Questionable Comparisons, Questionable Conclusions
The Commonwealth Fund published a study comparing the health care system in America to the systems of six other developed nations, and found it lacking in a few of the categories. Many Americans believe that the health care system needs some sort of reform, although they conflict on what type is necessary. While there is definitely room for improvement within the U.S. system, I take issue with some of the Commonwealth Fund’s analysis and conclusions that call for a more centralized, universal system.
First, some of the data relies on physician and patient surveys. Individuals in different countries have different expectations for their health care systems, an important factor that the study’s authors admit might have affected the ratings:
Patients’ and physicians’ assessments might be affected by their experiences and expectations, which could differ by country and culture.
One of the categories I find most objectionable is “long, healthy, and productive lives,” which has a rather ambiguous meaning. The authors used three indicators to determine what constituted a “long, healthy and productive life.” (Table data excerpted from the study):
Exhibit 8. Long, Healthy, and Productive Lives Measures
| Raw Scores | Ranking Scores | |||||||||||||
| AUS | CAN | GER | NETH | NZ | UK | US | AUS | CAN | GER | NETH | NZ | UK | US | |
| Overall Ranking | – | – | – | – | – | – | – | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
| Mortality Amenable to Health care (per 100,000) | 71 | 77 | 90 | 82 | 96 | 103 | 110 | 1 | 2 | 4 | 3 | 5 | 6 | 7 |
| Infant mortality | 4.7 | 5 | 3.8 | 4.4 | 5.2 | 5 | 6.7 | 3 | 4.5 | 1 | 2 | 6 | 4.5 | 7 |
| Healthy life expectancy at age 60 (average of women and men) | 24.6 | 23.8 | 23 | 22.8 | 23.7 | 22.5 | 22.6 | 1 | 2 | 4 | 5 | 3 | 7 | 6 |
These three indicators do not fully capture “productive” or “healthy” lives. There are more relevant measures of productivity and quality of life, such as statistics about morbidity, the amount of time spent ill, or disability-adjusted life years (DALYs), which account for degree of sickness as well as length of life. These are sometimes difficult to calculate, but they are standard measures used by the World Health Organization (WHO) and far more relevant for a category about “healthy” and “productive” lives.
The indicators used do not capture the fact that someone waiting 18.3 weeks for surgery in Canada may also be losing four months of work productivity, as well as spending a long time with an impaired quality of life. The United States ranked first in wait times for specialists and nonemergency surgeries. When one includes those factors, a different story emerges from the data.
For the indicator “Health life expectancy at age 60″ the United States ranks sixth, but a closer look at the raw percentages shows a very small range from first to last; whether these differences are even statistically significant was not addressed in the study. Nor does the category capture that Americans work longer — both in their work week and in their lifespan — than the other countries listed, which could explain the slight difference in the raw percentages. American work ethic is a cultural issue, not an implication of the health care system.
Also, infant mortality is a contentious indicator for the success of a health care system. Different countries use different measurements to calculate the statistic. The United States strictly follows WHO guidelines by counting all babies that have shown any sign of life, whereas Germany, for instance, only counts babies that weigh at least one pound at birth. Other countries do not count births earlier than 26 weeks. This disparity in measures of reporting artificially skews the rates, without factoring in cultural differences, like teen births, that also contribute to higher infant mortality.
In developed countries, a large portion of the increase in life expectancy is not attributable to the health care system. During the past century, the average life expectancy in the United States has increased by 30 years; modern medicine can only account for five of those years, while public health measures account for the other 25. Attributing small changes in mortality to medical care is very tricky. Lifestyles can affect health outcomes as much — if not more — than health care. The obesity rates in the United States are much higher than the other countries listed. Holding health care systems equal, that one factor would lead the United States to have lower health outcomes. Again, this is a cultural issue, and not an indication that a universal system would improve U.S. results.
A conclusion some may reach after reading the study is that universal health care is the solution to perceived disparity; this seems to be the conclusion the authors hoped to make. In fact, the study actually suggests that the new federal health care legislation will improve U.S. outcomes:
Newly enacted health reform legislation in the U.S. will start to address these problems by extending coverage to those without and helping to close gaps in coverage—leading to improved disease management, care coordination, and better outcomes over time.
Incentives need to be realigned, but that has more to do with the disconnect between patient and physician — the health care wedge, explained in the Show-Me Institute study “Prognosis for National Health Insurance: A Missouri Perspective.”
The Commonwealth Fund study admits that none of the other nations considered have “ideal” health care systems, and makes some questionable comparisons in order to “prove” that universal health care is the best way to solve problems in health care. Show-Me Institute staff and scholars have discussed better solutions for health care reform in blog entries, op-eds, and policy studies.
The Commonwealth Fund study notes that the largest problem in the U.S. system is affordability of health care; the study thus concludes that universal health care is the solution, rather than making health care more affordable. The Congressional Budget Office has calculated that the recent legislation, lauded in this study, will actually increase the cost of health care. The Commonwealth Fund study suggests a solution that will bring the exact opposite of the problem it anticipated: Health care will become too expensive for some people.
Just because a few countries are getting (questionably) better results by some carefully selected measures under universal health care systems does not negate the fact that market-based solutions are a better solution for Missouri and the whole United States.





This study is very interesting. There are just enough data presented to make you think the results might be real. However:
the surveys in these countries were supposedly random samples of a few thousand adults, some of whom were deemed “sicker.” However, the demographic data for those sampled is not presented – age, sex, marital status, smoking history, obesity, etc, so we can’t really evaluate those claims. Those who are deemed “sicker” in the US might not be alive in the UK, because certain diseases aren’t treated there! (It is hard to sample patients who are already in a cemetery.) Studies of people diagnosed with cancer consistently show the highest survival rates in the US, and lowest (among US/Canada/northern Europe) in the UK; renal dialysis is often not offered in the UK beyond a certain age. And you are right – the lifestyle choices in the US mean that the overall population is skewed toward higher risk / less healthy.
The mission and goals of the Commonwealth Fund are listed on the first few pages of this “study.” It looks as though they reached their conclusions first, then went looking for the data to support those conclusions.
Comment by Bill H — June 24, 2010 @ 12:07 a.m.
Is there any data out there comparing Massachusetts v. the rest of the country on like metrics? Mass. has universal coverage of a kind and there should be some some data point differences that start to show themselves over the few years they have had it. I am thinking things like cancer survivor/heart attach survivor/whatever in MA vs. USA before and after it went into effect.
The data would probably have talking points for both sides, but the trends should be starting to show. At we would know what we were trading in/trading for.
Comment by Papillon — June 24, 2010 @ 11:51 a.m.
Great post. It relates to a lecture I attended yesterday by Michael Tanner, Senior Fellow at Cato, on “How To Fix Health Care.”
Just as you describe in your post, Tanner said that it is difficult to compare health care systems across countries because the definitions are often inconsistent. He also gave the example of infant mortality.
He also said that life expectancy is a poor measure of health care systems because it is affected by many other things, such as murder rates, suicide, car accidents, and lifestyle choices. (I’ve blogged before that personal responsibility is the best medicine.) This explains why there are differences in life expectancy within single health care systems. For example, people in Utah live 3 years longer than people in Nevada, and they have the same system.
Comment by Christine Harbin — June 24, 2010 @ 6:52 p.m.