Exercise to cure depression: why moderated effects are so helpful

One of the areas of research I’m interested in is exercise. Most of us are aware that exercise benefits our well-being. It helps keeps our heart strong, maintain a healthy weight, and combat stress. However, is exercise so good it works as well as pharmaceuticals? The more radical exercise evangelicals claim it can. Some assert that exercise can cure depression and stop cancer. Are any of these claims true? 
Turns out that yes, exercise can (somewhat) cure depression. Several meta-analyses have revealed that healthy individuals who start an exercise program have significant decreases in depressive symptoms. The trickiness however, is in the detail. How much exercise do we need to do to see results? What type of exercise?  Conducting meta-analyses can help us answer some of these questions because we have enough statistical power to explore moderators of the relationship between exercise and our outcome. Moderators are things that influence the relationship between a predictor and an outcome. Here’s an example…

Cancer survivors often have higher depressive symptoms than the normal population as they struggle with fears of recurrence, returning to their pre-cancer “normal” life, and coping with treatment related side effects. A meta-analysis and review published this month in Cancer Epidemiology Biomarkers & Prevention by Craft and colleagues at Northwestern University concluded that exercise interventions decrease depressive symptoms in cancer survivors. This finding was consistent with what has been shown in healthy samples. The most interesting and relevant finding in my opinion, was the exploration of the moderators of this effect. They looked at whether certain aspects of the exercise program helped participants’ depression scores decrease.

Exercise is a very general term. It’s likely that certain types or exercise or amounts of exercise will help mental health more than others. Craft found that exercise programs that took place in group settings such as community centers decreased depressive symptoms, while programs that were home-based actually increased depressive symptoms.  Exercise routines that consisted of more than 30 minutes of activity decreased depression more than exercise routines that were shorter than 30 minutes per session. Other moderators tested showed that some aspects of the intervention did not impact the results. For example, aerobic vs. non-aerobic exercise programs did not act differently on depressive symptoms. It also did not seem to matter what type or stage of cancer the participant had previously had.

These nuanced findings are what will help empower patients and care providers the most. It’s not helpful to tell patients they need to exercise more – we all know that. But giving more specific advice may help. For example, based on this article, a doctor may provide the specific advice to exercise with a friend or a group for longer than thirty minutes and to perform any type of exercise they enjoy. Testing moderated effects may seem less important than testing the primary hypothesis (does exercise decrease depressive symptoms), but they provide the findings that are the most translatable to patient care.