Questioning Common Advice

My wife and I have had our two kids and are officially out of the pregnancy business.

Nevertheless, after having to endure the laundry lists of rules my wife was subjected to while pregnant, I must say that I found economist Emily Osler's Wall Street Journal article about the same to be a breath of fresh air. Not one to confuse correlation with causation or to be bullied by precautionary thinking, she put her data analysis skills to work to consider first-hand what the actual evidence was for and against some of these rules.

Regarding correlation vs. causation, Osler's thinking is very clear:

The key problem lies in separating correlation from causation. The claim that you should stop having coffee while pregnant, for instance, is based on causal reasoning: If you change nothing else, you'll be less likely to have a miscarriage if you drink less coffee. But what we see in the data is only a correlation--the women who drink coffee are more likely to miscarry. There are also many other differences between women who drink coffee and those who don't, differences that could themselves be responsible for the differences in miscarriage rates.

This realization caused Osler to look deeper into the literature on the risks posed by coffee drinking during pregnancy. She eventually concluded that it was fine for her to continue her three-to-four cups-a-day habit.

Osler similarly also concluded that small amounts of wine were okay. Here is a look at part of that reasoning:

One big worry about drinking during pregnancy is that it will result in child behavior problems later. One of the best studies of this issue was published in 2010 in the British Journal of Obstetrics and Gynecology. What makes it a reliable study? The sample group was large (3,000 women), and the researchers collected information about maternal drinking during pregnancy--not afterward. The study also followed the children of these women through the age of 14 and looked at behavior problems starting at age 2.

The other thing I liked about this study was that it was run in Australia, where recommendations on drinking during pregnancy are more lax than in the U.S. Because the rules are more permissive, Australian women who drink occasionally aren't necessarily the kind of women who go against medical advice; it's more likely that differences in drinking levels there are just random variation. Drinkers in the study were classified in five groups: no alcohol, occasional drinking (up to one drink a week), light drinking (2-6 drinks a week) and moderate drinking (7-10 drinks a week). [bold added]

Notice the awareness of the risks. the critical eye cast on the literature, and the insistence on having a causal link between evidence, goal, and rule.

Most people can see that completely ignoring medical advice is lazy, but so is accepting it uncritically. But it is also difficult to evaluate such advice really thoroughly. Although this article probably has enjoyed great popularity for questioning the current conventional wisdom on pregnancy and diet, it really deserves attention as an example of how difficult it is to weigh scientific evidence. We don't all have the time to do this with everything we hear, but this is what it would take. If you can't approximate what Osler did, you should be very careful about whom you trust to do so.

-- CAV