AAMC Reporter: June 2009

Variations in Health Spending Spark Interest, Debate

As director of the federal Office of Management and Budget, Peter Orszag left his imprint on every line of President Barack Obama's first budget. So when Orszag talks about health care reform—an issue he has called "key to our fiscal future" in various media accounts— one can reasonably assume his ideas carry serious weight.

With some reports indicating that health care spending reached $2.4 trillion in 2008, the administration proposing a $635 billion "down payment" on reform, and various health care leaders pledging $1.2 trillion in spending cuts, reducing costs and promoting efficiency is a near-obsession for health policymakers.

Several initiatives, including health IT, quality incentives, and comparative effectiveness research, could cultivate efficiency. However, a main tenet of Orszag's cost-cutting plan—and other similar plans—is reducing variations in health care delivery, or the notion that health care is overused in some geographic regions, mainly in the form of longer hospital stays and more treatment through certain specialists or procedures, without necessarily leading to better health outcomes.

The primary source of this assertion is the Dartmouth Atlas of Health Care, a project created to track these variations. A 2008 Dartmouth Atlas report estimated that one-third of total health care spending is "wasted," and that Medicare costs could be up to 43 percent lower if the system were more organized.

"There is a huge amount of care that is provided that is unnecessary," Orszag said in a January report from The American Prospect. "The Dartmouth folks say as much as 30 percent, others say between 15 percent or 10 percent, and fine, that's huge. The question is how we get out of that."

On the surface, reducing geographic spending variations may present a tantalizing path to a leaner, cheaper system. But some health policy experts believe that a deeper exploration of these variations reveals a far murkier picture.

"Some people think that if you spend 30 percent more in one area than another, it automatically means it's waste," said Atul Grover, M.D., Ph.D., AAMC's chief advocacy officer. "But these studies raise more questions than they answer. Maybe a proposal sounds good at the national level, but at the community level and the individual patient level, the perspective may be very different. How do those health care decisions really get made? We don't know enough about why differences in spending occur."

Many variables can drive up the use of certain health care services but may be difficult to control, Grover said.

"It's easy to call it wasteful if a CAT scan is done on every patient who comes into the ER complaining of a headache," Grover said. "But there are a number of factors that increase utilization. ER doctors do not want to miss anything. From the patient perspective, they want care that is as thorough as possible, sometimes even when the doctor tells them their risk factor is very low. Plus, if the patient is insured, they are not paying for the scan. So cost doesn't matter as much to them."

Dartmouth researchers suggested in an April report that lawmakers could reduce variations by limiting hospital growth in areas deemed to be overusing health care resources, and enacting reforms that encourage more integrated or team-based care.

Although there is widespread support for enhancing efficiency, Grover said, hospitals that streamline care may be financially penalized.

"[Greater efficiency] in a system may not be rewarded under the current reimbursement system," Grover said. "Is there any incentive to keep a hospital bed empty?"

Some studies offer different insights into the causes and effects of spending variations. A 2008 study in the journal Health Affairs by Richard A. Cooper, M.D., professor of medicine at the Hospital of the University of Pennsylvania and a senior fellow at the university's Leonard Davis Institute of Health Economics, argued that more available physicians and health care resources does mean better outcomes, and that patients should "expect better quality in states with higher per capita spending overall." Dartmouth Atlas officials have disputed Cooper's claims.

Another study led by researchers from Boston University School of Medicine and published in the March issue of the Archives of Internal Medicine suggested that a region's racial or ethnic makeup may play a role in spending variations. According to the study, in the final six months of life, black and Hispanic patients spent an average of 32 percent and 57 percent more on health care, respectively, than whites.

A study in the March/April Health Affairs found that adults who typically use a family or general practice physician spent $1,201 fewer than adults for whom that source was a general internist, with no significant differences in health status or patient satisfaction. This raised the question, according to study coauthor George E. Fryer Jr., Ph.D., M.S.W., of the University of Arkansas for Medical Sciences College of Medicine, of whether differences in medical education methods might play some role in variation.

"I think family medicine training tends to be a little less hospital-based," Fryer said. "Internists may be quicker to make referrals or be more procedure-oriented. That's not a bad thing, but it is a style difference that can act upon cost."

Lead study author Robert L. Phillips Jr., M.D., M.S.P.H., director of the Robert Graham Center of the American Academy of Family Physicians, said incentives to provide primary care services could potentially help.

"It is about helping primary care physicians behave like primary care physicians," Phillips said.

—By Scott Harris