 trillion worth of goods and services this year—truly an astonishing amount. But equally astonishing is that one out of every six of these dollars will go to health care. By comparison, in 1960 Americans spent only $1 out of every $20 on health care.

This ever-increasing share of national output going to health care is the source of much hand-wringing by policy makers. They worry that we cannot afford to spend so much, and that our national output will suffer as a result. They have it backwards.

Part of the reason we spend so much more on health care is because we are wealthier. As incomes rise, there is only so much value we can get from adding more square footage to our houses, putting more horsepower in our cars or buying finer wine. What really makes a rich society happy is to have more time to enjoy all these goods and services. So we spend more on health care so that we can live longer and healthier lives to enjoy the fruits of all our labor.

Put another way, if you had the choice between buying 1960s medicine at 1960s prices, or today's medicine at today's prices, which would you prefer?

This doesn't mean we couldn't spend our money better. The primary driver of rising health care costs has been medical technologies. We continue to develop expensive treatments and, equally importantly, apply them to broader swaths of the population. For example, the development of magnetic resonance imaging (MRI) has revolutionized our ability to diagnosis brain tumors and other serious musculoskeletal problems. But when patients with insurance clamor for an MRI every time they sprain their knee, that raises costs for all of us. It also is what makes insurance premiums unaffordable for many Americans.

So, going forward, the policy goal should be to get the right drugs and treatments to the patients who need it. But it also sometimes means withholding treatment in cases where it might do only very little good.

There are two distinct ways to ration medical technology. Here the term ‘ration' is used in the economic sense, not pejoratively. For example, we rely on the private marketplace to ration televisions. Suppliers who can sell televisions more cheaply are rewarded with more market share, and consumers decide what they want to buy. This is downstream rationing—and the key is that we rely on prices to determine who gets a good or service.

The problem with downstream rationing in health care is that insurance completely insulates the patient from the full price. There is little incentive for the supplier to lower costs. There are ways to make the patient more sensitive to price. For example, we could restructure health benefits so that consumers are asked to pay more out-of-pocket for services providing little clinical benefit. Patients who insist on an MRI for their ankle sprain will think twice if they have to pay several hundred dollars for it.

The alternative approach is upstream rationing—that is, the government decides what makes it to the market in the first place. We already do this a lot in health care. The Food and Drug Administration is responsible for approving many new treatments and medical devices. Right now they make this decision solely on the basis of medical necessity. So if your drug is somewhat better than standard care, it gets approved regardless of whether it costs $100 or $200,000 per year.

The FDA could tilt the playing field to encourage cost savings. This could be done by offering expedited approval—and immediate coverage by Medicare—if the manufacturer can demonstrate that its treatment will reduce overall health care costs. This judgment would be made by independent experts, as the FDA does now, and would be monitored through post-marketing studies. Such decisions are made routinely in other countries with national health systems.

Unfortunately, upstream rationing sounds better in theory than practice. It is hard to collect to data on outcomes, and it is even harder to use it to make policy decisions. Mainly, this is because for any medical treatment, there is usually a population who gets enormous benefit, and then there is a (usually much larger) population for whom the benefits are unclear.

This is the promise of comparative effectiveness—to try and identify the relative value of treatments. The data could then be used for both upstream rationing (through the FDA) or downstream (by insurers when they decide how much patients have to pay).

While this sounds good, whether the government can implement comparative effectiveness effectively is open to debate. We are not very good at estimating the health care benefits (and even costs) for many patient populations, and the thought of using the current methods to estimate gains should give everyone pause.

This essay was first published online as part of KQED Radio's special project, "Healthy Ideas: Californians Weigh In on Health Care Reform."

This commentary originally appeared on KQED online on April 29, 2009. Commentary gives RAND researchers a platform to convey insights based on their professional expertise and often on their peer-reviewed research and analysis.