David H. Freedman

A new study claims it shows that the children of pregnant mothers who put on lots of weight are much more likely to be obese years later than the children of mothers who don’t put on as much weight in pregnancy. The announcement of the findings has triggered comments from other researchers and experts suggesting that this study is simply the latest in a series of studies supporting the notion that women who eat too much during pregnancy are placing their children at greater risk of obesity.

Nope. This new study tells us very little, and none of the others tells us much, either. The only sort of study that would have at least a small chance of giving us some real insight into the effect of eating during pregnancy on children’s lifelong tendency toward obesity would be one that randomly assigned many, many pregnant women to either a group that was somehow made to put on a relatively large amount of weight during the pregnancy, or a group that was somehow made to put on a relatively small amount of weight. Then the study would have to follow the weight gain of the children in the two groups for a few decades. Such a study would still be plagued with potential confounders, since, for example, it wouldn’t be blinded (the women and the researchers would likely know which group each woman was in), and the two groups of mothers might start behaving very differently from one another because of the particular way in which their weight had been controlled. But it doesn’t matter, because no one is likely to do that study–I can’t imagine researchers being able to line up enough pregnant women willing to participate, and even if they did that they’d be able to effectively control the two group’s weight gains during pregnancy, or even if they could that the whole project would make it through ethics review boards. So all researchers can do is track the weight gain of a bunch of pregnant women, wait to see what happens to their children, and then go back and look for links. This approach almost always leads to utterly unreliable results, because we have no way of knowing what causes what in the resulting observed links, and indeed the links themselves are often merely artifacts of bad data and/or bad analysis.

All these pregnant-weight-gain studies really show at best is that mothers who tend to put on too much weight are more likely to have children who put on too much weight. Not only is that painfully unsurprising, but it’s tremendously useless information. It doesn’t tell us whether that’s happening because of a genetic predisposition for weight loss that gets handed down from parent to child, or because mothers tend to pass healthy or unhealthy eating habits onto their children in some way, or because exposing very young children or possibly even fetuses to excessive intakes of food or to less-healthy foods somehow changes their metabolisms or alters their appetite mechanisms. (That first study’s conclusions seem to vaguely imply the latter, for what that’s worth, which I think is very little.) The studies give even fewer clues as to what ought to be done about the situation. If it’s genetic, there’s probably little point in addressing what pregnant mothers eat, which seems wrong. If the mother’s eating is programming the child in the womb in some way, then what the mother eats during pregnancy becomes critical–and what and how the child eats after birth is not so important, a conclusion that defies everyday observation and common sense. If the mother’s weight gain is “transferred” to the child via the food attitudes and habits that the child picks up from a mother who simply doesn’t do a good job of controlling her weight, then genes or what happens during pregnancy is largely irrelevant. I think there’s a pretty good chance that what we’re seeing here is some combination of the three–genes, changes that take place during pregnancy, and (probably especially) family eating habits passed on to the child later on. But the studies don’t help pin any of this down in the least.

So what should pregnant mothers take away from all this? Come on, you know the answer to that. They should eat sensibly (and as per doctor’s advice), avoid either excessive or insufficient weight gain during pregnancy (unless the doctor orders otherwise), and most important (in my opinion and in the opinion of most of the many experts I’ve interviewed on this subject) work hard to transmit generally accepted good eating habits to their children, starting from birth and continuing on through all of childhood. (If your doctor disagrees with that last bit, get a new doctor.) I doubt we’ll ever see a study that provides convincing evidence that doing anything else makes sense.