View Full Version : Infant-mortality myths

09-10-2009, 11:59 AM
Wednesday, March 16, 2005

Infant-mortality myths
U.S. rate seems high because it actually counts all babies; other nations don't

By Michael Arnold Glueckand Robert J. Cihak
Both writers are physicians; Glueck is a Newport Beach resident.

Statistics, even at their best, don't tell a whole story. A variety of people employ medical statistics dubiously to push pet causes.

A perfect example: infant mortality statistics. The officially reported U.S. infant mortality rate has been indisputably high compared with similarly industrialized countries since at least the 1920s.

That fact has led to public health officials in the U.S. to conclude the rates are "caused" by poorly distributed health care resources and can be "solved" with a socialized, government-run health care system.

However, there's a basic problem with the numbers: Different countries count differently.

According to the World Health Organization (WHO) definition, all babies showing any signs of life - such as muscle activity, a gasp for breath or a heartbeat - should be counted as a live birth. The U.S. strictly follows this definition. But many other countries do not.

Switzerland doesn't count the death of very small babies, less than 30 centimeters (11.8 inches) in length, as a live birth, according to Nicholas Eberstadt, a former visiting fellow at Harvard's Center for Population and Developmental Studies. So comparing the 1998 infant mortality rates for Switzerland and the U.S. (4.8 and 7.2,respectively, per 1,000 live births) is comparing apples and oranges.

In other countries, such as Italy, definitions vary depending on where you are in the country.

Eberstadt notes "underreporting also seems apparent in the proportion of infant deaths different countries report for the first 24 hours after birth. In Australia, Canada and the United States, over one-third of all infant deaths are reported to take place in the first day."

In contrast, "Less than one-sixth of France's infant deaths are reported to occur in the first day of life. In Hong Kong, such deaths account for only one-twenty-fifth of all infant deaths."

As UNICEF has noted, "Under the Soviet-era definition ... infants who are born at less than 28 weeks, weighing less than 1,000 grams [35.3 ounces] or measuring less than 35 centimeters [13.8 inches] are not counted as live births if they die within seven days. This Soviet definition still predominates in many [formerly Soviet] countries. ... The communist system stressed the need to keep infant mortality low, and hospitals and medical staff faced penalties if they reported increases. As a result, they sometimes reported the deaths of babies in their care as miscarriages or stillbirths."

Since the United States generally uses the WHO definition of live birth, in their 2004 book "Lives at Risk," economist John Goodman and his colleagues conclude, "Taking into account such data-reporting differences, the rates of low- birth-weight babies born in America are about the same as other developed countries" in the Organization for Economic Cooperation and Development. Likewise, infant mortality rates, adjusted for the distribution of newborns by weight, are about the same.

American advances in medical treatment now make it possible to save babies who would have surely died only a few decades ago. Until recently, very low birth-weight babies - less than 3 pounds - almost always died. Now, some of these babies survive. Whilesuch vulnerable babies may live with advanced medical assistance and technology, low birth-weight babies (weighing less than 5.5 pounds) recently had an infant mortality rate 20 times higher than heavier babies, according to WHO. Ironically, U.S. doctors' ability to save babies' lives causes higher infant mortality numbers here than would be the case with less advanced treatment.

Because of varying standards, international comparisons of infant mortality rates are improperly used to create myths about how the United States should allocate local or national resources.

Finally, a good summary of some of the problems with comparisons among countries involving health care. Similar types of problems arise when U.S. life expectancy is compared to other countries.

OC Register (http://www.ocregister.com/ocr/sections/commentary/orange_grove/article_443950.php)

09-10-2009, 12:40 PM
The most honest way to portray US infant mortality rates is to do it on a state-by-state basis. The states break down the statistics by county or city, after that. I've seen the way Michigan does it.

When I was a new worker (late 80s), Detroit had the highest infant mortality rate for any city in the US. I don't think that's still the case, I'd have to find the statistics to be sure. The phrase used then was "infant mortality rates at third-world levels". At the same time, the rest of the counties and cities in Michigan were not experiencing those same levels.

There were some outside factors contributing to this number in the 80s-one being that Children's Hospital in Detroit was one of only 2 such hospitals in the state at the time, so if an infant died there, it was counted as a Detroit death even if the family was from somewhere else, because that's where the death was recorded. There's a children's hospital in West Michigan, now, because the population there has grown and Ann Arbor and Detroit are very far away from some of the places the hospital in Grand Rapids serves.

But even without factors like that, the numbers in the 80s were high enough to scare the crap out of people around here.