The Pregnant Couch Potato, or, How Rumors Get Started

If you exercise before or during pregnancy, will you decrease your risk of depression? Yes, a new report seems to say — but before you head for the gym, you might want to glance at the fine print.

Readers of this blog will recall an amusing study out of Holland that cast the relationship between workouts and mood stability in a new light. The investigators identified a genetic correlation between, on the one hand, a willingness or tendency to exercise voluntarily, and on the other, relative freedom from anxiety and depression. In an identical twin pair, if a twin exercises more, he or she does not experience less depression. The study authors wrote, “Longitudinal analyses showed that increases in exercise participation did not predict decreases in anxious or depressive symptoms.” Exercise correlated with decreased depression — but only because naturally upbeat people exert themselves more.

That finding changes the ground rules for research on depression and exercise. We’re no longer speaking of two independent variables. We simply expect moody people to exercise less, just as we expect them to become depressed.

New research out of Pennsylvania State University has the bad fortune to emerge in the context of the new genetic perspective. The Penn State researchers gave questionnaires to 203 pregnant women and asked about exercise, body image, and depression. The three proved to be correlated. Sedentary habits, negative body image, and low mood travel together. In particular, “Body image dissatisfaction” and depressive symptoms are significant predictors of further depressive symptoms during or after pregnancy. Not surprising. If nothing else, women falling into depression will criticize their own appearance.

But despite dozens of analyses, in the Penn State study, exercise barely showed up at all. For instance, first trimester depressive symptoms predicted 36% of the variance in second trimester depressive symptoms; adding in body image and exercise contributed only an additional 2%. When exercise was treated as a discrete variable, it made no contribution to outcome — and similar findings held across all the trimesters. Only one modest correlation emerged: in a model in which the three factors explained 38% of the variance in first trimester depressive symptoms, an interaction between depressive symptoms and high pre-pregnancy exercise contributed 3% of the variance. I don’t know enough about statistics to put that conclusion into ordinary English, but my guess is that what we’re talking about here is not as big a correlation as the Dutch researchers found on a genetic basis.

In other words, the Penn State study offers no evidence that pushing yourself to exercise before pregnancy prevents depression — and there is room in the data for the opposite conclusion. Given that there’s a natural moderate correlation between willingness to exercise and stable mood, the weak correlations found here (really, no correlation at all for exercise while pregnant) suggest that actually exercising before or during pregnancy may increase a woman’s vulnerability to depressive symptoms. I don’t suggest this conclusion seriously. I would like to see my pregnant patients exercise. But proper reporting on the Penn State study ought to say, there’s no news here.

The UPI did not see matters this way. Their summary is headlined, “Exercise may lessen depression.” To be fair to the reporter, he or she was faced with a lead author of the monograph who said, "Our study supports the psychological benefits of exercise to improve body image and lessen depressive symptoms." Of course, when other outlets picked up the wire report, they moved further in the Pollyannaish direction. The site Health Day: News for Healthier Living, features a headline and bullet point that both read: “Exercise Decreases Risk of Pregnancy-Related Depression.” The Times of India goes further: “Exercise a must for mums.” That’s how rumors get started!

There’s another problem: The Penn State investigators looked only at depressive symptoms, not clinical depression. So did the Dutch researchers, but at least they used measures known to correlate with actual diagnosis. So the new study tells us nothing about the effect of exercise before or during pregnancy on depression, the disease. In light of the genetic findings, the current report may even give us cause to worry. The take-home message is that major depression is hard to influence. We just don’t know how much benefit there is in the feel-good solutions, the ones that represent the way we’d prefer the world to work.