The prevalence of obesity and its associated conditions, such as hypertension, dyslipidemia, and insulin resistance, has proliferated worldwide over the past 2 decades (16). Increases in metabolic risk profile can lead to metabolic syndrome, type II diabetes mellitus (T2DM), and cardiovascular disease (CVD). Obesity and metabolic risk factors may not only have physiological and metabolic consequences (17) but may also have psychological effects (5). Depression is more prevalent in obese individuals (11) and patients with diabetes (1), compared to in the general population. Depression may also be a major risk factor for obesity and its related complications (such as T2DM) because it may lead to behavioral changes such as reduced physical activity and increased energy intake (24). In addition, in people with chronic physical illness, depression is associated with increased health care use and increased functional disability and work absence, compared to in individuals with chronic physical illness without depression (23). Finally, individuals with T2DM who also suffer from depression have an increased risk for developing diabetic complications(1).
Questionnaires to quantify depression (such as the Beck Depression Inventory) have commonly been developed for psychiatric populations, but they produce skewed score distributions in other populations (2). The cardiac depression scale (CDS) was specifically developed, originally in cardiac patients, to assess the wide range of depressed moods seen in nonpsychiatric populations, to encompass “adjustment disorder with depressed mood” and “minor depression” and “major depression” on the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition classification (9). Because many obese and middle-aged individuals have mildly elevated levels of depressed mood, the CDS was thought to be a potentially suitable tool for assessing depression in the range found in these individuals who are at high risk for developing T2DM and CVD.
Lifestyle modifications are considered as important interventions for obese individuals and those with metabolic risk factors for T2DM and CVD. Interventions including diet with behavioral modifications (10), aerobic exercise training (18), and resistance exercise training (RT) (12) have been shown to improve quality of life (QoL) in obese populations and those with metabolic risk factors for T2DM and CVD. In addition, Fox (8) has suggested that exercise may be useful in the treatment of depression. As such, the aim of this study was to determine whether RT reduces the level of depressed mood in middle-aged individuals with risk factors for developing T2DM and CVD. To our knowledge, no study has examined the effect of RT on the depressed mood of people with a cluster of metabolic risk factors. The primary hypothesis was that RT would significantly reduce depressed mood, as measured by the CDS, in individuals with high numbers of metabolic risk factors (HiMF).
Experimental Approach to the Problem
Participants with varying numbers of metabolic risk factors were allocated to HiMF and LoMF groups, and then these 2 groups were each randomly allotted to either the exercise training or nonexercise control group. Levels of depressive symptoms were analyzed before and after the 10 weeks of interventions of either exercise or nonexercise group for both HiMF and LoMF.
A total of 55 (men = 28, women = 27) untrained middle-aged individuals (50.8 ± 0.9 years, range = 40-69 years; mean ± SEM) took part in the study. Participants' anthropometric measurements were as follows: height = 168.7 ± 1.3 cm (range = 152-186 cm), mass = 79.4 ± 2.3 kg (range = 40-116 kg), body mass index = 27.7 ± 0.7 kg·m−2 (range = 17-40 kg·m−2), and waist circumference = 92.2 ± 1.9 cm (range = 59-121 cm). Participants with 2 or more metabolic risk factors, according to the International Diabetes Federation criteria (IDF) (28), were classified as having HiMF and those with one or no metabolic risk factors were classified as having a low number of metabolic risk factors (LoMF). The rationale for the HiMF group allocation is that individuals with 2 or more risk factors are at a high risk of developing metabolic syndrome and T2DM (19). The IDF criteria include the following: waist circumference ≥94 cm for men and ≥80 cm for women, triglycerides ≥1.7 mmol·L−1, high-density lipoprotein <1.03 mmol·L−1 for men and <1.29 mmol·L−1 for women, systolic blood pressure ≥ 130 mm Hg or diastolic blood pressure ≥ 85 mm Hg (or hypertensive medications) and fasting blood glucose level ≥5.6 mmol·L−1. As described previously (12), after the allocation into HiMF and LoMF groups, participants were randomly allocated (by a person who was not involved in the study, using sealed envelopes) to 1 of 4 groups: HiMF training (HiMFT, men = 8, women = 5), HiMF nonexercise control (HiMFC, men = 10, women = 5), LoMF training (LoMFT, men = 3, women = 8), and LoMF nonexercise control (LoMFC, men = 4, women = 9). Randomization was stratified according to sex. Participants were on a range of medications including beta-blockers (n = 2), calcium channel blockers (n = 2), angiotensin-converting enzyme inhibitors (n = 4), diuretics (n = 1), statins (n = 2), metformin (n = 1), and hormone replacement therapy (n = 6). Participants were excluded if they had documented incidence of cardiac disease or they were involved in regular physical activity in the previous 6 months.
Participants were given written and verbal information on the nature of the study including the experimental risks and then signed an informed consent document before the investigation. The investigation was approved by the Victoria University and Austin Health Human Research Ethics Committees.
Assessment of the Number of Metabolic Risk Factors
The method of assessing the number of metabolic risk factors has been described previously (12). In brief, plasma glucose, triglyceride, and high-density lipoprotein levels were analyzed (SYNCHRON LX® System/Lxi725, Beckman Coulter Inc, Carlsbad, CA, USA) after a 12-hour fast. Blood pressure was measured using a mercury sphygmomanometer after participants had rested in a seated position for 15 minutes. Systolic and diastolic blood pressures were recorded to the nearest 2 mm Hg. Waist circumference was measured with a steel tape and taken as the smallest circumference between the iliac crest and the lower border of the ribs.
Cardiac Depression Scale
The CDS contains 26 items on a Likert scale from 1 to 7, 4 items being reverse scored, and a higher score indicating a more severe depressed mood (9). The CDS has excellent receiver operating characteristics with an area under the curve of 0.94 for any depression and 0.96 for major depression (20). Although originally developed in cardiac patients, it measures core aspects of depression (e.g., depressed mood, anhedonia, and sleep disturbance) measured by commonly used depression scales such as the Beck Depression Inventory, Hospital Anxiety Depression Scale, and the Center for Epidemiologic Studies Depression Scale (22). In addition to measurement of the severity of core depressive symptoms, it measures hopelessness-related cognitions associated with depression in persons adjusting to a chronic illness such as diabetes. The CDS has been shown to be a sensitive, reliable, and responsive tool for assessing changes in depression in both English speaking (3) and non-English speaking populations (25).
Questionnaires were administered by a single investigator. The internal construct validity of the CDS in this population was tested in all 55 participants, at baseline. The test-retest reliability of the CDS was assessed in the 28 participants randomly allocated to the nonexercise controls. The external CDS validity was confirmed using the generic Short Form 36 (SF-36) Health Survey, which is although not designed to measure depression, includes dimensions that reflect depressed mood. The SF-36 contains 36 items comprising 8 subscales. Four subscales evaluate the physical health dimension. The remaining 4 subscales constitute the mental health dimension (15,26). A higher score represents a higher level of function and health-related QoL.
Resistance Training Protocol
The training protocol was as described by Levinger et al. (12). In brief, the RT was conducted 3 d·wk−1 for 10 weeks. Training included 7 exercises: chest press, leg press, lateral pull-down, triceps pushdown, knee extension, seated row, and biceps curl. Training intensity was determined according to the 1 repetition maximum (1RM) method. This method has been shown to have high reliability for assessing muscle strength (13). In the first week, training consisted of 2 sets of 15-20 repetitions at 40-50% of the 1RM for that particular exercise. From weeks 2-10, participants performed 3 sets of 8-20 repetitions at 50-85% 1RM for each exercise. The wide range of repetitions is because of the stages of progression, 15-20 repetitions in week 2, 12-15 repetitions during weeks 3-6, and 8-12 repetitions during weeks 7-10. At each session, weights were adjusted according to the capacity of the individual, with weights increased if the participant was able to achieve the maximum number of prescribed range of repetitions for that week and decreased if the minimum number of range of repetitions was not able to be achieved.
Training data were analyzed for the 52 participants who completed the study. Multivariate analysis of variance was used to examine the differences in anthropometric and metabolic risk factors after the allocation to groups, that is, HiMFT vs. HiMFC and LoMFT vs. LoMFC (Table 1). One-way analysis of covariance (ANCOVA) was used to examine the effect of training on the CDS score as the HiMFT group had a significant higher CDS score at baseline. The dependent variable was the change (Δ) from pre-to-post in the CDS score, the fixed factor (independent variable) was the intervention group (training or control), and the covariate was the baseline (pretraining) score. The relationship between the change in CDS and the change in muscle strength was assessed using Spearman correlation with gender as a covariate.
The relationship between the CDS and the SF-36 was assessed using Spearman correlation between the total CDS score, separately with both the physical and mental dimensions of the SF-36. The baseline data of the 55 participants were used for the internal construct validity for the CDS in this particular population using standard methods as used by Birks et al. (3). Cronbach's α was calculated using the 26 individual items of each CDS questionnaire. Test-retest reliability was assessed by comparing the total CDS score at baseline and the score after 10 weeks for the 28 participants who were randomly allocated to the control group. These values were compared using Spearman correlation, intraclass correlation coefficient (ICC), and Bland-Altman plots (4).
Data are reported as mean ± SEM, and all statistical analyses were conducted at the 95% level of significance.
Validity and Reliability of the Cardiac Depression Scale
The internal reliability of the CDS score (n = 28) was high, with Cronbach's α = 0.84. The test-retest reliability was satisfactory with a Spearman correlation = 0.77 (p < 0.01) and ICC = 0.84. Bland-Altman plots revealed a mean and SEM of difference, between repeat CDS scores, of 2.1 ± 3.3. A significant correlation was found between the CDS scores and the physical (r = −0.78, p < 0.01) and mental (r = −0.69, p < 0.01) health dimensions of the SF-36. In addition, the distribution of scores in the CDS demonstrated greater normality, compared to the physical and mental health dimensions of the SF-36 (Figure 1).
Participants' anthropometric characteristics are shown in Table 1. At baseline, there were no significant depression score differences between LoMFC and LoMFT (67.9 ± 6.2 vs. 65.5 ± 7.2, respectively, p = 0.78). By chance, the HiMFT group had higher depression scores at baseline, compared to the HiMFC group (82.6 ± 5.9 vs. 62.9 ± 4.6, p = 0.01). There was also a trend toward higher depression scores in the HiMFT group, compared to in the LoMFT group (p = 0.07).
Adherence to Training
Three participants from the training groups (1 from the LoMFT group and 2 from the HiMFT group) did not complete the study, and their data were excluded from the training analyses. These 3 individuals did not complete the study because of medical reasons not related to the study (1 from the HiMFT group and 1 from the LoMFT group) or because of work-related reasons (1 person from the HiMFT group). The adherence to training was high in both training groups (HiMFT = 88%, and LoMFT = 96%).
The Effect of Training on Cardiac Depression Scale
After training, the depression score for HiMFT was reduced (improved) by 14.8 ± 4.9 points on the CDS, which was a significant improvement compared to both baseline (p = 0.01) and the HiMFC (p = 0.049) values (Figure 2). No significant change was observed for the LoMFT or LoMFC group (all p > 0.05) (Figure 2). As reported previously (12), muscle strength improved for both HiMFT training groups (by 25%, p < 0.01) and the LoMFT (by 23.7%, p < 0.01), compared to their controls. In the HiMF group only, the percent change in absolute muscle strength and relative muscle strength (total muscle strength/body mass) was correlated with the Δ change in the CDS score (r = −0.045, p = 0.009, and r = −0.46, p = 0.008, respectively).
The main finding of this study is that RT may alleviate depression in individuals at high risk of developing T2DM and CVD. It also confirms that the CDS is a robust measure of depressed mood in this population.
It is widely reported that exercise training (6) can improve QoL in middle-aged and elderly individuals. It has also been reported that both aerobic (18) and resistance (12) training regimens can improve QoL in individuals at high risk of developing T2DM and CVD. Previous studies have shown that exercise can improve depression in elderly individuals with major or minor depression (21). There are, however, limited data with regard to the effect of RT on symptoms of depression in middle-aged individuals at a high risk for developing T2DM and CVD. We hypothesized that RT can improve depressive symptoms in this clinical population. We found that short-term RT improved depressive mood in the experimental cohort. The implication is that depression in these clinical populations may be alleviated by the application of RT, commencing at moderate intensity and progressing over several weeks to include high-intensity training. Further studies are needed to investigate the sustainability of these early benefits. Previously, we have shown that, in this population, RT can improve muscle strength, functional capacity, and self-perceived QoL (12) without changing metabolic risk factors such as fasting glucose levels, lipid profile, blood pressure, and waist circumference (14). This suggests that the improvement in the depression score for the HiMFT group occurred independent of changes to metabolic risk profiles. This also shows that the CDS is sensitive to exercise training-induced changes in depression. It is important to note that no change was observed in the depression score for the LoMFT group, compared to baseline and LoMFC. The different effects of exercise on HiMFT and LoMFT may be related to higher depression scores for HiMFT at baseline. This suggests that it is more difficult to improve depression score in individuals with relatively lower scores of depression or that a longer or more intense training protocol may be needed to improve depression scores in individuals with LoMF. It is unlikely that the improvement in the depression score in the HiMFT group was simply because of “regression toward the mean,” as (a) this bias phenomenon is of more relevance in nonrandomized studies (27), (b) we have used ANCOVA analysis to take the higher baseline score under consideration in the statistical analysis and (c) the percent changes in absolute muscle strength and relative muscle strength were correlated with the Δ change in CDS in the HiMF group only. This suggests that the improvement in CDS score was related to the increase in muscle strength as a result of the training.
Depression, especially mild depression, is more common in middle-aged individuals and those with metabolic risk factors or CVD, compared to the general population (7). As such, to identify and to treat individuals with depression, it is important to have a valid, reliable, sensitive, and simple tool for assessing depression. The results from this study indicate that, although designed for patients with overt cardiac disease, the CDS would appear to be a valid and reliable tool for middle-aged individuals without overt cardiac disease. The intercorrelations between the CDS and the SF-36 in this study are similar to the intercorrelations reported previously between the 2 questionnaires in cardiac patients (3). In addition, in both this study and that of Birks et al. (3), the CDS demonstrated a more normal distribution of scores compared to the SF-36, suggesting that the CDS is more sensitive to extreme (lower and higher) scores (3). Another advantage of the CDS is that it is written in relatively simple language and takes only minutes to complete.
This study has 3 potential limitations: (a) the possibility for sex bias between HiMF (male:female 20:10) and LoMF (8:17) groups. It is important to note that of the total sample of 55 participants, 28 were men and 27 were women. The allocation to HiMF or LoMF groups was based on objective IDF criteria, and participants were randomly allocated and stratified according to sex to ensure near equal numbers of men and women in each study subgroup (i.e., LoMFC and LoMFT and HiMFC and HiMFT, Table 1). (b) Despite the randomization, HiMFT had higher depression scores at baseline, compared to HiMFC. We have stratified the randomization to sex and used ANCOVA to take the higher baseline depression severity score into consideration. However, it is recommended that future studies stratify the randomization to groups according to baseline depression scores and not only sex. (c) We did not assess the long-term effect of RT on depressive symptoms and as such future studies should include a follow-up measure to identify for how long the benefits of RT on depressive moods last.
In conclusion, RT programs that consist of 7-8 exercises for the major muscle groups at both low-moderate and moderate-high intensities appear to alleviate the depressed mood in individuals who have multiple numbers of metabolic risk factors associated with T2DM and CVD. The CDS, as a measure of depressed mood, is a responsive tool for assessing exercise intervention in these individuals.
Resistance exercise training appears to alleviate depressed mood in individuals who have multiple numbers of metabolic risk factors associated with T2DM and CVD. Implications from this study are that RT programs consisting of both low-moderate and moderate-high intensities can have positive effects on the depressive mood of people with clusters of metabolic risk factors. Furthermore, based on the data that we presented here, we recommend the following: low-moderate intensity training of 6-8 exercises covering all major muscle groups with 2-3 sets of 15-20 repetitions each and at approximately 50-65% of 1RM; and moderate-high intensity training consisting of the same exercises with 2-3 sets of 8-15 repetitions, up to 85% of 1RM. In summary, RT is a simple and effective method to improve depressed mood in this population and should form an important part of the exercise training regimens for people at a high risk of developing T2DM and CVD.
1. Astle, F. Diabetes and depression: A review of the literature. Nutr Clin Nor Am
7: 67-78, 2007.
2. Beck, AT, Ward, CH, Mendelson, M, Mock, J, and Erbaugh, J. An inventory for measuring depression. Arch Gen Psych
4: 561-571, 1961.
3. Birks, Y, Roebuck, A, and Thompson, DR. A validation study of the Cardiac Depression Scale
(CDS) in a UK population. Br J Health Psychol
9: 15-24, 2004.
4. Bland, JM and Altman, DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet
1: 307-310, 1986.
5. Bray, GA. Medical consequences of obesity. J Clin Endocrinol Metabol
89: 2583-2589, 2004.
6. Ellingson, T and Conn, VS. Exercise and quality of life in elderly individuals. J Gerontol Nutr
26: 17--25, 2000.
7. Fontaine, KR, Cheskin, LJ, and Barofsky, I. Health-related quality of life in obese persons seeking treatment. J Fam Prac
43: 265-270, 1996.
8. Fox, KR. The influence of physical activity on mental well-being. Public Health Nutr
2: 411-418, 1999.
9. Hare, DL and Davis, CR. Cardiac depression scale
: Validation of a new depression scale for cardiac patients. J Psychosom Res
40: 379-386, 1996.
10. Kaukua, J, Pekkarinen, T, Sane, T, and Mustajoki, P. Health-related quality of life in obese outpatients losing weight with very-low-energy diet and behaviour modification—A 2-y follow-up study. Int J Obes Relat Metabol Dis
27: 1233-1241, 2003.
11. Kottke, TE, Wu, LA, and Hoffman, RS. Economic and psychological implications of the obesity. Mayo Clinic Proc
78: 92-94, 2003.
12. Levinger, I, Goodman, C, Hare, DL, Jerums, G, and Selig, S. The effect of resistance training on functional capacity and quality of life in individuals with high and low numbers of metabolic risk factors
. Diabetes Care
30: 2205-2210, 2007.
13. Levinger, I, Goodman, C, Hare, DL, Jerums, G, Toia, D, and Selig, S. The reliability of the 1RM strength test for untrained middle-aged individuals. J Sci Med Sport
12: 310-316, 2009.
14. Levinger, I, Goodman, C, Matthews, V, Hare, DL, Jerums, G, Garnham, A, and Selig, S. BDNF, risk factors for metabolic syndrome and resistance training in middle-aged individuals Med Sci Sports Exerc
40: 535-541, 2008.
15. McHorney, CA, Ware, JE, and Raczek, AE. The MOS 36-item short-form health survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care
31: 247-263, 1993.
16. Miranda, PJ, DeFronzo, RA, Califf, RM, and Guyton, JR. Metabolic syndrome: Definition, pathophysiology, and mechanisms. Am Heart J
149: 33-45, 2005.
17. Must, A, Spadano, J, Coakley, EH, Field, AE, Colditz, G, and Dietz, WH. The disease burden associated with overweight and obesity. J Am Med Assoc
282: 1523-1529, 1999.
18. Nishijima, H, Satake, K, Igarashi, K, Morita, N, Kanazawa, N, and Okita, K. Effects of exercise in overweight Japanese with multiple cardiovascular risk factors. Med Sci Sports Exerc
39: 926-933, 2007.
19. Reaven, GM. The metabolic syndrome: Is this diagnosis necessary? Am J Clin Nutr
83: 1237-1247, 2006.
20. Shi, WY, Stewart, AG, and Hare, DL. Both major and minor depression can be accurately assessed using the cardiac depression scale
. Eur Heart J
29: S745, 2008.
21. Singh, NA, Clements, KM, and Singh, MA. The efficacy of exercise as a long-term antidepressant in elderly subjects: A randomized, controlled trial. J Gerontol
56: M497-M504, 2001.
22. Snaith, P. What do depression rating scales measure? Br J Psychiatry
163: 293-298, 1993.
23. Stein, MB, Cox, BJ, Afifi, TO, Belik, SL, and Sareen, J. Does co-morbid depressive illness magnify the impact of chronic physical illness? A population-based perspective. Psychol Med
36: 587-596, 2006.
24. Stunkard, AJ, Faith, MS, and Allison, KC. Depression and obesity. Biol Psychiatry
54: 330-337, 2003.
25. Wang, W, Thompson, DR, Chair, SY, and Hare, DL. A psychometric evaluation of a Chinese version of the Cardiac Depression Scale
. J Psychosom Res
65: 123-129, 2008.
26. Ware, JE. SF-36 Physical and Mental Health Summary Scales: A Manual for Users of Version 1
. Boston, MA: The Health Institute, New England Medical Center, 1997.
27. Weeks, DL. The regression effect as a neglected source of bias in nonrandomized intervention trials and systematic reviews of observational studies. Eval Health Prof
30: 254-265, 2007.
28. Zimmet, PZ, Alberti, KG, and Shaw, JE. Mainstreaming the metabolic syndrome: A definitive definition. This new definition should assist both researchers and clinicians. Med J Aus
183: 175-176, 2005.