Waist circumference, abdominal obesity, and depression among overweight and obese U.S. adults: national health and nutrition examination survey 2005-2006

Obesity continues to be a cause of public concern in the United States and worldwide [1]. The prevalence of obesity, defined as a body mass index (BMI) of ≥ 30 kg/m², was 32% in the United States during 2001-2004 and increased slightly to 34% during 2005-2006 [2]. The health impact of obesity is tremendous, as shown by an increased risk for multiple chronic diseases and conditions including hypertension, diabetes, hypercholesterolemia, coronary heart disease, asthma, arthritis, cancers, and many others [3‐5]. In addition, obesity, especially abdominal obesity, is associated with increased all-cause, cardiovascular, or cancer mortality [6].

In addition to the broad range of obesity-related physiologic outcomes, obesity is associated with an increased risk for a number of mental disorders (i.e., depression, bipolar disorder, panic disorder, anxiety, or many others) [7‐15] that have a substantial impact on public health (e.g., associated with great burden of diseases and increased mortality, disability, and reduced quality of life) [16, 17]. However, other studies showed that BMI was not [18‐22] or was even inversely associated with some forms of mental illness [20, 23]. A possible explanation for these inconsistent results is that BMI as a measure of overall obesity does not account for varying proportions of muscle mass, bone, and fat, or the distribution of body fat. In fact, studies have consistently shown that abdominal visceral fat is more pathogenic than subcutaneous fat on metabolic risk profiles [24, 25] and that fat distribution (central adiposity vs general obesity, or visceral vs subcutaneous fat) is differentially associated with depressive symptoms [20, 26, 27]. Waist circumference, frequently used as a simple, inexpensive measure of central obesity in population-based studies, has been shown to be associated with depression in some studies [28, 29] but not in others [21, 23]. By using a large nationally representative sample, we sought to examine whether abdominal obesity measured by waist circumference was associated with depressive symptoms among overweight and obese adults after taking into consideration multiple risk factors including demographic characteristics, lifestyle factors, and coexistence of multiple chronic conditions. In this study, we only focused on adults who were overweight and obese because abdominal obesity is more physiologically or psychologically relevant in this population than in people who are underweight or normal weight, among whom depressive symptoms are unrelated to visceral fat [27]. Our study makes a unique contribution to the literature by using a larger, population-based, and nationally representative sample including both men and women with objective measures of overall and central obesity, which is rare for epidemiologic research.

Overall, the unadjusted and age-adjusted prevalence of having major depressive symptoms among adults who were overweight or obese was 2.5% (95% CI: 1.7-3.7%) and 2.3% (95% CI: 1.6-3.4%), respectively, and was 5.6% (95% CI: 4.4-7.0%) and 5.4% (95% CI: 4.3-6.7%), respectively, for having moderate-to-severe depressive symptoms (PHQ-9 score ≥ 10). Participants' socio-demographic characteristics differed significantly by depressive symptom status except for race/ethnicity (Table 1). Notably, the percentages of adults who were middle-aged (40-< 60 years), female, and obese were significantly higher, whereas the percentages of adults who attained an educational level of > high school diploma or had a poverty-income ratio of ≥ 3 were significantly lower, among participants with major depressive symptoms or moderate-to-severe depressive symptoms than among those without (P < 0.05 for all comparisons). Overall, the percentages of adults with ≥ 3 chronic conditions were higher among participants with major depressive symptoms or moderate-to-severe depressive symptoms than among those without (Table 1). The mean waist circumference among participants with major depressive symptoms or moderate-to-severe depressive symptoms was significantly higher compared to those without depression (P < 0.05).

The unadjusted and age-adjusted prevalence of having major depressive symptoms or moderate-to-severe depressive symptoms was significantly higher among participants with abdominal obesity than among those without abdominal obesity (Figure 1A and 1B, P < 0.001). Stratified analyses on overweight and obese adults yielded similar results (Figure 1). In unadjusted models (Model 1), waist circumference was significantly associated with the presence of both major depressive symptoms and moderate-to-severe depressive symptoms (P < 0.01, Table 2); the relationships persisted after adjusting for socio-demographic variables and lifestyle factors (Model 2). After further adjusting for the coexistence of multiple chronic diseases (Model 3), waist circumference remained significantly associated with the presence of major depressive symptoms (P = 0.031) but was only marginally associated with the presence of moderate-to-severe depressive symptoms (P = 0.074). When abdominal obesity was entered in the models, significant associations between abdominal obesity and depressive symptoms (by all definitions) existed after adjusting for all potential confounders (P ≤ 0.01, Table 2). No significant interactions between sex and waist circumference or between sex and abdominal obesity were observed in fully adjusted models.

When data analyses were further stratified by BMI and abdominal obesity status, abdominal obesity remained significantly associated with having major depressive symptoms (OR: 1.67, 95% CI: 1.12-2.50) and marginally associated with having moderate-to-severe depressive symptoms (OR: 2.03, 95% CI: 0.98-4.20) among participants who were overweight. However, the odds ratios between obese people with abdominal obesity and those without were not significant (Table 3).

Using a large, population-based sample from NHANES, we found that, among overweight and obese adults, waist circumference and/or abdominal obesity was significantly associated with increased prevalence and likelihood of having major depressive symptoms or moderate-to-severe depressive symptoms, suggesting abdominal obesity is a strong correlate of depression, particularly for adults who were overweight by their BMI status.

The relationship between depression and waist circumference or abdominal obesity as a component of metabolic syndrome has been explored previously in studies examining the associations of metabolic syndrome with mental illness [39‐43]. The inconsistent results of these studies may have resulted from differences in the populations being studied, in the measures of depression used, or in the number and type of covariates controlled for across studies. For example, three studies that were conducted in participants aged 35-55 years in London using the 4-item depression subscale of the General Health Questionnaire [39], in participants aged 25-84 years in Australia using the Hospital Anxiety and Depression Scale [40], and in Japanese men aged 20-67 years using the Profile of Mood States of the Likert-scale questionnaire [43] showed a significant association between waist circumference and depression. However, two studies conducted in Finland, one in participants aged 31 years using the Hopkins Symptom Checklist-25 questionnaire [41] and the other in participants aged 36-55 years using the Beck Depression Inventory [42], failed to observe a significant association as did studies conducted in Chinese elderly (aged ≥ 55 years) using the Geriatric Depression Scale-15 items [23] and in participants aged ≥ 25 years in New Zealand using self-reported, physician-diagnosed depression [21]. Moreover, two studies conducted in middle-aged women (mean age 50.4 years) using the Center for Epidemiological Studies Depression scale [27] and in overweight premenopausal women using the Zung's Self-Rating Depression Scale [44] reported that central obesity measured as visceral fat (but not subcutaneous fat) was significantly associated with an increased likelihood of having depression; surprisingly, waist circumference as an indicator of central obesity was not associated with depression in the study conducted by Everson-Rose et al [27]. A recent study using the PHQ-9 reported that waist circumference in the third and fourth quartiles was significantly associated with an increased likelihood of moderate-to-severe depression but not major depression [45], however, that study was conducted in U.S. adult women only, and only age- or age- and BMI-adjusted odds ratios were reported [45]. Our study using the data from both men and nonpregnant women who were overweight and obese further demonstrated that waist circumference and abdominal obesity were significantly associated with both major depressive symptoms and moderate-to-severe depressive symptoms after adjusting for multiple potential confounders. However, we did not conduct sex-stratified analyses because interactions between sex and waist circumference or abdominal obesity on outcome measures were not significant in the present study. In addition, we did not include BMI as a covariate because we only focused on people who were overweight and obese and also because of the high correlation between waist circumference and BMI [46, 47]. Whether or not BMI should be included as a covariate in studies like ours or in studies dealing with metabolic syndrome [39‐43] remains controversial at present.

Our stratified analyses by BMI and abdominal obesity revealed that overweight adults with abdominal obesity were more likely to have depressive symptoms (by both definitions) than overweight adults without abdominal obesity; however, there were no differences in obese adults with and without abdominal obesity. The fact that about 96% of obese adults have abdominal obesity may explain this observation. Nevertheless, our finding is consistent with previous research in obese women reported by Ma and Xiao [45]. Moreira et al.[29] reported that increasing in waist circumference was significantly associated with an increased prevalence of depressive symptoms and mood disorders in obese women; however, in that study, only simple correlation analysis was conducted and potential confounders were not taken into account. Taken together, the negligible differences in the prevalence and the odds ratios of having depressive symptoms between overweight and obese adults with abdominal obesity in the present study further suggest that waist circumference or abdominal obesity may be a preferred predictor of depression in this population.

Our study is subject to several limitations. First, the causal relationship between waist circumference or abdominal obesity and having depressive symptoms cannot be established based on the nature of our cross-sectional study. A growing body of evidence has shown that a bidirectional relationship may exist. Obesity in adolescence was associated with later depression in young adulthood [22]. The poor social relationships, low socioeconomic status, and the multiple chronic diseases associated with obesity may have predisposed obese people to impaired mental health. On the other hand, longitudinal studies have shown that baseline depression is a significant predictor of visceral fat accumulation and obesity [48‐50] and is associated with increased adrenal gland volume [49]. The latter suggests a long-term increased production of stress hormones from the hypothalamic-pituitary-adrenal (HPA) axis is involved in depression, which contributes to body fat accumulation [51, 52]. Second, our study was conducted only in overweight and obese participants representing a high risk population; this may have affected the generalizability of our results. Third, we conducted our analyses from combined data from both men and women mainly due to lack of interactions between sex and waist circumference or abdominal obesity on outcome measures and due to relatively small sample size. Future studies using sex-stratified data analyses are warranted to further explore sex disparities in the associations of depression with waist circumference and abdominal obesity and to study the potential effects of menopause on the associations. Fourth, we used PHQ-9 as a measure of depressive symptoms rather than a clinical diagnosis of depression. Although the PHQ-9 depression assessment has been validated in the general population including people who are overweight and obese as well as in patients with diabetes, coronary artery disease, or chronic heart failure, research on specific validation of this instrument in obese adults is currently not available. Thus, studies on clinical diagnosed depression and its association with abdominal obesity are warranted. Finally, antidepressant treatments, which are associated with weight gain [53], were not taken into account in the present study.

Our study from a large nationally representative sample demonstrated that waist circumference or abdominal obesity was associated with an increased likelihood of having major or moderate-to-severe depressive symptoms among overweight and obese adults. The continuing increases in BMI and waist circumference in the United States [54, 55] and the projected increases in the prevalence of overweight and obesity [1, 56] suggest that mental health status should be screened, monitored, and evaluated especially in people with abdominal obesity. A routine anthropometric measure of waist circumference as a simple and practical measure of abdominal obesity may be useful for providing information on depression risk in this population.