EXERCISE IS MEDICINE

Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

The Role of Exercise in Preventing

CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC4/OAVpDDa8KKGKV0Ymy+78= on 06/03/2024

and Treating Depression

Felipe Barreto Schuch, PhD1 and Brendon Stubbs, PhD2,3

The core features of depression symp-

Abstract toms include low mood, decreased inter-
Depression is a leading cause of global burden. The mainstay of treatment est or pleasure in most or all activities of
is pharmacological and psychological interventions. While effective, not all the day, decreased motivation, increases
people will respond to those treatments and alternative approaches for or decreases in appetite and weight, in-
preventing and treating depression are required. Recent literature has somnia or hypersomnia, psychomotor
demonstrated that higher physical activity (PA) levels and exercise confer agitation or retardation, fatigue, cogni-
protective effects on incident depression. Also, exercise has demonstrated tive impairments, such as memory deficit,
efficacy on reducing symptoms for people with depression. Despite its and suicidal thoughts with or without
effectiveness, similar to other treatments, some people may benefit more suicidal plans or attempts (5). In addition
from exercise and identifying these potential predictors of response is to the profound burden on mental health
necessary to deal with patients’ and professionals’ expectations. Dropout and well-being, there is a growing body
from exercise interventions is comparable to dropout from other treatments of evidence to suggest that people with
for depression and similar to dropout from exercise in other clinical popula- MDD experience substantially poorer
tions. However, some strategies to increase adherence are important. In physical health (6). For example, people
the present article, we provide an updated overview of the use of PA and with depression present increased preva-
exercise for the prevention and treatment of depression. lence of cardiometabolic disease (7), dia-
betes (8), and cardiovascular disease (9),
and experience premature mortality by 10 yr compared with
Introduction the general population (10). While suicide accounts for a part
Major depressive disorder (MDD) is a highly prevalent dis- of the premature deaths among people with depression, it is
order in most cultures across the world with a point prevalence well established that the higher levels of cardiovascular and
ranging from 6% to 18% across different countries (1). When metabolic disease when compared with the general population
considering the prevalence of people with subsyndromal/ play a significant role to the premature mortality gap (10).
subthreshold depression, or those that have significant depres- The current focus of treatment for people with confirmed
sive symptoms but do not meet the criteria for a formal diagno- MDD consists of antidepressants and psychotherapies. While
sis of MDD, the prevalence rate is approximately 15% to 20% antidepressants are typically more efficacious than placebo
(2). MDD is one of the top 10 causes of years lived with disabil- (11), some evidence suggests that only about half of the people
ity across the world and a leading global cause of burden (3), taking antidepressants achieve a clinically significant response
and the economic costs are considerable. For instance, the costs (a decrease of 50% or more on depressive symptoms) (12).
associated with the days lost of work due to depression and Also, dropout rates are considerable, ranging from 15% to
anxiety is estimated in US $ 1.15 trillion per year worldwide, 132% higher than placebo (11). Another factor that influences
and this amount is expected to increase twofold by 2030 (4). adherence is the side effects of antidepressant medication which
can include weight gain, increased diabetes risk, and sexual dys-
1
Department of Sports Methods and Techniques, Federal University of Santa function among others. Psychological therapies, such as cogni-
Maria, Santa Maria, BRAZIL; 2Institute of Psychiatry, Psychology and tive behavioral therapy, have a small-to-moderate effect for people
Neuroscience, King’s College London, De Crespigny Park, London, UNITED with depression (13), the impact of psychotherapies on the poor
KINGDOM; and 3South London and Maudsley NHS Foundation Trust,
physical health and premature-associated mortality is unclear.
Denmark Hill, London, UNITED KINGDOM
Given the considerable individual and societal burden of
Address for correspondence: Felipe Barreto Schuch, PhD, Federal University depression, there is a pressing need to identify modifiable risk
of Santa Maria, Av. Roraima, 1000 - CEFD, Cidade Universitária, Santa factors which may be amenable to change. To this end, there is
Maria - RS, 97105-900, Brazil; E-mail: [email protected]. growing recognition that lifestyle behaviors, such as physical
1537-890X/1808/299–304
activity (PA) and exercise partially contribute to the risk of de-
Current Sports Medicine Reports veloping depression and can be useful strategies for treating
Copyright © 2019 by the American College of Sports Medicine depression, reducing depressive symptoms, improving quality

www.acsm-csmr.org Current Sports Medicine Reports 299

Copyright © 2019 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 of life, and improving physical health outcomes. In the present levels of PA at baseline, including a total of 49 unique studies,
article, we provide a brief overview of the current evidence for: and a sample of 266,939 participants, accounting for a total
(1) the role of PA and exercise in the potential prevention of of 1,837,794 person-years (21). To avoid reverse causality
incident depression; and (2) the use of PA and exercise as ther- only studies with people free from depression at baseline
apeutic strategies for depression, including the use of exercise were included (21). We found that PA reduced the risk of de-
as a strategy for acute management of symptoms, the effects of pression 17% in studies adjusting the odds for potential co-
exercise training, the potential predictors/moderators of re- variates and a reduced odds of 41% in studies that did not
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

sponse, the neurobiological mediators, prescription, adherence adjust their analysis for potential covariates and calculated
and dropout from exercise, and the translation of the evidence using only raw numbers.
by current guidelines of depression treatment. We performed several subgroup analyses exploring the
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC4/OAVpDDa8KKGKV0Ymy+78= on 06/03/2024

potential differences across different countries, ages, and the

PA and Exercise as Protective Factors for variables included for adjustments in the regression models.
Incident Depression According to our subgroups analyses, the protective effects
Cross-sectional studies have shown that people with higher were significant across all ages: children and adolescents
levels of PA present decreased depressive symptoms, and these (10% decreased odds), adults (12% decreased odds), and
results are consistent across different countries and cultures. older adults (21% decreased odds); all the countries where we
For example, recent evidence using data from the Brazilian have found studies: Asia (24% decreased odds), Europe (17%
National Health Survey, accounting for 59,399 individuals, decreased odds), North America (14% decreased odds),
demonstrated that a lack of PA for leisure was associated with and Oceania (35% decreased odds); and the potential covar-
depression in young males (odds ratio [OR], 1.45; 95% con- iates included in the models for adjustment: age and sex (17%
fidence interval [CI], 1.02–2.06), middle age (OR, 2.38; decreased odds), body mass index (13% decreased odds),
95% CI, 1.4–4.03), and older adults (OR, 5.35; 95% CI, smoking (26% decreased odds), and the combination of these
2.14–13.37) (14). A similar pattern is seen in older Japanese three factors (17% decreased odds). The odds for those com-
adults, where individuals with lower PA have a higher risk pleting the 150 min of moderate to vigorous PA per week
of depressive symptoms (15), and in the United States, where were decreased by about 22% (21).
people 20 yr and older, who engaged in only light PA, were
more likely to have experienced depression than those who en- PA as a Treatment for Depression
gaged in vigorous PA with OR of 3.18 (95% CI, 1.59–6.37)
(16). Also, a study across 36 countries demonstrated that lower Acute Management of Symptoms (Effects of a
levels of PA (defined as less than 150 min of moderate-vigorous Single-exercise Bout)
PA per week) were consistently associated with elevated depres- People with MDD can benefit from a single bout of exercise.
sion (OR, 1.42; 95% CI, 1.24–1.63) (17). Clearly, this associa- A study by Meyer and colleagues (22) compared the effects of a
tion has the potential to be bidirectional. Previous studies have 20-min cycling bout at three different intensities: light (RPE,
shown that people with depression have lower levels of PA (17) 11), moderate (RPE, 13), or hard (RPE, 15) on 24 women with
and higher levels of sedentary behavior (18). In fact, a meta- MDD versus a control group (sitting quietly). The comparative
analysis of all of the published data demonstrated that people analyses showed that all intensities were equally effective and
with MDD have about 50% higher chance of not meeting the better than the control in promoting well-being at 10 and
150 min of moderate to vigorous PA (19) as recommended 30 min after the exercise. Also, in a following analysis of the
by general public health guidelines. same trial, Meyer and colleagues (23) investigated the role of
A limitation of cross-sectional studies is the inability to infer self-selected or preferred intensity on acute well-being. For this,
directionality. Thus, prospective cohort studies which follow the patients exercised at light (RPE, 11), moderate (RPE, 13),
nondepressed people at baseline and measure PA and future or hard (RPE, 15) intensities, or chose the workload themselves
depression can provide a better indication of whether PA is across a 20-min cycling session. Interestingly, no differences
truly a modifiable risk factor for depression. In this regard, a were found between the self-selected/preferred intensity and
systematic review by Mammen and Faulkner (20) reviewed the closest intensity, suggesting that all intensities can equally
30 prospective cohort studies looking at whether PA con- promote acute well-being in people with MDD, regardless if
ferred protective effects on incident depression. In their narra- it is self-selected or determined.
tive review, the authors reported that 25 of the 30 studies
found that PA is prospectively associated with reduced inci- Effects of Exercise Training
dent depression. Mammen and Faulkner (20), however, did A large body of trials has been performed over the last
not perform a meta-analysis; they just counted the number 40 yr, evaluating the role of exercise as a therapy for depres-
of studies showing a protective effect and those that did not sion. These results have been summarized in several meta-
show it. Therefore, the consistency and magnitude of the pro- analyses. Despite this enormous body of evidence, some
tective effect were uncertain. The authors have not evaluated discussion on the magnitude of the antidepressant effect,
whether some other factors like adjustments for potential co- and even if the effect really occurs, are still present. A meta-
variates, potential cultural differences, and ages could modify analysis that we have performed in 2016 (24), including 25
the effects. Trying to update the literature and fill these gaps, studies and more than 1487 people with depression (757 ran-
we performed a meta-analysis including only prospective co- domized to exercise and 730 to control conditions) identified
horts evaluating the effects of PA and incident depression in a very large and significant antidepressant effect favors exercise
studies of at least 1 yr follow up. For that, we compared the (standardized mean difference [SMD], 0.98; 95% CI, 0.68–1.28;
incident depression in people with higher levels versus lower P < 0.001). When adjusted for potential publication bias

300 Volume 18  Number 8  August 2019 Exercise and MDD: Prevention and Treatment

Copyright © 2019 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 according to the Duval and Tweedie trim and fill technique, the to exercise. A further study analyzing data from the
effect was equal to 1.11 (95% CI, 0.79–1.43). Also, when we REGASSA trial identified that higher levels of IL-6 pre-
restricted the analyses to the studies identified as having a dicts greater response rates (34).
lower risk of bias, the effect remained significant (SMD, 2) Clinical: better global functioning, as well as more se-
0.88; 95% CI, 0.22–1.54; P = 0.009). This result is consistent vere physical symptoms, predict response to exercise.
with the direction of a meta-analysis published some months Later, data from the TREAD study revealed that atypi-
later that corroborated our findings (25) and with the overall
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

cal depression benefits more from exercise than other

analysis published by Krogh et al. (26). However, in the subtypes (35). Also, lower cardiorespiratory fitness is
Krogh et al. study (26), a subgroup analysis including only linked to a greater chance of nonresponse (36).
four trials that were considered of “low risk of bias” did not
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC4/OAVpDDa8KKGKV0Ymy+78= on 06/03/2024

3) Psychological: higher self-esteem and life satisfaction

find a significant effect of exercise (SMD, −0.11; 95% CI,
predict better outcomes (33).
−0.41 to 0.18, P = 0.45). Of note, two of the four trials in-
4) Social: social support was the only moderator/predictor
cluded in this analysis compared exercise versus a control
group that performed “light-intensity” exercise (27,28). Com- that was consistently associated with better outcomes (31).
paring “exercise versus exercise” is a significant limitation of 5) Composed: it is reasonable to expect that a better-
the analysis since exercise, even when it is of lower intensity, described group and limited group, including a greater
exhibits a significant reduction in depressive symptoms (29). number of identifiable traits or characteristics, may pres-
Therefore, an effective “control” group may mask the effects ent a greater chance of response. For example, when we
of a comparison intervention. This limitation is even more im- say that men respond more to exercise, in this group of
portant in this case since one of the included studies, the men, there will be subgroups that may respond more
DEMO trial, found greater increases on V̇O2 in the control while others will respond less, or some won’t respond
group than in the exercise group, suggesting that the physio- at all. Therefore, it is expected that models including
logical intensity of the control group was even greater in this more variables will have a better predictive value.
group than in the exercise group (27).
Some studies have evaluated whether two or more traits or
characteristics are linked to a greater response to exercise in-
Moderators of Response terventions. In this regard, data from the TREAD US study re-
Exercise can improve depressive symptoms in people with vealed that people with higher BDNF levels and higher BMI
depression. However, similar to other treatments, exercise is present greater response rates than people with lower BDNF
not a panacea and may not work equally for all. A seminal and BMI (37). Also, men, regardless of the family history of
study by Dunn et al. (30), the Depression Outcomes Study mental illness, and women, without family history of mental ill-
of Exercise, found a response rate of about 40% in depressed ness, are more likely to benefit from exercise (32). A reanalysis
people free from other treatments. These results are compara- of this study data, accounting a greater number of potential of
ble with more recent data found in the REGASSA trial, the variables (BDNF, IL-1B, depressive symptom severity, postex-
largest study with exercise and depression, which found a re- ercise positive affect, cardiorespiratory fitness, and IL-6),
sponse rate of about 50% (31). The remission rate (people achieved predictive values greater than 70%, suggesting that
who no longer meet criteria for MDD diagnosis) was evalu- people with higher BDNF levels, higher IL-1B levels, lower de-
ated in the Treatment with Exercise Augmentation for Depres- pressive symptoms, and higher positive affect to exercise are
sion (TREAD) study (32) using and adapting (16 kilocalories more likely to achieve positive outcomes from exercise (36).
per kilogram of weight spent in exercise per week) the pioneer-
ing idea of Andrea Dunn in determining the exercise dose Neurobiological Mediators of the
based on the energetic expenditure did find a remission rate Antidepressant Response
of about 28%. To maximize the benefits to the patients and The neurobiological mechanisms underpinning the antide-
to deal with patients’ and health professionals’ expectations, pressant effects of exercise are largely unclear. However, some
matching the “right patient for the right treatment,” or under- hypotheses involving inflammation, oxidative stress, and neu-
standing who are the patients that are more likely to benefit ronal regeneration are speculated (38).
the most regarding depressive symptoms reduction, is neces- Inflammation and oxidative stress markers are altered in
sary. For doing that, understanding the potential predictors people with depression. For example, IL-6 and IL-1B (39), in-
and moderators of the antidepressant effects of exercise is flammation markers, and thiobarbituric acid reactive species
required. (40) are increased in people with depression. Exercise training,
A previous systematic review identified some potential 1) bi- however, is able to promote increases in anti-inflammatory
ological, 2) clinical, 3) psychological, 4) social factors, as well and anti-oxidant enzymes, referred to as an hormesis response
as, 5) the interaction between two or more factors (composed) (38,41), and subsequently decrease IL-6 levels (34). This effect
that were associated with greater response/remission rates was demonstrated in the REGASSA trial, where decreases in
(33). Since the publication of this study, some studies have im- IL-6 serum levels were associated with reductions in depres-
proved the state of the art in the field. We will briefly mention sive symptoms (34).
the findings below: Substantial evidence demonstrates that people with depres-
sion have decreased levels of BDNF (42), a marker of neuronal
1) Biological: higher levels of brain-derived neurotrophic growth and plasticity. Potentially, these reduced levels of neuro-
factor (BDNF), interleukin (IL)-1B, and of tumor ne- nal regeneration are linked to a decreased volume and activity
crosis factor-alpha are associated with greater response of certain brain regions, including hippocampus, orbitofrontal

www.acsm-csmr.org Current Sports Medicine Reports 301

Copyright © 2019 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 cortex, anterior and posterior cingulate, insula, and temporal the chance of success of treatment and the subsequent adop-
lobes observed in people with depression (43). Exercise, in- tion and maintenance of exercise.
versely, can promote brain plasticity, increasing hippocam-
pus volume (44). However, there is not enough evidence on Exercise for Improving Physical Health of People
changes in brain volume due to regular exercise in people with Depression
with depression (38). People with mental illness have a reduced life expectancy of
about 10 yr when compared with the general population
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

(55). Much of this reduction is attributable to an increased

Prescription, Adherence, and Dropout to rate of cardiometabolic diseases in this population. For exam-
Exercise Interventions ple, people with MDD have a higher risk (relative risk [RR],
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC4/OAVpDDa8KKGKV0Ymy+78= on 06/03/2024

Starting and sustaining an exercise program is a challenge 1.36; 95% CI, 1.29–1.72) of having type II diabetes (56), met-
for any clinical population, and naturally, this also is true abolic syndrome (RR, 1.54; 95% CI, 1.21–1.97) (57), and of
for people with depression. For example, the dropout rates cardiovascular disease (OR, 2.08; 95% CI, 1.51–2.88) (9)
to exercise interventions is about 20% for people with type than age and sex-matched controls.
II diabetes mellitus (45), 29% for people with HIV (46), 26% Exercise and PA are key factors for preventing and manag-
for people with schizophrenia (47), and 17% for children with ing cardiovascular disease and associated mortality (58) in the
attention deficit hyperactivity disorder (48). The dropout rate general population, and this is not different for people with
in exercise trials for people with depression is not greater than MDD. Evidence has shown that PA and exercise should be
that for other clinical populations, being of about 18% (49). prescribed to people with MDD to promote cardiometabolic
Also, it is important to note that adherence to other treatments health, as shown in improvements on cardiovascular and
for depression is equally challenging. For example, dropout metabolic markers as fitness (59,60), high-density lipoprotein
rate from psychotherapies, in general, is about 19% (50) while (60), and weight (60), epicardial and subcutaneous adipose
the dropout rate from selective serotonin reuptake inhibitors tissue (61). For example, exercise can increase aerobic capac-
are about 26% and tricyclics about 28% (51). ity of about 3.05 mL·kg−1·min−1 from baseline to postinter-
Some strategies may be useful for clinicians and exercise vention in people with MDD. This increase is associated
professionals to help people with depression to engage in PA with significant reductions on the risk of premature mortality
and to prevent dropout. First, Vancampfort et al. (52) suggests and can help to reduce the mortality gap (62).
that autonomous motivation may “hold the key” to keep peo-
ple with mental illness active. Autonomous motivation is the Treatment Guidelines for Depression
motivation that leads someone to do something for its own The evidence of the use of exercise is substantial and grow-
sake, for example, finding exercise enjoyable or challenging. ing fast. Despite this substantial evidence, the incorporation of
Therefore, adapting exercise prescription for people with depres- exercise as a key component in treatment is inconstant. Some
sion should account for personal preferences and previous ex- guidelines have incorporated PA and exercise as recommended
periences in terms of making it the most enjoyable experience therapeutic strategies for depression (63–66) while others have
possible. In this line, Brand and Ekkekakis (53) have suggested not. Despite this acknowledgement, PA still appears to not re-
that self-selected intensities or intensities above the ventilatory ceive the deserved attention and its use in clinical practice is not
threshold can be used as an appropriate option for public of equitable value to the more dominant strategies such as
health promotion. Self-selected intensities and intensities above pharmacotherapy and psychotherapies (67). According to
the ventilatory threshold are linked to positive core affective va- Ekkekakis and Muri (67), this can be potentially attributable
lence, whereas higher intensities are usually linked to a negative to the lack of awareness, incredulity, skepticism, or even a re-
core affective valence in sedentary and low-active groups (53), luctance to the existent evidence. Addressing this issue and
which is the case for people with depression (19). It should be the current reliance on the two-pronged approach of talking
noted, however, that higher intensities have demonstrated therapies and medication is important in going forward.
greater effects on reducing depressive symptoms (24); therefore,
a progression of the exercise intensity should be considered. Conclusions
Second, some strategies have been shown useful in other clinical PA can confer protection from the development of depres-
samples and can be used by physicians and health professionals sion in children, adults, and older adults. These effects are ev-
to help people with depression to engage in regular PA (54). ident in all continents. Also, among people with depression,
For example, Green prescriptions constituted by written exer- exercise can be used for acutely managing symptoms. Also, a
cise prescription by health professionals, defined according to robust body of evidence from randomized controlled trials
patients’ state of change and individual goals and supported by demonstrates that exercise is effective in treating depression.
exercise professionals’ calls and face-to-face meetings have re- Exercise has multiple benefits to several domains of physi-
sulted in an increase in 34 min·wk−1 of leisure exercise over cal and mental health and should be promoted to everyone.
12 months in patients consulting general practitioners (54). However, the use of moderators/predictors (e.g., biological,
Lastly, supervision provided by trained exercise professionals, clinical, psychological, social) and composed in response
as such as physiotherapists, exercise physiologists, physical ed- should be considered to deal with patients’ and professionals’
ucators, and others are a protective factor for dropout in people expectations and to maximize success chance. Dropouts to ex-
with depression, showing the clear relevance of these profes- ercise are a challenge for all clinical populations, which is not
sionals in the field (49). Social support is a potential moderator different from people with depression. However, adherence
for symptom improvements from exercising (33), thus encour- imposes a challenge to all other treatments. To keep exercise
aging patients to exercise with friends or family may increase adherence, autonomous motivation may play a central role.

302 Volume 18  Number 8  August 2019 Exercise and MDD: Prevention and Treatment

Copyright © 2019 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 Social support can be critical, and the supervision of exercise 17. Stubbs B, Koyanagi A, Schuch FB, et al. Physical activity and depression: a large
cross-sectional, population-based study across 36 low- and middle-income coun-
professionals can increase the chance of adherence and success tries. Acta Psychiatr. Scand. 2016; 134:546–56.
to the treatment. 18. Stubbs B, Vancampfort D, Firth J, et al. Relationship between sedentary behav-
ior and depression: A mediation analysis of influential factors across the lifespan
among 42,469 people in low- and middle-income countries. J. Affect. Disord.
The present study was carried out with the support of the 2018; 229:231–8.
Coordination of Improvement of Higher Education Person- 19. Schuch F, Vancampfort D, Firth J, et al. Physical activity and sedentary behavior
nel - Brazil (CAPES). B.S. is supported by Health Education
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

in people with major depressive disorder: a systematic review and meta-analysis.

England and the National Institute for Health Research J. Affect. Disord. 2017; 210:139–50.
HEE/NIHR ICA Programme Clinical Lectureship (ICA-CL- 20. Mammen G, Faulkner G. Physical activity and the prevention of depression: a
systematic review of prospective studies. Am. J. Prev. Med. 2013; 45:649–57.
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC4/OAVpDDa8KKGKV0Ymy+78= on 06/03/2024

2017-03-001). B.S. also is supported by the National Institute

21. Schuch F, Vancampfort D, Firth J, et al. Physical activity and incident depres-
for Health Research (NIHR) Collaboration for Leadership in sion: A meta-analysis of prospective cohort studies. Am. J. Psychiatry. 2018;
Applied Health Research and Care South London (NIHR 175:631–48.
CLAHRC South London) at King’s College Hospital NHS 22. Meyer JD, Koltyn KF, Stegner AJ, et al. Influence of exercise intensity for
Foundation Trust. The views expressed in this publication improving depressed mood in depression: a dose–response study. Behav. Ther.
2016; 47:527–37.
are those of the author(s) and not necessarily those of the
23. Meyer JD, Ellingson LD, Koltyn KF, et al. Psychobiological responses to pre-
NHS, the National Institute for Health Research or the ferred and prescribed intensity exercise in major depressive disorder. Med. Sci.
Department of Health and Social Care. Sports Exerc. 2016; 48:2207–15.
24. Schuch FB, Vancampfort D, Richards J, et al. Exercise as a treatment for depres-
The authors declare no conflict of interest. sion: a meta-analysis adjusting for publication bias. J. Psychiatr. Res. 2016; 77:
42–51.
25. Kvam S, Kleppe CL, Nordhus IH, et al. Exercise as a treatment for depression: a
References meta-analysis. J. Affect. Disord. 2016; 202:67–86.
1. Andrade L, Caraveo-Anduaga JJ, Berglund P, et al. The epidemiology of major
26. Krogh J, Hjorthøj C, Speyer H, et al. Exercise for patients with major depression:
depressive episodes: results from the International Consortium of Psychiatric Ep-
a systematic review with meta-analysis and trial sequential analysis. BMJ Open.
idemiology (ICPE) Surveys. Int. J. Methods Psychiatr. Res. 2003; 12:3–21.
2017; 7:e014820.
2. Ayuso-Mateos JL, Vázquez-Barquero JL, Dowrick C, et al. Depressive disorders
in Europe: prevalence figures from the ODIN study. Br. J. Psychiatry. 2001; 179: 27. Krogh J, Saltin B, Gluud C, et al. The DEMO trial: a randomized, parallel-group,
308–16. observer-blinded clinical trial of strength versus aerobic versus relaxation train-
ing for patients with mild to moderate depression. J. Clin. Psychiatry. 2009; 70:
3. Murray CJ, Vos T, Lozano R, et al. Disability-adjusted life years (DALYs) for 790–800.
291 diseases and injuries in 21 regions, 1990–2010: a systematic analysis for
the Global Burden of Disease Study 2010. Lancet. 2012; 380:2197–223. 28. Krogh J, Videbech P, Thomsen C, et al. DEMO-II trial. Aerobic exercise versus
stretching exercise in patients with major depression—a randomised clinical
4. Chisholm D, Sweeny K, Sheehan P, et al. Scaling-up treatment of depression and trial. PLoS One. 2012; 7:e48316.
anxiety: a global return on investment analysis. Lancet Psychiatry. 2016; 3:
415–24. 29. Stubbs B, Vancampfort D, Rosenbaum S, et al. Challenges establishing the effi-
cacy of exercise as an antidepressant treatment: a systematic review and meta-
5. Association AP. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. analysis of control group responses in exercise randomised controlled trials.
Arlington (VA): American Psychiatric Association; 2013. Sports Med. 2016; 46:699–713.
6. Moussavi S, Chatterji S, Verdes E, et al. Depression, chronic diseases, and decre- 30. Dunn AL, Trivedi MH, Kampert JB, et al. Exercise treatment for depression: ef-
ments in health: results from the World Health Surveys. Lancet. 2007; 370: ficacy and dose response. Am. J. Prev. Med. 2005; 28:1–8.
851–8.
31. Hallgren M, Kraepelien M, Ojehagen A, et al. Physical exercise and internet-
7. Vancampfort D, Stubbs B, Mitchell AJ, et al. Risk of metabolic syndrome and its based cognitive-behavioural therapy in the treatment of depression: randomised
components in people with schizophrenia and related psychotic disorders, bipo- controlled trial. Br. J. Psychiatry. 2015; 207:227–34.
lar disorder and major depressive disorder: a systematic review and meta-
analysis. World Psychiatry. 2015; 14:339–47. 32. Trivedi MH, Greer TL, Church TS, et al. Exercise as an augmentation treatment
for nonremitted major depressive disorder: a randomized, parallel dose compar-
8. Vancampfort D, Correll CU, Galling B, et al. Diabetes mellitus in people with ison. J. Clin. Psychiatry. 2011; 72:677–84.
schizophrenia, bipolar disorder and major depressive disorder: a systematic re-
view and large scale meta-analysis. World Psychiatry. 2016; 15:166–74. 33. Schuch FB, Dunn AL, Kanitz AC, et al. Moderators of response in exercise treat-
ment for depression: a systematic review. J. Affect. Disord. 2016; 195:40–9.
9. Correll CU, Solmi M, Veronese N, et al. Prevalence, incidence and mortality
from cardiovascular disease in patients with pooled and specific severe mental ill- 34. Lavebratt C, Herring MP, Liu JJ, et al. Interleukin-6 and depressive symptom se-
ness: a large-scale meta-analysis of 3,211,768 patients and 113,383,368 con- verity in response to physical exercise. Psychiatry Res. 2017; 252:270–6.
trols. World Psychiatry. 2017; 16:163–80. 35. Rethorst CD, Tu J, Carmody TJ, et al. Atypical depressive symptoms as a predic-
10. Charlson FJ, Diminic S, Lund C, et al. Mental and substance use disorders in tor of treatment response to exercise in major depressive disorder. J. Affect.
Sub-Saharan Africa: predictions of epidemiological changes and mental health Disord. 2016; 200:156–8.
workforce requirements for the next 40 years. PLoS One. 2014; 9:e110208. 36. Rethorst CD, South CC, Rush AJ, et al. Prediction of treatment outcomes to ex-
11. Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and acceptabil- ercise in patients with nonremitted major depressive disorder. Depress. Anxiety.
ity of 21 antidepressant drugs for the acute treatment of adults with major de- 2017; 34:1116–22.
pressive disorder: a systematic review and network meta-analysis. Lancet. 37. Toups MS, Greer TL, Kurian BT, et al. Effects of serum brain derived neuro-
2018; 391:1357–66. trophic factor on exercise augmentation treatment of depression. J. Psychiatr.
12. Pigott HE. The STAR*D Trial: It is time to reexamine the clinical beliefs that Res. 2011; 45:1301–6.
guide the treatment of major depression. Can. J. Psychiatry. 2015; 60:9–13. 38. Schuch FB, Deslandes AC, Stubbs B, et al. Neurobiological effects of exercise on
13. Cuijpers P, Turner EH, Mohr DC, et al. Comparison of psychotherapies for major depressive disorder: a systematic review. Neurosci. Biobehav. Rev. 2016;
adult depression to pill placebo control groups: a meta-analysis. Psychol. Med. 61:1–11.
2014; 44:685–95. 39. Miller AH, Raison CL. The role of inflammation in depression: from evolutionary
14. de Oliveira GD, Oancea SC, Nucci LB, et al. The association between physical imperative to modern treatment target. Nat. Rev. Immunol. 2016; 16:22–34.
activity and depression among individuals residing in Brazil. Soc. Psychiatry 40. Sowa-Kućma M, Styczeń K, Siwek M, et al. Are there differences in lipid perox-
Psychiatr. Epidemiol. 2018; 53:373–83. idation and immune biomarkers between major depression and bipolar disorder:
15. Jung S, Lee S, Lee S, et al. Relationship between physical activity levels and effects of melancholia, atypical depression, severity of illness, episode number,
depressive symptoms in community-dwelling older Japanese adults. Geriatr. suicidal ideation and prior suicide attempts. Prog. Neuropsychopharmacol. Biol.
Gerontol. Int. 2018; 18:421–7. Psychiatry. 2018; 81:372–83.
16. Liu Y, Ozodiegwu ID, Yu Y, et al. An association of health behaviors with de- 41. Schuch FB, Vasconcelos-Moreno MP, Borowsky C, et al. The effects of exercise
pression and metabolic risks: data from 2007 to 2014 U.S. National Health on oxidative stress (TBARS) and BDNF in severely depressed inpatients. Eur.
and Nutrition Examination Survey. J. Affect. Disord. 2017; 217:190–6. Arch. Psychiatry Clin. Neurosci. 2014; 264:605–13.

www.acsm-csmr.org Current Sports Medicine Reports 303

Copyright © 2019 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 42. Polyakova M, Stuke K, Schuemberg K, et al. BDNF as a biomarker for successful 55. Chesney E, Goodwin GM, Fazel S. Risks of all-cause and suicide mortality in
treatment of mood disorders: a systematic & quantitative meta-analysis. J. Affect. mental disorders: a meta-review. World Psychiatry. 2014; 13:153–60.
Disord. 2015; 174:432–40. 56. Vancampfort D, Mitchell AJ, De Hert M, et al. Type 2 diabetes in patients with
43. Suh JS, Schneider MA, Minuzzi L, et al. Cortical thickness in major depressive major depressive disorder: a meta-analysis of prevalence estimates and predic-
disorder: a systematic review and meta-analysis. Prog. Neuropsychopharmacol. tors. Depress. Anxiety. 2015; 32:763–73.
Biol. Psychiatry. 2019; 88:287–302. 57. Vancampfort D, Correll CU, Wampers M, et al. Metabolic syndrome and met-
44. Firth J, Stubbs B, Vancampfort D, et al. Effect of aerobic exercise on hippocam- abolic abnormalities in patients with major depressive disorder: a meta-analysis
pal volume in humans: a systematic review and meta-analysis. Neuroimage. of prevalences and moderating variables. Psychol. Med. 2013; 44:2017–28.
2018; 166:230–8.
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

58. Naci H, Ioannidis JP. Comparative effectiveness of exercise and drug interven-
45. Umpierre D, Ribeiro PB, Kramer CK, et al. Physical activity advice only or struc- tions on mortality outcomes: metaepidemiological study. BMJ. 2013; 347:
tured exercise training and association with hba1c levels in type 2 diabetes: a sys- f5577.
tematic review and meta-analysis. JAMA. 2011; 305:1790–9. 59. Stubbs B, Rosenbaum S, Vancampfort D, et al. Exercise improves cardiorespira-
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC4/OAVpDDa8KKGKV0Ymy+78= on 06/03/2024

46. Vancampfort D, Mugisha J, Richards J, et al. Dropout from physical activity in- tory fitness in people with depression: a meta-analysis of randomized control tri-
terventions in people living with HIV: a systematic review and meta-analysis. als. J. Affect. Disord. 2016; 190:249–53.
AIDS Care. 2017; 29:636–43. 60. Kerling A, Tegtbur U, Gutzlaff E, et al. Effects of adjunctive exercise on physio-
47. Vancampfort D, Rosenbaum S, Schuch F, et al. Prevalence and predictors of logical and psychological parameters in depression: a randomized pilot trial. J.
treatment dropout from physical activity interventions in schizophrenia: a Affect. Disord. 2015; 177:1–6.
meta-analysis. Gen. Hosp. Psychiatry. 2016; 39:15–23. 61. Kahl KG, Kerling A, Tegtbur U, et al. Effects of additional exercise training on
48. Vancampfort D, Firth J, Schuch FB, et al. Dropout from physical activity interven- epicardial, intra-abdominal and subcutaneous adipose tissue in major depressive
tions in children and adolescents with attention deficit hyperactivity disorder: a sys- disorder: a randomized pilot study. J. Affect. Disord. 2016; 192:91–7.
tematic review and meta-analysis. Ment. Health and Phys. Act. 2016; 11:46–52. 62. Kodama S, Saito K, Tanaka S, et al. Cardiorespiratory fitness as a quantitative
49. Stubbs B, Vancampfort D, Rosenbaum S, et al. Dropout from exercise random- predictor of all-cause mortality and cardiovascular events in healthy men and
ized controlled trials among people with depression: a meta-analysis and meta women: a meta-analysis. JAMA. 2009; 301:2024–35.
regression. J. Affect. Disord. 2016; 190:457–66. 63. Ravindran AV, Balneaves LG, Faulkner G, et al. Canadian Network for Mood
50. Cooper AA, Conklin LR. Dropout from individual psychotherapy for major de- and Anxiety Treatments (CANMAT) 2016 Clinical Guidelines for the Manage-
pression: a meta-analysis of randomized clinical trials. Clin. Psychol. Rev. 2015; ment of Adults with Major Depressive Disorder: Section 5. Complementary and
40:57–65. Alternative Medicine Treatments. Can. J. Psychiatry. 2016; 61:576–87.
51. Undurraga J, Baldessarini RJ. Direct comparison of tricyclic and serotonin- 64. Malhi GS, Bassett D, Boyce P, et al. Royal Australian and New Zealand College
reuptake inhibitor antidepressants in randomized head-to-head trials in acute of Psychiatrists clinical practice guidelines for mood disorders. Aust. N. Z. J. Psy-
major depression: systematic review and meta-analysis. J. Psychopharmacol. chiatry. 2015; 49:1087–206.
2017; 31:1184–9. 65. Scottish Intercollegiate Guidelines Network. Non-pharmacological management
52. Vancampfort D, Stubbs B, Venigalla SK, et al. Adopting and maintaining phys- of depression in adults. [cited 2019 June 13]. Available from: http://sign.ac.uk/
ical activity behaviours in people with severe mental illness: the importance of pdf/qrg114.pdf.
autonomous motivation. Prev. Med. 2015; 81:216–20. 66. NICE: Depression: the treatment and management of depression in adults, 2009.
53. Brand R, Ekkekakis P. Affective-reflective theory of physical inactivity and exer- [cited 2019 June 13]. Available from: https://mentalhealthpartnerships.com/
cise. Ger. J. Exerc. Sport Res. 2018; 48:48–58. resource/nice-guidelines-cg90-depression-the-treatment-and-management-of-
54. Elley CR, Kerse N, Arroll B, et al. Effectiveness of counselling patients on phys- depression-in-adults/.
ical activity in general practice: cluster randomised controlled trial. Br. Med. J. 67. Ekkekakis P, Murri MB. Exercise as antidepressant treatment: time for the tran-
2003; 326:793. sition from trials to clinic? Gen. Hosp. Psychiatry. 2017; 49:1.

304 Volume 18  Number 8  August 2019 Exercise and MDD: Prevention and Treatment

Copyright © 2019 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.