It's been a grand week for couch potatoes. First, the Chicago Tribune ran a round-up piece that cast doubt on the proposition that physical exertion staves off dementia. Then the Archives of General Psychiatry weighed in with research suggesting that exercise might not lessen anxiety or depression.
Looking at almost 6000 twins from the Netherlands and 2600 near relatives, researchers found that yes, working out correlates with untroubled mood, especially in women — but that this association is probably genetic. People inclined to exercise are also people protected from mood disruption; unknown “common genetic factors” may underly both tendencies.
This breathtaking — and, once proposed, reasonably intuitive — finding casts at least a light shadow on research that has found exercise training to be a (sometimes modestly effective, sometimes powerful) treatment for depression.
In a recent discussion on National Public Radio about alternative approaches to mood disorder, I raised the “intention to treat” issue. Most studies of interventions like exercise are “completer” trials. They look at people who follow through on the regimen and see how many get better. But of course, people disinclined to exercise might drop out of the “active intervention” arm of a trial of say, jogging. If those are the same people who carry a predisposition for low mood, then any positive research outcomes become suspect. Perhaps it’s not the exercise but the pressure for vulnerable subjects to quit the study that makes running look better than placebo. The Archives analysis makes clear why we should hold alternative treatments to the same research standards that apply to psychotherapy and medication.
I recommend at least a trial of exercise for all my depressed patients. But a good number are already working out vigorously — exercise is no panacaea.
The new evidence notwithstanding, I’m back to my summer regimen of swimming and bicycling. I’ll return to the intention-to-treat issue presently, in my promised second posting on cognitive behavior therapy in the treatment in anxiety.
Addendum: Overnight I found an "exercise versus medication for depression" study that uses an "intention to treat"analysis. The two interventions look equally good, but both have low effect sizes. Even so, depending on how exercise interacts with resilience, the bias problem arising from selective dropouts may not be entirly eliminated.
It is nice to see such a
It is nice to see such a large study really show the holes in the tendency for us to infer causation from correlation just because it "seems" right. It is kind of a jolt to suddenly doubt the seemingly obvious conclusion that exercise helps elevate mood, but this study really catches the attention.
That said, I personally feel the effects of elevated mood after and during exercise. But if I were depressed, it would be kind of begging the question to ask whether exercise lifted my depression or whether the lifting of my depression helped me have the will to get out there and exercise!
I doubt that exercise is as great a prescription as the correlational studies seem to suggest. It helps you feel better, as long as you feel better enough already to do it! But suggesting exercise as something that may help is a good idea, because it might help, and if you're already on the way up, it might help lift you higher. It does that for me.