I’ve been meaning to reply to Trent McBride and Tyler Cowen (see here and here) on my doubts about the effects of medicine on health.

McBride poses three challenges:

Challenge #1:

Don’t you have to recognize the non-mortality benefits of medical care? Medicine doesn’t always make us live longer, but it can make us live better. Preventing a lot of non-fatal heart attack may not change average mortality that much, but it sure makes life quality better… When you actually sit down and separate all the fields of medicine and health care, very few sectors are actually even aiming for improved quantity of life.

Great question. I focus on the link between medicine and life expectancy because it’s so widely discussed, but obviously quality of life is important too. There is admittedly less research on this dimension, but what exists reinforces my skepticism. The canonical RAND corporation experiment looked at many measures of health and found little effect on the quality of life as well as quantity. Hanson summarizes:

Those with free care consumed on average about 25-30% more health care, as measured by spending, obtained more eyeglasses, and had more teeth filled. They had more appropriate medical visits, and as a result suffered one more restricted activity day per year. Beyond this, there was no significant difference in mortality, a general health index, physical functioning, physiologic measures, health practices, satisfaction, or the appropriateness of therapy. Blood pressure may have been reduced, but the point estimate was that this produced a 1% reduction in future mortality rates, which translates to roughly seven weeks of life.

Again, this is the effect of the last 25-30% of medical dollars spent. But it is particularly striking that expert judgments’ of the “appropriateness of therapy” were not lower for these marginal treatments, suggesting that the marginal and the average may be more similar than we would expect.

Challenge #2:

[W]hat about gains in real mortality that could be eaten by a shift in choices? Steve Landsburg argues that the development of cholesterol-lowering drugs has helped usher in the increase in obese and overweight individuals. He says that by lowering the price of being fat, you get an increase in fat people (and any increase in mortality that come from obesity). It makes sense to me. The measured benefit could either be a decrease in mortality or an increase in unhealthy eating with unchanged mortality or anywhere in between. (It’s like the classis economic problem with seat belt laws).

This is a good point. I’m not aware of any data that shows this is wrong. It seems to me, though, that in most respects (obesity is a good counter-example), people today live more cautiously than they used to, not less, so it’s not clear where we are dissipating our gains.

Challenge #3:

And finally, with relation to the expensive treatments above that offer little measured marginal benefit in mortality – don’t you have to count some of the potential benefits down the road that will come of today’s low benefit treatments?

This seems like double-counting to me. After all, previous studies should have picked up the long-term benefits of earlier treatments. Studies in the 70’s should have picked up the benefits of treatments introduced in the 50’s, for example. Your story only works if the lag between the discovery of treatments and their useful application is increasing. Is it?

Now on to Tyler’s “bald asseverations.” He says:

4. It is true that many regressions show a zero positive effect for health care once you introduce a variable for income. This mainly shows that income is a better proxy for real health care than many of our highly imperfect measures for health care.

It sounds like Tyler is changing the meaning of “health care” so that it explains all health improvements by definition. I don’t like to fight about words, but this is pretty misleading. Suppose the government says it wants to raise taxes to spend more on “health care.” A plain language response based on the data is “Since income improves health a lot more than health care, it won’t work.” In Tyler-speak, we have to say “You’re actually cutting spending on health care.” That’s awfully confusing.

5. Almost every family in my Mexican village has lost a kid or two before the kid reaches age five. Few of these deaths would have occurred if a) a doctor rather than a shaman were around, b) they had a ready antidote for scorpion bites, c) they knew to take the right pills for diarrhea and fever and to stay hydrated. These variables will be more closely correlated with measured income than with whatever screwy figure the Mexican government provides for expenditures on medical care. Health care still matters, even though it won’t show up as significant in the regression.

I can believe this. But again this basically redefines “health care” to equal “the cheap health care that actually works.” And in terms of policy advice to the Mexican government, the upshot seems to be cut its “screwy figure” and refund the difference to raise after-tax incomes. This makes slogans like “Health care is very, very good for you” deeply misleading.

7. It is obvious that health care leads to greater longevity, and this is the greatest good of all. Just ask yourself, how much money would you have to receive to give up health care for the rest of your life? For me no sum of money would suffice.

This is a tough question for me, because I also don’t think money buys much happiness. The few benefits I get from medicine – like my allergy pills – are still worth at least a $1,000,000 to me (that’s willingness to accept, not willingness to pay!). But anyone who doesn’t share my lack of interest in more material possessions should set a much lower amount.

Tyler has a second post that cites new evidence on the link between IQ and longevity:

High IQ is correlated with compliance with doctors’ instructions, good choice of doctor, adequate medical attention, and so on.

Maybe he’s right. But there are lots of other explanations. The simplest is genetics. Across species, smarter animals live longer, and no one claims it’s because whales get better medical care than plankton – or pay more attention to their vet’s instructions. Why shouldn’t the same hold within species?

In any case, the study Tyler cites emphatically does not show that “High IQ thus appears to be doing some of the work that should be credited to health care, as properly defined and measured.” The study appears to show the opposite. If Tyler’s interpretation is 100% right, then, contrary to appearances, the study doesn’t show this after all, and he’s back at square one.

Last gasp: A new study once again confirms that annual check-ups are a waste of time. But that’s not stopping doctors from recommending them:

[M]any among the 783 doctors queried said routine exams should include tests that the task force says haven’t been proven to prevent disease in healthy adults, including urine tests, blood-sugar tests for diabetes, and thyroid tests.

The most frequently recommended tests included complete blood counts, or CBCs, which check for conditions including anemia. Nearly 40 percent of doctors said those tests should be part of routine physicals, despite studies showing “that there is very little yield” from routine CBC testing, the researchers said.

How do doctors rationalize their position? One Dr. Goyal sounds like a Hansonian caricature:

Still, yearly physicals have value even for healthy patients with no risk factors, he said.

“It’s an opportunity to get to know your patients” and make sure they’re adopting lifestyle habits that will keep them healthy, he said.

It other words, it’s a chance to show patients you care about them, and nag them for their own good. You really need a medical degree to do that, don’t you?!