Abstract
Although it has been hypothesized that the association of physical activity with depressive and anxiety symptoms is bidirectional, few studies have examined this issue in a prospective setting. We studied this bidirectional association using data on physical activity and symptoms of anxiety and depression at three points in time over 8 years. A total of 9,309 participants of the British Whitehall II prospective cohort study provided data on physical activity, anxiety and depression symptoms and 10 covariates at baseline in 1985. We analysed the associations of physical activity with anxiety and/or depression symptoms using multinomial logistic regression (with anxiety and depression symptoms as dependent variables) and binary logistic regression (with physical activity as the dependent variable). There was a cross-sectional inverse association between physical activity and anxiety and/or depressive symptoms at baseline (ORs between 0.63 and 0.72). In cumulative analyses, regular physical activity across all three data waves, but not irregular physical activity, was associated with reduced likelihood of depressive symptoms at follow-up (OR = 0.71, 95 % CI 0.54, 0.99). In a converse analysis, participants with anxiety and depression symptoms at baseline had higher odds of not meeting the recommended levels of physical activity at follow-up (OR = 1.79, 95 % CI 1.17, 2.74). This was also the case in individuals with anxiety and/or depression symptoms at both baseline and follow-up (OR = 1.70, 95 % CI 1.10, 2.63). The association between physical activity and symptoms of anxiety and/or depression appears to be bidirectional.
Keywords: Common mental disorders, Physical activity, Bidirectional association, Longitudinal studies
Introduction
The beneficial effects of physical activity on somatic health outcomes have been recognized for a long time [1–4]. In addition, a positive influence of physical activity on mental health has been reported [5–9].
Systematic reviews of observational and interventional studies [10–12] have assessed the relationship between physical activity and depression. Although there is some heterogeneity in the results, these reviews generally conclude that physical activity is associated with a reduced likelihood of experiencing depression or depressive symptoms. A dose–response association between physical activity and depressive symptoms has been shown in several studies [13, 14]. A number of investigations have also examined the association between physical activity and anxiety symptoms [8, 15–17]. A recent meta-analysis of 40 studies concluded that regular physical activity is associated with a reduction in anxiety symptoms [18].
Taken together, these results suggest an inverse association between physical activity and symptoms of depression and anxiety. However, many of these studies have important methodological and conceptual shortcomings. First, most of these studies were cross-sectional, making it difficult to determine the temporal sequence between physical activity, depression and anxiety symptoms [19]. Second, the prospective studies that examined the effect of physical activity, generally assessed at one point in time, on subsequent depressive or anxiety symptoms several years later; made the assumption that physical activity levels remain constant over time [14, 20]. Third, potential confounders, such as comorbidity, stressful life events [21], satisfaction with work [22, 23] are rarely taken into account in these studies.
Finally, and perhaps more importantly, very few studies have examined the hypothesis [24] that the relationship between physical activity, depressive and anxiety symptoms may be bidirectional—that is, symptoms of depression and anxiety may also affect the likelihood of undertaking physical activity. Longitudinal studies with repeated measures on both physical activity and depression/anxiety symptoms are required to address the two causal directions.
In this study, we examined the bidirectional association between physical activity at recommended levels and depression and/or anxiety symptoms, all measured three times over 8 years, using cross-sectional, prospective and cumulative exposure analyses.
Methods
Participants
The Whitehall II study, established in 1985, is a longitudinal study based on 10,308 civil servants (6,895 men and 3,413 women) [25]. All civil servants aged 35–55 years in 20 London based departments were invited to participate by letter, and 73 % agreed. Baseline examination (phase 1) took place between 1985 and 1988 and involved a clinical examination and a self-administered questionnaire. The University College London Medical School Committee on the Ethics of Human Research approved the protocol and informed consent was gained from all participants.
Measures
Physical activity
Physical activity was assessed using a standard self-administered questionnaire at phases 1, 2 (1989–1990), and 3 (1991–1993). Participants were asked two questions on the average number of hours per week spent in “moderately energetic” (e.g., dancing, cycling, leisurely swimming) and “vigorous” (e.g., running, hard swimming, playing squash) physical activity. They were then categorized into two groups based on whether they met the current recommendations on physical activity (i.e., at least 30 min a day of at least moderate intensity on at least 5 days of the week) [26]. Thus, participants reporting 2.5 h or more of at least moderately energetic physical activity per week were considered to have a sufficient level of physical activity and assigned the value 0. Those with levels of physical activity under the recommended levels were assigned the value 1 for the analysis.
Depression and anxiety symptoms
Depression and anxiety symptoms were also assessed at phases 1, 2, and 3 using the 30-item General Health Questionnaire (GHQ) [27, 28]. A depressive symptom score [29] was derived from the 30-item General Health Questionnaire using the following four items: ‘been thinking of yourself as a worthless person’, ‘felt that life is entirely hopeless’, ‘felt that life isn’t worth living’, and ‘found at times you couldn’t do anything because your nerves were too bad’ (Cronbach’s a = 0.88). Responses to these items were on a four-point Likert scale, range 0–3. A sum score was calculated and, as previously, a total score of 4 or more was used to define the presence of depressive symptoms [29].
An anxiety symptom score (‘lost much sleep over worry’, ‘felt constantly under strain’, ‘been getting scared or panicky for no good reason’, ‘found everything getting on top of you’, ‘been feeling nervous and strung up all the time’; Cronbach’s a = 0.86) was derived using five items of GHQ. Participants were asked to rate these items on a four-point scale (range 0–3) and scores in the top decile (8 or more points of the total of 15 points) were used to define anxiety cases [30]. The prevalence of anxiety cases of the Whitehall II study has been found to be close to that of anxiety disorders in the general UK population [31].
We created a four-category variable based on the cross-classification of depression (yes vs. no) and anxiety symptoms (yes vs. no) at the three waves of data collection. At each wave, this variable was assigned a value of 0 for participants without symptoms of depression and anxiety, one for those with anxiety but not depression symptoms, two for those with depression but not anxiety symptoms, and three for those with both anxiety and depression symptoms.
Covariates
We used 10 baseline (phase 1) covariates as potentially confounding variables. Sociodemographic characteristics included sex, age, ethnicity (white vs. other), marital status (married/cohabitant vs. other), and occupational position. In the Whitehall II study, occupational position reflects education, income and job responsibilities; it was categorised as high (administrators), middle (executives, professionals and technical staff) and low (clerical and office support staff). Two psychosocial variables were included: stressful life events (0, 1 and ≥2 events) and satisfaction with work (low, intermediate and high). Health-related behaviours included were alcohol intake (high [>14 units/week for women and >21 units/week for men] vs. other; where a unit is 8 g of alcohol), and smoking status (never, ex, and current smokers). Self-reported respiratory disease (yes vs. no) was included as health-related covariate.
Statistical analysis
Differences in baseline characteristics of participants as a function of physical activity and anxiety and/or depression symptoms were assessed using a Chi-square test. To explore associations between physical activity and anxiety and/or depression symptoms, described below, we undertook two logistic regressions, one multinomial (with anxiety and depression symptoms as dependent variable) and one binary (with physical activity as dependent variable). Model 1 shows crude odds ratios (ORs). In model 2, ORs were adjusted for sociodemographic characteristics (age, sex, ethnicity, marital status and occupational position), psychosocial variables (stressful life events and satisfaction with work), health-related behaviours (smoking, alcohol intake) and self-reported respiratory disease.
Physical activity as a predictor of anxiety and/or depression symptoms
Cross-sectional, prospective and cumulative exposure analyses based on multinomial logistic regression were conducted. The cross-sectional analysis aimed to assess the association between physical activity and anxiety/depression symptoms at baseline (phase 1). The prospective analysis, undertaken among participants with no anxiety and depression symptoms at phase 1, examined the association between physical activity at phase 1 and incident anxiety and/depression symptoms at phases 2 or 3. The cumulative exposure analysis examined the association between cumulative measures of physical activity, calculated by summing the values at the three data points, and anxiety/depression symptoms at the last follow-up (phase 3). The physical activity score ranged from 0 to 3 and was recoded into 3 categories: regular if physical activity level corresponded to the recommended levels at least at two phases (score 0–1), irregular if physical activity level corresponded to the recommended levels at one phase only (score 2), and none if the physical activity level did not correspond to the recommended levels at any phase (score 3).
Anxiety/depression symptoms as predictor of physical activity
Prospective and cumulative exposure analyses based on binary logistic regression were conducted. The prospective analysis examined the association between anxiety/depression symptoms at baseline (phase 1) and subsequent practice of physical activity at phases 2 or 3 among participants with physical activity level under the recommended levels at phase 1. The cumulative exposure analysis examined the association between the cumulative score of anxiety/depression symptoms from phases 1 to 3 and physical activity at the last follow-up (phase 3). For this analysis, a cumulative anxiety score ranging from 0 to 3 was obtained by summing values at each phase. The value 0 corresponded to participants without anxiety symptoms at any phase and values 1, 2, and 3 represented participants with anxiety symptoms at one, two and three phases, respectively.
A similar score was also calculated for depressive symptoms. Then, a new score was created by summing the cumulative scores of anxiety and depression scores. This score ranged from 0 to 6 and was re-categorized as follows: 0 for participants without anxiety and depression symptoms at any phase, 1–3 for participants who experienced anxiety and/or depression symptoms less frequently between phases, 4–6 for participants who experienced anxiety and/or depression symptoms more frequently; the score of 6 corresponding to participants who experienced both anxiety and depression symptoms at all three phases.
Results
Among the 10,308 participants of the Whitehall II cohort study, 9,309 had complete data on all variables at baseline and were included in the cross-sectional analysis. Among these participants, 13.4 % reported practising physical activity at the recommended levels and 19.9 % experienced anxiety and/or depression symptoms. Tables 1 and 2 present the baseline characteristics of the participants as a function of physical activity and anxiety/depression symptoms.
Table 1.
Characteristics of participants in phase 1 (1985–1988) of the Whitehall II study by physical activity level
| Variables | N total (9,309) | Physical activity at recommended levels | ||
|---|---|---|---|---|
| Yes N (%) | No N (%) | P value | ||
| Sex | ||||
| Men | 6,374 (68.5) | 1,056 (83.9) | 5,318 (66.1) | <0.001*** |
| Women | 2,935 (31.5) | 202 (16.1) | 2,733 (33.9) | |
| Age groups (years) | ||||
| ≤39 | 2,582 (27.7) | 386 (30.7) | 2,196 (27.3) | <0.001*** |
| 40–44 | 2,436 (26.2) | 375 (29.8) | 2,061 (25.6) | |
| 45–49 | 1,892 (20.3) | 240 (19.1) | 1,652 (20.5) | |
| ≥50 | 2,399 (25.8) | 257 (20.4) | 2,142 (26.6) | |
| Ethnicity | ||||
| White | 8,487 (91.2) | 1,180 (93.8) | 7,307 (90.8) | <0.001*** |
| Other | 822 (8.8) | 78 (6.2) | 744 (9.2) | |
| Marital status | ||||
| Married/cohabiting | 6,943 (74.6) | 1,018 (80.9) | 5,925 (73.6) | <0.001*** |
| Other | 2,366 (25.4) | 240 (19.1) | 2,126 (26.4) | |
| Occupational grade | ||||
| High | 2,828 (30.4) | 427 (33.9) | 2,401 (29.8) | <0.001*** |
| Middle | 4,573 (49.1) | 682 (54.3) | 3,891 (48.4) | |
| Low | 1,908 (20.5) | 149 (11.8) | 1,759 (21.8) | |
| Satisfaction with work | ||||
| Low | 3,299 (35.4) | 473 (37.6) | 2,826 (35.1) | 0.33 |
| Intermediate | 2,363 (25.4) | 293 (23.3) | 2,070 (25.7) | |
| High | 3,647 (39.2) | 492 (39.1) | 3,155 (39.2) | |
| Stressful life events score | ||||
| 0 | 2,693 (28.9) | 351 (27.9) | 2,342 (29.1) | 0.24 |
| 1 | 3,054 (32.8) | 407 (32.4) | 2,647 (32.9) | |
| ≥2 | 3,562 (38.3) | 500 (39.7) | 3,062 (38.0) | |
| Smoking | ||||
| Non smoker | 4,578 (49.2) | 598 (47.5) | 3,980 (49.4) | 0.75 |
| Ex-smoker | 3,054 (32.8) | 462 (36.7) | 2,592 (32.2) | |
| Current smoker | 1,677 (18.0) | 198 (15.7) | 1,479 (18.4) | |
| High alcohol intake | ||||
| No | 7,808 (83.9) | 1,050 (83.5) | 6,758 (86.6) | 0.67 |
| Yes | 1,501 (16.1) | 208 (16.5) | 1,293 (16.1) | |
| Self-reported respiratory disease | ||||
| No | 8,573 (92.1) | 1,173 (93.2) | 7,400 (91.9) | 0.10 |
| Yes | 736 (7.9) | 85 (6.8) | 651 (8.1) | |
Table 2.
Characteristics of participants in phase 1 (1985–1988) of the Whitehall II Study by anxiety and/or depression symptoms
| Variables | Anxiety and/or depression symptoms | ||||
|---|---|---|---|---|---|
| None (N = 7,454) |
Anxiety (N = 644) |
Depression (N = 739) |
Combined anxiety and depression (N = 472) |
P value | |
| Sex | |||||
| Men | 5,182 (69.5) | 390 (60.6) | 522 (70.6) | 280 (59.3) | <0.001*** |
| Women | 2,272 (30.5) | 254 (39.4) | 217 (29.4) | 192 (40.7) | |
| Age groups (years) | |||||
| ≤39 | 2,060 (27.6) | 192 (29.8) | 182 (24.6) | 148 (31.4) | 0.09 |
| 40–44 | 1,927 (25.9) | 173 (26.9) | 196 (26.5) | 140 (29.7) | |
| 45–49 | 1,518 (20.4) | 129 (20.0) | 162 (22.0) | 83 (17.5) | |
| ≥50 | 1,949 (26.1) | 150 (23.3) | 199 (26.9) | 101 (21.4) | |
| Ethnicity | |||||
| White | 6,799 (91.2) | 603 (93.6) | 652 (88.2) | 433 (91.7) | 0.39 |
| Other | 655 (8.8) | 41 (6.4) | 87 (11.8) | 39 (8.3) | |
| Marital Status | |||||
| Married/cohabiting | 5,701 (76.5) | 474 (73.6) | 477 (64.5) | 291 (61.7) | <0.001*** |
| Other | 1,753 (23.5) | 170 (26.4) | 262 (35.5) | 181 (38.3) | |
| Occupational grade | |||||
| High | 2,330 (31.3) | 205 (31.8) | 189 (25.6) | 104 (22.0) | <0.001*** |
| Middle | 3,613 (48.5) | 328 (50.9) | 384 (52.0) | 248 (52.5) | |
| Low | 1,511 (20.3) | 111 (17.2) | 166 (22.5) | 120 (25.4) | |
| Satisfaction with work | |||||
| Low | 2,386 (32.0) | 278 (43.2) | 353 (47.8) | 282 (59.7) | <0.001*** |
| Intermediate | 1,910 (25.6) | 159 (24.7) | 196 (26.5) | 98 (20.8) | |
| High | 3,158 (42.4) | 207 (32.1) | 190 (25.7) | 92 (19.5) | |
| Stressful life events score | |||||
| 0 | 2,328 (31.2) | 100 (15.5) | 202 (27.3) | 63 (13.3) | <0.001*** |
| 1 | 2,536 (34.0) | 169 (26.2) | 228 (30.9) | 121 (25.6) | |
| ≥2 | 2,590 (34.7) | 375 (58.2) | 309 (25.6) | 288 (61.0) | |
| Smoking | |||||
| Non smoker | 3,704 (49.7) | 280 (43.5) | 381 (51.6) | 213 (45.1) | 0.004** |
| Ex-smoker | 2,457 (33.0) | 229 (35.6) | 230 (31.1) | 138 (29.2) | |
| Current smoker | 1,293 (17.3) | 135 (17.3) | 128 (17.3) | 121 (25.6) | |
| High alcohol intake | |||||
| No | 6,285 (84.3) | 530 (82.3) | 618 (83.6) | 375 (79.4) | 0.01* |
| Yes | 1,169 (15.7) | 114 (17.7) | 121 (16.4) | 97 (20.6) | |
| Self-reported respiratory disease | |||||
| No | 6,933 (93.0) | 581 (90.2) | 652 (88.2) | 407 (86.2) | <0.001*** |
| Yes | 521 (7.0) | 63 (9.8) | 87 (11.8) | 65 (13.8) | |
Cross-sectional association between physical activity, anxiety and/or depression symptoms
Table 3 presents the odds ratios (OR) estimating the association of physical activity, anxiety and/or depression symptoms at baseline. Model 1 shows that participants who reported practicing physical activity at recommended levels were less likely to have anxiety symptoms (OR = 0.71, 95 % CI, 0.54–0.0.91), depressive symptoms (OR = 0.63, 95 % CI, 0.48–0.81) and both (OR = 0.72, 95 % CI, 0.54–0.97). After adjustment for sociodemographic characteristics, psychosocial factors, health-related behaviours, and self-reported respiratory disease these inverse associations remained fairly unchanged.
Table 3.
Cross-sectional associations between physical activity at recommended levels and anxiety and/or depression symptoms at phase 1 (1985–1988) (N = 9,309)
| OR (CI 95 %) | P value | |
|---|---|---|
| Anxiety symptoms | ||
| Model 1 | ||
| Physical activity | ||
| Yes | 0.71 (0.54, 0.91) | 0.01 |
| No | 1 (reference) | |
| Model 2 | ||
| Physical activity | ||
| Yes | 0.71 (0.55, 0.93) | 0.01 |
| No | 1 (reference) | |
| Depression symptoms | ||
| Model 1 | ||
| Physical activity | ||
| Yes | 0.63 (0.48, 0.81) | <0.001 |
| No | 1 (reference) | |
| Model 2 | ||
| Physical activity | ||
| Yes | 0.63 (0.49, 0.82) | 0.001 |
| No | 1 (reference) | |
| Combined anxiety and depression symptoms | ||
| Model 1 | ||
| Physical activity | ||
| Yes | 0.72 (0.54, 0.97) | 0.03 |
| No | 1 (reference) | |
| Model 2 | ||
| Physical activity | ||
| Yes | 0.74 (0.54, 1.01) | 0.05 |
| No | 1 (reference) | |
Model 1: Crude ORs
Model 2: ORs adjusted for sex, age, ethnicity, marital status, occupational grade, satisfaction with work, stressful life events score, smoking, alcohol intake and respiratory disease
Physical activity as a predictor of anxiety and/or depression symptoms
Figure 2 presents ORs for the associations between physical activity and incident anxiety and/or depressive symptoms at either phase 2 or 3 among participants without anxiety and/or depression symptoms at phase 1. In the model unadjusted for covariates (model 1), physical activity at phase 1 was not associated with the incidence of anxiety and/or depressive symptoms over the follow-up. A similar pattern was observed in the fully adjusted model (Fig. 1).
Fig. 2.
Prospective association between physical activity at recommended levels at phase 1 (1985–1988) and anxiety and/or depression symptoms at phase 2 (1989–1990) and/or 3 (1991–1993) (N = 7,454). Adjusted ORs were for sex, age, ethnicity, marital status, occupational grade, satisfaction with work, stressful life events score, smoking, alcohol intake and respiratory disease
Fig. 1.
Study design
Figure 3 shows the associations between the physical activity score (cumulative score from phase 1 to 3) and anxiety and/or depression symptoms at phase 3. In model 1, participants who reported practicing physical activity at recommended levels regularly had reduced odds of depressive symptoms (OR = 0.72, 95 % CI, 0.52–0.99) compared to those who did not practice physical activity at recommended levels at any of the three phases. These associations were unchanged after adjustment for all potentially confounding variables.
Fig. 3.
Association of cumulative physical activity at recommended levels at phase 1 (1985–1988), 2 (1989–1990) and 3 (1991–1993) and anxiety and/or depression symptoms at phase 3 (N = 6,280). Adjusted ORs were for sex, age, ethnicity, marital status, occupational grade, satisfaction with work, stressful life events score, smoking, alcohol intake and respiratory disease. Reg and Irreg represent regular and irregular physical activity at recommended levels, respectively
Anxiety and/or depression symptoms as predictor of physical activity practice
Figure 4 shows ORs for the association between anxiety and/or depression symptoms at phase 1 and physical activity levels at phases 2 and 3 among participants who reported physical activity below recommended levels at phase 1. Participants with elevated anxiety and depression symptoms at phase 1 were more likely not to engage in physical activity at the recommended levels over the follow- up (OR = 1.79, 95 % CI, 1.17–2.74) compared to those without such symptoms.
Fig. 4.
Prospective association between anxiety and/or depression symptoms at phase 1 (1985–1988) and physical activity practice at recommended levels at phase 2 (1989–1990) or 3 (1991–1993) among participants with physical activity level under the recommended levels at phase 1 (N = 5,385). Adjusted ORs were for sex, age, ethnicity, marital status, occupational grade, satisfaction with work, stressful life events score, smoking, alcohol intake and respiratory disease
To examine this association further, we repeated the analysis in a subgroup of participants who undertook physical activity at the recommended levels at phase 1. The results suggested that baseline depressive symptoms were associated with higher probability of not meeting the recommended levels of physical activity at follow-up (adjusted OR = 1.52, P > 0.05), even though this association did not reach statistical significance at the 5 % level, perhaps because it was based on small numbers (N = 901).
Finally, Fig. 5 presents the associations between the cumulative score (phases 1–3) of anxiety and/or depression symptoms and physical activity at phase 3. In model 1, participants who experienced anxiety and/or depression symptoms less frequently (cumulative score of 1–3) were more likely to be physically inactive at phase 3 (OR = 1.44, 95 % CI, 1.21–1.70) compared to those without such symptoms (cumulative score of 0). The OR was even higher in participants who experienced anxiety and/or depression symptoms more frequently (cumulative score of 4–6) (OR = 1.82, 95 % CI, 1.19–2.79). When all covariates were included in the model 2, ORs remained statistically significant but were slightly reduced.
Fig. 5.
Prospective association between anxiety and/or depression symptoms at phase 1 (1985–1988) and physical activity practice at recommended levels at phase 2 (1989–1990) or 3 (1991–1993) among participants who (N = 5,385). Adjusted ORs were for sex, age, ethnicity, marital status, occupational grade, satisfaction with work, stressful life events score, smoking, alcohol intake and respiratory disease. None (score 0), episodic (score 1–3), frequent (score 4–6)
Discussion
Key findings
We examined the bidirectional association of physical activity with anxiety and/or depressive symptoms in a large cohort of men and women. Results of cross-sectional analyses showed that physical activity at recommended levels was associated with a reduced probability of anxiety and depressive symptoms. In cumulative exposure analysis, we found that regular physical activity during the follow-up was associated with a reduced likelihood of depressive symptoms. In the converse analysis, exposure to anxiety and depressive symptoms, once or more times, was associated with an increased probability of not meeting the recommended levels of physical activity over the follow-up.
Results in the context of previous studies
Our cross-sectional results are consistent with previous studies showing an inverse association of physical activity with anxiety and depressive symptoms [7, 8, 13–16, 18]. The prospective findings of our study showing no significant associations between physical activity at baseline with incident anxiety and/depression symptoms are also consistent with those from previous studies. For example, a study conducted in a community-based sample of older men and women in southern California found that exercise did not protect against future depressed mood in those not clinically depressed at baseline [32]. In the same vein, a study conducted among physicians found that the incidence of depression 15 years later was similar among those who were physically active and those who were physically inactive at baseline [33]. In contrast, some other studies have found physical activity to be associated with the reduced incidence of depression [6, 9, 14]. However, in these studies physical activity was generally assessed at one point in time with the assumption that it remains unchanged over time.
Compared to previous prospective studies, the present study is novel in its use of repeated measures of physical activity, anxiety and/or depressive symptoms. The use of repeated measures in our study allowed us to examine the association between regular physical activity practice at recommended levels and anxiety and/or depressive symptoms. Indeed, it appears that it is only regular practice of physical activity which is associated with reduced risk of depression symptoms.
In contrast to the results of the cross-sectional and prospective analyses, the use of repeated measures provides further insight into the nature of the physical activity-depressive and/or anxiety symptoms link. Based on the results of our cross-sectional and prospective analyses only, we would have had mixed conclusions about this association; not being able to conclude whether the associations we observed cross-sectionally were due to a higher prevalence of physical inactivity in depressed or anxious participants. Cumulative exposure analysis was vital to determine whether the lack of associations in prospective analysis is likely to be true or an artefact owing to the fact that physical activity was assessed only at one point in time at baseline.
Furthermore, our study adds to the literature in this domain by considering simultaneously anxiety and depression symptoms and their co-occurrence. Epidemiological studies have documented the co-occurrence of anxiety disorders and depression symptoms [34], but the relationship between physical activity and co-occurring anxiety and depression symptoms has not been examined previously.
Finally, our study design also allowed us to show that the association is likely to be bidirectional: physical activity decreases the risk for anxiety and/or depression symptoms, but these symptoms are also likely to be associated with poor levels of physical activity. A further strength was the opportunity to take into account a wide range of potential confounders in the analysis.
What are the mechanisms behind the observed associations?
Several hypotheses have been posited to explain the protective association of physical activity on mental health outcomes. The distraction hypothesis [35] suggests that the diversion of individuals from negative stimuli may lead to an improvement in mood during and subsequent to the practice of physical activity. The self-efficacy/mastery hypothesis [36] suggests that physical activity could be seen as a challenging activity, the ability of individuals to be physically active on a regular basis may lead to improvements in their mood and self-confidence. Finally, the social interaction hypothesis [37] views social relationships resulting from physical activity practice and associated social support among individuals involved in physical activity to play a role in the effects of physical activity on mental health.
Physiological mechanisms have also been proposed to explain this inverse relationship. The monoamines hypothesis [37, 38] based on animal models suggests that physical activity increases the synaptic transmission of monoamines, which are assumed to have antidepressant effects. Another hypothesis [37] highlights the role of endorphins, released upon physical activity, on the central nervous system to create feeling of calm and improved mood after physical activity [39].
We were not able to test these mechanistic hypotheses, but our findings raise other possibilities, particularly in relation to the observed reciprocal association between physical activity, depressive and anxiety symptoms. Poor health may prevent or reduce the ability to undertake physical activity [40, 41] and it may also be associated with the development of symptoms of anxiety and depression [42], which in turn may decrease the possibility of physical activity. A further possibility lies in the fact that symptoms of anxiety and depression are associated with increased risk of gaining weight [43], which then could reduce the likelihood of undertaking physical activity at recommended levels [44]. Finally, it is possible that individuals with anxiety and depression are more socially isolated than those free of these symptoms, and therefore less motivated to engage in physical activity [45]. In the same vein, individuals with anxiety and depressive symptoms may be less likely to engage in physical activity simply because of lower energy levels or apathy [8]. In sum, the precise mechanisms underlying the current observations remain to be examined.
Limitations
In interpreting the present results, it is important to note some limitations. First, this cohort of civil servants did not include blue collar workers, unemployed or individuals with precarious jobs and younger adults; thus it is not representative of the general population of working age, which may limit the generalisability of our findings. In addition, our data were collected between 1985 and 1993 when the participants were 39–64 years of age. This may also limit the applicability of the present findings to today’s society or other age groups such younger and older adults.
Second, we assessed anxiety and depressive symptoms rather than clinical anxiety and depression, even though it has been suggested that significant anxiety and depressive symptomatology are a risk factor for clinical anxiety and depression [46]. Thus, we cannot completely exclude misclassification bias as a source of error that explains the current observed associations. Third, self-reported physical activity is problematic; a declaration bias for instance cannot be ruled out. This implies that the relationship between physical activity and symptoms of anxiety and depression may have been underestimated. Objective measures of physical activity and an assessment of mental health based on psychiatric interview are required for more accurate estimates. Fourth, although we were able to control for a wide range of potentially confounding variables, we cannot exclude the possibility of residual confounding from unmeasured variables, such as personality traits, social isolation or genetic factors. Finally, experimental studies with longer follow-up would be needed to assess the real impact of physical activity on symptoms of anxiety and depression, because the few existing studies have limited follow-up [10].
Implications
Despite these potential limitations, the results of this study suggest that the association between physical activity and symptoms of anxiety and/or depression is likely to be bidirectional. The practice of physical activity on a regular basis reduces the risk of developing symptoms of anxiety and/or depression, but having symptoms of anxiety and/or depression increases the probability of not meeting the recommended levels of physical activity. These findings suggest that health promotion strategies should consider including physical activity not only to improve physical health but because it has the potential to improve mental health as well. Physical activity can also be considered an important part of treatment to alleviate depression and anxiety symptoms, in particular when the symptoms are moderate rather than severe.
Acknowledgments
We thank all participating civil service departments and their welfare personnel, and establishment officers; the Occupational Health and Safety Agency; the Council of Civil Service Unions; all participating civil servants in the Whitehall II study; all members of the Whitehall II study team. The Whitehall II study team comprises research scientists, statisticians, study coordinators, nurses, data managers, administrative assistants and data entry staff, who make the study possible. The Whitehall II study is supported by grants from the Medical Research Council; British Heart Foundation; National Heart Lung and Blood Institute (R01HL036310), US, NIH and the National Institute on Aging (R01AG013196 and R01AG034454), US, NIH. MJS is supported by a grant from the British Heart Foundation. MK is supported by the BUPA Foundation, UK, the Academy of Finland and the EU New OSH ERA Research Programme. AS-M is supported by a “European Young Investigator Award” from the European Science Foundation and the National Institute on Aging, NIH (R01AG013196, R01AG034454).
Abbreviations
- BMI
Body mass index
- CI
Confidence interval
- GHQ
General Health Questionnaire
- OR
Odds Ratios
Footnotes
Conflict of interest None.
Contributor Information
Marine Azevedo Da Silva, Centre for Research in Epidemiology and Population Health, Epidemiology of Occupational and Social Determinants of Health, INSERM, U1018, 94807 Villejuif, France; Université de Versailles St. Quentin, UMRS 1018, 94807 Villejuif, France.
Archana Singh-Manoux, Centre for Research in Epidemiology and Population Health, Epidemiology of Occupational and Social Determinants of Health, INSERM, U1018, 94807 Villejuif, France; Université de Versailles St. Quentin, UMRS 1018, 94807 Villejuif, France; Centre de Gérontologie, Hôpital Ste Périne, AP-HP, Paris, France; Department of Epidemiology and Public Health, University College London, London, UK.
Eric J. Brunner, Department of Epidemiology and Public Health, University College London, London, UK
Sara Kaffashian, Centre for Research in Epidemiology and Population Health, Epidemiology of Occupational and Social Determinants of Health, INSERM, U1018, 94807 Villejuif, France; Université de Versailles St. Quentin, UMRS 1018, 94807 Villejuif, France.
Martin J. Shipley, Department of Epidemiology and Public Health, University College London, London, UK
Mika Kivimäki, Department of Epidemiology and Public Health, University College London, London, UK; Finnish Institute of Occupational Health, University of Helsinki, Helsinki, Finland.
Hermann Nabi, Email: hermann.nabi@inserm.fr, INSERM, U1018, Hôpital Paul Brousse, Bâtiment 15/16 avenue Paul Vaillant Couturier, 94807 Villejuif Cedex, France.
References
- 1.Morris JN, Heady JA, Raffle PA, Roberts CG, Parks JW. Coronary heart-disease and physical activity of work. Lancet. 1953;265(6796):1111–1120. doi: 10.1016/s0140-6736(53)91495-0. [DOI] [PubMed] [Google Scholar]
- 2.Paffenbarger RS, Jr, Brand RJ, Sholtz RI, Jung DL. Energy expenditure, cigarette smoking, and blood pressure level as related to death from specific diseases. Am J Epidemiol. 1978;108(1):12–18. [PubMed] [Google Scholar]
- 3.Paffenbarger RS, Hale WE. Work activity and coronary heart mortality. N Engl J Med. 1975;292(11):545–550. doi: 10.1056/NEJM197503132921101. [DOI] [PubMed] [Google Scholar]
- 4.Hu FB, Sigal RJ, Rich-Edwards JW, et al. Walking compared with vigorous physical activity and risk of type 2 diabetes in women: a prospective study. JAMA. 1999;282(15):1433–1439. doi: 10.1001/jama.282.15.1433. [DOI] [PubMed] [Google Scholar]
- 5.Fox KR. The influence of physical activity on mental well-being. Public Health Nutr. 1999;2(3):411–418. doi: 10.1017/s1368980099000567. [DOI] [PubMed] [Google Scholar]
- 6.Wiles NJ, Haase AM, Gallacher J, Lawlor DA, Lewis G. Physical activity and common mental disorder: results from the Caerphilly study. Am J Epidemiol. 2007;165(8):946–954. doi: 10.1093/aje/kwk070. [DOI] [PubMed] [Google Scholar]
- 7.Harvey SB, Hotopf M, Overland S, Mykletun A. Physical activity and common mental disorders. Br J Psychiatry. 2010;197(5):357–364. doi: 10.1192/bjp.bp.109.075176. [DOI] [PubMed] [Google Scholar]
- 8.Goodwin RD. Association between physical activity and mental disorders among adults in the United States. Prev Med. 2003;36(6):698–703. doi: 10.1016/s0091-7435(03)00042-2. [DOI] [PubMed] [Google Scholar]
- 9.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):1340–1351. doi: 10.1093/oxfordjournals.aje.a115087. [DOI] [PubMed] [Google Scholar]
- 10.Brosse AL, Sheets ES, Lett HS, Blumenthal JA. Exercise and the treatment of clinical depression in adults: recent findings and future directions. Sports Med. 2002;32(12):741–760. doi: 10.2165/00007256-200232120-00001. [DOI] [PubMed] [Google Scholar]
- 11.Saxena S, Van Ommeren M, Tang KC, Armstrong TP. Mental health benefits of physical activity. J Ment Health. 2005;14(5):445–451. [Google Scholar]
- 12.Craft LL, Perna FM. The benefits of exercise for the clinically depressed. Prim Care Companion J Clin Psychiatry. 2004;6(3):104. doi: 10.4088/pcc.v06n0301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Dunn AL, Trivedi MH, O’Neal HA. Physical activity dose-response effects on outcomes of depression and anxiety. Med Sci Sports Exerc. 2001;33(6):S587–S597. doi: 10.1097/00005768-200106001-00027. [DOI] [PubMed] [Google Scholar]
- 14.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):265–272. doi: 10.1016/j.amepre.2005.06.009. [DOI] [PubMed] [Google Scholar]
- 15.StrOhle A, Hofler M, Pfister H, et al. Physical activity and prevalence and incidence of mental disorders in adolescents and young adults. Psychol Med. 2007;37(11):1657–1666. doi: 10.1017/S003329170700089X. [DOI] [PubMed] [Google Scholar]
- 16.De Moor MHM, Boomsma DI, Stubbe JH, Willemsen G, de Geus EJC. Testing causality in the association between regular exercise and symptoms of anxiety and depression. Arch Gen Psychiatry. 2008;65(8):897. doi: 10.1001/archpsyc.65.8.897. [DOI] [PubMed] [Google Scholar]
- 17.Smits JAJ, Berry AC, Rosenfield D, Powers MB, Behar E, Otto MW. Reducing anxiety sensitivity with exercise. Depress Anxiety. 2008;25(8):689–699. doi: 10.1002/da.20411. [DOI] [PubMed] [Google Scholar]
- 18.Herring MP, O’Connor PJ, Dishman RK. The effect of exercise training on anxiety symptoms among patients: a systematic review. Arch Intern Med. 2010;170(4):321–331. doi: 10.1001/archinternmed.2009.530. [DOI] [PubMed] [Google Scholar]
- 19.Ströhle A. Physical activity, exercise, depression and anxiety disorders. J Neural Transm. 2009;116(6):777–784. doi: 10.1007/s00702-008-0092-x. [DOI] [PubMed] [Google Scholar]
- 20.Farmer ME, Locke BENZ, Mo Cicki EVEK, 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. doi: 10.1093/oxfordjournals.aje.a115087. [DOI] [PubMed] [Google Scholar]
- 21.You S, Conner KR. Stressful life events and depressive symptoms: influences of gender, event severity, and depression history. J Nerv Ment Dis. 2009;197(11):829. doi: 10.1097/NMD.0b013e3181be7841. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Gimeno D, Elovainio M, Jokela M, De Vogli R, Marmot MG, Kivimäki M. Association between passive jobs and low levels of leisure-time physical activity: the Whitehall II cohort study. Occup Environ Med. 2009;66(11):772. doi: 10.1136/oem.2008.045104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kouvonen A, Kivimäki M, Elovainio M, Virtanen M, Linna A, Vahtera J. Job strain and leisure-time physical activity in female and male public sector employees. Prev Med. 2005;41(2):532–539. doi: 10.1016/j.ypmed.2005.01.004. [DOI] [PubMed] [Google Scholar]
- 24.Roshanaei-Moghaddam B, Katon WJ, Russo J. The longitudinal effects of depression on physical activity. Gen Hosp Psychiatry. 2009;31(4):306–315. doi: 10.1016/j.genhosppsych.2009.04.002. [DOI] [PubMed] [Google Scholar]
- 25.Marmot M, Brunner E. Cohort profile: the Whitehall II study. Int J Epidemiol. 2005;34(2):251. doi: 10.1093/ije/dyh372. [DOI] [PubMed] [Google Scholar]
- 26.Department of Health. At least five a week: Evidence on the impact of physical activity and its relationship to health. A report from the Chief Medical Officer. 2004 ( www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4080994).
- 27.Goldberg DP, Hillier VF. A scaled version of the General Health Questionnaire. Psychol Med. 1979;9(01):139–145. doi: 10.1017/s0033291700021644. [DOI] [PubMed] [Google Scholar]
- 28.Goldberg DP. The detection of psychiatric illness by questionnaire. Oxford University Press; 1972. [Google Scholar]
- 29.Stansfeld SA, Head J, Marmot MG. Explaining social class differences in depression and well-being. Soc Psychiatry Psychiatr Epidemiol. 1997;33(1):1–9. doi: 10.1007/s001270050014. [DOI] [PubMed] [Google Scholar]
- 30.Virtanen M, Ferrie JE, Singh-Manoux A, et al. Long working hours and symptoms of anxiety and depression: a 5-year follow-up of the Whitehall II study. Psychol Med. 2011;18:1–10. doi: 10.1017/S0033291711000171. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Jenkins R, Lewis G, Bebbington P, et al. The National Psychiatric Morbidity surveys of Great Britain—initial findings from the household survey. Psychol Med. 1997;27(4):775–789. doi: 10.1017/s0033291797005308. [DOI] [PubMed] [Google Scholar]
- 32.Kritz-Silverstein D, Barrett-Connor E, Corbeau C. Cross-sectional and prospective study of exercise and depressed mood in the elderly: the Rancho Bernardo study. Am J Epidemiol. 2001;153(6):596–603. doi: 10.1093/aje/153.6.596. [DOI] [PubMed] [Google Scholar]
- 33.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):670–673. doi: 10.2105/ajph.87.4.670. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry. 1994;51(1):8–19. doi: 10.1001/archpsyc.1994.03950010008002. [DOI] [PubMed] [Google Scholar]
- 35.Morgan WP. Affective beneficence of vigorous physical activity. Med Sci Sports Exerc. 1985;17(1):94–100. [PubMed] [Google Scholar]
- 36.North TC, McCullagh P, Tran ZV. Effect of exercise on depression. Exerc Sport Sci Rev. 1990;18:379–415. [PubMed] [Google Scholar]
- 37.Ransford CP. A role for amines in the antidepressant effect of exercise: a review. Med Sci Sports Exerc. 1982;14(1):1–10. doi: 10.1249/00005768-198201000-00001. [DOI] [PubMed] [Google Scholar]
- 38.Dunn AL, Dishman RK. Exercise and the neurobiology of depression. Exerc Sport Sci Rev. 1991;19:41–98. [PubMed] [Google Scholar]
- 39.Allen M. Sport psychiatry: theory and practice. New York: WW Norton & Company; 2000. The psychobiology of athletic training; pp. 22–44. [Google Scholar]
- 40.Trost SG, Owen N, Bauman AE, Sallis JF, Brown W. Correlates of adults’ participation in physical activity: review and update. Med Sci Sports Exerc. 2002;34(12):1996. doi: 10.1097/00005768-200212000-00020. [DOI] [PubMed] [Google Scholar]
- 41.Devoogdt N, Van Kampen M, Geraerts I, et al. Physical activity levels after treatment for breast cancer: one-year follow-up. Breast Cancer Res Treat. 2010;123(2):417–425. doi: 10.1007/s10549-010-0997-6. [DOI] [PubMed] [Google Scholar]
- 42.Moussavi S, Chatterji S, Verdes E, Tandon A, Patel V, Ustun B. Depression, chronic diseases, and decrements in health: results from the World Health Surveys. Lancet. 2007;370(9590):851–858. doi: 10.1016/S0140-6736(07)61415-9. [DOI] [PubMed] [Google Scholar]
- 43.Kivimaki M, Lawlor DA, Singh-Manoux A, et al. Common mental disorder and obesity: insight from four repeat measures over 19 years: prospective Whitehall II cohort study. BMJ. 2009;339:b3765. doi: 10.1136/bmj.b3765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Ekelund U, Brage S, Besson H, Sharp S, Wareham NJ. Time spent being sedentary and weight gain in healthy adults: reverse or bidirectional causality? Am J Clin Nutr. 2008;88(3):612–617. doi: 10.1093/ajcn/88.3.612. [DOI] [PubMed] [Google Scholar]
- 45.Kaplan GA, Lazarus NB, Cohen RD, Leu DJ. Psychosocial factors in the natural history of physical activity. Am J Prev Med. 1991;7(1):12–17. [PubMed] [Google Scholar]
- 46.Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psychol Meas. 1977;1(3):385. [Google Scholar]