Doctor (Image: Thinkstock)

Health care costs too much for what we get in return. Is it more reasonable to hope that doctors will curb unnecessary spending, or consumers? I have long believed that while both are useful, our primary focus should be on influencing what doctors recommend.

Recent evidence shows that should indeed be where we put most of the emphasis. Consumer-driven health care may have some benefits, but it doesn’t come close to a doctor-driven approach.

The motivation for focusing on doctors is that most health care is, in the end, what the doctor orders. As I argued seven years ago, “in high-cost and chronic cases, which account for the bulk of overall costs, the patient typically agrees to the care recommended by the provider — so that the provider’s recommendation is most often the care that winds up being delivered. In the end, therefore, fundamentally reducing health care costs requires that providers alter their recommendations.” At the time, that view was quite controversial and hopes were high for a consumer-driven alternative, in which patients would seek out higher-value care.

(Related: Out-of-Pocket Health Costs Aren’t That High)

A fascinating new study suggests the limits of such consumerism in health care. Economists Michael Chernew of Harvard Medical School, Zack Cooper of Yale University, Eugene Larsen-Hallock of Columbia University and Fiona Scott Morton of Yale study the market for lower-limb magnetic resonance imaging (MRI) scans. Such scans are often used to examine fractures, arthritis and other problems affecting the hip, knee, leg or foot.

MRI scans would appear to be an almost ideal setting for consumer shopping: their prices are high and vary substantially, the scan itself doesn’t vary much from provider to provider, and it can usually be scheduled in advance. Using data from a large insurance company, the economists limited their analysis to MRIs that were not associated with emergency care or an inpatient experience (to maximize the applicability of a consumer-driven framework) and those that were ordered by orthopedic surgeons (who would typically review the results themselves, so that the results are not contaminated by differential quality of reviewing radiologists at different providers).

The motivation for shopping under these conditions should be particularly strong because the researchers document significant out-of-pocket costs: more than a fifth of patients getting a scan paid for the whole cost themselves, and another 54% faced some cost sharing. The prices charged, furthermore, average $850 and vary widely: It is not uncommon to find some providers charging more than twice as much as others in the same area, so that substantial savings would be possible. Finally, there seems to be little difference in the quality of the scan (as measured, for example, by whether the scan needs to be repeated), so consumers could basically just shop for the lowest-cost provider.

And yet despite all these auspicious factors for the wonders of consumer shopping, the team found shockingly little of it. Under 15% of people went to the lowest-cost MRI provider within 30 minutes’ drive of their home, and on average patients passed a half-dozen lower-priced providers between their homes and where they got their scan. Total spending on the MRIs would be cut by more than a third if each patient chose the lowest-priced provider within the same drive time as where they actually went; as part of that total reduction, out-of-pocket spending would be cut by more than a quarter.

So if patients are not actively shopping for the MRIs, what determines where they are done? The answer appears to be the doctor. On average, each referring orthopedic surgeon sends almost 80% of patients to the same MRI provider. The economists conclude: “Because of the weight patients appear to place on the advice of their referring physicians, it is unlikely that a significant number of patients will use information from an app or from a pricing webpage, or to diverge from where their physicians typically send patients for imaging studies. … It is even less likely that patients will actively price shop for more complex and differentiated services.”

If the key to the health care delivered is what the doctor orders, how do we influence those recommendations while respecting the professional knowledge of doctors and the variability of patients? Two broad approaches seem promising. One involves the ongoing shift toward paying for value rather than paying for volume (including more bundled payments for all the care associated with different procedures), so that doctors benefit when they deliver better care rather than just more care. The other involves focusing on improving clinical decision support, to narrow unwarranted variation but with an ability for doctors to override the recommended treatment courses to reflect the needs of their patients.

Let’s hope these approaches help doctors steer toward better value in health care, because expecting health care consumers to take the wheel isn’t going to work.

— Read Fast Changes in U.S. Health Careon ThinkAdvisor.

Peter Orszag

Peter R. Orszag is a Bloomberg Opinion columnist. He is a vice chairman of investment banking at Lazard. He was director of the Office of Management and Budget from 2009 to 2010, and director of the Congressional Budget Office from 2007 to 2008.

Copyright 2018 Bloomberg. All rights reserved. This material may not be published, broadcast, rewritten, or redistributed.