Depression is one of the most debilitating mental disorders and the leading cause of disability in the developed world (19). It is responsible for social and organizational problems facing families and workplaces alike. Depression contributes to the economic burden and has been shown to reduce quality of life (8,20). Physical activity (PA) has been shown to protect the body from many physical ailments, including cardiovascular disease and mortality (3,30). Attention has also focused on other health benefits of PA such as its effect on cognitive decline and mental health, which are addressed in recent PA guidelines (30). Some studies have suggested that inactive behavior is associated with a higher risk of depressive symptoms (7,28,29). Many observational studies have reported that even a small amount of PA has a protective effect against the likelihood of depression (14,28), whereas others say a larger dose of PA is warranted (14,28,33). Furthermore, several studies have suggested that the dose needed for mental health benefits may be lower than that needed for other health benefits (28,30). Therefore, additional research is needed to clarify what dose of PA is associated with a reduced risk of depressive symptoms.
Most of the PA–depression evidence thus far has reported on measures that examine participation in leisure time activities (7,28). These studies identify the type of activity (e.g., participating in a walking program, jogging, sports) but may not capture the volume of activity in MET-minutes per week. Minimal research has examined the association between depression and the volume of PA (4). In fact, to date, no study has used the US Department of Health and Human Services’ 2008 Physical Activity Guidelines for Americans (30) to quantify the volume of PA. The purpose of this study was to examine the association between depressive symptoms and PA (using the 2008 PA guidelines) in a large cohort of men from the Aerobics Center Longitudinal Study (ACLS). Secondary analyses examined the relationship between depressive symptoms and participation in different types of leisure time activities and the PA–depression relationship across age and body mass index (BMI) groups.
The ACLS is an observational study that investigates health effects of PA on chronic diseases in patients from the Cooper Clinic in Dallas, TX. The current analysis included 9580 men age 20–87 yr who provided complete data on the 10-item Center for Epidemiological Studies Depression Scale (CES-D-10) during 1996–2006. They were self-, employer-, or physician-referred to the clinic for preventative care and medical examinations and to receive general health counseling. The study protocol was reviewed and approved annually by the Cooper Institute Institutional Review Board. All participants provided written consent to participate in the study. Most participants were white, well educated, and from middle or upper socioeconomic strata (17).
Each individual underwent a thorough preventive medical examination, including a physical examination, body composition assessment, blood chemistry analyses, blood pressure measurement, health and lifestyle questionnaires, and history of mental health including histories of depression, anxiety, suicidal thoughts, and prior psychiatric counseling. Several covariates were chosen because of their possible effect on mental health, which could confound the association between PA and depressive symptoms. To take into account the moderate correlation of past mental health conditions, a single variable was created to account for past mental health history (yes or no). Other covariates that were included were age, examination year, current smoking (yes or no), alcoholic intake (drinks per week), and BMI calculated using weight (kg)/height (m2) then categorized into normal weight (BMI < 25), overweight (BMI = 25–29.9), and obese (BMI ≥ 30). Detailed descriptions of the covariates are available elsewhere (12,27).
All participants completed the CES-D-10 by Kohout et al. (15). The CES-D (24) is a well-validated tool used to rate depressive symptoms in epidemiologic research (33). The CES-D-10 is a shortened form of the CES-D-20, and the two questionnaires have been shown to have similar predictive accuracy (κ = 0.97, P < 0.001) (2). Participants scoring 10 or higher on the CES-D-10 were classified as exhibiting a high level of depressive symptoms (2).
A more detailed description of the PA assessment used in this study has been previously reported (17). Briefly, type, frequency, and duration of several leisure time activities (e.g., walking, running, biking, swimming, dancing, and other types of sports such as softball and racquet sports) performed in the preceding 3 months were assessed. Each activity was assigned a MET value on the basis of the Compendium of Physical Activities (1). Total MET-minutes per week were obtained by accumulating MET-minutes from each reported activity by multiplying the frequency by the duration. Four MET-minute categories were created to be comparable to the 2008 Physical Activity Guidelines for Americans (30): inactive (0 MET·min·wk−1), low (1–499 MET·min·wk−1), medium (500–999 MET·min·wk−1), and high (≥1000 MET·min·wk−1).
Demographic and other health-related characteristics were summarized across the four PA categories, using chi-square analysis to test for differences across categories. Logistic regression was used to model the relationship between depressive symptoms (CES-D-10 score ≥10 vs <10) and categories of PA. Odds ratios (ORs) and 95% confidence intervals (CI) are presented along with P values for trends. Three models were used in the primary analysis. Model 1 controlled for age and examination year. Model 2 additionally adjusted for any history of several mental health conditions (yes or no) such as clinical depression, clinical anxiety, suicidal thoughts, or psychiatric counseling. Model 3 additionally adjusted for BMI, current smoking status (yes or no), and alcohol intake (number of drinks per week). To examine if any specific leisure time PA contributed to lower odds of depressive symptoms, separate logistic regressions were used to compare nonmutually exclusive types of PA (e.g., biking, sports, running) with depressive symptoms. Thus, a participant reporting walking may also participate in other reported types of activities. The analysis was adjusted for all covariates from the primary analysis. The potential modifying effects of BMI and age on the depressive symptoms–PA relationship were examined by stratifying the data into the following groups: BMI (<25, 25–29.9, and ≥30) and age (20–39, 40–59, and ≥60 yr old). The stratification analyses were adjusted for the full model used in the primary analysis. SAS statistical software version 9.2 (SAS, Inc., Cary, NC) was used to run all analyses.
Demographic and other health-related characteristics are shown in Table 1. Participants with higher levels of PA had lower BMIs, had a more favorable blood lipid profile and lower fasting glucose levels, and were less likely to be current smokers (all P < 0.0001). There was also a significant difference with mental health histories (P < 0.02) between PA categories.
Table 2 shows the ORs and the corresponding 95% CI for each of the three models tested. Of the sample, 727 men exhibited a high level of depressive symptoms (i.e., ≥10 on CES-D-10). There was a significant positive relationship between PA and depressive symptoms for all models (P for trend < 0.0001). Generally, higher categories of PA were associated with lower odds of depressive symptoms. Men in the low PA category showed a 22% reduction in risk for depressive symptoms compared with inactive participants after adjustment for age and examination year (model 1). Model 2 showed a small increase in the effect after adjustment for mental health histories. Subsequent modeling adjusting for BMI, smoking status, and alcohol intake (model 3) remained significant. Men in the medium and high PA categories maintained significance from model 1 through 3 and were both 51% less likely to have depressive symptoms when compared with inactive men after adjusting for all confounders (model 3). When investigating if prior mental health conditions had an effect over PA categories, we noticed that having a history of mental health conditions with depressive symptoms was 23.8% compared with 3.7% for those with no prior history. We then stratified our primary analysis by mental health history and found no significant differences after adjustment for the full model (data not shown), implying that both groups improved their odds proportionately to have depressive symptoms.
Table 3 shows the association between types of nonmutually exclusive PA and depressive symptoms. Participation in vigorous PA had the largest reduction in risk for depressive symptoms. Compared with inactive men, men who participated in vigorous sports (e.g., racquetball, singles tennis, and skating) were 64% less likely to have depressive symptoms (OR = 0.36, CI = 0.21–0.61). The reduction in risk for depression symptoms was similar for all other types of leisure time activities, such as biking and swimming (OR range = 0.47–0.58).
The associations between depressive symptoms and PA, stratified by BMI and age, are shown in Table 4. There was a significant negative relationship between PA categories and depressive symptoms for normal-weight (P for trend = 0.013), overweight (P for trend < 0.0001), and obese men (P for trend = 0.002), where higher levels of PA were associated with less depressive symptoms. Significantly lower odds of depressive symptoms were identified at 500 MET·min·wk−1 of activity or greater for the normal-weight and overweight groups and at any amount of PA for obese men.
When stratified by age, the 20- to 39-yr-old group showed significantly lower odds of depressive symptoms within and across all PA categories (P trend = 0.023). The 40–59 age group also showed lower odds with ≥500 MET·min·wk−1 of PA, when compared with the inactive group, having an overall inverse association (P trend < 0.0001). No associations were found for the 60+ age group within or across the PA categories (P trend = 0.27).
Our cross-sectional analysis found an inverse association between higher categories of PA and the presence of depressive symptoms identified from the CES-D-10. Higher categories of PA were associated with lower odds of depressive symptoms. Although a reduction of odds was found in the low PA category (OR = 0.76, CI = 0.60–0.95), the association seemed to plateau at the moderate (500–999 MET·min·wk−1) category. In general, a marked reduction in the odds of depressive symptoms was significantly lower in men who accumulated ≥500 MET·min·wk−1 of leisure time PA. The amount of 500 MET·min·wk−1 is comparable to approximately 150 min of moderate-intensity activity such as walking, which meets the current PA guidelines recommendation (30). This implies that benefits are seen below the current PA guidelines, but a maximum reduction in odds is reached at the 500-MET·min category. Secondary analysis showed that all types of leisure time PA were associated with lower odds of depressive symptoms. Although these activities are not exclusive, this represents a more realistic approach where cross-training or multiple types of activities may be used. Lastly, inverse relationships between depressive symptoms and PA were seen across BMI and age categories, with the exception of men age 60 yr or older. It also seems that men with a BMI ≥ 30 may need lower levels of PA (i.e., 1–499 MET·min·wk−1) to experience a reduction in the odds of depressive symptoms.
The 2008 PA guidelines committee reported that most cross-sectional studies for active Americans had approximately 30% lower odds of depression (31). An association with habitual exercise and lower depressive scores has been consistently reported in the literature (25,26). In NHANES I, Farmer et al. (10) found that little or no recreational PA was cross-sectionally associated with depressive symptoms. In a sample of 3403 (1547 men) Finnish adults age 25–64 yr, Hassmen et al. (14) found that individuals who exercised two or three times a week had a significantly lower risk of depression as measured by the Beck Depression Inventory. In contrast to our findings, a study by Lindwall et al. (18) (n = 860 Swedish older adults age 60–96 yr) found that inactive older adults had higher depression scores than more active individuals. The difference in findings could be due to the lower proportion of older men that were sampled in our study or to the higher socioeconomic status in the ACLS. We did find that the oldest category had a 4.1% prevalence of depressive symptoms compared with 8.2% in the middle-aged category and 12.3% in the youngest age category. This is contradictory to some other studies that have shown that depression is more prevalent in older age populations. This implies that the oldest age category may have some unique characteristics compared with other studies.
When examining cohort studies, the 2008 PA guidelines committee found that the odds remained 15%–25% lower for depressive symptoms among active people after adjusting for confounders (31). In the Alameda County Study, which examined both men and women, Camacho et al. (5) reported that those who had low activity levels at baseline were at a significantly greater risk for depression at follow-up. In the Harvard alumni study, 10,201 men were followed for 23–27 yr. Men who played ≥3 h of sports each week had a 27% lower risk of physician-diagnosed depression than men playing <1 h or none at all (21). These findings are similar to ours where men who participated in vigorous activities or sports and who had a higher volume of activity had the lowest odds of depressive symptoms. Cooper-Patrick et al. (6) examined 973 physicians from medical school until midlife and found no association between depression and strenuous activity. Few studies have examined the relationship between depression and PA in a male population; most of the literature has focused on combined populations of men and women (5,10,14,18) or women only (4,28,33).
Although much of the literature suggests an association between depressive symptoms and being active, the volume of PA needed to experience such benefits is not well established. Our study showed an inverse relationship between PA and depressive symptoms, with findings suggesting that additional benefits may not be experienced after 500–999 MET·min·wk−1 of PA. Because of the cross-sectional nature of our study, we cannot suggest that PA caused a decrease in the number of men who had depressive symptoms. It is possible that men who were less depressed exercised more. To better understand the causal dose–response relationship between PA and depressive symptoms, future studies should explicitly measure the volume and intensity of PA using a prospective methodology. It is already plausible that exercise can be an adjunct or primary treatment option for depressive symptoms (9,11), although the quantity needed is still being investigated. Our study agrees with other reviews that PA has a positive influence on depressive symptoms as it has with other aspects of mental health, such as cognitive function (25,28).
Depression is a multifaceted disorder encompassing symptoms such as sadness, despair, or a clinical syndrome where a depressed mood presents with symptoms of fatigue, decreased energy, difficulty in sleeping, decreased interest for pleasure, and altered appetite (16). There are many plausible biological/physiological mechanisms that might explain the relationship between PA and depression. A traditional explanation is the “monoamine hypothesis,” which suggests that reductions in serotonin, norepinephrine, or a combination of the two result in depressive symptoms (22). These neurochemical systems have been the primary targets for antidepressant pharmacological intervention for more than 50 yr through drug effects that increase serotonin or norepinephrine levels in the brain. Similarly, PA acutely increases serotonin, norepinephrine, and their metabolites (13,23). However, as with the time lapse observed after the administration of antidepressants, a more complex pathway must exist to explain the relationship between increased concentrations of these neurotransmitters after exercise and antidepressive effects. Thus, more recent studies have examined potential structural changes in the CNS elicited by antidepressants and PA. Studies in humans and animals show exercise-induced increases in neurotrophic factors and enhanced energy metabolism pathways that can explain neurogenerative effects of PA. The significance of these findings arise from data indicating a decreased hippocampal and forebrain volume in patients during episodes of depression and the time delay required for antidepressants to induce neurogenesis in these regions (22). In summary, PA induces many changes to the structure and function of the nervous system including elevated serotonin and norepinephrine levels and stimulation of neuronal growth. These adaptations, in addition to numerous other changes, could together or in combination reduce depressive symptoms as observed after administration of antidepressants or participating in a program of PA.
Although the CES-D-10 does not provide a clinical diagnosis of depression, many epidemiological studies use this tool to rate depressive symptoms in large populations. Strengths of this study include the large sample of men, the wide range of ages studied, the different PA groups identified from a series of activity questions, and the ability to control for demographic variables and other histories of mental illnesses.
We also recognize study limitations. First is the use of self-reported questionnaires, which are not always accurate and may lead to misclassification (32). However, when examining the self-report data for PA, the participants’ fitness (exercise tolerance in maximal METs, from a graded treadmill test) showed incrementally higher fitness levels across the PA categories. It was also not possible to distinguish between walkers, joggers, and runners in our study sample because of the formatting of the questionnaires in Table 3. Although we acknowledge that the intensity of PA may affect our analyses, these groupings were noninclusive and should be considered noting this limitation. Second, our estimate of MET-minutes per week used reported leisure time activities and did not include activities of daily living, which would lead to additional energy expenditure (e.g., occupational, housework, active commuting). These light and possibly moderate activities may play an important role in reducing depressive symptoms but were outside the context of our data collection. Because we observed an association with data from leisure time PA, we can speculate that large amounts of activities of daily living may add an additional protective effect against depressive symptoms via accumulation of PA. In addition, we acknowledge that because of the large age range of the study population, the effort to accumulate 500 MET·min for a younger man may not be the same for a person 50 yr older. This implies that older men accumulating higher amounts of MET-minutes may have few depressive symptoms because they are healthier in general. Third, we also did not have information regarding current medications for depression and could not adjust for this potential confounder. However, we did adjust for the history of mental health conditions. Fourth, the majority of the study participants were college-educated non-Hispanic whites from middle to upper socioeconomic strata. Although the nature of the jobs and stress level of the participants were not measured, this could imply that our cohort is unique and was probably employed in white-collar jobs, which influences the generalizability of our findings. Finally, the cross-sectional nature of the data limits our ability to examine the cause–effect association between PA and depressive symptoms. Future longitudinal studies need to confirm our findings and should address a more exhaustive list of variables that may have a relationship with depressive symptoms such as sleep quality, stress, socioeconomic status, and nutrition.
Our findings suggest an association between PA and depressive symptoms. Lower odds of depressive symptoms were found across all categories of leisure time PA, which seemed to plateau around 500 MET·min·wk−1. This corresponds to the current PA guidelines for preventing other health conditions such as cardiovascular disease (30). Our findings support the need for behavioral interventions aimed at increasing PA; interventions targeting inactive men may be particularly important. Those meeting recommendations would not only experience physical health benefits but may also reduce the likelihood of depressive symptoms. Such reductions could have a significant public health effect in terms of disability, quality of life, and cost.
Funding for this research was supported by a grant from the National Swimming Pool Foundation, an unrestricted research grant from the Coca-Cola Company, and National Institutes of Health grants AG06945, HL62508, and DK088195. None of the sponsors had any role in the creation of this article.
The authors thank the Cooper Clinic physicians and technicians for collecting the baseline data and staff at the Cooper Institute for data entry and data management.
The authors declare no conflicts of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Ainsworth BE, Haskell WL, Whitt MC, et al.. Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc. 2000; 32 (9 suppl): S498–516.
2. Andresen EM, Malmgren JA, Carter WB, Patrick DL. Screening for depression in well older adults: evaluation of a short form of the CES-D (Center for Epidemiologic Studies Depression Scale). Am J Prev Med. 1994; 10 (2): 77–84.
3. Blair SN, Cheng Y, Holder JS. Is physical activity or physical fitness more important in defining health benefits? Med Sci Sports Exerc. 2001; 33 (6 suppl): S379–99.
4. Brown WJ, Ford JH, Burton NW, Marshall AL, Dobson AJ. Prospective study of physical activity and depressive symptoms in middle-aged women. Am J Prev Med. 2005; 29 (4): 265–72.
5. Camacho TC, Roberts RE, Lazarus NB, Kaplan GA, Cohen RD. Physical activity and depression: evidence from the Alameda County Study. Am J Epidemiol. 1991; 134 (2): 220–31.
6. Cooper-Patrick L, Ford DE, Mead LA, Chang PP, Klag MJ. Exercise and depression in midlife: a prospective study. Am J Public Health. 1997; 87 (4): 670–3.
7. Daley A. Exercise and depression: a review of reviews. J Clin Psychol Med Settings. 2008; 15 (2): 140–7.
8. Diener E. Subjective well-being. The science of happiness and a proposal for a national index. Am Psychol. 2000; 55 (1): 34–43.
9. Dinas PC, Koutedakis Y, Flouris AD. Effects of exercise and physical activity on depression. Ir J Med Sci. 2011; 180: 319–25.
10. Farmer ME, Locke BZ, Moscicki EK, Dannenberg AL, Larson DB, Radloff LS. Physical activity and depressive symptoms: the NHANES I Epidemiologic Follow-up Study. Am J Epidemiol. 1988; 128 (6): 1340–51.
11. Fox KR. The influence of physical activity on mental well-being. Public Health Nutr. 1999; 2 (3A): 411–8.
12. Galper DI, Trivedi MH, Barlow CE, Dunn AL, Kampert JB. Inverse association between physical inactivity and mental health in men and women. Med Sci Sports Exerc. 2006; 38 (1): 173–8.
13. Hand GA, Phillips KD, Wilson MA. Central regulation of stress reactivity and physical activity. In: Acevedo EO, Ekkekakis P, editors. Psychobiology of Physical Activity. Champaign, IL: Human Kinetics; 2006. p. 189–201.
14. Hassmen P, Koivula N, Uutela A. Physical exercise and psychological well-being: a population study in Finland. Prev Med. 2000; 30 (1): 17–25.
15. Kohout FJ, Berkman LF, Evans DA, Cornoni-Huntley J. Two shorter forms of the CES-D (Center for Epidemiological Studies Depression) depression symptoms index. J Aging Health. 1993; 5 (2): 179–93.
16. Kruijshaar ME, Hoeymans N, Spijker J, Stouthard ME, Essink-Bot ML. Has the burden of depression been overestimated? Bull World Health Organ. 2005; 83 (6): 443–8.
17. Lee DC, Sui X, Ortega FB, et al.. Comparisons of leisure-time physical activity and cardiorespiratory fitness as predictors of all-cause mortality in men and women. Br J Sports Med. 2011; 45: 504–10.
18. Lindwall M, Rennemark M, Halling A, Berglund J, Hassmen P. Depression and exercise in elderly men and women: findings from the Swedish national study on aging and care. J Aging Phys Act. 2007; 15 (1): 41–55.
19. Lopez AD, Mathers CD. Measuring the global burden of disease and epidemiological transitions: 2002–2030. Ann Trop Med Parasitol. 2006; 100 (5–6): 481–99.
20. Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause 1990–2020: Global Burden of Disease Study. Lancet. 1997; 349 (9064): 1498–504.
21. Paffenbarger RS Jr, Lee IM, Leung R. Physical activity and personal characteristics associated with depression and suicide in American college men. Acta Psychiatr Scand Suppl. 1994; 377: 16–22.
22. Pittenger C, Duman RS. Stress, depression, and neuroplasticity: a convergence of mechanisms. Neuropsychopharmacology. 2008; 33 (1): 88–109.
23. Post RM, Kotin J, Goodwin FK, Gordon EK. Psychomotor activity and cerebrospinal fluid amine metabolites in affective illness. Am J Psychiatry. 1973; 130 (1): 67–72.
24. Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977; 1: 385–401.
25. Salmon P. Effects of physical exercise on anxiety, depression, and sensitivity to stress: a unifying theory. Clin Psychol Rev. 2001; 21 (1): 33–61.
26. Strohle A. Physical activity, exercise, depression and anxiety disorders. J Neural Transm. 2009; 116 (6): 777–84.
27. Sui X, Laditka JN, Church TS, et al.. Prospective study of cardiorespiratory fitness and depressive symptoms in women and men. J Psychiatr Res. 2009; 43 (5): 546–52.
28. Teychenne M, Ball K, Salmon J. Physical activity and likelihood of depression in adults: a review. Prev Med. 2008; 46 (5): 397–411.
29. Teychenne M, Ball K, Salmon J. Sedentary behavior and depression among adults: a review. Int J Behav Med. 2010; 17 (4): 246–54.
31. US Department of Health and Human Services. Physical Activity Guidelines Advisory Committee Report, 2008 [Internet]. Washington (DC): Department of Health and Human Services; 2008 [cited 2011 Jan 10]. Available from: http://www.health.gov/PAGuidelines/Report/
32. Walsh MC, Hunter GR, Sirikul B, Gower BA. Comparison of self-reported with objectively assessed energy expenditure in black and white women before and after weight loss. Am J Clin Nutr. 2004; 79 (6): 1013–9.
33. Wise LA, Adams-Campbell LL, Palmer JR, Rosenberg L. Leisure time physical activity in relation to depressive symptoms in the Black Women’s Health Study. Ann Behav Med. 2006; 32 (1): 68–76.