1. INTRODUCTION
Mental health lies on a continuum from having no symptoms and being fully functioning to having a
severe mental disorder. Mental disorders are not categorically distinct, but rather label the part of the
continuum where symptoms are longer lasting and cause disability. They are characterized by
problems in thinking, emotional state, and behavior. Mental disorders are common in Australia.
Recent studies show that the 12-month prevalence in adults was 14.4% for anxiety disorders, 6.2%
for affective disorders (such as depression) and 5.1% for substance use disorders (37). In addition,
the 12-month treated prevalence rate of psychotic disorders such as schizophrenia is 0.45% (26).
The population impact of mental disorders is principally on disability rather than on mortality. In
Australia, mental disorders are the third largest source of disease burden after cancers and
cardiovascular diseases, but the largest source of disability burden (4). This disability burden arises in
part because mental disorders often begin early in life and have a relapsing course. For anxiety and
affective disorders, the major treatment options are psychological therapy and antidepressant
medication, with the latter being appropriate for more severe cases (9). For psychotic disorders,
antipsychotic medication is a key component of treatment, but needs to be accompanied by
psychosocial intervention and practical support (33). The present review looks at the evidence for
exercise as a treatment option, in particular for depression, anxiety, and psychotic disorders.
There is a strong relationship between physical activity and mental health. Cross-sectional studies
show that regular physical activity is associated with better mental health and emotional well-being
(11) and lower rates of mental disorders (12). Longitudinal studies also show an association between
physical activity and reduced risk of developing a mental disorder (41). For example, a populationbased
study of 7076 Dutch adults found that engaging in physical exercise reduced the risk of
developing a mood or anxiety disorder over the 3-yr follow-up period, even when controlling for
sociodemographic characteristics and physical illnesses (41).
However, the relationship between physical activity and mental health is likely to be complex and
bidirectional. Physical inactivity may be the cause and/or the consequence of poor mental health, and
there may be common factors (such as overlapping genetic vulnerabilities) that predict both (6).
Consistent with the relationship between physical activity and mental health is that individuals with a
mental disorder are at higher risk of chronic physical conditions such as heart disease, diabetes,
arthritis, and asthma (40,45).
66
2. ROLE OF EXERCISE IN TREATMENT
Exercise has been investigated for its potential to improve mental health outcomes in a variety of
mental disorders, but there is a paucity of high-quality research. Apart perhaps from depression,
exercise is not considered an established treatment for mental disorders, and it is typically evaluated
as an adjunctive treatment to pharmacotherapy or psychological therapy. Nevertheless, the Royal
Australian and New Zealand College of Psychiatrists recommends that exercise may complement
other treatments and be used as a stress management strategy to improve recovery, help prevent
recurrences, to manage the side effects of some medications, and to improve lifestyle practices and
overall health (42,43).
Most research to date has been conducted on the benefits of exercise as a treatment for depression.
Several recent meta-analyses of randomized controlled trials indicate that compared to controls,
exercise has a moderate (21,24) to large (24,31) effect size for individuals with a depressive disorder.
The benefits from exercise may not be long-lasting however, as research suggests no difference at
follow-up between exercise and control groups (21), or a reduced effect size compared to effects at
the end of the exercise intervention (24). Participants in these studies typically have mild to moderate
depression rather than severe depression. Exercise is also mildly effective at improving depressive
symptoms in individuals with a chronic physical illness (e.g., cardiovascular disease and chronic pain)
(18).
Though exercise has been evaluated in over 25 randomized controlled trials, there are few wellconducted
studies on which to make conclusions about how effective it is and how to optimize its
effectiveness for depression. Despite these concerns, it does appear reasonable to recommend
exercise as a treatment for depression. Indeed, the National Institute for Health and Clinical
Excellence in the United Kingdom recommends that a structured group program of exercise should
be considered as a first step in the treatment and management of people with persistent ‘subthreshold’
depressive symptoms or mild to moderate depression (27). Exercise is also highly rated by
people who have experienced depression (29).
Exercise is also moderately effective for anxiety, though most studies have evaluated its effect on
anxiety symptoms in non-clinical populations (47). One RCT showed that exercise may be helpful for
panic disorder. It found that 10 wks of aerobic exercise was more effective than a placebo pill but less
effective than medication (5). However, a more recent study failed to show a benefit for exercise
when completed as an adjunct to medication (46). Although panic symptoms improved in the group
undergoing exercise training, exercise was no more beneficial than a relaxation training control.
Research in adults with schizophrenia shows that exercise programs can improve certain kinds of
mental health symptoms (e.g., blunted emotions, loss of drive, and thinking difficulties), but are less
effective for other symptoms (e.g., delusions and hallucinations) (13). Exercise programs may also
improve other psychological outcomes, such as social competence, self-esteem, and well-being (19).
Exercise is also important in improving the physical health of individuals with serious mental disorders
(e.g., schizophrenia), as these individuals die 16 to 20 yrs earlier than the general population, with
this differential mortality gap growing in recent decades (34). Two-thirds of these premature deaths
are related to physical health risk factors resulting from poor access to medical care, poor diet, little
exercise, and medication-induced weight gain (34). Indeed, up to 86% of patients treated with
antipsychotic medication experience significant gain in weight (2).
Likewise, metabolic syndrome, a combination of medical disorders associated with an increased risk
for developing cardiovascular disease, diabetes, and early death occurs in approximately 40% of
67
patients receiving antipsychotic treatment (23). As a result, the prevalence of type 2 diabetes mellitus
among people with schizophrenia more than doubles that of the general population (16). Given the
benefits of exercise on weight control, preventing the onset of type 2 diabetes mellitus, and improving
glycemic control in those with pre-diabetes (20), there is a growing interest in the interventions
focused on increasing physical activity as an adjunctive therapy for this population (8,13). Most
studies have evaluated structured, supervised exercise interventions lasting between 4 and 16 wks
that includes predominantly aerobic exercise (i.e., walking, cycling, and swimming) (8,13). These
studies indicate that exercise therapy is well received, and is associated with improved physical
health (3,8,13).3. EXERCISE PRESCRIPTION – BOUNDARIES OF EVIDENCE
There is a lack of rigorous research on which to guide decisions about optimal exercise parameters
(e.g., dose, type of exercise) for individuals with mental disorders. Meta-analyses of exercise
interventions for mental health have examined this issue by evaluating which factors explain
heterogeneity in effect sizes from different studies (e.g., 17,18,21,24,31,47). But these analyses can
be limited by the small number of studies to compare and the variety of confounding parameters and
study characteristics. Another strategy is to compare different kinds of exercise and doses head-tohead
within a clinical trial (7,22,36,44). A third approach has been to summarize the most common
parameters in exercise interventions shown to be helpful in clinical trials (30).
For treating depression and anxiety, most studies have evaluated aerobic exercise rather than
resistance or mixed exercise interventions. One meta-analysis (24) suggested that aerobic exercise
may be less effective than resistance or mixed exercise, but effect size confidence intervals
overlapped, and other meta-analyses have not shown a moderating effect from type of exercise
(21,31,47). Only a handful of studies have directly compared the effectiveness of aerobic and
resistance exercise for mental disorders. These studies have indicated no difference in effectiveness
for depression (39). The effect of exercise supervision is also unclear (24), though the majority of
studies have involved some degree of supervision (30) and it may be helpful to improve adherence.
There are too few studies to determine whether individual- or group-based exercise is better (24).
Both have been shown to be effective (30), but for individuals with serious mental disorders,
encouraging other people to be involved in the physical activity will lead to a higher chance of
success (1).
Duration of the exercise intervention may have an effect on mental health outcomes. One metaanalysis
of exercise to treat depression showed that interventions lasting 10 wks or longer had
weaker effects than durations less than 10 wks (21). However, the reverse was found in a different
meta-analysis of exercise for clinical depression with interventions lasting 10 to 16 wks more effective
than shorter durations (31). It is possible that other study characteristics may be responsi