Although the inexorable increases in healthcare costs are to blame for Medicare’s financial woes, certain parts of the country consume considerably more Medicare services than others.


For more than two decades, the Dartmouth Atlas Project has used small-area analysis to demonstrate geographic variation in Medicare expenditures. For example, in 2009, average age-, gender-, and race-adjusted Medicare Part A and Part B expenditures per beneficiary aged 65 to 99 varied by more than 60 percent, from $6,763 in North Dakota to $10,859 in Florida. That year, 10th percentile annual expenditures per Medicare beneficiary were $7,240, 50th percentile expenditures were $8,272, and average expenditures for the nation were $9,021.

Thus, although taxpayers in North Dakota and Florida are subject to the same premiums and federal payroll tax rates to support Medicare, beneficiaries in Florida consumed services that cost about 60 percent more in 2009 than did those in North Dakota. 

This finding is not an anomaly. There has been longstanding subsidy of high-consumption states by low-consumption states. Each year, during the period of 2003 through 2009, the average Medicare beneficiary who lived in North Dakota consumed $2,119 less of Part A and Part B healthcare services than did the average U.S. Medicare beneficiary. Each year,on average, an average Medicare beneficiary living in Florida consumed $999 more than the national average, and those living in New Jersey consumed $1,267 more. 

In essence, taxpayers in North Dakota subsidized health care received by Medicare beneficiaries living in Florida and New Jersey. In the aggregate, from 2003 through 2009, taxpayers in states that experienced lower-than-average Medicare expenditures subsidized $71.9 billion worth of Medicare services for patients in states that experienced higher than average Medicare expenditures. That subsidy represented a substantial proportion of total annual Medicare expenditures in a number of states: North Dakota’s subsidization represented 26 percent of average annual spending while New Jersey’s subsidy receipt represented 16 percent of average annual spending.

Just as American colonists rebelled against being asked to bear taxes on sugar and various paper products because they were not directly represented in distant Britain, North Dakotans may feel compelled to question why they pay taxes to subsidize healthcare services in distant Florida or New Jersey, when they have no direct input on the level of consumption of those services. Although revolution would be counterproductive, low-utilization states that subsidize high-utilization states should enter into honest discussions about whether Floridians and New Jerseyans, for instance, really need the additional health care that they receive. 

We already know the answer. In 2003, Elliott Fisher and colleagues found that additional Medicare consumption was not associated with patients getting more care that they need (effective care, such as reperfusion within 12 hours of admission for a heart attack or getting an annual flu shot) or that they want (preference-sensitive care, such as knee replacement surgery or back surgery).a Higher Medicare consumption was associated with obtaining more supply-sensitive care, such as tests, X-rays, physician visits, hospital admissions, and days in intensive care. Higher spending was associated with higher risk-adjusted mortality rates and lower patient satisfaction. Sadly, North Dakotans, who are known for their neighborliness, could not even feel good about helping distant communities to the South and East; their subsidy actually appears to have had an inverse impact on their neighbors’ health.



Short of revolution, representatives of low Medicare utilization states should chat with representatives of high Medicare utilization states about the error of their ways, and perhaps raise some expectations for future performance. In 2009, if high-utilization states had consumed the national average amount of care, Medicare Part A and Part B expenditures for patients aged 65 to 99 would have been about $11.6 billion, or 4.9 percent, lower than they were; if the median reimbursement was achieved, cost savings would have increased to $22.5 billion, or 9.5 percent of total Medicare Part A and Part B expenditures.

On its current course, healthcare expenditure growth will dwarf the problems represented by the impending, albeit delayed, fiscal cliff that the nation faces. Instead of temporary fixes, we need to create long-term solutions to solve a looming problem. But it might make sense to start where there is plenty of room for improvement: States that have been the most profligate in their use of federal healthcare services should be expected to make the greatest reductions. Such states could partner with North or South Dakotans or Washingtonians to learn how they are able to provide low-cost, high-quality care. But they should seek that advice soon, before low-expenditure states revolt—by demanding either that the subsidies stop or that there be adequate representation of Dakotans in Florida. Who knows: They might like the weather and bring their conservative medical consumption ways with them. 

William B. Weeks, MD, FHFMA, is professor, The Geisel School of Medicine, Hanover, N.H., and a member of HFMA’s New Hampshire-Vermont Chapter (




a. Fisher, E.S., Wennberg, D.E., and Stukel, T.A., et al., “The Implications of Regional Variations in Medicare Spending. Part 1: The Content, Quality, and Accessibility of Care,” and “Part 2: Health Outcomes and Satisfaction with Care,” Annals of Internal Medicine, Feb. 18, 2003.

Publication Date: Friday, March 01, 2013

Login Required

If you are an existing member, please log in below. Username and password are required.



Forgot User Name?
Forgot Password?

If you are not an HFMA member and would like to access portions of our content for 30 days, please fill out the following.

First Name:

Last Name:


   Become an HFMA member instead