In the ongoing struggle over wasteful health care spending, geographic cost variation is a veteran hot topic. The fact that public programs and private health plans spend more for providers in some regions, even after accounting for normal differences in costs by location, has irked policymakers and researchers who see the discrepancy as evidence of health care misuse and overuse. Yet a new Institute of Medicine report argues against Medicare payment reforms based on geography, saying other factors instead could be to blame for unnecessary health spending.
Geographic Differences Long Seen as a Source of Waste:
Policy experts for many years have recognized that health costs and spending vary with geography. Some variation is to be expected: Most businesses selling goods and services in New York City have higher prices than their counterparts in smaller cities. Physicians, hospitals, nursing homes, pharmacies, and other providers are no exception.
But health care costs vary even after accounting for differences in business expenses across regions, which has left many policymakers to suspect that overuse and misuse of health care services are behind the geographic health cost variations.
Medicare and Medicaid account for more than a third of national health expenditures, so unexplained cost differences add up to potentially large amounts of wasteful spending and deserve scrutiny. The Centers for Medicare and Medicaid Services (CMS) has gathered a lot of data on Medicare geographic cost variations, which you can explore here. In 2012, Congress asked the Medicare Payment Advisory Commission (MedPAC) to look into Medicare physician payment adjustments by region, which you can read about here. And the Government Accountability Office (GAO) recently recommended a series of revisions to the Federal Medical Assistance Percentages (FMAP) rate formula for Medicaid, including a factor to account for geographic health cost differences.
IOM Report Finds Geographic Variations Exist but are Not the Whole Story:
Not all experts are convinced that health plan payment policies should adjust for geography. Factors other than simply where a provider is located are likely to be associated with higher health costs and spending. A policy that reduced provider reimbursement in high-cost regions, regardless of outcomes or other measures, would punish providers who were saints among sinners.
Despite all of the research on geographic health cost variations to date, more was needed. Shortly before the Affordable Care Act (ACA) became law, members of the U.S. House of Representatives from Midwestern and Northwestern states formed the Quality Care Coalition to push for Medicare payment reforms tied to geographic health cost disparities. In response to the coalition’s requests, Secretary Kathleen Sebelius pledged to commission studies and recommendations on the issue from the Institute of Medicine (IOM).
Removing variation in post-acute care spending would reduce Medicare spending discrepancies across regions by 73 percent.
But the IOM included important caveats to its findings. Its statistical model could not explain a large amount of geographic health spending differences. The model included variables for health status, factors related to market differences, and health plans.
Yet there are likely many other variables – such as patient preferences or provider decisions – that were not in the model but could explain cost differences by region. That is a common problem of inferential statistics, called omitted variable bias. In this case, it means there could be factors as yet unaccounted for that are creating the appearance of a relationship between region and health spending where none exists. In other words, something other than location could be the culprit.
IOM Recommends Against Geographically Based Payment Reforms:
Based on its findings, the IOM concluded that a geographically based Medicare payment index would be ill-advised. Its overarching reason for saying so was that significant variations in spending and health care use exist among providers within regions, and even within smaller geographic units.
Specifically, the IOM found that:
physician behavior can differ among doctors in the same physician practice.
Rather than focus on geographically based payment reforms, the IOM said, CMS should continue to promote its payment reform innovation models. Most of that work has come from the Center for Medicare and Medicaid Innovation (CMMI), which is promoting bundled payments, patient-centered medical homes, and accountable care organizations (ACO) to reduce cost and increase quality.
Response by Dartmouth Experts:
In a thoughtful post on the Health Affairs blog, Elliott S. Fisher, MD, MPH and Jonathan Skinner, PhD, both of the Dartmouth Institute for Health Policy and Clinical Practice, critique the IOM report and its limitations. They agree with the IOM’s continued call for improved financial incentives for health care providers. However, they remind readers “both health and health care are local” and criticize the IOM committee for “turn(ing) its back on regional initiatives to improve both health and health care.”