To determine whether obesity, inflammation, or conventional risk factors are related to depressive symptoms (DeprSy) in the general population. Responses to 3 questions served to assess sense of depression. Body mass index (BMI), C-reactive protein (CRP), and other epidemiologic data of participants were available. In 1940, individuals who consulted a psychiatrist in the preceding year, or felt depressed (together DeprSy), 248 were female. Logistic regressions for adjusted associations of BMI with DeprSy were not significant as was serum CRP level. Diabetes and, in men, fasting glucose concentrations were associated with DeprSy. Systolic blood pressure (SBP) was robustly inversely associated with DeprSy in diverse models at ORs of 0.74 (95%CI 0.63; 0.89) independent of confounders, including antihypertensive and lipid-lowering medication. The use of antidepressants could not explain the reduced BP. Women are predisposed to depression with which, not BMI and CRP, but SBP is inversely associated. Anti-inflammatory substances produced in depressed persons might explain the slightly lower BP.
Evidence is available for the association of both obesity and depressive disorder with increased risk of coronary heart disease (CHD) incidence [
Underlying mechanisms such as chronic inflammation have been extensively investigated as candidate pathways that subsequently link obesity and depression in an attempt to explain how each confers vulnerability to the other and subsequently elevate as the risk for physical illness [
We aimed to investigate whether obesity, systemic inflammation, or a conventional cardiovascular risk factor is related to DeprSy in a representative sample of a middle-aged and elderly general population by using body mass index (BMI), serum CRP as markers, and other risk factors including blood pressure. Such information could serve at least a twofold purpose, namely, (a) regarding utility for screening depressed individuals in the population at large and (b) as a reference information of possible use in future trials with weight-reducing drugs in differentiating potential side effects of depression.
This study sample was composed of participants of the 2008/09 follow-up survey of the Turkish Adult Risk Factor (TARF) Study. The TARF is a longitudinal study on the prevalence of cardiac disease and risk factors in a representative sample of adults in Turkey carried out periodically almost biennially since 1990 in 59 communities throughout all geographical regions of the country [
Following 3 questions were directed to participants during face-to-face interview by a physician at his 4th year of specialization. Questions were to be replied by Yes/No; the second query included a subsidiary question. Clarification was given by interviewer, when necessary. Did you consult in the past year a physician due to a change in your mood or spirits? Do you feel depressed? If so, do you use antidepressant drugs? Have there been periods that life is not worth living or that you thought of seriously harming yourself?
Blood pressure was measured with an aneroid sphygmomanometer (Erka, Bad Tölz, Germany) in the sitting position on the right arm, and the mean of two recordings 5 min apart was recorded. Height was measured without shoes using a measuring stick and weight in light indoor clothes using scales. BMI was computed as weight divided by height squared (kg/m2).
In regard to cigarette smoking, nonsmokers, former smokers, and current smokers formed the categories. Anyone who drank alcoholic drinks once a month or more frequently was considered as alcohol consumer.
Descriptive parameters were shown as mean ± standard deviation, estimated mean ± standard error, or in percentages. Due to the skewed distribution, values derived from log-transformed (geometric) means were used for CRP and insulin.
1940 men and women who were interviewed in the months of September of 2008 and of 2009 formed the sample population; 966 (49.8%) were female, and 974 (50.2%) male. 1692 people (87.2%) negated all questions; these are referred to as “healthy” (Figure
Diagrammatic representation of the responses of participants to the interview.
Females were significantly more often DeprSy than males (19.8% versus 6/2%,
Baseline characteristics of the sample population (
Women ( | Men ( | |||||||
Healthy ( | Depres. disorder* ( | Healthy ( | Depres. disorder* ( | |||||
Mean | SD | Mean | SD | Mean | SD | Mean | SD | |
Age, years | 57.5 | 11 | 9.8 | 56.4 | 11 | 56.3 | 11.7 | |
Systolic blood pressure, mmHg | 129.0 | 21 | 21.3 | 121.9 | 19 | 118.6 | 16.4 | |
Diastolic blood pressure, mmHg | 77.0 | 10.3 | 10 | 74.7 | 10 | 74.1 | 9.6 | |
Current smoker, | 92 | 13 | 26 | 15.4 | 306 | 38 | 18 | 36 |
Alcohol user, | 2 | 0.3 | 2 | 1.2 | 102 | 12.5 | 7 | 13.7 |
Body mass index, kg/m2 | 31.2 | 6 | 31.3 | 5.6 | 28.2 | 4.3 | 27.8 | 4.1 |
Fasting glucose, mg/dL | 102.3 | 40.2 | 104.8 | 48 | 102.7 | 39.7 | 116.1 | 60.6 |
Total cholesterol, mg/dL | 199.7 | 44.4 | 42.1 | 198 | 41.8 | 192.6 | 37.7 | |
HDL cholesterol, mg/dL | 48.1 | 12.4 | 49.6 | 13.5 | 40.6 | 11 | 40.9 | 8 |
Fast. triglycerides, mg/dL | 177.2 | 117 | 190.7 | 168 | 177.2 | 117 | 190.7 | 168 |
Fasting insulin, mIU/L | 8.64 | 2.02 | 8.74 | 2.00 | 8.28 | 2.05 | 9.17 | 2.04 |
C-reactive protein, mg/L | 2.50 | 2.88 | 2.35 | 2.92 | 2.04 | 2.73 | 1.73 | 2.9 |
Presence of CHD, | 80 | 10.3 | 19 | 10.2 | 122 | 13.4 | 9 | 15.3 |
Presence of diabetes, | 119 | 15.5 | 33 | 17.6 | 135 | 15.3 | 13 | 23.2 |
*includes those who consulted a physician or felt depressed without having consulted a physician.
Table
Logistic regression analysis for the association of certain parameters with depressive disorder, by gender.
OR 95% CI | OR 95% CI | OR 95% CI | ||||
241/1919† | 57/963† | 184/956† | ||||
Female gender | 2.87; 5.49 | |||||
Age, 11 years | 0.91 | 0.77; 1.06 | 1.07 | 0.82; 1.41 | 0.70; 1.02 | |
Body mass index, 5 kg/m2 | 1.08 | 0.94; 1.23 | 0.97 | 0.69; 1.35 | 1.10 | 0.96; 1.28 |
Systolic b. pressure, 20 mmHg | 0.63; 0.87 | 0.80 | 0.58; 1.10 | 0.60; 0.89 | ||
189/1594† | 49/822† | 140/772† | ||||
Female gender | 2.20; 5.52 | |||||
Age, 11 years | 0.90 | 0.76; 1.07 | 1.04 | 0.77; 1.44 | 0.84 | 0.68; 1.03 |
Body mass index, 5 kg/m2 | 1.01 | 0.87; 1.18 | 0.99 | 0.85; 1.43 | 1.03 | 0.87; 1.22 |
Systolic b. pressure, 20 mmHg | 0.62; 0.89 | 0.75 | 0.52; 1.06 | 0.60; 0.92 | ||
Fasting glucose, 40 mg/dL | 1.00; 1.32 | 1.00; 1.55 | 1.08 | 0.89; 1.27 | ||
Current versus never smoking | 0.83 | 0.45; 1.53 | 0.92 | 0.42; 2.00 | 0.99 | 0.30; 3.28 |
Antihypert. drug use, yes/no | 1.19 | 0.83; 1.73 | 1.20 | 0.73; 1.81 | 1.17 | 0.76; 1.79 |
Lipid-lower. drug use, yes/no | 1.08 | 0.63; 1.88 | 1.07 | 0.54; 2.08 | 1.03 | 0.54; 1.97 |
†number of depressive disorder/total number.
201/61 antihypertensive/lipid-lowering drug usage in men, 322/75 usage in women, respectively.
Significant values are highlighted in boldface.
When the association between DeprSy and SBP was scrutinized by stratifying for individuals using antidepressive drugs, participants reporting the use of antidepressants (
When only consulters were analyzed by multiple logistic regression again in the 2 similar models (Table
Logistic regression analysis for the association of certain parameters with individuals consulting a physician, by gender.
OR 95% CI | OR 95% CI | OR 95% CI | ||||
184/1919† | 42/963† | 142/956† | ||||
Female gender | 2.76; 5.76 | |||||
Age, 11 years | 0.95 | 0.79; 1.12 | 1.18 | 0.88; 1.61 | 0.86 | 0.70; 1.06 |
Body mass index, 5 kg/m2 | 1.07 | 0.93; 1.24 | 0.92 | 0.62; 1.36 | 1.10 | 0.94; 1.29 |
Systolic b. pressure, 20 mmHg | 0.63; 0.92 | 0.80 | 0.56; 1.17 | 0.62; 0.94 | ||
147/1594† | 36/822† | 111/772† | ||||
Female gender | 2.02; 5.68 | |||||
Age, 11 years | 0.91 | 0.75; 1.10 | 1.09 | 0.76; 1.56 | 0.84 | 0.66; 1.06 |
Body mass index, 5 kg/m2 | 1.05 | 0.89; 1.23 | 0.94 | 0.61; 1.46 | 1.07 | 0.89; 1.28 |
Systolic b. pressure, 20 mmHg | 0.62; 0.94 | 0.75 | 0.51; 1.13 | 0.62; 1.00 | ||
Fasting glucose, 40 mg/dL | 0.96 | 0.82; 1.17 | 1.08 | 0.79; 1.43 | 0.92 | 0.76; 1.17 |
Current versus never smoking | 0.80 | 0.40; 1.60 | 0.97 | 0.40; 2.35 | 0.95 | 0.25; 3.59 |
Antihypert. drug use, yes/no | 1.23 | 0.82; 1.85 | 0.91; 4.34 | 1.05 | 0.65; 1.68 | |
Lipid-lower. drug use, yes/no | 0.92 | 0.48; 1.75 | 0.90 | 0.26; 3.15 | 0.90 | 0.43; 1.90 |
†number of consulters/total number.
Antihypertensive/lipid-lowering drug usage in 523/136 men and women, respectively.
Significant values are highlighted in boldface.
In a further model, we examined in 1745 subjects the association of CRP with DeprSy, along with BMI, SBP, and other variables (Table 4, model 2). CRP, BMI, and smoking status persisted not to be associated whereas female sex (4-fold), and, inversely, SBP was highly significantly associated [OR 0.72 (95%CI 0.60; 0.85]. Serum CRP disclosed for a 3-fold increment an OR of 0.98 (95%CI 0.88; 1.08).
Salient findings in this population-based study seeking the relationship between depression and obesity/inflammatory markers and conventional cardiovascular risk factors was that, while BMI and serum CRP level were not associated, female sex and fasting glucose were significantly associated with DeprSy. Noteworthy was that SBP was inversely and robustly associated with DeprSy in the total study sample.
Our finding relative to a current depressive disorder in this middle-aged population to prevail in 6.2% in men, and 19.8% of women is coarsely in line with the reported prevalence of a major lifetime depression among 6914 young US adults from the NHANES-III survey as 5.7% in men and 11.7% in women, corresponding to a two-fold risk in women as in men [
Mean BMI in the NHANES-III survey was not significantly different in persons with than without a major depression (25.8 versus 25.4 kg/m2,
Among young US adults from the NHANES-III survey, an elevated CRP (≥2.14 mg/L) was observed in 13.7% of men and 27.3% in women. Elevated levels were not associated in multivariably adjusted with current or lifetime depression in females with which our findings are in agreement. No difference in BMI, nor in TNF-
Level of IL-6 was not independently associated with elevated depressive symptoms also in a study on 416 Finns, but levels of the anti-inflammatory IL-1 receptor antagonist were exhibited an OR over 2-fold compared to those not belonging to the group with elevated depressive symptoms [
The surprising and main finding in the current study was that BP was lower in depressed subjects (in the order of 5/2 mmHg) which could not be accounted for by age, smoking status, BMI, glucose, or CRP. The odds for DeprSy in multivariable analyses were consistently lower by one-quarter for a 1-SD increment in SBP in each gender (though failing to reach significance in men because of limited statistical power). Although reasons for it may be unclear, this association was robust, exhibiting a narrow confidence interval and preserving the OR in variously adjusted models in each gender. In subgroup analyses, we could exclude the use of antidepressants as possibly accounting for this finding.
This finding does not seem to be specific for the current study. Though in elderly Chinese, not detected hypertension but undetected severe hypertension (≥160/95 mmHg) was reported to be associated with depression (OR 1.78; [95%CI 1.05; 3.01]) (17) that may have been related to a high BMI or to diabetes for which adjustment was not made. In young US adults from the NHANES-III survey, mean SBP was lower by 1.5 mmHg in persons with than without a major depression (
It has been proposed that anti-inflammatory substances (including anti-inflammatory cytokines IL-4 and IL-10) or IL-1 receptor antagonist [
The assessment of depression herein is limited by using the indicators of such a disorder rather than a diagnosed depression. The reliability and validity of our questionnaire has not been formally tested in community samples, but it is congruent with the essence of the Beck Depression Inventory commonly utilized in clinical [
We
The financial support for the Turkish Adult Risk Factor survey over the years by the Turkish Society of Cardiology and the various pharmaceutical companies in Istanbul, Turkey, is gratefully acknowledged.