Arnold Kling  

Another Reason To Doubt Infant Mortality Data

Should Steven White Brand Hims... A Difficult Book...

Cheryl Miller writes

Prematurity is now the leading cause of infant mortality in the United States, in part due to the “epidemic” of multiple births to IVF patients. Multiples are twenty times more likely to die in the first month of their lives than singletons; those multiples that survive are more likely to have respiratory difficulties, learning disabilities, and other problems. Cerebral palsy, for instance, has become more common in the United States, even as its major cause, jaundice, has been all but eliminated. And even IVF singletons are less healthy than non-IVF children: they tend to be smaller and are more likely to be born with birth defects, including bowel and genital deformations and eye cancer.

(emphasis added). As you can tell, the article takes a dim view of in vitro fertilization. But I'm not jumping on that bandwagon.

What struck me was the potential for IVF to affect our infant mortality numbers. I believe that something like 1 percent of births in the U.S. these days are IVF. Suppose that for natural pregnancies, the infant mortality rate is 0.5 percent, but the IVF infant mortality rate is twenty times that, or 10 percent. Then our 1 percent IVF rate would add 0.1 percentage points to our infant mortality rate. In the game of international infant mortality comparisons, that would be a huge handicap.

The bottom line is that if you are trying to have good health care statistics in the sense of low infant mortality and low cost, you want to do less IVF. But that is not necessarily a more humane or better health system.

Comments and Sharing

COMMENTS (16 to date)
General Specific writes:

The paper "The economic impact of the assisted reproductive technologies" (link) says that:

"In the US, multiple births occur after 39% of IVF cycles, whereas in Europe, the figure is 26%. Indeed, multiple births are a major factor in the costs attributable to IVF."

Another paper says that on average both the US and Europe have rates of around 1%.

In addition, the economist 26 July 2007 article "In Vitro Veritas" (link) says that:

"Denmark's native population is more or less stable, but some 3.9% of babies born there in 2003 (the most recent year for which figures are available) were the result of IVF. The comparable figure for another northern European country, Britain, was 1.5%."

Basically, this data indicates that, on average, European IVF rates are similar to the US (e.g Massachusettes has a rate of almost 4%) and that we are on average dealing with 13% more multiple births.

So while IVF should be taken into account, it's not clear that it will prove that the higher US mortality rate is due to IVF.

Maybe a better reason is that we have a large percentage of the population--including pregnant women--without any healthy coverage.

Getting to the bottom of issues like this is important. It's just not clear that IVF is the solution, though it may play a role. Until the specifics are known, and given the above numbers, it doesn't seem that pointing at IVF for the higher mortality is warranted.

Lauren writes:

I agree with General Specific that looking at IVF to explain the higher infant mortality rate in the US is not the right solution. As General Specific noted the US and Europe have similar IVF rates yet Europe has less multiple births. This information clearly illustrates that multiple births can not be directly correlated with IVF's. General Specific was right in looking at other explanations of the increased multiple births occuring within the US. While IVF rates are similiar between the US and European citizens with health care is not as similar. The percentage of Americans without healtcare coverage in 2004 was 15.7% ( while Europe has a universal healthcare program. Therefore could the high percentage of multiple births not be attributed to the fact that less citizens in the US recieve proper healthcare during pregancy based on the fact that 15.7% of individuals are unable to recieve healthcare assistance? General Specific is correct in stating that other causes of high infant mortality rate should be examined rather than focusing on IVF births in the US.

reason writes:

I'm always amused when people treat empirical questions as though they are questions of opinion.

Buzzcut writes:

As you can tell, the article takes a dim view of in vitro fertilization. But I'm not jumping on that bandwagon.

Why not? Doesn't this article just SCREAM for government regulation of IVF?

If this were any other issue, the governistas would be all over demanding more regulation.

General Specific writes:

Empirical questions warrant empirical answers.

Here's an empirical question: I suggest looking at the higher rate of out of wedlock births in the United States, combined with the lack of health care amongst the lower economic classes (for whatever reason). Combine those two issues with problems such as drug abuse and a higher mortality rate amongst new borns is almost guaranteed.

Buzzcut writes:

You commenters pushing for socialized medicine have me confused.

How does the lack of health insurance lead to more multiple births?

I know one answer: "selective reduction". Is that a provided service in socialized countries? Are Britons and Danes more likely to abort their excessive embryos than Americans are? Is this because the government is picking up the tab?

That's the only explanation I can think of, and I hardly think that it is a persuasive argument in favor of socialized medicine (selective reduction is absolutely barbaric, even moreso than straight up abortion, if that's possible).

Buzzcut writes:

Or maybe the socialized countires only pay for one embryo at a time to be implanted? That would certainly make multiples less likely.

MT57 writes:

Lauren writes "could the high percentage of multiple births not be attributed to the fact that less citizens in the US recieve proper healthcare during pregancy based on the fact that 15.7% of individuals are unable to recieve healthcare assistance?"

It would be an extremely remarkable scientific development if, after fertilization, there was an inverse relationship between the pregnant woman's healthcare and the number of her fertilized eggs that went to term - i.e., less healthcare, more successful pregnancies. I could imagine the argument that there would be a direct correlation - i.e., less healthcare, less successful pregnancies - but I think it is really quite improbable that the relationship is inverse. Respectfully, I suggest the thesis quoted is simply evidence that its proponent has decided, no doubt for sincerely well intentioned political motives, that all problems with health in the US are to be attributed to the extent of healthcare insurance. That is bad analysis, it will lead to raising false expectations, grievous confusion and disppointment because after all, insured people get sick too, and if we lived in an intelligent country, it would be bad politics as well.

Floccina writes:

“The highest rate of twinning is found among women of African decent. For example, here in the United States, the rate of DZ twins among the African American community is 25.8 per 1,000 live births. Furthermore, the rate of twinning among the Yoruba tribe of the West African country Nigeria is an incredible 45 per 1,000 live births. Remaining twinning rates here in the United States, from highest to lowest are: Alaskan natives 24.9 per 1,000 live births; Caucasians 19.6 per 1,000 live births, Native American descent 18.8 per 1,000 live births, Japanese descent 17 per 1,000 live births, Hawaiian descent 15.3 per 1,000 live births, Filipino descent 13.2 per 1,000 live births and Chinese descent 11.2 per 1,000 live births.”

Is it any wonder that the USA has higher infant mortality than Japan!

Carter writes:

General Specific writes:

Two interesting comments in the second link provided by Carter (above):

"most of the rise [in infant mortality] was due to an increase for babies born in single deliveries."

So this says the issue is not multiple births.


"More intensive monitoring of at-risk pregnancies may have resulted in an increased likelihood that a Caesarean delivery will take place and a low birth weight baby will be born."

Essentially, what I'm hearing is a big not sure.

Buzzcut writes:

Essentially, what I'm hearing is a big not sure.

What you are hearing is that technology (whether it be fetal monitoring or IVF) leads to more low birth weight babies being "born". The US, where technology is embraced, will have more of these "births", which eventually lead to death and an infant mortalist statistic.

Countries with socialized medicine don't have this technology. There, these babies die in the womb, and are not counted in the statistics of infant mortality.

Anybody who has had a baby in the US recently knows that technology in the delivery room is pervasive. It's everywhere. At the slightest sign of trouble, you're in the operating room having a C section. In most cases, it's a good thing (my daughter had the umbillical cord wrapped around her neck, which wasn't seen on the ultrasound. The fetal heart rate monitor showed that something was seriously wrong, and my wife had the C section. If not for the technology, maybe my daughter would have had brain damage, or worse).

But you can see how it could have a negative impact to the infant mortality statistics.

General Specific writes:

To resolve the issue of technology in the delivery room, simply look at infant mortality and still birth rates. I started investigating it, but was fortunate that someone already beat me to it: "US infant mortality rate" (link)

This data clearly proves that the high US infant mortality rate cannot be explained by survival of babies that would be considered still births in other countries.

The problem, as I stated above, is that the data does not support any position very well, but seems to point towards poor prenatal care in the US, leading to increased still births and infant mortality.

We should not let our biases lead us to argue along lines that are not supported by the data.

Buzzcut writes:

Interesting link. The link more or less explains the discrepancy in infant mortality: the underclass.

Why does the underclass (african-americans, principally) have such a higher infant mortality rate? Is it simply a lack of access to health care? Is there a behavioral aspect? After all, crack whores, with or without health care, are going to have a higher rate of infant mortality. Or is it just institutional racism as the root cause?

I suppose you could look at the infant mortality statistics for people on Medicaid, broken down by race.

Floccina writes:

Considering that native Canadians, Aborigines in Australia and Maoris in New Zealand all have very high infant mortality rates while Mexicans in the usa have low rates of infant mortality I think that we can all but rule out lack of socialized medicine in the USA causing the higher rate problem.

Cheryl Miller writes:

Arnold, thanks for linking! There's actually an article about just this point at Slate:

Also, re: Europe and IVF: This is discussed in Mundy's book. Many European countries limit the number of embryos that can be implanted, thereby reducing multiple births. In the US, there are guidelines, but they are voluntary. (Last year, the American Society for Reproductive Medicine issued guidelines suggesting that no more than two embryos at a time be transferred into women under age 35.)

There are various reasons for this. Europe is already much more restrictive about ART (e.g., no sex selection in the U.K., no surrogate mothers or PGD in Germany, open donor registries in many European countries) (

There's also the factor of cost. IVF is very expensive, and not always covered by insurance. So U.S. patients are--as Mundy puts it--"buying in bulk." Many specifically ask for twins so that they can avoid paying for another round of IVF. (

Comments for this entry have been closed
Return to top