Exercise and Depression


Major Depressive Disorder (MDD) presents a considerable challenge to public health in the United States. Estimates of lifetime prevalence are 22.9% in females and 15.1% in males and the condition is likely increasing among younger cohorts. Successful treatment does not seem to increase with rising costs of care (over $20 billion annually). A thorough review of existing literature suggests that exercise is a valid treatment for moderate and mild depression and comparable to antidepressants and Cognitive Behavioral Therapy (CBT). However, the precise mechanisms explaining the efficacy of exercise for depression have yet to be elucidated. It is likely the many of the same pathways are involved that coincide with those of antidepressants and CBT. Recent studies on the biomarkers of these relationships suffer from poor methodology making conclusions premature.


Major Depression Disorder (MDD) is a public health crisis in the United States. For example, the Blue Cross Blue Shield Health Index, “a unique measurement of America’s health that quantifies how more than 200 diseases and conditions impact longevity and quality of life”, sampled 41 million commercially insured Americans and reported that nine million Americans are currently diagnosed with MDD representing a 33% increase (47% for millennials) since 2013. The authors of the report comment that this figure is rivaled only by hypertension for overall impact on health and indicates about ten years of healthy life lost for affected men and women. (Blue Cross/Blue Shield Health Index, 2018). Further, employing data that did not distinguish between causes, The National Institute for Mental Health estimated that 16.2 million (6.7%) adults in the United States had at least one major depressive episode in 2016. The prevalence was higher among adult females (8.5%) compared to males (4.8%) and highest among individuals aged 18-25 (10.9%). Overall, 64% of depressed individuals had severe impairment for that year. (National Institute of Mental Health, n.d.) Additionally, of the 36, 309 adults who participated in the 2012-2013 National Epidemiologic Survey on Alcohol and Related Conditions III (NESARC-III), the 12-month and lifetime prevalence of MDD were 10.4% and 20.6%, respectively. (Hasin, DS, Sarvet, AL, Meyers, JL, Saha, TD, Ruan, WJ, Stohl, M, Grant, BF, 2018) Consistent with these findings, the US National Comorbidity Survey-Replication findings indicate an adult lifetime prevalence of Major Depressive Disorder (MDD) of 22.9% in females and 15.1% in males. Moreover, the lifetime risk among younger cohorts appears to be increasing. (Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, et al. 2003; Kessler R.C., Birnbaum H., Bromet E, Hwang I, Sampson N, Shahly V, 2010)

Investigators in the field, however, caution that the increasing rates may be due to methodological issues such as recall bias and changes in diagnostic criteria within retrospective studies and that there are often conflicting results in longitudinal studies. (Hidaka BH, 2012) Wittchen opines that depression within the younger cohorts, while more frequent, are typically of a shorter duration than that found in older individuals. Older individuals, he argues, tend to experience greater chronicity which may skew a cursory impression of the data. (Wittchen, H, 2010) Scott et. al further found that age related depression is not well correlated with physical ailments. (Scott KM, Von Korff M, Alonso J, et al., 2008) In general, however, evidence from longitudinal studies remain suggestive of an epidemic (Wittchen, H, 2010; Hidaka, BH, 2012) and MDD is considered the “leading cause of disability among adults in high-income countries” (Siu, 2016).

The condition also presents a tremendous economic cost. For example, in comparing economic costs of MDD between 2005 and 2010, Greenberg et al. found that the “incremental economic burden of individuals with MDD increased by 21.5% (from $173.2 billion to $210.5 billion, inflation-adjusted dollars)” with 62% of the total costs due to comorbid conditions. (Greenberg PE, Fournier AA, Sisitsky T, Pike CT, Kessler RC, 2015). The National Institute of Mental Health summarizes how attempts for treatment were made in 2016:

The NIMH defines Health professional as “any of the following types of medical doctors or other professionals: general practitioner or family doctor, other medical doctor (e.g., cardiologist, gynecologist, urologist), psychologist, psychiatrist, psychotherapist, social worker, counselor, occupational therapist, or other mental health professional (e.g., mental health nurse or other therapist where type is not specified).” (National Institute of Mental Health, n.d.) Hundreds of psychotherapeutic interventions exist for the treatment of depression and although the superiority of CBT is debated among investigators, CBT, with a medium effect size (d = .67) relative to a variety of control conditions ranging from the absence of treatment to non-specific controls, is perhaps the leading psychotherapeutic intervention. CBT effect sizes tend to be larger when the modality is compared to wait-list controls (d = 0.88) than when CBT is compared to care-as-usual (d = 0.38) or non-specific controls (d = 0.38). Translated into numbers needed to treat (NNT), this effect size corresponds to an NNT of 2.75. (Driessen, E, Hollon, SD, 2010) Dreisen identifies the critical importance of highly skilled therapists. For example, she states that for one large study “Ratings conducted by experts at the Beck Institute suggested that the less experienced cognitive therapists at Vanderbilt were not performing at the same level of competence as the more experienced cognitive therapists at Penn. Therefore, the Vanderbilt therapists were provided with additional training through the extra-mural training program at the Beck Institute during the early years of the trail.” (Driessen, E, Hollon, SD, 2010) This forces the question of whether the typical “therapist” is more akin to the Penn or early Vanderbilt practitioners as even being “somewhat” less experienced may have a large impact on outcomes.

Moreover, in a review by DeRubeis et. al, anti-depressant medications (ADMs) and CBT were found to be about as equally efficacious. The authors state that a “mega-analysis of data pooled across the major CT-ADM comparisons showed that the two types of treatment are comparably effective in severely depressed patients”. (DeRubeis, RJ, Siegle, GJ, Hollon, SD, 2008). Among the studies reviewed, one found the comparisons represented below:

Note, however, the rather large placebo effect at 8 weeks. If we may ascribe about half of the efficacy due to placebo effects in the comparison, the outcomes of each become less impressive.

To sum, MDD imparts a very significant human and economic burden. The best available treatments, ADMs and CBT, are moderately effective in comparison studies, are associated with significant costs (and in the case of the former, adverse reactions), and may have large placebo effects at play. Thus, more investigation into other or supplemental modalities is warranted, especially given a likely rise in lifetime prevalence. Research into these other possible modalities over last several decades has produced a rather large body of literature supporting exercise as a treatment for depression. Arguably, the two most prominent analyses were published by Cooney et. al in a 2013 Cochrane Review and by Schuch et. al in a 2016 meta-analysis which corrected for publication bias. Cited by 732 related articles including JAMA and BMJ, and as an update to a 2009 Cochran Review, Cooney et. al found that exercise is only moderately more effective than no therapy for reducing symptoms of depression, is no more effective than antidepressants for reducing symptoms of depression, is no more effective than psychological therapies for reducing symptoms of depression, and that the evidence about whether exercise for depression improves quality of life is inconclusive. (Cooney GM, Dwan K, Greig CA, Lawlor DA, Rimer J, Waugh FR, McMurdo M, Mead GE, 2013) However, Schuch et. al concluded much differently: “compared to non-active interventions, exercise has a large and significant antidepressant effect, and it would require over 1000 negative studies to nullify this result.” (Schuch, FB, Vancampfort, D, Richards, J, Rosenbaum, S, Ward, PB, and Stubbs, B, 2016). The purpose of this investigation is, therefore, to: 1) summarize and add insights to the case each of these investigators has made on the question of whether exercise is an appropriate treatment for depression and 2) review plausible mechanisms whereby exercise may contribute to the treatment of MDD.

Review of Efficacy

Large epidemiological studies dating back to the late 1980’s found a significant relationship between physical activity and mental health. In exemplum is the 1988 follow up study to the first National Health and Nutrition Examination Survey (NHANES I) examined 1,900 adult Americans and found that physical inactivity is a likely risk factor depression. (Farmer ME, Locke BZ, Moscicki EK, Dannenberg AL, Larson DB, Radloff LS, 1988) A decade later, in a study of 1,536 subjects, Weyer et al. reported a higher odds ratio (3.15) for depression among physically inactive individuals compared to regular exercisers. (Weyer S., 1992) The trend indicating a preventative and ameliorating role for depression continued into the early 2000’s. (Bui, K, Fletcher, A., 2000; Sale, C, Guppy, A, El-Sayed, M, 2000; Wyshak, G., 2001; Dunn, AL, Trivedi, MH, Kampert, JB, et al., 2005) Notably, Goodwin found in 2003 that in a sample size of 8,098 adults, physical activity significantly decreased the prevalence of both depression and anxiety. (Goodwin, RD, 2003; Carek, PJ, Laibstain, SE, Carek, SM, 2011) A series of reviews and meta-analyses in the last five years demonstrated that, in general, more recent investigations support the earlier findings. In 2013, Danielson et al. concluded that, in the nine studies of sufficient quality analyzed, “exercise appears to be beneficial in the treatment of depression when used in combination with medication”. (Danielsson, L, Noras, AM, Waern, M, & Carlsson, J, 2013).

Also publishing in 2013, Cooney et. al reviewed and performed a metanalysis on 35 trials (n=1,356) comparing exercise (defined according to guidelines set by the American College of Sports Medicine) with controls or no treatment. To normalize the data which employed different depression scales, his team combined the results using the Standard Mean Difference (SMD) where an SMD of .20, 0.50, and 0.80 represent small, medium, and large effects respectively. The SMDs were then converted into the 21 item, self-reporting Beck Depression Inventory (BDI) scores. The BDI questions are worth 3 points each for a maximum of 63 where scores above 30 are understood to represent severe depression. For the pooled 35 trials, the team found a medium reduction in depression, a pooled SMD of −0.62 [95% CI, −0.81 to −0.42], representing about a five-point reduction in BDI scores. When the six high quality studies were evaluated (n=464), the effect size dropped dramatically to a low SMD of −0.18 [95% CI, −0.47 to 0.11]. Further, seven studies (n = 189) found no difference between exercise and psychological therapy and four trials (n = 298) found no difference between exercise and antidepressant therapy. Though no studies provided head to head comparisons, resistance exercise was found to have a large effect (SMD, -1.03) whereas aerobic exercise was found to have a moderate effect (SMD, -0.55). According to the authors, the optimal parameters of exercise (intensity, frequency, and duration) are not known, nor is the relationship between the benefits of exercise and the severity of depression. They also identify difficulties in blinding subjects and small study sizes as important limitations. Cooney et. al, again, concluded that “The evidence about whether exercise for depression improves quality of life is inconclusive.” (Cooney GM, Dwan K, Greig CA, Lawlor DA, Rimer J, Waugh FR, McMurdo M, Mead GE, 2013) Most notably, Ekkekakis states of Cooney et al.: “updates of this review published over a period of only 12 years have resulted in the remarkable stepwise reduction of the pooled standardized mean difference by 44%, from -1.10 in 2001, to -0.82 in 2009, to -0.67 in 2012, to -0.62 in 2013.” (Ekkekakis, P, 2015).

In 2015, however, Ekkekakis published an exceptionally glaring, six pronged critique of the methodology employed by the Cochrane Review focusing on: 1) the choice of inclusion and exclusion criteria, 2) the uniform-versus-selective application of rules, 3) the rationale behind protocol changes, 4) the lesser known implications of the random-effects meta-analytic model, 5) the complexities involved in appraising the methodological quality of randomized controlled trials (RCTs), and 6) reporting errors. (Ekkekakis, P, 2015) The more salient points of Ekkekakis’ critique will be summarized here.

Ekkekakis argues that for both the 35 trials yielding a medium effect size and the six high quality trials yielding a small effect size, the level of heterogeneity was substantial with an I2 of 63% and 57% respectively and points out that mixing comparisons of different treatments with different comparators is “nonsensical” and renders a meta-analysis unhelpful. (Ekkekakis, P, 2015) For example, in six studies Cooney et. al compared exercise combined with either medication or psychotherapy to these two treatments alone. This does not represent controlled studies but a comparison of exercise to an established form of therapy and it is fallacious to assume the effect sizes may be added algebraically. That is, effect sizes of, say, (exercise + medication)- medication alone ≠ exercise. Medication may even hinder the effects of exercise on depression as several participants in one study noted. For instance, in animal studies, SSRI’s have been found to reduce locomotor activity and spontaneous running behavior and exercise is thought to have similar neurological and psychological mechanisms as anti-depressants and cognitive therapy. (Ekkekakis, P, 2015) Moreover, while Cooney et al. reportedly excluded all studies that had “no non-exercising comparison group”, six trials compared different forms of exercise and lacked a nonactive control. (Cooney GM, Dwan K, Greig CA, Lawlor DA, Rimer J, Waugh FR, McMurdo M, Mead GE, 2013) Examples of different forms of exercise include yoga, flexibility exercises, and contraction-relaxation exercise that rated 12 on a perceived exertion scale that maxed at 20. With the discovery of these errors of inclusion, Ekkekakis concludes that “of the studies included in the main analysis by Cooney et al. (2013), six of the eight effect sizes that mostly weaken the pooled SMD, and all five effect sizes favoring the so-called “control” groups, belong in the aforementioned categories (i.e., studies without control groups, studies with exercising comparison groups). Removing the six studies without control groups raises the pooled SMD to -0.66 (95% CI from -0.85 to -0.47)” while “removing the additional six studies with exercising comparison groups raises the pooled SMD further, to -0.72 (95% CI from -0.91 to -0.53)” (Ekkekakis, P, 2015) Doing so also reduces the heterogeneity in both cases. Further, of the six “high quality” studies, one did not have a control group at all and two had exercise comparators. This leaves three studies (two of which were placebo controlled and one employed health education as a control in an investigation of subjects 53-91 years old) with a pooled SMD of -0.33, 95% CI from -0.59 to -0.07, a significantly better outcome than the SMD of -.18 calculated by Cooney et. al. (Ekkekakis, P, 2015).

Ekkekakis also raises concerns about defining terms accurately. By employing the ACSM definition of exercise, studies may erroneously include physical activity as comparators that do not count as “exercise” by the ACSM definition. ACSM considers exercise as “a type of physical activity consisting of planned, structured, and repetitive bodily movement done to improve and/or maintain one or more components of physical fitness” whereas physical activity is “any bodily movement produced by the contraction of skeletal muscles that results in a substantial increase in caloric requirements over resting energy expenditure”. (ACSM, 2013) The difference proved to be substantial in outcomes and underscores the necessity of accurately defining terms for inclusion and exclusion as in this case activities such as walking, Tai Chi, and Yoga were not considered “exercise” but were presumably sufficient physical activities to trigger helpful biochemical cascades. According to Ekkekakis, “Inclusion of the excluded studies described in this section (i.e., assisted walking, tai-chi, qigong, yoga, postnatal depression) in the main analysis raises the pooled SMD to -0.77 (95% CI from -0.98 to -0.57). Including these studies while also excluding studies with questionable comparators (i.e., studies without control groups, studies with exercising comparison groups) results in a large pooled SMD of -0.90 (95% CI from -1.11 to -0.69)”. (Ekkekakis, P, 2015).

Such errors of inclusion and exclusion were not limited to exercise comparators but would also include the definition of depression itself. For example, Cooney et. al included a 2012 study by Blumenthal et. al (Blumenthal et. al, 2012) based on elevated BDI II scores (≥ 7), amounting to none or minimal, with less than half of the subjects meeting the criteria for depression. However, exercise is not as likely to lower depression in those who do not meet reasonable criteria for being depressed; it lowers the SMD significantly. But if one were to only consider those subjects who meet the criteria for depression, the SMD is raised from -0.67 (95% CI from -1.23 to -0.12) to -0.94 (95% CI from -1.82 to -0.07). (Ekkekakis, P, 2015)

With such criticisms in mind, Ekkekakis proposed the following changes to the Cochrane Review: (a) studies without a control group should be excluded; (b) studies with active treatments as comparators should be excluded or considered separately; (c) studies with exercising groups as comparators (e.g., stretching and toning, yoga) should be excluded; (d) studies of postnatal depression should be included since there is no scientific basis for their exclusion; and (e) studies of tai-chi, qigong, and yoga should be included as long as they satisfy the ACSM definition of “exercise”. If such rules were applied, he argues, the pooled SMD from the Cochrane Review would be raised from “medium” (-0.62, 95% CI from -0.81 to -0.42) to “large” (-0.90, 95% CI from -1.11 to -0.69). The effect size, he adds, would remain large even after removing the two studies with the strongest effects in favor of exercise. (Ekkekakis, P, 2015)

While recognizing the unjustified lowering of pooled SMDs in the Cochrane Review provided by Ekkekakis, Schuch et. al also note that a meta-regression on available studies has not been performed in about a decade and that no previous meta-analyses have adjusted for publication bias. His team therefore set out to address existing limitations in the literature by aiming: “(1) to establish the updated effects of exercise on depression comparing exercise versus non-active control groups, (2) to identify moderators through meta-regression analyses, including sample characteristics (sex, use of medication and severity of baseline symptoms) and exercise intervention variables (length of the trial, frequency) that could impact the effects of exercise on depression, (3) to investigate, through subgroup and sensitivity analyses, the magnitude of the effects of exercise considering study quality, group format, setting, intensity, type, supervision, presence of clinical co-morbidities, type of publication and diagnosis of MDD, (4) to assess the influence of publication bias on the reported effects of exercise on depression, and (5) to quantify the strength of the existing evidence by calculating the number of negative studies required to nullify the pooled ES of the analyses performed.” (Schuch, F. B., Vancampfort, D., Richards, J., Rosenbaum, S., Ward, P. B., & Stubbs, B., 2016). Such et. al performed a meta-analysis on subjects demonstrated to have depression (including dysthymia) based on validated measures such as DSM-IV, ICD, BDI, HAM-D, etc. The inclusionary criteria were RCTs that included a non-active control, published in peer reviewed journals or doctoral dissertations, and which wereinvestigating exercise “as planned, structured, repetitive and purposive physical activity, in the sense that improvement or maintenance of one or more components of physical fitness was an objective, in the active arm of the trial”. RCTs that employed yoga, tai chi or qi going, were excluded given that they involve such mind-body behavioral techniques. (Schuch, F. B., Vancampfort, D., Richards, J., Rosenbaum, S., Ward, P. B., & Stubbs, B., 2016) The team measured effect size with a SMD and 95% confidence intervals (CIs), performed a random effects meta-analysis to account for heterogeneity, evaluated a number of moderators, and assessed for publication bias. The search yielded the 35 studies examined in Cooney et. al (Cooney GM, Dwan K, Greig CA, Lawlor DA, Rimer J, Waugh FR, McMurdo M, Mead GE, 2013) as well as an additional 41 papers published papers appearing after the Cooney et. al review. Of these 76, 25 met all the inclusionary criteria and had sufficient data to perform the analysis, but only 4 were regarded as high quality. The findings from the 25 studies found a significant benefit of exercise on depression with an SMD of -1.11 and a failsafe number of 1057. Alternatively, benefits of -4.52 points (95% CI 2.03 to 7.01, p < 0.001) and of -6.46 points (95% CI 4.18 to 8.41, p < 0.001) were associated with the HAM-D and the BDI scales, respectively. Moreover, among the diagnosed MDD subjects, the SMDs of high (N=3) and low-quality studies (N=6) were SMDs of -1.133 and -1.176 respectively. The authors therefore concluded that “compared to non-active interventions, exercise has a large and significant antidepressant effect, and it would require over 1000 negative studies to nullify this result. Publication bias is evident in exercise RCTs, but this has largely resulted in an underestimation of the ES of exercise. Our novel ES, calculated adjusting for publication bias, confirms and strengthen the case that exercise is an evidence-based treatment for depression.” With regards to intensity, moderate exercise was found to be most beneficial for MDD subjects followed by vigorous then light intensity. Only aerobic exercise was examined in this subgroup. (Schuch, F. B., Vancampfort, D., Richards, J., Rosenbaum, S., Ward, P. B., & Stubbs, B., 2016)

Two recent meta- analyses were performed subsequent to the investigation by Schuch et al. The first was a 2016 meta-analysis by Kvam et al. which found that “physical exercise had a moderate to large significant effect on depression compared to control conditions (-.68)” and “exercise compared to no intervention yielded a large and significant effect size (-1.24).” (Kvam, S, Kleppe, C.L, Nordhus, IH, Hovland, A., 2016) The team also found that when compared to care as usual, exercise had a moderate effect of (-o.48). Compared to anti-depressant medications (ADMs) and psychotherapies, the difference was not significant, but when exercise was combined with medication, the difference was moderately in favor of exercise with an effect size of (-0.50). The authors concluded that exercise is an effective treatment for depression and that exercise may be “a viable adjunct treatment in combination with antidepressants.” (Kvam, S, Kleppe, C.L, Nordhus, IH, Hovland, A., 2016). Further, Krogh et al. analyzed thirty-five trials totaling 2,498 participants and the effect size was moderate (−0.66, 95% CI −0.86 to −0.46; p<0.001). However, when the team examined only studies deemed to be of sufficient quality, the effect size diminished to non-significance. (Krogh J, Hjorthøj C, Speyer H, et al., 2017). Krogh et al. cited the following sources of bias according to the Cochrane Handbook: “allocation sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessors, incomplete outcome data, selective outcome reporting, for-profit bias and other bias.” (Krogh J, Hjorthøj C, Speyer H, et al., 2017) More specifically, the authors state that “Sequence generation was adequate in 15/35 (43%), allocation concealment was adequate in 13/35 (37%) trials, blinding of participants and trial personnel was adequate in 0/35 (0%), blinded outcome assessment was performed in 16/35 (46%), low risk of bias in the ‘incomplete outcome data’ domain was found in 12/35 (34%) trials, selective outcome reporting domain was adequate in 31/35 (89%), for-profit bias domain was adequate in 19/35 (54%) and 25/35 (71%) were free of other bias. Accordingly, all trials were at high risk of bias.” (Krogh J, Hjorthøj C, Speyer H, et al., 2017) However, the authors’ assessment of bias was the same as that found in Cooney et al. (Cooney GM, Dwan K, Greig CA, Lawlor DA, Rimer J, Waugh FR, McMurdo M, Mead GE, 2013) and has largely been addressed by Schuch and Ekkekakis (Schuch, FB, Vancampfort, D, Richards, J, Rosenbaum, S, Ward, PB, Stubbs, B, 2016; Ekkekakis, P, 2015). Moreover, of the four studies that Krogh found to be adequately free from bias, two were published by himself violating his own rules of publication bias (Krogh J, Saltin B, Gluud C, et al., 2009; Krogh J, Videbech P, Thomsen C, et al., 2012) and may have significantly affected outcomes.

Plausible Mechanisms

While the recommendation of exercise for the treatment of depression is now well grounded, it remains unclear precisely why and the underlying mechanisms have not been conclusively elucidated. In general, investigators have directed their attention toward the putative causes of depression. That is, given that depression is associated with changes in the Hypothalamic Pituitary Axis (HPA), Brain Derived Neurotrophic Factor, the 5-HT System (i.e. serotonin), and psychoneuroimmunological changes as well as genetic and epigenetic conditions, it was reasonable to investigate exercise as a normalizer of this states. The model depicted below may be helpful, but one must be mindful that the theoretical demarcations are artificial given the high degree interconnectivity between the systems. Also, the physiological effects of exercise and the causes of depression are complicated, multifaceted and perhaps due to biological cascades not represented in the following model. However, this where is the lion’s share of research efforts have been placed:

To date, Schuch et. al produced the only systematic review examining the putative effects of the neurological biomarkers of exercise on MDD. (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016) Schuch et. al looked at both the acute and chronic neurological effects in four different categories: The Neuroendocrine Hypothesis, The Inflammatory Hypothesis, The Neurotrophic Hypothesis, The Oxidative Stress Hypothesis, and The Changes in Cortical Structure and Activity Hypothesis. In tabulated fashion:

Dysregulation of the HPA Axis has traditionally been thought to play a major role in MDD. (Frodl, T., 2016) Rising levels of cortisol during periods of high stress were thought to have a detrimental impact on limbic system function which governs emotional behavior. In particular, the hippocampus is especially rich in cortisol receptors and the hormone has been implicated in decreasing tissue volume. There is now strong evidence, however, that MDD is “not in any simple way a disorder of too much cortisol” but is now also associated with low available levels of cortisol, largely via a feedback loop glucocorticoid receptor insufficiency. Thus, depression is best thought of as being correlated with cortisol dysregulation in general rather than merely high levels. (Maletic, V, Raison, CL, 2017). Thus, though the analysis by Schuch et. al reviewed two studies examining the chronic effect of exercise on cortisol levels and found no significance, it must be kept in mind that the feedback loops are complicated, and more investigations are warranted to reach conclusions. As stress is considered to have a prominent role in the causal nexus of depression, investigating other physiological attempts at reducing the stress response may prove more fruitful. (Hackney AC, 2006) Ekkekakis cites several studies whereby ANP was demonstrated to reduce anxiety as its levels rise- having a dampening effect on the HPA Axis- and further showing that exercise raises ANP levels. (Ekkekakis, P, 2015) This, however, was the opposite finding for Schuch et. al. The two evaluated studies reported that exercise results in a large reduction in exercise levels of ANP (SMD = −1.22, p = 0.0002) and BNP (SMD = −0.88, p = 0.02). Theoretically, ANP and BNP levels should have risen in depressed individuals. Therefore, the authors may have confused a reduction in depression with a reduction in these biomarkers because they later state: “Exercise promotes a large acute effect in atrial natriuretic peptide in MDD” and “ANP plays a role in the inhibition of HPA axis activity”. (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016) Copeptin, a metabolite of Arginine vasopressin (also termed Antidiuretic Hormone) is a biomarker for stress and HPA activity (Katan et al., 2008). This biomarker was found by Schuch et. al to increase moderately in acute exercise and decrease moderately in chronic exercise and its relationship to exercise and depression remains unknown. Thus, within the boundaries set by Schuch et. al, ANP and BNP may be drivers behind any physiological effects of exercise on depression via stress reduction.

In this analysis, the serotonergic system did not prove itself to be a likely candidate to explain the effects of exercise on depression. One speculation was that L- Tryptophan is normally bound to albumin in the peripheral circulatory system and during acute exercise free fatty acids displace bound L- Tryptophan making it available to form serotonin in the brain. Additionally, a 2015 study demonstrated that PGC-1-alpha released during exercise enables the break down L-kynurenine, a biomarker linked to depression. (Agudelo, L, Femenía, T, Orhan, F, Porsmyr-Palmertz, M, Goiny, M, Martinez-Redondo, V, . . . Ruas, J, 2015). Schuch et. al reviewed one study (Hennings et al., 2013) where “no changes were found for kynurenin (p = 0.22), tryptophan (p = 0.6) and 5-hydroxyindoleacetic acid (p = 0.88) –precursors of serotonin – following one week of increased physical activity.” (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016) Nor were levels of prolactin, an indicator of serotonergic function, changed in the one study reviewed. (Krogh et al., 2010) These findings lead the authors to conclude “the present results do not support the hypothesis that exercise promotes acute or chronic effects on serotonergic system in people with MDD.” (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016)

The findings on the serotonergic system were impactful as both anti-depressants and exercise were thought to raise levels of brain-derived neurotrophic factor (BDNF) at least in part by a serotonergic mechanism. (Ekkekakis, P, 2015; Meyer, JD, Koltyn, KF, Stegner, AJ, Kim, J, & Cook, DB, 2016). When BDNF in turn binds to Tyrosine kinase (Trk) receptors, the Trk undergoes a conformational change (dimerization) and phosphorylation. This sets in motion a biochemical cascade that ends in gene transcription supporting neuron growth, differentiation, and survival. However, in the three studies reviewed by Schuch et. al, no increases of exercise brain-derived neurotrophic factor were found. The same held true for insulin like growth factor 1 (IGF-1) and vascular endothelial growth factor (VEGF). A moderate increase in Growth Hormone was found in the acute, resistance exercise studies and may play a helpful role in neurogenesis for subjects with MDD. (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016) However, this does not well explain how aerobic exercise seems to ameliorate depressive symptoms. The Neurotrophic Hypothesis, then, does not seem to have significant explanatory power.

The analysis also did not have any significant findings with regards to the Inflammatory Hypothesis. No significant changes were found in serum amyloid (SAA) (p = 0.58), soluble vascular cell adhesion molecule (S-VCAM) (p = 0.45) or soluble inter-cellular adhesion molecule (S-ICAM) (p = 0.42). Nor were any significant changes found in the pro-inflammatory cytokines (IFN-γ IL-2 IL-5, IL-6, IL-8, IL-12, TNF-α, C-reactive protein (CRP), or Neopterin) and anti-inflammatory cytokines (IL-4, IL-10, and IL-13). Thus, no conclusions may be drawn from the data. (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016)

Thiobarbituric acid reactive substances (TBARS) is a standard measurement of oxidative stress, lipid peroxidation more specifically. Exercise, while raising oxidative stress in the short term, produces long term adaptations to ameliorate oxidative insults to the body via nitric oxide production’s stimulation of BDNF synthesis. (Ekkekakis, P, 2015) Here, Schuch et. al did find a large effect for TBARS in MDD subjects who exercised (SMD = −1.08, p = 0.01). (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016)

In summary, ANP and reduction of oxidative stress seem to have the most explanatory power for exercise mediated effects on depression. No strongly notable effects for the other biomarkers were found in this review and the reason exercise abates depressive symptoms remains to be elucidated. Moreover, the authors were challenged by the weak methodology in the few studies available for analysis. For example, concomitant antidepressant use was widespread in the studies reviewed. Schuch et. al noted that “most of evaluated studies used samples with concomitant antidepressant use (n = 6/10) of neuroendocrine markers, (n = 3/5) of neurogenesis biomarkers, (n = 1/1) of oxidative stress, (n = 3/5) of inflammation, (n = 2/3) of cortical thickness and activity)”. (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016) This is a tremendous confounder given that exercise and antidepressants have common pathways (i.e. increased monoamine and BDNF production) and make relevant conclusions tenuous. (Ekkekakis, P, 2015) Moreover, all the studies on cytokines had low sample sizes and only one study performed on each. Nor was the definition of “exercise” tightly controlled. In exemplum, is the 2014 study by Krogh (Krogh et al., 2014) where participants in a hippocampal study only exercised once per week. Schuch et. al further identify a pattern: “An important issue is that two of the largest trials (Krogh et al.,2009, 2012) which evaluated multiple biological biomarkers (Krogh et al., 2010, 2011b, 2013, 2014a, 2014b) had an attendance rate of approximately one time per week and failed to find any antidepressant effect of exercise.” (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016) Another methodological concern is that, due in part to the blood brain barrier, physiological parameters in the peripheral system do not always model conditions in the CNS and most studies measure peripheral biomarkers. (Schuch, FB, Deslandes, AC, Stubbs, B, Gosmann, NP, da Silva, CB, & de Almeida Fleck, MP, 2016) Therefore, while this review seemed to upset the general paradigm outlined above, it would seem premature to jettison the body of literature that gave rise to these theories- especially given the limitations described above.

Conclusions and Recommendations

Due to the rising prevalence and increasing costs of management, Major Depressive Disorder in the United States may be characterized as a public health crisis. To date, Cognitive Behavioral Therapy (CBT) and Antidepressant medications are the treatments of choice. Research suggests the beneficial effect size of exercise on depression is comparable to CBT and ADMs with lower costs and, in the case of ADMs, less prone to adverse reactions. (Uher, R, Farmer, A, Henigsberg, N, Rietschel, M, Mors, O, Maier, W, . . . Aitchison, K, 2009).

Among numerous analyses and reviews finding exercise as having a substantial positive effect on depression outcomes, two remarkable investigations found otherwise. (Cooney GM, Dwan K, Greig CA, Lawlor DA, Rimer J, Waugh FR, McMurdo M, Mead GE, 2013; Krogh J, Hjorthøj C, Speyer H, et al., 2017) The 2013 study by Cooney et al. was adequately addressed in favor of exercise (Ekkekakis, P, 2015; Schuch, F. B., Vancampfort, D., Richards, J., Rosenbaum, S., Ward, P. B., & Stubbs, B., 2016) but no papers have been published critiquing the 2017 study by Krogh et al. who employed roughly the same methodology as Cooney et al. Taken as a whole, the entire body of literature suggests a strong effect in favor exercise and, again, comparable outcomes to CBT and ADMs. Therefore, it is concluded here that exercise should be strongly considered in treatment plans for at least mild to moderate depression. Notably, serious adverse effects do not seem to be significant across the body of literature. (Krogh J, Hjorthøj C, Speyer H, et al., 2017) Further, depression is a condition with multiple etiologies. Mixed results among studies may reflect a significant effect for depression of a given origin or subtype while not for others. For example, Medina et al. suggested tracing pro-inflammatory cytokines and deficits in BDNF as biomarkers to predict response to exercise outcomes. (Medina, JL, Jacquart J, Smits, JA, 2015) Schuch et al. also examined possible biological (2), clinical (3), psychological (2), social (2), and two composite individual moderators of response to exercise. His team found that only tumor necrosis factor alpha (TNF-α), brain derived neurotrophic factor (BDNF) levels, and the BDNF-BMI interaction were significant moderators. Family history of mental illness and gender were not significant moderators. (Schuch, FB, Dunn, AL, Kanitz, SC, Delevatti, RS, Fleck, MP, 2016) Increasing levels of BDNF is most likely optimized with low-intensity exercise. (Schuch, F, Morres, I, Ekkekakis, P, Rosenbaum, S, & Stubbs, B, 2017) Moreover, Rethrost et al. found that subjects suffering from atypical depression are more responsive to exercise than those with the melancholic variety. (Rethorst CD, Tu J, Carmody TJ, Greer TL, Trivedi MH, 2016)

The optimum frequency and intensity of exercise to maximize the beneficial effects remains unclear. However, in the study by Schuch et. al, moderate aerobic exercise was found to be most beneficial for MDD subjects followed by vigorous then light intensity in study participants. (Schuch, F. B., Vancampfort, D., Richards, J., Rosenbaum, S., Ward, P. B., & Stubbs, B., 2016) Additionally, Stanton et. al recommend “supervised aerobic exercise, undertaken three times weekly at moderate intensity for a minimum of nine weeks in the treatment of depression.” (Stanton, R, Reaburn, P, 2014). Both resistance training and aerobic exercise are recommended as each may associated different physiological cascades. Precisely why exercise is effective has not been elucidated. More high-quality studies need to be performed and include biomarkers not represented in these studies. These might include MTor (mechanistic target of rapamycin), a kinase regulator of protein synthesis which “might be important in the exercise-induced increase in hippocampal BDNF and neurogenesis.” which in turn may abate MDD. (Watson, K, Baar, K, 2014) The production of endogenous opiates and endocannabinoids during bouts of exercise may also play a role in the acute benefits. (Boecker, H, Sprenger, T, Spilker, ME, Henriksen, G, Koppenhoefer, M, Wagner, KJ, . . . Tolle, TR, 2008; Dietrich, A, McDaniel, WF, 2004; Sparling, P, Giuffrida, A, Piomelli, D, Rosskopf, L, Dietrich, A, 2003) Another emerging area of investigation is the beneficial role of epigenetic mechanisms on brain plasticity and neuronal flourishing that include the upregulation of BDNF. (Fernandes, J, Arida, RM, Gomez-Pinilla, F, 2017).



American College of Sports Medicine. (2013). ACSM’s guidelines for exercise testing and prescription (9th ed.). Philadelphia, PA: Lippincott Williams & Wilkins.

Agudelo, L., Femenía, T., Orhan, F., Porsmyr-Palmertz, M., Goiny, M., Martinez-Redondo, V., . . . Ruas, J. (2015). Skeletal Muscle PGC-1α1 Modulates Kynurenine Metabolism and Mediates Resilience to Stress-Induced Depression. Cell, 160(1-2), 351. doi:10.1016/j.cell.2014.12.025

Archer T, Josefsson T, Lindwall M. 2014. Effects of physical exercise on depressive symptoms and biomarkers in depression. CNS & Neurological Disorders 13(10):10-15

Blue Cross/Blue Shield. (Ed.). (2018, May 10). Major Depression: The Impact on Overall Health. Retrieved May 20, 2018, from https://www.bcbs.com/the-health-of-america/reports/major-depression-the-impact-overall-health

Blumenthal JA, Doraiswamy PM. Exercise to Combat Depression. JAMA. 2014;312(20):2166–2167. doi:10.1001/jama.2014.14334

Blumenthal, J. A., Babyak, M. A., Doraiswamy, P. M., Watkins, L., Hoffman, B. M., Barbour, K. A., . . . Sherwood, A. (2007). Exercise and Pharmacotherapy in the Treatment of Major Depressive Disorder. Psychosomatic Medicine, 69(7), 587-596.

Blumenthal, J. A., Sherwood, A., Babyak, M. A., Watkins, L. L., Smith, P. J.,Hoffman, B. M., et al. (2012). Exercise and pharmacological treatment of depressive symptoms in patients with coronary heart disease: results from the UPBEAT (understanding the prognostic benefits of exercise and antidepressant therapy) study. Journal of the American College of Cardiology, 60(12), 1053e1063

Bui K, Fletcher A. Common mood and anxiety states: Gender differences in the protective effect of physical activity. Social Psychological and Psychiatric Epidemiology 2000; 35:8-35.

Boecker, H., On the emerging role of neuroimaging in determining functional and structural brain integrity induced by physical exercise: impact for predictive, preventive, and personalized medicine, EPMA Journal (2011) 2:277–285

Boecker, H., Sprenger, T., Spilker, M. E., Henriksen, G., Koppenhoefer, M., Wagner, K. J., . . . Tolle, T. R. (2008). The Runners High: Opioidergic Mechanisms in the Human Brain. Cerebral Cortex, 18(11), 2523-2531. doi:10.1093/cercor/bhn013

Carek, P. J., Laibstain, S. E., & Care, S. M. (2011). Exercise for the treatment of depression and anxiety. International Journal of Psychiatry in Medicine, 41(1), 15-28. doi:10.2190/PM.41.1.c

Carey, N. (2012). The epigenetics revolution: How modern biology is rewriting our understanding of genetics, disease, and inheritance. New York: Columbia University Press.

Chen, C., Nakagawa, S., An, Y., Ito, K., Kitaichi, Y., Kusumi, I., (2017) The exercise-glucocorticoid paradox: How exercise is beneficial to cognition, mood, and the brain while increasing glucocorticoid levels Frontiers in Neuroendocrinology, Volume 44, 2017, pp. 83-102

Chen, C., Nakagawa, S., Kitaichi, Y., An, Y., Omiya, Y., Song, N., & … Kusumi, I. (2016). The role of medial prefrontal corticosterone and dopamine in the antidepressant-like effect of exercise. Psychoneuroendocrinology, 691-9. doi:10.1016/j.psyneuen.2016.03.008

Cooney G, Dwan K, Mead G. (2014) Exercise for Depression. JAMA. 2014;311(23):2432–2433. doi:10.1001/jama.2014.4930

Cooney GM, Dwan K, Greig CA, Lawlor DA, Rimer J, Waugh FR, McMurdo M, Mead GE. (2013) Exercise for depression. Cochrane Database of Systematic Reviews 2013, Issue 9. Art. No.: CD004366. DOI: 10.1002/14651858.CD004366.pub6.

Danielsson, L., Noras, A. M., Waern, M., & Carlsson, J. (2013). Exercise in the treatment of major depression: A systematic review grading the quality of evidence. Physiotherapy Theory & Practice, 29(8), 573-585. doi:10.3109/09593985.2013.774452

DeRubeis RJ, et al. Cognitive therapy vs. medications in the treatment of moderate to severe depression. Arch Gen Psychiatry. 2005;62:409–416.

DeRubeis, R. J., Siegle, G. J., & Hollon, S. D. (2008). Cognitive therapy vs. medications for depression: Treatment outcomes and neural mechanisms. Nature Reviews. Neuroscience, 9(10), 788–796.

Diagnostic and statistical manual of mental disorders: DSM-5. (5th ed.). (2013). Washington, D.C.: American Psychiatric Association

Dietrich, A. (2004). Endocannabinoids and exercise. British Journal of Sports Medicine, 38(5), 536-541. doi:10.1136/bjsm.2004.011718

Dietrich, A, McDaniel, WF (2004) Endocannabinoids and exercise   Br J Sports Med 2004;38: 536–541. doi: 10.1136/bjsm.2004.011718

Ding, Q., Ying, Z., & Gómez-Pinilla, F. (2011). Exercise influences hippocampal plasticity by modulating brain-derived neurotrophic factor processing. Neuroscience, 192, 773-780. doi:10.1016/j.neuroscience.2011.06.032

Driessen, E., Hollon, S. D. (2010). Cognitive Behavioral Therapy for Mood Disorders: Efficacy, Moderators and Mediators. The Psychiatric Clinics of North America, 33(3), 537–555. http://doi.org/10.1016/j.psc.2010.04.005

Driessen, E., Smits, N., Dekker, J. J., Peen, J., Don, F. J., Kool, S., . . . Van, H. L. (2016). Differential efficacy of cognitive behavioral therapy and psychodynamic therapy for major depression: A study of prescriptive factors. Psychological Medicine, 46(04), 731-744. doi:10.1017/s0033291715001853

Ekkekakis, P. (2015). Routledge handbook of physical activity and mental health. London: Routledge.

Ekkekakis, P., (2015) Honey, I shrunk the pooled SMD! Guide to critical appraisal of systematic reviews and meta-analyses using the Cochrane review on exercise for depression as example. Ment. Health Phys. Act 8, 21e36.

Elkin I, Shea MT, Watkins JT, et al. Treatment of Depression Collaborative Research Program: general effectiveness of treatments. Arch Gen Psychiatry. 1989;46(11):971–982.

Eyre, H., & Baune, B. T. (2012). Neuroimmunological effects of physical exercise in depression. Brain, Behavior, And Immunity, 26(2), 251-266. doi:10.1016/j.bbi.2011.09.015

Farmer ME, Locke BZ, Moscicki EK, Dannenberg AL, Larson DB, Radloff LS. Physical activity and depressive symptoms: The NHANES I epidemiologic follow-up study. American Journal of Epidemiology 1988; 128: 1340-1351.

Fernandes, J., Arida, R. M., & Gomez-Pinilla, F. (2017). Physical exercise as an epigenetic modulator of brain plasticity and cognition. Neuroscience & Biobehavioral Reviews, 80, 443-456. doi:10.1016/j.neubiorev.2017.06.012

Frodl, T. (2016). Systems neuroscience in depression. Amsterdam: Elsevier, Academic Press. P.183

Greenberg PE, Fournier AA, Sisitsky T, Pike CT, Kessler RC (2015). The economic burden of adults with major depressive disorder in the United States (2005 and 2010). Journal of Clinical Psychiatry 76, 155–162.

Goodwin RD. Association between physical activity and mental disorders among adults in the United States. Preventive Medicine 2003; 36: 698-703.

Hackney AC. (2006) Stress and the neuroendocrine system: the role of exercise as a stressor and modifier of stress. Expert Rev Endocrinol Metab. 2006;1(6):783–792. doi: 10.1586/17446651.1.6.783.

Hallgren, M., Herring, M. P., Owen, N., Dunstan, D., Ekblom, Ö., Helgadottir, B., & … Forsell, Y. (2016). Exercise, physical activity, and sedentary behavior in the treatment of depression: Broadening the scientific perspectives and clinical opportunities. Frontiers in Psychiatry, 7

Hasin, D.S., Sarvet, A.L., Meyers, J.L., Saha, T.D., Ruan, W.J., Stohl, M, Grant, B, F., (2018) Epidemiology of Adult DSM-5 Major Depressive Disorder and Its Specifiers in the United States. JAMA Psychiatry. Published online February 14, 2018. doi:10.1001/jamapsychiatry.2017.4602

He, S.B., Tang, W.G., Tang WJ, Kao XL, Zhang CG, Wong XT. (2012) Exercise intervention may prevent depression. Int J Sports Med 2012;33:525-530

Hearing C, Chang W, Szuhany K, Deckersbach T, Nierenberg A, Sylvia LG. (2016)

Physical exercise for treatment of mood disorders: a critical review. Current Behavioral Neuroscience Reports. 2016;3(4):350–9.

Hidaka B. H. (2012). Depression as a disease of modernity: explanations for increasing prevalence. J. Affect. Disord. 140, 205–214. 10.1016/j.jad.2011.12.036

Hill, A. S., Sahay, A., & Hen, R. (2015). Increasing Adult Hippocampal Neurogenesis is Sufficient to Reduce Anxiety and Depression-Like Behaviors. Neuropsychopharmacology, 40(10), 2368-2378. doi:10.1038/npp.2015.85

Josefsson, T., Lindwall, M., & Archer, T. (2014). Physical exercise intervention in depressive disorders: Meta-analysis and systematic review. Scandinavian Journal of Medicine & Science in Sports, 24(2), 259-272.

Kandel, E. R., Schwartz, J. H., Jessell, T. M., Siegelbaum, S. A., & Hudspeth, A. J. (2013). Principles of Neural Science. New York: McGraw Hill.

Katan, M., Morgenthaler, N., Widmer, I., Puder, J.J., Konig, C., Muller, B., Christ-Crain, M., (2008) Copeptin, a stable peptide derived from the vasopressin precursor, correlates with the individual stress level. Neuroendocrinol Lett 29, 341–346.

Kerling, A., Kück, M., Tegtbur, U., Grams, L., Weber-Spickschen, S., Hanke, A., Stubbs, B., Kahl, K.G., Exercise increases serum brain-derived neurotrophic factor in patients with major depressive disorder, Journal of Affective Disorders, Volume 215, 2017, pp. 152-155

Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, et al. (2003) The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R) JAMA. 2003; 289: 3095–3105.

Kessler, R. C. (2012). The Costs of Depression. The Psychiatric Clinics of North America, 35(1), 1–14. http://doi.org/10.1016/j.psc.2011.11.005

Kessler, R. C., Birnbaum, H., Bromet, E., Hwang, I., Sampson, N., & Shahly, V. (2010). Age Differences in Major depression: Results from the National Comorbidity Surveys Replication (NCS-R). Psychological Medicine, 40(2), 225.

Krogh, J., Rostrup, E., Thomsen, C., Elfving, B., Videbech, P., Nordentoft, M., (2014) The effect of exercise on hippocampal volume and neurotrophines in patients with major depression—a randomized clinical trial. J Affect Disord 165,24–30.

Krogh J, Hjorthøj C, Speyer H, et al. (2017) Exercise for patients with major depression: a systematic review with meta-analysis and trial sequential analysis. BMJ Open 2017; 7:e014820. doi: 10.1136/bmjopen-2016-014820

Kvam, S., Kleppe, C. L., Nordhus, I. H., & Hovland, A. (2016). Review article: Exercise as a treatment for depression: A meta-analysis. Journal of Affective Disorders, 20267-86. doi: 10.1016/j.jad.2016.03.063

Loftis JM, Huckans M, Morasco BJ. (2010) Neuroimmune mechanisms of cytokine-induced depression: current theories and novel treatment strategies. Neurobiol Dis 37: 519–533.

Loughridge, A. B., Greenwood, B. N., Day, H. W., McQueen, M. B., & Fleshner, M. (2013). Microarray analyses reveal novel targets of exercise-induced stress resistance in the dorsal raphe nucleus. Frontiers in Behavioral Neuroscience, 7doi:10.3389/fnbeh.2013.00037

Maass, A., Duzel, S., Brigadski, T., Goerke, M., Becke, A., Sobieray, U., Neumann, K., Lovden, M., Lindenberger, U., Backman, L., Braun-Dullaeus, R., Ahrens, D., Heinze, H.J., Muller, N.G., Lessmann, V., Sendtner, M., Duzel, E., 2016. Relationships of peripheral IGF-1, VEGF and BDNF levels to exercise-related changes in memory, hippocampal perfusion and volumes in older adults. Neuroimage 131, 142–154.

Maletic, V., & Raison, C. L. (2017). The new mind-body science of depression. New York: W.W Norton & Company.

Medina J. L., Jacquart J., Smits J. A. (2015). Optimizing the exercise prescription for depression: the search for biomarkers of response. Curr. Opin. Psychol. 4 43–47. 10.1016/j.copsyc.2015.02.003

Meyer, J. D., Koltyn, K. F., Stegner, A. J., Kim, J., & Cook, D. B. (2016). Influence of Exercise Intensity for Improving Depressed Mood in Depression: A Dose-Response Study. Behavior Therapy, 47(4), 527-537. doi: 10.1016/j.beth.2016.04.003

Meyer, J. D., Koltyn, K. F., Stegner, A. J., Kim, J., & Cook, D. B. (2016). Relationships between serum BDNF and the antidepressant effect of acute exercise in depressed women. Psychoneuroendocrinology, 74286-294. doi: 10.1016/j.psyneuen.2016.09.022

Miller, A. H., & Raison, C. L. (2016). The role of inflammation in depression: From evolutionary imperative to modern treatment target. Nature Reviews Immunology, 16(1), 22-34. doi:10.1038/nri.2015.5

National Institute of Mental Health, Major Depression. (n.d.). Retrieved March 16, 2018, from https://www.nimh.nih.gov/health/statistics/major-depression.shtml#part_155029

Rethorst CD, Tu J, Carmody TJ, Greer TL, Trivedi MH, (2016) Atypical Depressive Symptoms as A Predictor of Treatment Response to Exercise in Major Depressive Disorder. J Affect Disord 2016; 200:156–158.

Rippe, J. M. (2013). Lifestyle medicine. Boca Raton, FL: CRC Press.

Schuch, F. B. (2014). Progress in the study of the effects of exercise on affective and anxiety disorders. Frontiers in Psychiatry, 5

Schuch, F. B., Deslandes, A. C., Stubbs, B., Gosmann, N. P., da Silva, C. B., & de Almeida Fleck, M. P. (2016). Neurobiological effects of exercise on major depressive disorder: A systematic review. Neuroscience and Biobehavioral Reviews, 611-11. doi: 10.1016/j.neubiorev.2015.11.012

Schuch, F.B., Dunn, A.L., Kanitz, S.C., Delevatti, R.S., Fleck, M.P., (2016) Moderators of response in exercise treatment for depression: A systematic review, Journal of Affective Disorders, Volume 195, 2016, pp. 40-49

Schuch, F., Morres, I., Ekkekakis, P., Rosenbaum, S., & Stubbs, B. (2017). Exercise works for depression: Bridging the implementation gap and making exercise a core component of treatment. Acta Neuropsychiatrica, 29(2), 124-126. doi:10.1017/neu.2017.1

Schuch, F. B., Vancampfort, D., Richards, J., Rosenbaum, S., Ward, P. B., & Stubbs, B. (2016). Exercise as a treatment for depression: A meta-analysis adjusting for publication bias. Journal of Psychiatric Research, 7742-51. doi: 10.1016/j.jpsychires.2016.02.023

Scott KM, Von Korff M, Alonso J, et al. (2008) Age patterns in the prevalence of

DSM-IV depressive/anxiety disorders with and without physical co-morbidity.Psychol Med. 2008; 38:1659-1669.

Silveira H, Moraes H, Oliveira N, Coutinho ES, Laks J, Deslandes A. Physical exercise and clinically depressed patients: a systematic review and meta-analysis. Neuropsychobiology 2013; 67(2): 61-8. [http://dx.doi.org/10.1159/000345160] [PMID: 23295766]

Siu AL, Screening for Depression in Adults US Preventive Services Task Force Recommendation Statement. JAMA. 2016;315(4):380–387. doi:10.1001/jama.2015.18392

Sparling, P. B.; Giuffrida, A.; Piomelli, D.; Rosskopf, L.; Dietrich, A. Exercise activates the endocannabinoid system Neuroreport: 2 December 2003 – Volume 14 – Issue 17 – pp 2209-2211

Stanton, R., & Reaburn, P. (2014). Exercise and the treatment of depression: A review of the exercise program variables. Journal of Science and Medicine in Sport, 17(2), 177-182. doi:10.1016/j.jsams.2013.03.010

Sweatt, J. D. (2013). Epigenetic regulation in the nervous system: basic mechanisms and clinical impact. London: Academic Press

Szuhany, K.L., Bugatti, M., Otto, M.W., 2015. A meta-analytic review of the effects of exercise on brain-derived neurotrophic factor. J. Psychiatr. Res. 60, 56–64. http://dx.doi.org/10.1016/j.jpsychires.2014.10.003.

Uher, R., Farmer, A., Henigsberg, N., Rietschel, M., Mors, O., Maier, W., . . . Aitchison, K. (2009). Adverse reactions to antidepressants. British Journal of Psychiatry, 195(3), 202-210. doi:10.1192/bjp.bp.108.061960

Watson, K., Baar, K., (2016) Exercise as a treatment for depression: A meta-analysis Journal of Affective Disorders, Volume 202, 2016, pp. 67-86

Watson, K., Baar, K. (2014) mTOR and the health benefits of exercise. Seminars in Cell & Developmental Biology, 36130-139.

Wegner M, Helmich I, Machado S, Nardi AE, Arias-Carrion O, Budde H. (2014) Effects of exercise on anxiety and depression disorders: review of meta- analyses and neurobiological mechanisms. CNS Neurol Disord Drug Targets 2014: 13: 1002–1014

Weyer S. Physical inactivity and depression in the community. International Journal of Sports Medicine 1992; 13: 492-496.

Wittchen, H. (2010, October 04). The timing of depression: An epidemiological perspective. Retrieved March 16, 2018, from https://www.medicographia.com/2010/10/the-timing-of-depression-an-epidemiological-perspective/

Zheng, H., Liu, Y., Li, W., Yang, B., Chen, D., Wang, X., & … Halberg, F. (2006). Beneficial effects of exercise and its molecular mechanisms on depression in rats. Behavioural Brain Research, 168(1), 47-55. doi: 10.1016/j.bbr.2005.10.007