Revisiting the Exercise Recommendations in the 2023 CANMAT Guidelines for Major Depressive Disorder

Level of Evidence

The Guidelines define Level 1 ⁱ evidence as “meta-analysis with narrow confidence interval”; and Level 2 evidence as “meta-analysis with wide confidence intervals.” However, the terms “narrow” and “wide” are not quantitatively defined by CANMAT, leaving its interpretation subjective. This subjectivity becomes problematic in the recommendations for exercise: It is not clear why Level 1 evidence is assigned to the consideration of exercise as a treatment for mild depression, but then Level 2 evidence is assigned to justify the use of exercise for moderate depression—especially when both conclusions draw from the same body of meta-analytic evidence.

Symptom Severity

The Guidelines recommend exercise as a first-line ⁱⁱ monotherapy for mild depression and a second-line adjunctive treatment for moderate depression. This distinction suggests that the effectiveness of exercise is believed to reduce as depressive symptoms become more severe. However, the evidence does not support this assumption. Five systematic reviews^A,C,E,F,G conducted analyses on whether the severity of depressive symptoms moderated the effect of exercise. Four^A,C,F,G found no effect of symptom severity, while one^E reported larger effects for greater symptom severity. While it is noteworthy that comparatively less research has examined exercise for people with severe depressive symptoms, there is a growing body of trials with this population. In short, exercise appears to be effective across a range of symptom severities (see also Noetel et al. ⁵ ). The evidence does not support the recommendation that exercise can only be used as a second-line adjunctive treatment for moderate symptoms, and not used at all for severe symptoms.

Comparison With Conventional Treatments

The Guidelines state, “There is stronger evidence for the efficacy and safety of pharmacotherapy and psychotherapy compared to exercise” (p. 654). While exercise-based treatment for depression is a comparatively young field, there is increasing evidence of exercise's impact. Meta-analyses cited by the Guidelines suggest that exercise is comparable^D,H to pharmacotherapy and psychotherapy in treating depression; that its number needed to treat^B is equivalent or superior to conventional treatments; and that it has relatively low attrition rates.^A,F Indeed, publication bias estimates suggest an under-estimate of exercise's effect.^B,G These results align with more recent publications, including a 2026 Cochrane review ⁶ which found exercise to produce meaningful albeit modest effects (SMD ≈ −0.46) in methodologically robust trials. Crucially, this effect is comparable to those often seen for conventional treatments for depression.

The Guidelines do not provide any references to indicate increased safety risk. With important exceptions (e.g., active eating disorder), mental illness does not present an absolute contraindication^7,8 to exercise. When compared to pharmacotherapy, ⁹ individuals receiving exercise-based treatment report fewer adverse events like dizziness, drowsiness, and agitation. Exercise tends to be associated with minor adverse events,^B,C such as muscle pain and fatigue; however, these transitory unpleasant experiences are common in exercise initiates, tend to reduce with time and physical conditioning, and can be mitigated with a tailored and supervised exercise program. While no health intervention is risk-free, the health benefits of exercise—such as improved cardiometabolic health and brain health—tend to outweigh potential risks.

Exercise Dose

The exercise prescription presented by the Guidelines is inconsistent with 7 comprehensive reviews.^{A,B,C,E,F,G,H} For instance, the Guidelines’ narrow focus on aerobic exercise disregards evidence showing that different types of exercise—including strength and balance/flexibility—are effective. Similarly, the recommendation for “light-moderate” intensity is puzzling, as the literature consistently finds that moderate-to-vigorous intensities are associated with greater benefits. ⁵ Lastly, while a minimum duration of 30 min is supported, there is no clear evidence favouring the specific 30 to 40 min per bout prescribed by the Guidelines.

In short, there is limited basis for the current recommendations in terms of type (aerobic), intensity (light-moderate), or bout (30-40 min). By presenting a rigid prescription that lacks a transparent evidence base, the Guidelines create a significant disconnect between its recommendation and the scientific literature. The critical issue is not to define an alternative dose, but rather to question the opaque process that resulted in such a specific and seemingly arbitrary conclusion.

Structure and Supervision

The Guidelines omit the term “structured” while retaining the concept of “supervision” in their recommendations. Exercise programs are said to be “structured” if they follow a plan, and “supervised” if they involve oversight from an exercise professional (e.g., kinesiologist, clinical exercise physiologist). Four meta-analyses^A,B,E,G identified supervised exercise as being more beneficial than unsupervised exercise. However, in the evidence base, supervision is almost always applied to a structured program, with participants asked to follow a specific exercise program (e.g., 3 × 30-min brisk walks/week). By selectively removing the qualifier of “structured,” the Guidelines have fundamentally misinterpreted the evidence. This omission of “structured” reframed exercise as a general lifestyle suggestion rather than an intentional, prescribed, evidence-based treatment, which may further undermine its clinical adoption.