Why Do U.S. Infants Die Too Often?

  • Share
  • Read Later
Michael Krasowitz / Getty

Following a dramatic fall through the 20th century, the U.S. infant mortality rate — the proportion of babies who die before they reach their first birthday — has leveled off at just under seven deaths per 1,000 live births. That's a much higher rate than in other parts of the developed world. Across the European Union, for example, fewer than five in 1,000 babies die before they turn one. And in some stand-out countries like Japan, Singapore, Sweden and Norway, the proportion of babies who die is less than half that in the United States. Marian MacDorman, a statistician at the U.S. National Center for Health Statistics (NCHS), helps explain.

Q: Why do U.S. babies die more often than babies in other wealthy countries?

A: The U.S. ranks rather poorly compared to most developed countries. NCHS publishes a book each year called Health, United States, which has international rankings, and in the latest [edition], which ranks data based on the 2004 data year, the U.S. ranked 29th in the world in infant mortality.

One of the main [factors] is whether the baby is delivered too small or too soon, which increases its chances of death. About two-thirds of all of our infant deaths occur among the 8.2% of babies that are born at low birth weight. Most developed countries have lower rates of preterm and low birth weight deliveries [than the U.S.] and that makes a difference in infant mortality rates.

I think the single most important thing we can do to lower the rate of infant mortality is to reduce the rate of preterm birth. But in fact the trend is going in the opposite direction — that rate [in the U.S.] is increasing. There are several causes of preterm birth. One is spontaneous preterm labor. But there's another component, which is medical intervention — for example, doing a preterm caesarean section or induction of labor. That component has been increasing as well, and I think that's worrisome.

There are a lot of doctors who say it's O.K. to take a baby out a little bit early because they're going to do well — and it's true. It's only seven per 1,000 that are dying. Most of them do well. But still I think it's important to note that the infant mortality rate for late-preterm infants is three times what it is for [full-]term infants. This is not a difference that may be perceptible to the average obstetrician/gynecologist doing 300 deliveries a year. But when you're a statistician like me and you're grouping millions of births and thousands of infant deaths, you can see a difference there. Basically the longer that baby can stay in utero, the better it's going to do.

There are also really large differences in infant mortality rates by race and ethnicity. For 2004, the overall [U.S.] infant mortality rate was 6.78; but for non-Hispanic black mothers it was 13.60. It was also fairly high for American Indian mothers: 8.45. For non-Hispanic whites, it was 5.66. For Hispanics it was 5.55. And the lowest was for Asian or Pacific Islander mothers, which was 4.67. So there's a huge range there. The rate for non-Hispanic black mothers was 2.4 times the rate for Hispanic or non-Hispanic white mothers.

Non-Hispanic black women have much higher rates of preterm and low-birth-weight delivery. Why that is we're not exactly sure. But if you look at any of a wide variety of risk factors related to infant mortality, you can find differences between blacks and whites and between different races and ethnic groups. For example, a higher percentage of births to non-Hispanic black and American Indian mothers occurred to mothers who did not receive adequate prenatal care, were unmarried or were teenagers, were having a fourth or higher order birth, or had not completed high school. Certainly, if you take into account these kind of factors, the racial disparity lessens, but it doesn't completely go away. Part of this racial disparity remains really unexplained. I think that's something that researchers have really been struggling with for decades, and there are no simple answers.