Arnold Kling  

John Goodman on the Health Care Olympics

A Surprise OWSer... Betty Friedman Nails It...

I refer to the attempt to measure national health care efficiency by comparing outcomes to expenditures as the international health care Olympics. The conventional wisdom is that the U.S. does poorly. John Goodman disagrees. On spending, he writes,

the U.S. has fewer doctors, fewer physician visits, fewer hospital beds, fewer hospital stays and less time in the hospital than the OECD average. We're not just a little bit lower. We are among the lowest in the developed world. In fact, about the only area where we "spend" more is on technology (MRI and CT scans, for example)

His point is that we use fewer health care inputs than other countries, even though we spend more on those inputs. This is the same observation that forms the basis of the classic piece by Uwe Reinhardt and others, "It's the Prices, Stupid." However, Reinhardt argues that our health care providers earn too much. Goodman argues that health care providers in other countries have their incomes artificially repressed. Either interpretation might be reasonable.

For what it's worth, I think that we utilize our inputs more intensively. I think our doctors work longer hours and perform more procedures per hour, although I do not have data on this. I have seen data (and included it in Crisis of Abundance) showing that we have a higher ratio of specialists to ordinary doctors than other countries. This may or may not be a better allocation of resources (I suspect that it is not). But simply counting up doctors is not the right way to measure inputs.

Goodman continues,

What about outcomes? Do we get more and better care for the resources we devote? Here the evidence is mixed. As the second table shows, we replace more knees per capita than any other country and it's hard to believe that any of these are unnecessary procedures.

Not so hard to believe, actually. My late mother-in-law was told by one orthopedist that she was not a good candidate for knee replacement. She found another surgeon who was willing to do it. The outcome was more consistent with the opinion of the first orthopedist.

Goodman is on stronger ground when he writes,

What about life expectancy statistics -- a favorite of the critics, since Americans don't score very high? It turns out that when you remove outcomes doctors have almost no impact on -- death from fatal injuries (car accidents, violent crime, etc.) -- U.S. life expectancy jumps from 19th in the world to number one!

To me, the real question in the health care Olympics is how well costs and benefits are aligned. I am fairly sure that in the United States, a lot of medical procedures are undertaken that have high costs and low benefits. I have little information about other countries, but it would not surprise me if it turns out that they undertake fewer of these high-cost, low-benefit procedures.

UPDATE: Joshua D. Gottlieb apparently has some evidence on the growing use of high-cost, low-benefit procedures.

Comments and Sharing

COMMENTS (10 to date)
Ted Levy, MD writes:

Goodman is quoted as saying: "it's hard to believe that any of these are unnecessary procedures."

That's so clearly not true I wonder if it's a typo, and he meant to say "it's hard to believe that many of these are unnecessary procedures.."

CCR writes:

Of course, if you remove those dying from violent crime and car accidents you take away the possibility of their dying from natural causes, thus potentially artificially inflating the life expectancy of the nation as a whole. It's easy to believe that those likely to die from violent crime are also likely to die younger. So we shouldn't be surprised the US is number 1 in life expectancy if you remove many from the dataset that would likely die young anyway.

John Goodman writes:

Point well taken on the joint replacements. But Arnold, surely you think 85 year olds should be able to ski and golf and play tennis. How are they going to do all that without new knees?

Frank in midtown writes:

Talk about a problem that needs a supply side solution. The Hospitals control the number of beds, the Doctors control the number of Doctors, and so it goes. This is why I laugh every time I hear the phrase "Heathcare market."

V writes:

To Frank in Midtown and others:

Increasing supply in an industry where supply-induced demand (i.e., physicians and hospitals can increase demand for their services) is dangerous.

You could conceivably reduce the average price per procedure (and average provider income) but likely will increase total costs (similar to what happened in the legal profession with partial deregulation and the subsequent explosion in overall legal spending). Since the total cost is what threatens the federal budget, I think limiting providers is one of the few ways we can feasibly ration in the US system (albeit at the cost of higher prices)...

Eelco Hoogendoorn writes:
To me, the real question in the health care Olympics is how well costs and benefits are aligned. I am fairly sure that in the United States, a lot of medical procedures are undertaken that have high costs and low benefits. I have little information about other countries, but it would not surprise me if it turns out that they undertake fewer of these high-cost, low-benefit procedures.

True; which of course does not point to a lack of insurance coverage in the US, but to an absurd overabundance of it. One can argue about the distribution thereof; but perhaps if healthcare insurance, or prepaid healthcare I should say, was not promoted as a cultural norm, that would help a lot with keeping prices reasonable to everybody.

mcarson writes:

One area of health cost differences between U.S. and others is the amount of money spent per person per decade of life. When you look at the statistics you will find that most of the U.S. 'overspending' occurs during the last 18 months of life. Too aggressive cancer treatments and heart failure treatments after age 70 really hike U.S. per capita spending without adding to any quality of life. The lack of good nursing home and in-home support causes heroic measures to be taken to insure the old can maintain their lifestyle, instead of transitioning them to visiting nurse care to help them adjust to failing abilities to care for themselves.

In effect, we throw everything we have at keeping people out of nursing homes, which we neglect and refuse to pay for, and keep a few people in their homes, while spending tons of money on people who will not stay in their homes anyway, and then leaving those with bad outcomes in inferior care, causing us to try even harder with the next failing grandparent.

I am not advocating not treating the elderly, I quit my job and spent 3 years caring for my cancer-ridden Father before he died. His last 18 months was a horrible ordeal of surgery and hospitalization, with short stops in nursing homes between 'treatments.' He would have been better served if 10% of that money was spent on maintaining him in his home, even if he died 6 months earlier.

England does the last 18 months of life much better than the U.S., and saves a bundle doing it.

Mark V Anderson writes:

Of course we do not have a free market in health care in the USA. The biggest reason for our high medical costs is the over-insurance of the population, as Eelco says. If all insurance benefits received from employers was taxable like other income, then we wouldn't associate insurance with one's employer, and individuals would get their own insurance, like everyone does now with car insurance. Presumably then we wouldn't have the over-preponderance of insurance we do now, as it would be more obvious to folks that insurance is for catastrophic care and not everyday doctor visits.

It is true that even if people only had medical insurance for catastrophic care, there would still be the issue of the sickest creating the overwhelming majority of costs, since that is usually catastrophic care. But it is a beginning.

How much do you want to bet that Arnold's mother got her knee replacement paid for by insurance? And would she have done that if it she had to pay it directly?

Mark V Anderson writes:

I meant mother-in-law, not mother in my post.

Yancey Ward writes:


Violent crime victims tend to be very, very young deaths. Even if 100% of them were African-American, their not being killed in a violent act would probably significantly affect the US life-expectancy in a huge way, and if 100% of them were Caucasian (which, of course, they aren't), I find it hard to believe the contribution to the effect would be much different (there isn't a large difference between the life expectancy of African Americans and Whites when you account for the differences in violent deaths).

Of course, transport accidents are probably weighted the other way with regards to race since whites drive more. However, a more granular analysis would have been nice.

Comments for this entry have been closed
Return to top