Improving the U.S. health-care system requires encouraging low-value doctors and hospitals to practice as well as high-value ones do. The gap between the two is wide, but that only shows how much room we have for improvement.

Costs vary wildly across regions, among hospitals within a region, and even among doctors within a given hospital. Because this variation doesn’t appear to be reliably correlated with differences in quality, value seems to be much higher in some settings than in others. What is causing this, and what might we do about it?

A new report from the Institute of Medicine provides some answers. A blue-ribbon panel, asked by Congress to examine whether Medicare should pay more in high-value areas of the country than in low-value ones, concluded that would be unwise; instead, the agency should reward value provided by individual doctors and hospitals. In studying this question, the panel also provided new answers to why spending varies across the country.

The report confirmed that Medicare spending varies widely, as Dartmouth College researchers had found earlier. In some areas, spending per beneficiary was found to be about 40 percent higher than in others. And the same amount of variation was found among commercial health insurers. This persisted over decades; regions that spent the most in 1992 tended to remain big spenders in 2010.

Market Power

There is one important difference between Medicare and commercial insurance, the institute found, and that is in the causes of spending variation. With commercial insurance, spending is higher in some areas because of markups — that is, the difference between the charge for a service and the cost of providing that service. Seventy percent of the variation in commercial spending was attributed to differences in markups, which in turn probably reflect local differences in market power among hospitals and other providers relative to insurance companies and beneficiaries.

The story for Medicare is much different. Here the variation is driven by use. In some regions and at some hospitals within a region, Medicare spends more because beneficiaries there use more services. (The greater use of services couldn’t be fully attributed to differences in the population’s health.)

That difference between Medicare and private insurance, by the way, poses a fundamental dilemma for policy makers. The Institute of Medicine panel, like most other health economists and experts, strongly supports the health-care system’s shift away from fee-for-service payment and toward paying for value — through bundled payments or accountable care organizations, for example — because this should discourage unnecessary care. At the same time, it also loads more risk onto doctors and hospitals, as if they were the insurers. Yet many of them lack the scale to spread the risks efficiently. This is one reason so many providers are looking to merge with one another.

As a result, there’s even more pressure for concentration in local hospital markets. That means even higher price markups for private insurers.

Ultimately, then, value-based payments, by discouraging unnecessary care, may narrow the spending variation in Medicare but, by encouraging consolidation and increased markups, widen price and spending variation in commercial insurance.

The institute thus raises one issue not mentioned in the report itself: a growing tension between the trend toward consolidation in health care and the effort to move away from fee-for-service payment.

That shouldn’t discourage Medicare from adopting value-based payments. It does suggest, however, that policy makers will need to watch for excessive local hospital concentration and markups and to stay alert to any violations of antitrust laws.

Our ideas can save democracy... But we need your help! Donate Now!

Peter Orszag

Peter Orszag , an occasional Bloomberg View columnist, is vice chairman of corporate and investment banking at Citigroup Inc. and a former director of the Office of Management and Budget in the Obama administration.