The Promise of Real Healthcare

Next in my series of optimistic posts: healthcare. There’s been a lot of bad news lately about healthcare. A new study just came out showing that life expectancy at birth has been declining in some counties since the 1980s. In particular “after 1983, life expectancy declined an average of 1.3 years in 11 counties for men, and in 180 counties for women” (from the NYT article) At the same time, health care expenditures continue to rise. For example, the latest numbers from the OMB show that in FY 2007, health care exceeded 25% of the federal budget for the first time, making it bigger than defense spending.

In this post, I’m going to argue that the combination of poor outcomes and rising costs is due to a lack of medical knowledge, not to a failure of institutional arrangements. To put it another way, doctors know far less about what works and what doesn’t than we think.

I choose to interpret this as good news. The implication is that improvements in medical knowledge have the potential to both improve outcomes and reduce costs. In particular, the maturing of biotechnology—now 25 years old—offers great possibilities of escaping the health care trip.

Why do I say that we have a lack of medical knowledge? Just look at the numbers. We’ve spent trillions of dollars on researching and treating diseases of the middle-aged and elderly—heart disease and cancer, in particular. Yet life expectancy for the middled-aged and elderly have moved far less than one would expect. In 1950 your life expectancy at age 50 was 24.4 years—that is, you could expect to live until 74.4 Today (actually as of 2005) your life expectancy at age 50 is now 30.9 years. In more than a half century, and endless medical interventions, we’ve gained a grand total of 6.5 years.

The situation is even worse if you are 75. In 1950 you could expect to live to 83, and now that’s up to 87. Whoop-de-doo.

This lack of progress is not a factor of sex or race, with one exception. Take a look at the table below.

Increase in life-expectancy, 1950-2005
At age 50 At age 75
(years) years)
All 6.5 3.6
male 6.3 3.0
female 6.3 3.9
white 6.4 3.6
black 6.5 1.9
Data: National Center for Health Statistics

All the groups show up with small gains in life expectancy over that 55 year period, with blacks at 75 showing up with teeny gains.

This 6 year gain is much less than most people realize. (When I give speeches and ask people whether they think life expectancy at age 50 has increase by 6, 11, or 17 years, usually roughly half the people pick 17)

Of course, some people would argue that medical science is fine--it's just that we are fatter and fatter, and living in a less healthy environment. And certainly that could be partly right.

But more and more, it looks like some of the major medical interventions we have tried are simply not supported by the evidence. Take statins such as Lipitor, which millions of Americans take. Other signs. John Carey of BusinessWeek recently wrote about Lipitor in a cover story, and he found that the drug, by any reasonable standards, was only marginally helpful in preventing heart attacks. He wrote:

The difference credited to the drug? One fewer heart attack per 100 people. So to spare one person a heart attack, 100 people had to take Lipitor for more than three years. The other 99 got no measurable benefit.

The researchers at the Dartmouth Institute for Health Policy and Clinical Practice have been making much the same point from a different angle. They find wide variances in clinical practice across different areas, suggesting that there is no consensus about what works and what doesn't.

Grant me for the moment the proposition that doctors know a lot less than we (and they) think they do. That would explain a lot about the current situation we find ourselves in. For one, it would explain why health care costs keep rising. It is the lack of effectiveness of medical science which is driving costs, not its effectiveness. People want to live longer, they don’t want to die. They are willing to spend for even a couple of extra months of life. And if one treatment doesn't work, they try something else, and then something else (just think about your own experiences with doctors).

It also explains why all sorts of institutional reforms over the past 20 years have not managed to rein in health care costs. We've tried tight control (health maintenance organizations) and the free market (health savings accounts) and everything in between, and nothing has managed to solve the problem. But of course, the real problem has to to do with the state of medical science, and not the institutional arrangements.

So why am I optimistic? The implication is that improvements in medical science could be a win-win proposition, giving us better outcomes at less cost. That would require biotechnology to finally fulfill its promise of targeted treatments--That is, really understanding the mechanism of the medical problem and the solution, rather than by trial and error.

This breakthrough hasn't happened yet...but it's about the right time. The first biotech drug came out in 1982, a quarter century ago. It was about 25 years between the first microprocessor (Intel, 1971) and the Information Revolution of the mid 1990s.

The first sign would be a surge in the price of biotech stocks, which isn't happening right now. That weighs against my optimism. In fact, I wouldn't blame people for being skeptical of my argument.

Still, just remember there was still a lot of skepticism about the economic value of information technology right up to the moment of the Netscape IPO in 1995. Technological change is inherently low visibility.