The ubiquitous pilot program

September 10, 2010

The Patient Protection and Affordable Care Act (PPACA, ObamaCare) purports to control healthcare spending. Is starting a pilot program the answer?

How do you know when the problem solvers have hit rock bottom? How do you know when they have exhausted every possible idea in the search for a solution to a nagging problem and come up empty? How do you know when they’re ready to throw in the towel and grasp at any last desperate hope - no matter how fleeting, no matter how far-fetched?

Answer:  They start a pilot program.

This is about the only idea for controlling costs in the Patient Protection and Affordable Care Act (PPACA, ObamaCare), and before I heap on it all the scorn and opprobrium it so richly deserves, let me say that on rare occasions pilot programs can generate very useful insights. Two examples come to mind.

Did you know that before there was a modern welfare state, the federal government actually did a test run? Called the Seattle-Denver Income Maintenance Experiment (SIME/DIME), they guaranteed the participants a minimum income, regardless of their wages or marital status. The results were stunning - at least stunning to the research community. Once couples realized they didn’t have to stay together for economic reasons, large numbers of them split up. Think about that. Before the welfare state created all the tragedy and the social pathology and dysfunctionality that we are living with today, federal government planners knew in advance what was going to happen!

The other interesting multimillion dollar experiment was conducted by theRAND Corporation. This project created another notable result. People with high deductible insurance (about $2,500 at today’s prices) spent about one-third less on health care without any adverse affects on their health. Think about that. Almost thirty years before

Medicare Part D created first-dollar coverage for drugs and PPACA created first-dollar coverage for preventive care we knew that first-dollar coverage for anything in health care creates huge amounts of waste!

Both these experiments reveal two important things about pilot programs: (1) it is possible to spend an enormous amount of money to learn something ordinary people with an ounce of common sense (people without PhDs) probably could have told us anyway and (2) no matter what we discover, politicians are likely to do what is politically expedient for them to do in any event.

Returning to the problem at hand, the pilot programs called for under PPACA have an explicit purpose: to discover ways of lowering the cost of care without reducing quality. Yet even if a project turns out to do just that, the experiment is of no value unless it can be replicated. As explained in a previous Health Alert, this implicitly assumes the engineering model will work.  That is, it assumes experimenters can discover the best way to practice medicine and that regulators can then force doctors all over the country to copy the model.

Will that work? It’s worth remembering that we have been trying this model in education for a quarter of a century with no success whatsoever.

Meanwhile, we have hundreds of natural experiments that have cropped up all over the country without the federal government spending a dime. These are the islands of excellence - examples of low cost, high-quality care that have emerged despite being penalized for doing so under the current third-party payment system. If the Dartmouth researchers are to be believed:

• If everyone in America went to the Mayo Clinic for his care, we could lower the nation’s health care bill by one-fourth.

• If everyone in America went to Intermountain Healthcare for her care, we could lower national health care costs by one-third.
Trouble is, not everyone can go to these two places. Of course, if every other medical establishment practiced the way doctors at Mayo and Intermountain practice, we could achieve the result that way. But, as Hamlet said, there’s the rub. We don’t know how to replicate what doctors do at the centers of excellence. We can’t even write it down on paper.

We have previously reported on a research project by Atul Gawande, Donald Berwick, Elliott Fischer and Mark McClellan that identified 10 hospital referral regions (HRRs) as health care islands of excellence. What did they learn from this exercise? Not very much. For example:

• Despite the conventional wisdom that ideal medicine requires salaried doctors, only two follow the Mayo Clinic in this respect.

• Two others pay on a traditional fee-for-service basis; and the rest have mixed-payment schemes.

• Despite the conventional wisdom that a greater ratio of primary care physicians to specialists is essential, the regions are all over the map in this regard as well.

• One is twice the national average; two are below it; and the others ranged from 14% to 52% above the national average.
So how do we get everyone else to practice medicine as successfully as these 10? As summarized by Gawande, the top performers have these characteristics:
1. Leadership.
2. Altering financial incentives.
3. Using measurement to provide a force for restraint.
4. Engaging with the community to help others see “how much high costs and poor quality are harming the greater good.”

And if that’s not the recipe you were hoping for, be aware that what you just read is as specific as it gets. All this is coming from someone who explicitly endorses the “engineering” approach to