Obesity and depression are important sources of disease burden, but the extent to which these two conditions are related to each other remains unclear. Although a recent meta-analysis of prospective cohort studies suggest that people with greater body mass index (BMI) have an increased risk of depressive symptoms,¹ several individual studies report no association between obesity and these symptoms² and another group of studies show greater BMI to be associated with reduced risk of future mental health problems³ and suicide⁴. Recent Mendelian randomization studies, using adiposity-related genetic variants as an unconfounded instrument variable for obesity have also produced inconsistent findings. For example, using the fat mass and obesity-associated (FTO) genotype as a proxy for higher BMI, obesity was associated with increased risk of depressive symptoms in men participating in the Whitehall II study,⁵ but lower likelihood of psychological distress and antidepressant use in a study of Danish adults.⁶ One possible explanation for these inconsistencies is that the association between obesity and depressive symptoms is context specific, such that the nature of the association differs in different populations. However, this is likely to explain only a small part of the discrepancies as conflicting findings have been reported between studies from the same country.

Obesity is typically accompanied by unfavourable metabolic profiles, such as high glucose, adverse lipid levels, systemic inflammation and elevated blood pressure, but it is increasingly recognised that this may not always be the case. ^7–9 An attempt to capture this heterogeneity is the concept of “metabolically healthy obesity”, used to describe individuals with a BMI≥ 30 kg/m² but an otherwise healthy metabolic profile.^7,8 There is convincing evidence to show adverse effects of obesity on health,^10,11 but some recent research suggests that metabolically healthy obesity is not associated with increased cardiovascular disease risk^12–17 although the findings are not entirely consistent.^18,19 Since metabolically healthy obesity might simply reflect the early stages of excess adiposity, it is difficult to separate the effects of time being obese from the impact of ageing on metabolism. In this regard it is useful to utilise cohort studies of older participants.

In the present study, that was based on a general population sample of older adults, we extend this line of research to examine, for the first time, whether the association of obesity with depressive symptoms is dependent on the individual’s metabolic health. Since some studies have demonstrated an association of depression with metabolic syndrome^20,21 and its individual components such as impaired glycaemic control^22–24 and inflammatory markers,²⁵ we hypothesized that metabolically healthy obesity would not be associated with risk of depression.

A total of 8,688 participants (82% of wave 1 participants) attended the wave 2 (baseline) clinical assessment. The present study reports only on those that consented and were eligible and able to give blood (n=5903); this excluded men and women with clotting and bleeding disorders, or taking anti-coagulant medication (this drug can sometimes cause problems in clotting after venepuncture). After excluding 875 participants that did not attend the wave 3 follow up, and a further 1177 because of missing data, the final analytic sample comprised 3851 individuals (aged 63.0 ± 8.9 yrs, 45.1% men). In comparison with the baseline sample, the sub-group used in the present analyses were slightly younger (63.0 vs. 63.8 yrs, p<0.001), had a lower prevalence of longstanding illness/disability (50.4% vs. 58.1%, p<0.001), and better health behaviours including lower rates of smoking (13.2% vs. 17.9%, p<0.001) and greater physical activity (32.6% vs. 23.4%, p<0.001, vigorously active ≥1/wk). While these differences are statistically significant, the absolute difference was small.

Obesity prevalence was 27.5%, but 34.3% of this group was categorized as metabolically healthy at baseline (Table 1). Compared with metabolically unhealthy obese and non-obese, metabolically healthy obese participants were on average younger, contained a lower proportion of smokers, and had intermediate levels of risk factors. Central adiposity was comparable across healthy and unhealthy obese groups and considerably greater compared with non-obese groups. However, the metabolically healthy obese had lower BMI than their unhealthy counterparts.

The aim of this study was to examine the association between metabolically healthy obesity and risk of depression over two years follow up. Although a recent meta-analysis of prospective cohort studies suggest obesity is associated with an increased risk of depressive symptoms,¹ there are inconsistencies in the literature.^2–4,6 Indeed, a previous analysis of ELSA data has demonstrated an association between waist circumference and depressive symptoms in women, but a nearly significant inverse association between BMI and depression.³² However, those analyses did not account for metabolic risk factors and did not examine the healthy obesity phenotype. The main findings from the present study show that metabolically unhealthy obese participants were at elevated risk of developing depression although the metabolically healthy obese were not. These results were independent of central adiposity. Also the findings do not simply reflect an effect of obesity chronicity because over two-thirds of the metabolically healthy obese participants had been obese for at least 4–5 years prior to baseline. We are not aware of any previous work that has examined the association between metabolically healthy obesity and risk of depression, and these results might partly explain the equivocal findings in this area.

A metabolically healthy phenotype was observed in 34.3% of the obese sample from the present study, which is comparable with data from some studies, although prevalence figures have ranged widely (from 10–30%) depending on the definition used.^7,8 For example, in a sample of 5440 participants of the National Health and Nutrition Examination Surveys (NHANES), 31.7% of obese adults were defined as metabolically healthy.⁸ Metabolically healthy overweight and obese individuals from NHANES were of younger age, non-Hispanic black race/ethnicity, had higher physical activity levels, and smaller waist circumference. The metabolically healthy obese participants from the present study were also younger, displayed higher physical activity levels, but comparable waist circumference compared with at-risk obese counterparts.

In previous studies metabolic syndrome has been associated with depressive symptoms,^{20,21, 33–35} although the findings are not entirely consistent.^36–38 However, individual components of the metabolic syndrome including impaired glycaemic control^22–24 and inflammatory markers²⁵ appear to be particularly important in driving the association, which was also observed in this study. The mechanisms underlying the association between metabolic abnormalities and depression remain unclear. One of the driving factors might be disturbances in key stress axes including the hypothalamic pituitary adrenal axis and sympathetic nervous system. Disturbances in these axes have been associated with depressive symptoms³⁹ and are linked to insulin resistance and the cascade of events in the metabolic syndrome.^40–42 Depression could also result from the biochemical changes directly caused by disturbances in metabolic abnormalities. For example, preliminary evidence found brain abnormalities, such as reduced white matter volume and enlarged cerebrospinal fluid space, in obese adolescents with type 2 diabetes, which might result from a combination of subtle vascular changes and glucose abnormalities.⁴³

The metabolically unhealthy non-obese participants were also at elevated risk of developing depressive symptoms. One of the striking features of this group was the high CRP concentration, indicative of systemic inflammation, which was in fact comparable to the levels seen in the unhealthy obese group. Although the brain is generally regarded as being protected from the damaging effects of an inflammatory immune response, signals of systemic inflammation and elevated levels of peripheral inflammatory markers can affect the levels of pro-inflammatory cytokines in the brain⁴⁴ and a recent meta-analysis confirmed higher concentrations of pro-inflammatory cytokines in depressed individuals.⁴⁵ Clinically, it is important to identify metabolically unhealthy non-obese individuals, as early intervention with exercise and diet may help prevent these participants from developing obesity and diabetes,⁴⁶ and delay the onset of overt disease.

There is presently no consensus for the definition of metabolically healthy obesity. In the present study metabolic risk was based on an adaptation of previous criteria⁸ according to availability of data, and the results were not largely different using measures of fasting glucose or HbA1C as an indicator of impaired glycaemic control. The decision to use a categorization of more than one metabolic risk factor in defining metabolically unhealthy was based on previously published work,⁸ and was justified in the present study as we observed a threshold effect in that risk of depression was evident only in participants with more than one metabolic risk factor (see online supplementary material; Table S1). We did not assess metabolic risk factors at follow-up, thus it is possible that some of the healthy participants at baseline did go on to develop metabolic abnormalities. Nevertheless, given the short follow up period of 2 years it is unlikely that this could have influenced our results. Many of the metabolically healthy obese participants had been obese for at least 4–5 years prior to baseline suggesting that short-term exposure to obesity is an unlikely explanation for the null association with depressive symptoms. Although ELSA is designed to be nationally representative, the present sample was younger and healthier that the overall cohort due to limitations placed on blood sampling protocols.

In summary, we demonstrated that metabolically healthy obesity is not associated with risk of depression. The association between obesity and depression appears to be dependent on metabolic profile, which may partially explain why previous findings in the area have been inconsistent.

FN2Author contributions

MH had full access to the data, and takes responsibility for the integrity and accuracy of the results. All authors contributed to the concept and design of study, drafting and critical revision of the manuscript.

FN3Conflict of interest

None of the authors have any competing interests to declare.

Funding sources

The data were made available through the UK Data Archive. The English Longitudinal Study of Ageing (ELSA) was developed by a team of researchers based at University College London, the Institute of Fiscal Studies and the National Centre for Social Research. The funding is provided by the National Institute on Aging in the United States (grants 2RO1AG7644-01A1 and 2RO1AG017644) and a consortium of UK government departments coordinated by the Office for National Statistics. MH is supported by the British Heart Foundation (RE/10/005/28296); GDB is a Wellcome Trust Career Development Fellow (WBS U.1300.00.006.00012.01); MK is supported by the National Heart, Lung, and Blood Institute (R01HL036310-20A2) and the National Institute on Aging (R01AG034454-01), NIH, US and the Academy of Finland. The funders had no role in the study design; in the collection, analysis and interpretation of data; in writing of the report; or in the decision to submit the paper for publication. The developers and funders of ELSA and the Archive do not bear any responsibility for the analyses or interpretations presented here.