Thursday, March 27, 2008

Their conclusions might be a bit premature (lousy pun intended)

theboy

A new study hit all the major presses recently, and hit them pretty hard. As of today Google News lists 447 news articles about it. You can read about it in US News & World Report, the LA Times, Reuters, USA Today, Time, and all the little guys who picked up the story off the wire.

The breaking story has to do with the results of a longitudinal study of Norwegians recently published in the Journal of the American Medical Association, entitled "Association of Preterm Birth With Long-term Survival, Reproduction, and Next-Generation Preterm Birth." (The folks at JAMA are kind enough to offer free access to the article, so you can read the whole thing if you like. It's pretty intelligible to the uninitiated, actually, for something in a professional journal; maybe that's why so many journalists picked it up.) Swamy and colleagues followed a little over 1.1 million people born between 1967 and 1988 in Norway up until 2002 (for survival rates) and 2004 (for educational achievement and "reproductive outcomes"). As they state clearly right up front, their objective is: "To determine the long-term effects of preterm birth on survival, reproduction, and next-generation preterm birth."

And what about those long-term effects? They conclude that preemies are more likely to die in early childhood, less likely to have their own children, and more likely to have less education. How's that for a trifecta of dread?

Except that it isn't. As I've mentioned before, we've heard alarming predictions from basically the first time we contemplated our son under glass in the Neonatal Intensive Care Unit, and just as we didn't buy the nurse practioner's line then, we're not buying now. Turns out it's almost too easy to poke holes in both the research and the reporting. After just a few minutes of talking, my lovely wife and I noticed:
  • They tracked those born up until 1988, which is to say nearly fifteen years before The Boy's birth. Needless to say, much has changed in pre- and postnatal care, including preemie care. (Props to the researchers and a handful of reporters who noted this.)

  • The above bullet really needs to be kept in mind especially when thinking about the conclusions the authors reached about lower levels of education and reproduction. Those claims depend upon only looking at those born between 1967 and 1976. I hate to say it, but 1976 was a long time ago, medically speaking.

  • The study (and reporting) focused on gestation period and not weight. Born at just under 36 weeks, Q technically qualifies as premature, but she was, relatively speaking, huge. Is she supposed to have the same kind of problems?

  • Most important, the researchers didn't control for basically anything. That is, they only looked at when Norwegians were born and when (or whether) they died or had kids or stopped studying. They didn't, for example, separate out infants born prematurely because of lousy or absent prenatal care or drugs or whatnot and those born prematurely for (presumably) genetic factors. Controlling for such variables is crucial, though, because if, say, you're born extra early to a sixteen-year-old single mother with little family support, my guess is that your shot at a healthy, well-educated life might be diminished just a smidge. So not only are the results at best correlational (this happens and then this happens) and not causal (this happens because this happens), it's not at all clear what's being correlated. Preemies and educational underachievement? Underbaked Norwegians and Scandinavian mating rituals?
For these and other reasons, my wife and I remain unmoved (though unavoidably worried) about The Boy's being born at just over 31 weeks.

These types of studies have to be done, of course; we still know so little about what causes premature birth, let alone what effects being shoved into the world early has at birth, at middle school, and at college (should one go). And some of the differences found between preemies and their full-term coevals are striking (e.g., that only 14% of men born between 22-27 weeks gestation had children, compared with 50% of those born on time). But the journalists who circle medical journals looking for copy should honor the science they report on by not giving birth to their stories prematurely — that is, by preserving the skepticism that sharpens empirical research. Otherwise, they too easily kick general worries of parents like us up a notch into full-on anxiety.

Now, if you'll excuse me, I need to go hold my son.

(Photo above: The Boy, my lovely wife, and myself the day he came home. He was one month old and weighed less than four pounds at the time.)

1 comment:

Nadine said...

First of all: what a beautiful photo. Though no faces, the photo really moves me. Thanks for posting that.

And I think you're right, probably a LOT of things have changed since 1976 . And it is absurd they include babies of teens and drug abusers in their statistics.

And than there is premature at 35 weeks.. 31 weeks or 22 weeks..

Did you hear about this story: http://www.guardian.co.uk/society/2007/feb/21/health.lifeandhealth?