Comparing U. S. Health Care With That of Other Countries

I was very much struck by this analysis from Peterson-Kaiser comparing the quality of U. S. health care with that of other countries. It’s full of graphs and charts. I would summarize it as that care is better here but health is better in many other OECD countries.

Consider this graph, for example:

What it seems to say is that age-adjusted mortality per 100K population tracked pretty closely between the U. S. and other OECD countries until 1985. Since then the statistic has improved for both the U. S. and other OECD countries but it has improved less in the U. S. Why?

13 comments… add one
  • steve Link

    First, mortality is not an especially good way to measure quality. It has some uses but only begins to capture quality. That said, in general if you manage to get care in the US it is usually pretty good. Getting access is the issue.

    Steve

  • That doesn’t answer the question. I look at trends and changes. We are trending in the wrong direction and there appears to have been a change in 1985. Is access decreasing? Is behavior changing? Is the population changing? Those would seem to be the most obvious potential explanations.

  • steve Link

    As I said, mortality is not the best metric. It could be the age of our population, drugs and alcohol, driving, whatever. It could very well be the effects of all the other OECD countries having adopted universal health care so that their entire population is living longer whereas in our population only those who can afford health care are the ones living longer. A current hot area in the literature is the social determinants of health. We know that poor people dont live as long in the US. If this is unique to the US (I really dont know) that could help explain.

    Steve

  • Gray Shambler Link

    How about this, in the late 1960’s, moral constraints against unhealthy behavior were loosened. More booze, more pot, more cocaine. By 1985, those new behaviors started to show results. More young deaths, bringing average life expectancy down. One angle.

  • steve:
    What in your opinion is a good metric? IMO any good metric should be an output rather than an input.

  • steve Link

    Would come up with a very complicated compilation of multiple measurements. Mortality could be in it but just one of many. I would look at infant and maternal mortality since I think they relate more directly to medical care. I would track outcomes from individual illnesses/diseases, but I would spend more time looking at days of increased (or decreased) functionality. Take cataracts which I have mentioned before. You used to spend a week in the hospital flat in bad after cataract surgery. Then home for six weeks with very restricted movement. Now you go home in 20 minutes and can work the next day. Gallbladder surgery. You used to have a 6-8 week recovery before you could go back to work, and it was painful. Now you can go back to work in a couple of days (longer if heavy labor). Knee replacements. Recovery is much faster and complication rate is much lower. For medical illnesses lets look at progression to ESRD (dialysis) for HTN or diabetic patients. Progression to joint replacement or immobility for arthritis. Increased functional days for COPD.

    I think this probably stems from my work with our pain team. What we emphasize is not so much controlling pain (that is a big part of it of course) but rather return to functionality. People dont just want to be fixed or cured, they want to return to be able to function the way they want. In general, getting people back to full function sooner seems to lead to more prolonged and happier recoveries.

    Needless to say this would be hard to track and expensive too. I would start with days out of work and days on disability and also years able to work as they are easier to track, but with confounders. Then get RWJ or someone similar to do the more detailed look.

    Steve

  • That’s exactly what the article I cited in the post does. Mortality is just their first graph. They cite 18 in all. In some the U. S. does well; in most not so well. The impression with which I was left was not of a health care system that was ineffective or overburdened but one that was rushed.

  • steve Link

    The article still looks mostly at mortality numbers, but for different diseases. I would include that, but I think that you need to also look at function. Here is one more thing I would suggest. Just compare our insured population with the insured populations of the OECD countries. I would predict we would look much better. Think that paper has been done so will look for it when i have the energy.

    Steve

  • bob sykes Link

    All social statistics for the US must be broken out by race, especially crime and health statistics. If that were done for mortality, it is likely that there would be no difference between White Americans and White Europeans, in fact since American per capita incomes are substantially higher than nearly all European countries, White Americans might have substantially lower mortality than White Europeans.

    PS Muslim and African immigration to Europe has become so great that European statistics probably should be broken out by race, too.

  • You’ll never get such a breakdown by race from France, at least. It would actually be against the law. The whole of French society and government is based on the idea of “Frenchness” which is independent of race. There are no official statistics in France showing the distribution of the population by race.

  • steve Link

    The data already exists by race and income for the US. Mortality rates are worse for blacks and for the poor (of all races).

    Steve

  • Generally, I think it’s more helpful when taking race into account to control for income and or urban/rural residence. So, for example, this CDC study of suicide:

    Suicide rates in nonmetropolitan/rural counties are consistently higher than suicide rates in metropolitan counties. These trends also are observed by sex, race/ethnicity, age group, and mechanism of death.

    That’s particularly significant when considering homicides. The homicide rates are about the same among rural whites and rural blacks. The homicide rate among urban blacks is considerably higher than that among urban whites. My interpretation of that is that decline of social support among urban blacks, particularly young black men, is particularly high.

  • Jeff Link

    The data would suggest that despite numerous ‘health reform’ initiatives, there is a fundamental weakness in US policy and programs. Improving health outcomes – including mortality (the ultimate outcome) – requires acknowledgment that health care is just one of a number of factors that influences these outcomes. Physical and social environment, individual behaviors, and genetics all play a role. I suspect Denmark and other OECD countries take a more wholistic, interdisciplinary approach to Population Health and thus have better outcomes overall. (Population health is defined as the health outcomes of a group of individuals, including the distribution of such outcomes within the group.) US policy makers should take note.

Leave a Comment