Monday, June 29, 2009

Infant mortality and health care

To make the case for a more pronounced government role in health care you often see statistics trotted out that purport to show that many other countries which use such systems have superior performance to the U.S. on a number of metrics. One such metric is life expectancy, which I have already discussed.

Another is a ranking of global health systems by the World Health Organization in 2000 that placed the U.S. at number 37. If you delve into the ranking, however, you discover that it was assigned because of political value judgments instead of more objective criteria.

The third statistic you often see cited is infant mortality, where the U.S. is behind 45 other countries according to the CIA. Infant mortality, however, is not simply a product of both pre- and postnatal care, with behavior of the mother also playing a significant role. One proxy for a mother's behavior is education as this report from the Center for Disease Control notes:
Maternal education has long been considered an important factor in fertility and maternal and infant health and has been shown to have a profound effect on the number of births and the risk of adverse birth outcome. Women with higher educational attainment are more likely to desire and give birth to fewer children, and are less likely to engage in behaviors detrimental to health and pregnancy (41,42).
In other words, less educated women are not only likely to engage in behavior not conducive to the baby's health but also have more babies, thus placing more of them at risk. Now, it stands to reason that this phenomenon should apply to other countries as well, with the less educated and more risky population having more children.

I would speculate, however, that in other countries that the differences in birthrates between more and less educated groups is not as pronounced as in the U.S. given the lower fertility rates found in those countries. Indeed, at 2.05 the U.S. has a higher fertility rate than 96 other countries and is especially anomalous as a developed country, comparing with the European Union rate at 1.50.

Anecdotally I have found that Europeans tend to have smaller families (1 or 2 children) regardless of income level. Large families simply aren't very common. I suspect, but cannot prove, that low European fertility rates are due more to people at lower educational levels having fewer children rather than higher educated populations having one or zero children. Thus, the demographics that correspond with riskier behavior during pregnancy are having fewer children than their U.S. equivalents. Again, I have to caution that this is very speculative.

I would, however, also note that even within the U.S. we see vast disparities between states as this article points out:
For the three-year period from 2002 through 2004, there were significant differences in infant mortality rates by state, ranging from a rate of 10.32 in Mississippi to 4.68 in Vermont.
These are substantial differences, with Vermont's ranking placing it between the Netherlands and Luxembourg while Mississippi is more akin to Russia and American Samoa. Now, is that a product of differences in the state's health care systems (which I do not believe were very pronounced during this time period, with Vermont only passing health care reform measures in 2006) or the type of mothers giving birth in each state? While we don't know for certain who is giving birth, it is worth noting some pretty substantial demographic differences between the two states:
Vermont:

Percent of high school grads age 25+, 2000: 86.4%
Percent of Bachelor's degree age 25+, 2000: 29.4%
Persons below poverty, percent, 2007: 10.1%

Mississippi:

Percent of high school grads age 25+, 2000: 72.9%
Percent of Bachelor's degree age 25+, 2000: 16.9%
Persons below poverty, percent, 2007: 20.7%
It seems to me that infant mortality could be a useful tool to compare health care systems if the populations being compared were controlled for things like education and income to ensure an apples to apples comparison. Lacking this, however, it doesn't seem like it can be used as fair evaluation of competing health care systems.

Update: Also note this article which looks at the high percentage of babies born in the U.S. which don't make it to full term. If one only counts babies born at full term in infant mortality statistics the U.S. places in the top 10 for mortality.

No comments: