Thoughts on Regional Variation

The Office of Management and Budget's Peter Orszag tells us that the growing costs of our ailing health care system present the greatest long-term threat to the U.S. budget deficit. In the context of on-going economic, energy, and foreign affairs crises, such a claim should be loud enough to wake up the political sideliners and health reform obstructionists who assume or hope our country's health care crisis will somehow fix itself.

For those afraid that cost control necessitates government suffocation of technology innovation and "rationing" of valuable care, today's Health Populi post highlights a recent NEJM article (Fisher et al., 2009;NEJM 360(9):849-852) that promotes a more optimistic view of the road to affordable, universal health care. The study compares vastly different growth rates in per capita Medicare expenditures among U.S. cities with, presumably, equal access to medical technology. Instead of blaming technology for increased cost growth from 1992 to 2006 in cities like Miami, FL (5.0%) and East Long Island, NY (4.0%) versus that in Salem, OR (2.3%), the authors found that regional differences were highly correlated with differences in physician propensity to recommend discretionary services in clinical "gray areas" (e.g., subspecialist referrals for common, low-risk symptoms such as typical gastroesophageal reflux). Because increased expenditure growth from city to city fails to correlate with any measure of improved health, the take-home message of the report is that comparative studies between high-cost, high-growth regions and low-cost, low-growth regions can influence policy that encourages the former to behave more like the latter, without reasonable fear of a subsequent fall in care quality and health.

The regional differences in health care cost efficiency have been noted before - see, for example, 2006 papers by David Cutler ("Making Sense of Medical Technology") and Jonathan Skinner et al. ("Is Technological Change in Medicine Always Worth It?"). However, the important data uncovered by the recent NEJM article indicate that "technology" and "payment systems" are insufficient explanations of regional spending differences, as all populations in the study have access to the same technologies, and are all in the fee-for-service system). What the study does not analyze is correlation of regional spending with other factors, such as prevalence of medical malpractice reports, physician work-load, patient education, or other information that might provide explanations for regional differences in physician behavior. And with these and other unknown cost-of-care determinants comes the necessity of government help.

Individual physician leadership in medical practices, communities, and the broader health care system will go a long way in improving the efficiency of care, but without data-driven guidelines based on comparative cost-efficacy measurements, there can be no unified progress toward improved, affordable care. Physicians and policy makers can work together to assure the right information is collected and properly acted upon. But, ultimately, if we wish to provide sustained, quality, universal care, we need fundamental and comprehensive payment reform through which payers and providers are accountable for costs, but incentivized to keep more patients healthy.