Supporters of Obamacare or those who want to go further and implement universal health care in the US often make the argument that America spends more than most countries on health care, but relatively speaking gets pretty lousy results. Having personally experienced universal health care systems in the UK, New Zealand and Australia this argument has always puzzled me. In comparing the service I received in those countries with what my family enjoys here, the argument never added up.
And while I’ve never experienced universal health care in Canada, here in Seattle we all too often watch reports of border crossing Canadians in our city’s hospital beds, because their universal health care system is incapable of treating them. It was a sad sight several Christmases ago when our local news broadcast distraught Canadians who had gone into labor and had to travel here to deliver their babies; they had relatively mild complications that the hospitals in British Columbia simply couldn’t handle.
Measuring a health care system
The two key metrics that are cited by those who argue US health care is inferior are life expectancy and infant mortality. Economist and blogger Paul Krugman often cites these, most recently in June defending the appalling scandal at the Department of Veteran’s Affairs, itself a universal health care system.
“[T]he United States actually ranks low on basic measures of performance; we have low life expectancy and high infant mortality”
The problem with using these two metrics, as Krugman should know, is that they aren’t at all useful in measuring the performance of a health care system. David Hogberg has written an excellent but academic article that examines these metrics in great depth. The essential point he makes is that when studying how well a health care system is performing, the statistics used must satisfy three criteria. He lists them as follows:
First, the statistic must assume actual interaction with the health care system.
Second, it must measure a phenomenon that the health care system can actually affect.
Finally, the statistic must be collected consistently across nations.
As Hogberg explains, life expectancy fails to satisfy the first criteria of interaction. If someone dies in a manner that precludes any possible contact with a health care professional, we shouldn’t count that as a strike against that country’s health care system. But using life expectancy to measure a health care system’s performance does just that. If you’re killed instantly by a bus, an ER doctor isn’t going to help, but your untimely death will adversely impact your country’s life expectancy statistic.
Life expectancy doesn’t satisfy the second criteria either. Critics who cite the poor US performance suggest that, “the high costs and poor outcomes seem to stem from inefficiencies that are unique to the U.S. health care system.” If health care systems are able to substantially impact life expectancy, those that spend relatively large amounts but don’t have the US’s so-called “unique inefficiencies” would be expected to perform better. As Hogberg notes amongst OECD countries:
“Greece, the country that spends the least per capita on health care, has higher life expectancy than seven other countries, including Belgium, Denmark, Finland, Germany, Netherlands, the United Kingdom and the United States. Spain, which spends the second least per capita on health care, has higher life expectancy than ten other countries that spend more.”
So what does explain the difference in life expectancy across countries? The National Center for Policy Analysis (NCPA) cites the following:
“Life expectancy – in all but the least developed countries – is primarily a result of genetic and social factors (e.g. lifestyle, environment, education, etc.) rather than the quality of medical care.”
If we could attribute the US’s poor performance on this metric to an inferior health care system, different ethnic groups ought to have a lower life expectancy than their counterparts elsewhere in the world. But as the NCPA points out, Japanese-Americans have a higher life expectancy than white Americans. In fact, their life expectancy is on par with the average Japanese life expectancy. This supports the conclusion that genetics and other factors play a much larger role than a country’s health care system.
Similar trends can be seen for other ethnic groups in America. And as Hogberg points out, one of the big reasons for the US’s lower life expectancy is a result of a much poorer life expectancy for African-Americans – 72.3 years vs. 77.7 years for whites.
While infant mortality does satisfy the first two criterion for being a useful measure of a health care system’s performance (it assumes interaction and is a phenomenon the system can affect), it’s a metric that isn’t collected consistently across nations. Moreover, drastic diversity in infant mortality rates across ethnic groups in the US also suggests something other than the health care system is at play.
In terms of different infant mortality statistics across nations, Hogberg cites examples such as Switzerland, which doesn’t include infant deaths in its metric if the baby is less than 30 centimeters long at birth. The United States does count these deaths and also happens to be well known for attempting to deliver and keep alive very premature babies. Canada also has its own set of peculiarities. Canadians who deliver in the US are included in Canada’s statistics, although those babies are treated in US hospitals, and it excludes Americans who give birth in Canada, which would conversely be deliveries in Canadian hospitals.
As for genetic disparities amongst infant mortality statistics, the NPCA notes the following:
“Americans of Japanese descent living in California had an infant mortality rate of 4.8 deaths per 1,000 live births through 1989 (the latest data available for these populations), while Chinese had 7.1 deaths, Filipinos 7.8 deaths, Hispanics 7.8 deaths, whites 7.7 deaths and blacks 18.0 deaths per 1,000 live births.”
The study notes that as the individuals from these ethnic groups are often treated at the same hospitals and by the same doctors, something other than system-wide failures must be the cause of these disparities.
None of this is to suggest that America has or had a perfect health care system, before or after Obamacare was implemented. But to criticize it for prevailing conditions that its health care system is not responsible for is to miss the point. On other health measures that the system can and does influence America performs as well and often better than many or most countries.
It’s in areas not directly related to health outcomes that America needs to focus, such as costs and insuring the uninsured. Obamacare isn’t the right answer to either of those and it adversely impacts some of the best aspects of the system such as choice and an absence of waiting lists. Health care costs for the previously insured have continued to rise, despite pledges to the contrary to ensure its passage. In fact, according to recent reports, Americans are spending more on medical services yet using fewer of them. Furthermore, the CBO estimates that tens of millions of Americans remain and will remain uninsured under Obamacare. And more than 71% of the uninsured who have received coverage since the passage of the Affordable Care Act have done so as a result of the expansion of Medicaid, not because of the individual mandate or plans on the health insurance exchanges. In fact, barely 1 out of 10 new enrollees in Obamacare were previously uninsured.
If we are really seeking ways to improve America’s health care system, first we need to understand what truly ails it. Life expectancy and infant mortality metrics aren’t symptoms of an underlying problem.