Exercise: A Prescription for Depression

Photo by Andy Beales | Unsplash.com
Photo by Andy Beales | Unsplash.com

Exercise is the adult equivalent of the toddler’s broccoli. The more we are told that exercise is good for us, the more entrenched we become in our corporeal stagnation. There hardly exists a form of media that does not participate in the admonishment. Even in our local grocery store, no sooner have we filled our carts with delicious calories than we are confronted by displays of magazines at checkout imploring us to burn off the still unpaid-for calories through various novel calisthenic routines.

Why then, you might ask, have we decided to write yet another article about exercise given our rather sarcastic introduction? Our goal is not to convince the reader to begin an exercise program (although this would be a happy, albeit unintended, consequence). We hope instead to approach exercise from an angle seldom examined.

Too often our cultural trappings muddy the proverbial waters, obscuring a clear discussion of exercise and diet. We either explicitly or implicitly focus only on the external benefits of exercise in regards to our physical appearance. And when we do discuss the internal psychological benefits of exercise we frequently fall into the circular trap of attributing psychological benefits to the changed external appearance (“Think how confident you’ll feel when you can fit into a size x!”).

Today we hope to discuss the internal science of exercise. We will start by reviewing the current body of research regarding the general medical benefits of exercise for the body. Then we will turn to the less-discussed benefits for the mind. It must be noted that separating mind and body creates a false dichotomy that exists only at the level of our current scientific understanding, but we will utilize this semantic distinction for conceptual purposes nonetheless.

Photo by Kalen Emsley | Unsplash.com
Photo by Kalen Emsley | Unsplash.com

First, let’s review exercise’s benefits for the body.

Individuals who exercise a total of 7 hours per week have a premature mortality rate 40% lower than those who exercise less than 30 minutes per week. Physical activity also appears to reduce your risk for colon and breast cancers. Furthermore, there is evolving evidence that physical activity may also reduce your risk for endometrial and lung cancers.1–3

Research also suggests that health benefits may be appreciated from even modest exercise programs. As little as 2.5 hours of exercise per week significantly reduces your risk of type 2 diabetes and cardiovascular disease. When it comes to exercise, half a loaf really is better than none. In fact physical inactivity is estimated to cause 1 in 25 deaths worldwide each year.1–3

And yet despite all that is known about the health benefits of exercise, a little more than 50% of Americans do not meet the current CDC recommendations of 2.5 hours of moderate-intensity (50-70% maximal heart rate) or 1.25 hours of vigorous intensity (70-85% maximal heart rate) exercise per week.1

Maximal heart rate can be calculated by taking 208 – 0.7 x age (an older, unvalidated version of this equation used 220 as the base).4 As an example, a 30-year old’s maximal heart rate is calculated to be 187 beats per minute (“bpm”). This means that in our 30-year old example, a moderate-intensity activity should achieve a heart rate of at least 94 bpm while a vigorous-intensity exercise should aim for a target of at least 131 bpm.

We will return to the concept of maximal heart rate and exercise intensity later in our discussion of exercise and the mind. But before we do we have to make a brief detour into statistical territory to prepare us for the ensuing analysis.

An effect size, more specifically a standardized mean difference, is a statistical value frequently employed in meta-analytic research to quantify the number of standard deviations difference between the control and the treatment groups.5 For example, an effect size of 0.9 in a study of a given treatment for depression would indicate that the average improvement in depression scores among the treatment group was 0.9 standard deviations above the average improvement in the control group.

For the non-statistician, it is challenging to make meaningful interpretations of a standardized mean difference so statisticians have developed various methods to aid in the interpretation of effect sizes. For example, Dr. Jacob Cohen categorized effect sizes of 0.2-0.49 as “small,” 0.5-0.79 as “medium,” and ≥0.8 as “large.”5

Fritz CO, Morris PE, Richler JJ. Effect size estimates: Current use, calculations, and interpretation. Journal of Experimental Psychology: General. 2012;141(1):2-18.
Fritz CO, Morris PE, Richler JJ. Effect size estimates: Current use, calculations, and interpretation. Journal of Experimental Psychology: General. 2012;141(1):2-18.

Another method of interpreting effect sizes is to utilize equivalent probability of superiority (“PS”) values.6 Returning to our example of depression treatment, a PS would describe the probability that a randomly selected individual from the treatment group would have a superior reduction in depression score when compared to the control group. For our example effect size of 0.9 the PS would be 74%,6 indicating that 74% of the time a randomly selected individual from the treatment group would have a superior reduction in depression scores when compared to the control group.

In other words, if an individual were to receive the studied treatment, 74% of the time he or she would appreciate a greater reduction in depression score than if they received no treatment in the control group. Importantly, this would not preclude improvement in depressive symptomatology occurring in the control group. It is important to be aware that up to 40% of patient response in antidepressant medication trials is derived from placebo effects.7 Before we conclude our introduction to PS, let us also note that a PS of 50% would indicate that there was no difference between the treatment group and the control group.

We will utilize both Dr. Cohen’s small-medium-large effect categorical distinctions as well as PS values to help us interpret the proceeding literature.

A meta-analysis examining well-controlled studies of exercise as an intervention for clinical depression found an effect size of 1.39 when compared to nonactive controls. Notably, previous work had demonstrated an effect size of 0.72-1.10 for study populations of undifferentiated clinical and non-clinical subjects with depressed mood.8 If we translate the effect size of 1.39 using our preceding interpretative values, we find that the antidepressant effects of exercise on depression are “large” and possess a PS of almost 84%. Even in populations of clinical and non-clinical depression, exercise demonstrated a large effect and a PS of about 74%.6

We wish to pause at this point to put the PS of 84% for exercise in perspective. Let’s turn briefly to effect sizes associated with various psychiatric and general medical pharmaceuticals and treatments. We will use the most optimistic estimates of efficacy for the various classes of interventions so as to level the playing field as much as possible. We fully acknowledge that we will not be comparing apples to apples. The following discussion is not meant to be a definitive statement regarding the efficacy of various treatments. Instead we hope that the comparisons will help place the magnitude of exercise’s effect size in context.

medicationsTo begin, let’s compare exercise’s large PS of 84% with antidepressant medication’s small PS of about 60% in acute depressive episodes.6,9 Psychotherapeutic interventions have similar effect sizes to psychopharmacologic medication in the treatment of depressive episodes. However, the combination of psychotherapy and psychopharmacologic medication yields a medium PS of about 64%; a value notable for its superiority to either intervention offered in isolation.6,7 Electroconvulsive therapy for an acute depressive episode has a large PS of 74%.6,10

There are numerous potential confounding variables in this very brief comparative overview. Despite every effort to control for a variety of confounds, it is likely that depressed cohorts who were able to exercise were qualitatively different in some ways from some of the populations included in electroconvulsive trials for example. Researchers have employed various techniques to try to eliminate these confounds and there are reasons to treat much of the data as valid; nonetheless, caution is certainly warranted when interpreting the data.

Antipsychotics for acute psychosis, antihypertensives for high blood pressure, and corticosteroids for the prevention of asthma exacerbations all have approximately the same medium PS of about 64%. Whereas, antibiotics for ear infections and metformin for diabetic mortality have a small PS of about 56%. To find a general medicine medication with a PS on par with exercise we have to turn to proton pump inhibitors with a large PS of about 84% for esophagitis.6,9 In fact, the mean PS for all general medical medications is about 62.5%, corresponding to a small magnitude effect.7

Photo by Austin Schmid | Unsplash.com
Photo by Austin Schmid | Unsplash.com

Research has revealed benefits of exercise in other domains of mental health as well. Meta-analytic reviews have found a PS for exercise on state or trait anxiety to be 60% (small magnitude effect).8 Perhaps unsurprisingly, upon closer examination of the data researchers found that exercise has a stronger effect on state anxiety than on trait anxiety.11

Evidence also suggests a broader application of exercise beyond strictly pathological states.12–14 One large non-clinical population-based study demonstrated that individuals who exercised more than 2 times per week experienced reductions not only in depressive symptomatology, but also in cynical distrust, anger, and stress when compared to individuals who exercised less often.14

There is also significant evidence for a dose-response relationship between exercise and mental health. Exercise regimens with higher intensities, greater frequencies, and longer durations tend to lead to greater response rates in depressed populations. Interestingly, depression remission rates seem to peak at moderate levels of intensity, frequency, and duration suggesting that sustainability of a regimen is an important ingredient to consider when developing a program.8

The setting that one exercises in also appears to play a role. Research has demonstrated that exposure to nature and so-called “green space” exerts powerful effects on mood and self-esteem. Exercising outdoors in a natural setting with trees and plants appears to be superior to exercising in an environment devoid of such “green” qualities. The positive effects rapidly develop with even just 5 minutes of outdoor time, offering a very achievable goal for even the busiest individuals. Interestingly, natural settings with bodies of water present (e.g. streams, rivers, lakes, etc.) appear to offer enhanced benefits over and above those seen in other natural settings.3

Photo by Austin Ban | Unsplash.com
Photo by Austin Ban | Unsplash.com

Natural settings seem to exert their positive effects on health through a variety of mechanisms; however it should be noted that the effects are not fully explained by the association of green space and exercise.15 Regardless of specific mechanism, exposure to natural environments appears to have significant effects on health. In fact, greater exposure to nature has been associated with as much as a 12% reduction in all-cause non-accidental mortality.16

How does exercise exert these far-ranging effects?

There are many gaps in our understanding of the mechanisms by which exercise exerts its anxiolytic and antidepressant effects. There is some evidence that exercise may increase turnover of serotonin, leading to an adaptive downregulation of the serotonergic 5-HT2C receptor. Activation of the 5-HT2C receptor seems to inhibit dopamine and norepinephrine release. Thus, a downregulation at the 5-HT2C receptor leads to an increase in availability of dopamine and norepinephrine. This effect is thought to be particularly important in the prefrontal cortex and is hypothesized to contribute to the anxiolytic and antidepressant effects associated with exercise.8

In addition to increasing serotonin turnover exercise seems to trigger a release of beta-endorphins. Endorphins are part of the brain’s endogenous opioid system and also tend to produce anxiolytic and antidepressant effects when released.8

From a more macroscopic scale exercise, like antidepressant medication, helps restore sleep patterns frequently disrupted in the setting of depression. Furthermore, evidence suggests that activity in the prefrontal cortex is reduced during exercise and that this modification of cognitive processing may correlate with the subjective anxiolytic and antidepressant effect of exercise.8

Finally, exercise engages an individual in an activation and approach set of behaviors that are diametrically opposed to passive and avoidant cognitive strategies classically found in depression and many other psychopathological states. In this way exercise seems to operate on a similar theoretical framework as the psychotherapeutic technique known as behavioral activation. Behavioral activation relies on changes in behavior to alter cognition, departing from the cognitive focus emphasized in many other therapeutic techniques.17 It must be noted that although exercise may be a component of a behavioral activation treatment regimen, the psychotherapeutic technique utilizes many other activation strategies to catalyze change.8

Photo by Joshua Sortino | Unsplash.com
Photo by Joshua Sortino | Unsplash.com

Let’s be optimistic and imagine that the preceding discussion helped you move from the contemplative to the preparatory stage of change and that you are preparing to make a change in your exercise habits.18 How much exercise do you need to do to before you can appreciate the mental health benefits?

Evidence suggests that an optimal exercise program is about 30 minutes in duration, has a frequency of 2-4 times per week, and is of such an intensity level that an individual achieves 70-80% estimated maximal heart rate. This target intensity would be classified as “vigorous” by the CDC guidelines outlined in the beginning of this article.8

Recall that the maximal heart rate from our 30-year old example was calculated to be 187 bpm. This means that the targeted intensity level of exercise for optimal mental health benefits should aim for a heart rate between 130-150 bpm.

Finally, the individual should commit to at least 4 weeks of the new exercise program to optimize the chances for long-term habit formation. Evidence suggests that while 70% of individuals maintain a short-term exercise program, only 50% maintain the program for six months.8

We have covered a lot of ground in our exploration of the varied health benefits associated with exercise.

Photo by Anton Repponen | Unsplash.com
Photo by Anton Repponen | Unsplash.com

We began by discussing the significant benefits of exercise for our general medical health. We learned that exercise reduces rates of mortality, some cancers, type 2 diabetes, and cardiovascular disease.

We next turned to exercise and mental health, studying depression as our archetype condition. If we compare the PS for exercise with the PS expected if there were no difference between treatment and control, we find that exercise is at least 48% more likely to result in greater improvement in depressive symptomatology when compared to an inactive control group (74% [PS of exercise for clinical and non-clinical populations] / 50% [PS if there was no difference between treatment and control groups]). When we compared exercise to antidepressant medications combined with psychotherapy we found that exercise was at least 15% more likely to result in greater improvement in depressive symptomatology.

And finally, let’s recall that exercising in natural outdoor settings, ideally in close proximity to a body of water, may enhance the health benefits associated with exercise.

As we discussed before, we are not comparing apples to apples and direct comparisons between techniques are not fair outside of a given trial. Our point is not to assert the unrivaled superiority of exercise to psychopharmacologic agents, psychotherapeutic techniques, or other treatments. Instead we wish to elevate exercise from a healthy lifestyle habit to an adjunct treatment for depression, anxiety, and other psychological conditions.

The most effective treatment for a given mental illness is almost certainly to be pluralistic rather than singular. A holistic treatment strategy that targets biological, psychological, and sociological substrates of disease offers significant synergistic advantage over a singular approach. Besides, wouldn’t we all be better off with more walks along a river?


  1. Centers for Disease Control and Prevention (CDC). Physical activity and health. CDC.gov. https://www.cdc.gov/physicalactivity/basics/pa-health/. Accessed November 12, 2016.
  2. Penedo FJ, Dahn JR. Exercise and well-being: a review of mental and physical health benefits associated with physical activity. Curr Opin Psychiatry. 2005;18(2):189-193.
  3. Barton J, Pretty J. What is the Best Dose of Nature and Green Exercise for Improving Mental Health? A Multi-Study Analysis. Environ Sci Technol. 2010;44(10):3947-3955. doi:10.1021/es903183r.
  4. Tanaka H, Monahan KD, Seals DR. Age-predicted maximal heart rate revisited. J Am Coll Cardiol. 2001;37(1):153-156. doi:10.1016/S0735-1097(00)01054-8.
  5. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, N.J: L. Erlbaum Associates; 1988.
  6. Fritz CO, Morris PE, Richler JJ. Effect size estimates: Current use, calculations, and interpretation. J Exp Psychol Gen. 2012;141(1):2-18. doi:10.1037/a0024338.
  7. Huhn M, Tardy M, Spineli LM, et al. Efficacy of Pharmacotherapy and Psychotherapy for Adult Psychiatric Disorders: A Systematic Overview of Meta-analyses. JAMA Psychiatry. 2014;71(6):706. doi:10.1001/jamapsychiatry.2014.112.
  8. Stathopoulou G, Powers MB, Berry AC, Smits JAJ, Otto MW. Exercise Interventions for Mental Health: A Quantitative and Qualitative Review. Clin Psychol Sci Pract. 2006;13(2):179-193. doi:10.1111/j.1468-2850.2006.00021.x.
  9. Leucht S, Hierl S, Kissling W, Dold M, Davis JM. Putting the efficacy of psychiatric and general medicine medication into perspective: review of meta-analyses. Br J Psychiatry. 2012;200(2):97-106. doi:10.1192/bjp.bp.111.096594.
  10. Lisanby SH. Electroconvulsive Therapy for Depression. N Engl J Med. 2007;357(19):1939-1945. doi:10.1056/NEJMct075234.
  11. Paluska SA, Schwenk TL. Physical Activity and Mental Health. Sports Med. 2000;29(3):167-180. doi:10.2165/00007256-200029030-00003.
  12. Stephens T. Physical activity and mental health in the United States and Canada: Evidence from four population surveys. Prev Med. 1988;17(1):35-47. doi:10.1016/0091-7435(88)90070-9.
  13. Taylor CB, Sallis JF, Needle R. The relation of physical activity and exercise to mental health. Public Health Rep. 1985;100(2):195-202. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1424736/. Accessed November 8, 2016.
  14. Hassmén P, Koivula N, Uutela A. Physical Exercise and Psychological Well-Being: A Population Study in Finland. Prev Med. 2000;30(1):17-25. doi:10.1006/pmed.1999.0597.
  15. Bowler DE, Buyung-Ali LM, Knight TM, Pullin AS. A systematic review of evidence for the added benefits to health of exposure to natural environments. BMC Public Health. 2010;10:456. doi:10.1186/1471-2458-10-456.
  16. James P, Hart JE, Banay RF, Laden F. Exposure to Greenness and Mortality in a Nationwide Prospective Cohort Study of Women. Environ Health Perspect. 2016;124(9). doi:10.1289/ehp.1510363.
  17. Cuijpers P, van Straten A, Warmerdam L. Behavioral activation treatments of depression: A meta-analysis. Clin Psychol Rev. 2007;27(3):318-326. doi:10.1016/j.cpr.2006.11.001.
  18. DiClemente CC, Prochaska JO, Fairhurst SK, Velicer WF, Velasquez MM, Rossi JS. The process of smoking cessation: An analysis of precontemplation, contemplation, and preparation stages of change. J Consult Clin Psychol. 1991;59(2):295-304. doi:10.1037/0022-006X.59.2.295.


You may also like