Wednesday, July 29, 2009

Orzag's Bad Idea - Ending Practice Variation

Prelude: The Obama team, and its main Budget man, Peter Orzag, are fixated, indeed, obsessed, with the Dartmouth Group’s idea that if you could reduce Medicare spending in the high spending regions to the level of the low spending regions, you could save the system 30% or $700 billion, which is not chicken change. The problem is, is you examine the matter closely, it doesn’t make any sense- given the mix of the rich and the poor, the vast differences in socio-economic status, and the supply-demand diequilibrium in the various regions.

Here John Goodman, the economist who founded the National Center of Policy Analysis, has his go at destroying the Orzag thesis.

In the course of his analysis, Goodman observes HR3200, the House Bill, uses the words "require," "report," "limit," "penalty," and "regulation" a total of 1335 times and the words "competition" and markets" only 6 times. This word use reflects a 99.6% bureaucratically locked mindset. It leaves little room for freedom of doctors to act or innovate.

I'm in favor of a Medicare Czar with complete authority to form contracts with providers. And no ability of Congress to override the decisions. But as I explained at the National Journal blog on Monday, I'm 100% against this idea if Office of Management and Budget Director Peter Orszag is the Czar.

I like Peter. He's smart. He's capable. He's competent. But his idea on how to reform Medicare is completely wrong. As the CBO has affirmed, it will save no money. Zip. Zero. Nada. And it may harm patients.

The Claim

But first things first.The Claim. Researchers at Dartmouth have documented wide disparities in Medicare spending that appear to have no impact on health outcomes. Further, if doctors and hospitals across the country practiced medicine as efficiently as the lowest spending regions, it appears that as much as one-third of Medicare's outlays could be saved. Orszag has generalized these findings for the health care system as a whole and concluded that national health care spending could be potentially reduced by $700 billion a year.

The Evidence

NCPA Senior Fellows Andrew J. Rettenmaier and Thomas R. Saving (also a former Medicare Trustee) have examined data from the Centers for Medicaid and Medicare Services in a Brief Analysis the National Center for Policy Analysis is releasing today [full study here]. Their initial results confirm the Dartmouth findings:

• Medicare spending per enrollee for residents in the five highest-spending states is about 40 percent higher than spending in the five lowest-spending states.

• Further, if every state were as "efficient" as the fifth lowest-spending state (that is, achieved the 10th percentile), Medicare spending could be reduced by about 25 percent.

However, when Rettenmaier and Saving looked at total health care spending they found that high Medicare spending often occurs in places where private sector spending is low, and vice versa. For example:

• Although Texas is fifth from the top in Medicare spending per capita, it is seventh from the bottom in per capita spending for the state's population as a whole.

• California is 11th from the top in Medicare spending, but eighth from the bottom in spending overall.

• North Dakota is seventh from the bottom for Medicare, but 11th from the top in overall spending.

These results are consistent with the idea that cost shifting between the public and private payers is at play, although there may be other explanations as well.
There are also other factors that affect differences in spending, including income, age, race and various characteristics of state economies. After adjusting for these factors, the authors ask and answer this question:

Suppose that we could transform the practice of medicine all across the country and induce doctors everywhere to practice medicine the same way it is practiced in the fifth lowest-spending state (10th percentile). Based on the experience over a 15 year period, the potential savings across all populations is only about 5 percent!
Bottom line: After all adjustments are made, the potential savings are quite meager. They are probably smaller than what purchasers would have to spend to try to realize them.

Case Study: McAllen, Texas. An article in The New Yorker by Atul Gawande singled out this Texas border city as one of the worst examples in the country of unjustifiably high Medicare spending. Yet as Greg Scandlen has previously pointed out at this blog, McAllen is a poor city with many health problems and very little private insurance. So Medicare is the only reliable source of much needed funds. Like much of the rest of Texas, it appears that McAllen is overbilling Medicare in order to subsidize the lack of funds from other sources. Overall, McAllen is not a high spending area at all. The city is much more of an example of cost shifting rather than inefficient medicine.

Islands of Excellence. None of this suggests that there is no fundamental problem. With consumers paying out-of-pocket no more than 13 cents out of each dollar they spend, it would be impossible not to have an enormous amount of waste and inefficiency in health care. And, as is the case with education, there appears to be a sea of mediocrity punctuated by islands of excellence — distributed here and there almost randomly. Dartmouth research suggests that if everyone went to the Mayo Clinic for his health care, the nation would reduce its health spending by one-fourth and quality of care would go up. If everyone went to Intermountain Healthcare in Salt Lake City, the nation would reduce its spending by one-third and quality would improve in the process.

Problem is, not everyone can go to Mayo. Nor do we know how to replicate Mayo. Even Mayo doesn't know how to replicate Mayo. And if we try to use the power of the purse to force other institutions to operate like Mayo, we could cause a lot of harm.
Supply-side versus Demand-side Changes. Everything that Peter Orszag is talking about and everything that is incorporated in legislation before Congress takes a demand-side approach. It seeks to use the purchasing power of the federal government to force doctors to change how they practice medicine. The National Taxpayers Union reports that:

• The House bill has the word "require" and its derivatives 494 times.

• It uses "report" 427 times, "limit" 167 times, "penalty" 156, "regulation" 91.

• By contrast, the words "marketplace" and "competition" are used 3 times each.

As I have argued here and here, what is needed is a supply-side approach — one that liberates doctors rather than seeking to order, manipulate and control them.
Lessons: There are many examples in our health care system of high-quality, low-cost care. They all originated on the supply side of the market. Not one of them was created by Blue Cross, Medicare, or any other payer.

Why can't we learn from this experience? Let the providers tell Medicare how to reform the system rather than the other way around. As proposed on my recent testimony, every provider should be encouraged to repackage and reprice his services, provided that (1) the cost to the government does not go up and (2) quality of care does not go down.

I'm in favor of a Czar who is willing to liberate the supply side of the market. I'm not in favor of a Czar who wants to tell doctors how to practice medicine.

No comments: