Geographic Variation in Fee-for-Service Medicare Beneficiaries’ Medical Costs Is Largely Explained by Disease Burden

By James D. Reschovsky, Jack Hadley and Patrick S. Romano
Medical Care Research and Review, May 28, 2013


Control for area differences in population health (casemix adjustment) is necessary to measure geographic variations in medical spending. Studies use various casemix adjustment methods, resulting in very different geographic variation estimates. We study casemix adjustment methodological issues and evaluate alternative approaches using claims from 1.6 million Medicare beneficiaries in 60 representative communities. Two key casemix adjustment methods—controlling for patient conditions obtained from diagnoses on claims and expenditures of those at the end of life—were evaluated. We failed to find evidence of bias in the former approach attributable to area differences in physician diagnostic patterns, as others have found, and found that the assumption underpinning the latter approach—that persons close to death are equally sick across areas—cannot be supported. Diagnosis-based approaches are more appropriate when current rather than prior year diagnoses are used. Population health likely explains more than 75% to 85% of cost variations across fixed sets of areas.


Health Differences Explain Most Geographic Variation in Medicare Costs

Center for Studying Health System Change, May 28, 2013

Wide geographic variation in Medicare costs is largely explained by health differences across communities rather than inefficient care delivery, according to a Center for Studying Health System Change (HSC) study published online today in the journal Medical Care Research and Review.

Earlier research has asserted that Medicare could reduce spending by as much as 30 percent without harming health if all providers adopted treatment patterns found in low-cost areas, but the new Medical Care Research and Review study questions how well earlier studies accounted for differences in Medicare beneficiaries’ health status.

HSC Senior Fellow James D. Reschovsky, Ph.D., along with Jack Hadley, Ph.D., of George Mason University; and Patrick Romano, M.D., M.P.H., of the University of California, Davis, Center for Healthcare Policy and Research, examined multiple ways of adjusting for patient health, finding that a fuller accounting of health status explained at least 75 percent to 85 percent of Medicare geographic cost differences between high- and low-cost areas.


Dartmouth Institute for Health Policy and Clinical Practice Responds to Reschovsky at al

By Jonathan Skinner, PhD
The Dartmouth Institute for Health Policy and Clinical Practice, May 29, 2013

There are three fundamental concerns with the Reschovsky et al (2013) geographic variations paper published online May 28, 2013 in Medical Care Research and Review.

The first and most important is that they include current HCCs – diagnostic billing codes – as “explanatory” factors for spending. … the authors’ use of the HCC billing codes would “explain” the more aggressive cardiologist’s behavior as worse health status, rather than attributing it to the more aggressive physician behavior.

But they do find that “non-discretionary” measures of disease, like hip fractures, are higher in high-cost regions.  Wouldn’t that lend credence to their claim that poor health explains higher spending?   I believe the answer is still no – because we cannot reproduce even those basic findings using population based Medicare claims data.

The final concern with this study is a more technical one, so bear with me.  Typically, HCC risk adjustment is predetermined by the HCC formula designed by CMS.  If someone has diagnosed COPD, Medicare would implicitly allow (say) 23% more expenditures as a consequence.  As far as I can tell from the methods section, this is not what they do. Instead, they throw the HCC diagnostic codes into the regression, and let the regression do its best to explain the regional variation.

Now of course, my views on this paper may be viewed as tainted, given my Dartmouth affiliation.  So it’s informative to consider the interim report by the IOM-convened panel of experts on regional variations – a panel that did not include any Dartmouth faculty.  The panel commissioned many studies of regional variation in Medicare, Medicaid, and in private insurance markets.  They read every paper in the literature, and conducted their own analysis of the data.   After sorting through all this evidence, they concluded, as we have:

Observation 4
Although a non-trivial amount of geographic variation can be explained by specific demographic and, potentially, health status variables, a substantial amount of variation remains unexplained.…


Medicare Spending Variations Mostly Due To Health Differences, Study Concludes

By Jordan Rau
Kaiser Health News, May 28, 2013

The idea that uneven Medicare health care spending around the country is due to wasteful practices and overtreatment—a concept that influenced the federal health law — takes another hit in a study published Tuesday. The paper concludes that health differences around the country explain between 75 percent and 85 percent of the cost variations.

“People really are sicker in some parts of the country,” said Dr. Patrick Romano, one of the authors.

That’s a sour assessment for those hoping to wring large savings out of the health care system by making it more efficient. Some, such as President Barack Obama’s former budget director, Peter Orszag, assert that geographic variations in spending could mean that nearly a third of Medicare spending may be unnecessary.

The new paper is one of the sharpest attacks yet on the work of the Dartmouth Institute for Health Policy & Clinical Practice, whose three decades of research has popularized the theory that the unexplained regional differences in spending are due to the aggressiveness of some physicians to do more, in large part because it enriches them.

“The trouble with Dartmouth is they were trying to spin a simple story from a world which is far more complex and far more nuanced,” said James Reschovsky, the lead author on the paper. “They are to be credited for highlighting that there’s a lot of inefficiency in the delivery of health care in the United States. They defaulted by hanging their hat on geographic determinants of efficiency, and I think that premise is fundamentally being torn down, not only by my research, but also by the IOM work and a bunch of other studies.”

Dartmouth strongly disputed the study, which they said they could not replicate. Jonathan Skinner, a Dartmouth economist, called the study “fatally flawed” in an email. He noted that the Institute of Medicine’s preliminary report stated that “although a non-trivial amount of geographic variation can be explained by specific demographic and, potentially, health status variables, a substantial amount of variation remains unexplained.”

Skinner also noted that Dartmouth researchers have never endorsed the idea of adjusting Medicare spending by region.

A 2008 “white paper” from Dartmouth directed at policy makers and titled “an agenda for change” implied the possibility of substantial savings if Medicare rooted out inefficiency and unnecessary treatments.

Dartmouth researchers have long maintained that experts shouldn’t trust Medicare diagnosis records, which the new study used for its analysis. They believe that doctors in some areas of the country are more aggressive in diagnosing people with serious ailments than doctors elsewhere.

Still, while faulting Dartmouth’s methods, the new paper left between 15 percent and 25 percent of geographic differences unexplained. And Romano said he agrees there are substantial differences in how medicine is practiced. “I really do believe there is huge practice variation, but I don’t think we see that variation at the level of these large geographic units,” he said. “We see those variations at the level of individual physician practices.”…

It has been well documented that there is considerable variation in Medicare spending both between and within geographical areas. What has been disputed is whether or not these differences are due to population health. This latest study indicates that population health explains about four-fifths of the differences in spending. If that is true, the differences in spending are largely warranted.

If the differences in spending are due to geography and physician behavior and not to how sick the patient is, then the onus of explaining how the excess spending could be precisely identified and recovered should be borne by those supporting this view.

Jonathan Skinner of the Dartmouth Institute has been a leading advocate of using accountable care organizations to ferret out this waste. It is difficult to conceptualize how providing care and making the providers “accountable” for it leads to recovery of this alleged waste, especially since accountability measures are so primitive and misdirected that they could never identify most of the excesses.

What is tragic is that this debate has distracted us from seriously considering changes that are well documented to be capable of recovering waste while providing greater value in our health care spending – changes that we could easily achieve by enacting and implementing a single payer national health program.

If there are crooks out there who are wasting our health care dollars, we can investigate them as outliers and then take appropriate measures as warranted. It would be much easier to do under a single payer system.