Stop the Pilot Programs

8 Min Read

The federal government has been spending billions of dollars on health care pilot programs and demonstration projects. President Obama has explained their purpose: “We need to find out what works and then go do it.” In other words, the exercises are supposed to discover how to lower costs and raise quality so that everyone else can copy them.

The federal government has been spending billions of dollars on health care pilot programs and demonstration projects. President Obama has explained their purpose: “We need to find out what works and then go do it.” In other words, the exercises are supposed to discover how to lower costs and raise quality so that everyone else can copy them.

healthcare reformThere are five problems. First, three Congressional Budget Office reports (see here, here and here) have found that these programs are not working. Second, even where there is evidence of success, the gains are usually too small to warrant much hope for meaningful change. Third, no matter how successful a project, it is not of much value if it cannot be replicated — which appears to be generally the case. Fourth, even in the rare instance where a pilot program is remarkably successful and there is every reason to think the results are replicable, Washington will ignore the project if it does not fit into the bureaucratic vision of how health care should be delivered. (I produce a stunning example below.)

Fifth, and this is the real killer, we don’t need pilot programs and demonstration projects in the first place. Why? Because we have hundreds of natural experiments where costs have been lowered and quality raised without any cost to the taxpayer at all. This is because of:

Goodman’s law of medical innovation: For whatever we are trying to do in medicine, there is someone, somewhere, who has found a way to do it 50% better.

For almost any kind of surgery — mastectomies, knee or hip replacements, spinal fusion, etc. — there is someone in the United States who has discovered how to cut the patient recovery time in half. Partly for that reason, there is someone who has discovered how to cut the cost in half. For infection rates, readmissions and other indicators of quality care, there is some institution, somewhere, that is chalking up rates that are half that of what the country as a whole is experiencing.

If you think we can copy excellence, don’t run a demonstration project. Just go copy what’s already working and working well.

We got to stop and
Think it over.

 

Of course, in a normal market, if someone discovered how to lower costs by half for a given level of quality or to increase quality by 50% for a given level of cost, that person would have a huge advantage over his competitors. The rivals would have to quickly discover how to emulate the innovator, lest they be priced out of the market. Only in health care, where normal market processes have been systematically suppressed for decades, can widely different levels of efficiency coexist, side by side, for year after year.

How do I know that Goodman’s law is true? Because I meet people every day who appear to affirm it. Many of them have been profiled at this blog. We have posted before about American Physician Housecalls, which appears to cut the cost of care in half for chronically ill Medicare patients. To my knowledge, Health and Human Services has made no effort whatsoever to even investigate this successful venture. (They’re not ACO? Forget it!) And then there is Jeffery Brenner, the “hotspots” doctor who is saving millions of dollars for Medicare and Medicaid, and getting nothing in return. The problem: Brenner is changing patient behavior mainly through “social work” and Medicare doesn’t pay for social work. (Besides, he’s not an ACO either!)

Here’s another fascinating example (sent to me by Dr. Brenner). Health Quality Partners, in Doylestown, PA, participated in a 10-year Medicare chronic care demonstration project, with a 1,700 patient randomized control trial run by Mathematica. They used a nurse outreach model to visit the homes of frail, elderly Medicare patients. As explained by Dr. Brenner:

The project showed a 25% sustained reduction in death rates, reduced cost, and reduced hospitalization (33%). It’s really groundbreaking [with] stunning results. There are no pills or treatments that come close to these kinds of results. For the frailest patients the death rate dropped 50% and the results were sustained. They essentially discovered the fountain of youth.

So what is the federal government doing with this information? CMS is about to end the demonstration project with no plans to replicate the results.

So, back to the title of this post. There are really three questions to be asked:

  1. Are there techniques that will substantially reduce cost and/or improve quality?
  2. Will entrepreneurs discover them and implement them, given market incentives?
  3. Can a centralized planner manipulate doctor behavior in this regard, with small penalties and rewards?

The answer to the first question is clearly “yes.” The answer to the second is also “yes,” but let’s be clear about what that means. If an entrepreneur saves a million dollars for the system, he is going to want to bank, say, $500K for himself. No one is going to take big risks for a normal rate of return. Right now, no one is out there trying to copy what Dr. Brenner is doing in Camden, New Jersey, or what American Physician Housecalls is doing in Dallas, Texas, because they can’t capture a significant part of the huge value such activities create for the system.

To solve our problems with entrepreneurship, we have to free the market and let entrepreneurship bloom.

That brings us to the third question, which I think is the real reason why so much money is being spent on pilot programs when we could learn so much more without spending any money at all.

It turns out that the answer to that question is apparently “no.” Big bureaucracies cannot manipulate doctor behavior with small penalties and rewards.

Darn. The world would be so much simpler if only socialism worked.

Share This Article
Exit mobile version