Tuesday, July 7, 2009

Limits of intervention - Obama, McNamara, and Limits of Data and Systems Engineering

Robert McNamara, who died yesterday at 93, will go down as a cautionary tale for the ages, and perhaps none more than for the Age of Obama. Whatever else distinguishes JFK's New Frontier or LBJ's Great Society from Barack Obama's "New Foundation," this too is an era of soaring rhetoric, big plans and boundless self-regard, issued by an administration convinced it can apply technocratic, top-down solutions to huge and unpredictable systems -- the banking, auto and health-care industries, for instance, or the climate. These are people deeply impressed by their own smarts, the ones for whom the phrase "the best and the brightest" has been scrubbed of its intended irony.

Bret Stephens, From McNamara to Obama, Wall Street Journal, July 7, 2009

If health care reform falls apart again in Congress, the most likely cause will be failure to agree on how to subsidize coverage for tens of millions of uninsured Americans. The cost will almost certainly be at least $1 trillion over the next decade and perhaps much more, depending on how generous the reform might be.

Our preference would be to extract savings from the bloated, inefficient health care system — but also to raise revenues from a wider pool, preferably from well-to-do Americans who could be taxed more for a badly needed reform that would benefit all Americans.

The first task is to find savings. Some respected analysts suggest that as much as 30 percent of all health care spending in this country — some $700 billion a year — may be wasted on tests and treatments that do not improve the health of the recipients. If even half that money could be recaptured, the amount saved would be more than enough to finance health care reforms.

Overspending, however, permeates the system and would be devilishly difficult to eliminate in any systematic manner that reaped savings within the decade. Most of the truly “game-changing” innovations that could slow the rate of increase in health care costs — electronic medical records, research comparing the effectiveness of treatments, restructuring the way doctors organize themselves — will take years to affect costs and quality.

Editorial, New York Times, Financing Health Reform, July 6

When the New York Times editorializes “If the health reform effort fails apart again in Congress,” it is a sign that the liberal elite is worried the Obama health proposal may be in trouble. But why should this be? Not enough money? Not enough savings? Too much overtreatment?

I have another explanation - too much reliance on “technocratic, top-down solutions to huge and unpredictable systems,” on data and systems engineering to fix the system and too little empathy with human expectations of the health system. The methods suggested by the Obama administration – electronic records, research based on comparative effectiveness, money “wasted” on practice variations in different regions in the country, doctors organized in large groups with enough volume to generate data to track continuous improvement – will not “save” or “fix” the system. Nor will competition based on wide population –based outcome data to reward “efficient” providers.

Instead we should rely on the good sense and the cost-consciousness of patients and doctors on the ground. This will require health consumers to spend part of their own money. It will require doctors and hospitals competing on the basis of price and publicly available data. It will require patients asking, ”How much is this going to cost? How many of these have you done, and what are your results? “ It will require catastrophic insurance to make sure no one goes bankrupt. It will require subsidies for those who cannot pay.

But it also requires common sense about the ingrained beliefs of Americans. They distrust big government, are leery of runaway federal debt, believe deeply in individualism, in equal opportunities for all but not necessarily equal outcomes, and access to the best care and technologies America has to offer. Recent polls indicate Americans consider themselves “conservative” rather than “liberal” by a ratio of two to one.

I’m not sure what this means for universal coverage, except that it likely to come slowly rather than all at once. In my book, Obama, Doctors, and Health Reform, I say there are four obstacle to sweeping heath reform.

President Obama faces four major reform obstacles. I call them the four “Cs.”

• Culture, American style, abhors the word “rationing.” Our health care culture cherishes unlimited choice, quick access to the latest and best in medical “cures,” and proven lifestyle restoring technologies. These traits conflict with a centralized, command-and-control, federal expansion of health care.

• Complexities, American health care is a whirling Rubik’s Cube, with millions of interrelated moving parts, institutions, and people, each with agendas, axes to grind, and oxen to gore. Though they may help to a limited extent, data-driven care and systems engineering will not resolve all these complexities. Resolution will require knowledge of the human condition and individual circumstances.


• Costs, Obama says prevention, electronic medical records, and paying only for what works, as established through comparative research, will save billions of dollars, yet scant evidence exists that these measures work. Proposed savings remain hypothetical.

• Consequences, of curtailing health costs, may be worse than the cure, because health care institutions and private practices in many communities are the biggest and fastest growing employer in town. Collectively, health care profoundly impacts most communities’ economies . Health care’s building blocks can’t be downsized quickly or dramatically.

It may be that President Obama and his team can overcome these obstacles through top-down planning, data dissection , and cool intellectual certainty. But President Obama might take a close look at McNamara's obituaries and note that McNamara was a whiz kid with numbers who overlooked what numbers could not tell him.

No comments: