Life course BMI trajectories from childhood to mid-adulthood are differentially associated with anxiety and depression outcomes in middle age

In our long-term population-based cohort, five distinct trajectories of BMI from ages 5 to 43 years were differentially associated with depression and to a lesser extent, anxiety at age 53. Specifically, the persistently high and child average-increasing trajectories were associated with a greater risk of current depression in mid-adult life, with participants in the child average-increasing trajectory more likely to experience a greater severity of symptoms. Furthermore, despite finding no association between BMI trajectories and the risk of current anxiety per se, participants in the child high-decreasing trajectory were less likely to experience severe symptoms of anxiety.

Current evidence on the longitudinal effects of life course obesity on depression in adulthood is limited [6,7,8] and only one other study has modelled BMI trajectories using prospectively collected data and only to early adulthood [8]. Consistent with what we observed in the child average-increasing trajectory, Perry et al. [8] found a greater risk of depression in young adults (24 years) who had a major increase in BMI following puberty. However, unlike our study, no greater risk of depression was found in the persistently high trajectory, possibly owing to differences in follow up duration (24 vs 43 years), cohort age (1991 vs 1961 birth cohort) and age at outcome assessment (early- vs. mid-adulthood).

Although less comparable, the studies that analyzed trajectories of recalled body size were conducted in specific sub-populations, namely middle aged university graduates [7] and post-menopausal women [6]. Sayon-Orea et al. [7] found that women with a large body size in childhood, which increased throughout adulthood, were at an increased risk of new-onset depression at age 40, but found no associations in males. Similarly, Perquier et al. [6] showed that a persistently large body size was associated with a greater risk of new onset depression post-menopause (age 65). And a marked increase in body size following puberty was associated with current depression post-menopause. When taken together, it seems excess weight that persists from childhood to adulthood, and excess weight gained across the adolescent transition, may increase vulnerability to depression in adulthood. This highlights the need to increase mental health screening and support among individuals with comparable weight trajectories.

Notably, our study extends the literature by demonstrating that in comparison to the persistently high BMI trajectory, belonging to the child average-increasing trajectory was associated with a greater risk of having more severe and frequent symptoms of depression and a greater risk of several anxiety symptoms. Although we would expect to see a dose-response relationship, where a greater duration of obesity worsens mental health, in actual fact it appears that participants belonging to the child average-increasing trajectory had the worst mental health profile. These novel insights suggest adolescence may be a period of increased sensitivity, during which, the onset of obesity has an apparent influence on long term mental health. The timely introduction of obesity prevention strategies during these formative years may aid in reducing the burden of depression. However, we cannot exclude the possibility that depression onset preceded weight gain in this group, as a bidirectional relationship is quite possible; [15] furthermore, earlier-onset of depression has also been associated with greater severity [16].

Although we couldn’t establish a statistically significant association between BMI trajectories and current anxiety, we did observe a positive trend in the two highest BMI trajectories. Despite limited longitudinal evidence on this topic, a meta-analysis of four prospective studies found a small, positive association between obesity and anxiety [4] and so it is possible that our study was insufficiently powered to pick up a small effect. This would not be surprising given the frequent clinical concurrence of anxiety and depression.

Encouragingly, the child high-decreasing trajectory was associated with a reduced likelihood of experiencing severe anxiety symptoms, sleep disturbance or depressed mood, and had no association with current depression. Previous studies have also found no association between excess weight confined to childhood only and adult depression risk; [8, 17,18,19] however, our study is the first to show resolving excess weight may promote mental health. These findings suggest that the potentially depressive effects of obesity may be reversible with its resolution and therefore amenable to intervention: highlighting the need for and value of early intervention strategies to prevent and treat excess weight. Moreover, as the reference trajectory maintained a lower BMI throughout the duration of the observation period, it is likely that methods of weight reduction such as exercise and other healthy lifestyle behaviours partially explain the results observed in the child high-decreasing trajectory [20, 21]. And so, further research should explore methods of weight reduction in at-risk trajectory groups as possible intervention strategies to promote mental health and wellbeing.

To date, several plausible mechanisms have been suggested to explain why obesity may increase vulnerability to depression and anxiety. From a psychosocial perspective, weight-based stigmatization and discrimination faced by individuals with obesity have been shown to mediate associations between weight status, anxiety and depression, by increasing body image dissatisfaction, psychological distress, and lowering self-esteem [22]. Several related behaviours also offer explanatory pathways, as poor dietary patterns, physical inactivity, and sedentary lifestyles are common among individuals with obesity and present as risk factors for depression and anxiety [23]. More recently, increasing research attention has focused on underlying physiological and genetic mechanisms. Most prominently, chronic low-grade systematic inflammation, induced by obesity, has been identified as a key catalyst for depression, and possibly anxiety, by triggering dysregulation of the hypothalamic-pituitary-adrenal axis, with activated immune-inflammatory pathways inducing neuroendocrine abnormalities [24].

Sex differences were apparent in our stratified analyses; however, evidence of effect modification was weak. We observed an increased risk of depression among women in the child average-increasing trajectory, which is consistent with the findings of Perry et al. [8]. However, our study is the first to report a greater risk of depression among males in the persistently high trajectory. Although obesity and depression are more prevalent in women, sex differences in the relationship between obesity and depression are lesser known. It has been shown that women experience greater body dissatisfaction and place greater value on their appearance regardless of age [25]. In this context, a consistently increasing BMI might be negatively affecting body image among women in the child average-increasing trajectory, causing an increased vulnerability to depression. Somewhat paradoxically, however, a greater magnitude in the risk of depression was observed among males in the persistently high trajectory. Although this finding is unexpected, it’s possible that women in this group have sought support for their mental health, as males are less likely to seek treatment [26].

We found evidence of effect modification by socioeconomic status, such that a lower occupational class predicted a greater risk of depression in the persistently high trajectory. Given the disadvantages faced by those from lower socioeconomic backgrounds, factors such as food insecurity, financial and social stress, and reduced health care access, may explain this association. However, suprisingly, participants in the persistently low trajectory were less likely to report anxiety if they were from a lower occupational class; and the reason for this finding is not well understood. Furthermore, we also found a greater risk of anxiety among participants in the persistently low trajectory if they were consumers of alcohol, suggesting alcohol screening may be pertinent in this group. However, further research is needed to replicate our findings.

Strengths and limitations

The main strength of this study is that our BMI trajectories were modelled using eight repeated measures of BMI, with five collected across the transition from childhood to adulthood. This enabled us to adequately characterize within-subject variations in obesity over the life course. Furthermore, data were prospectively collected from a large representative sample of the Tasmanian population, and we assessed a variety of mental-health outcomes to understand the point prevalence, severity and symptomatology of anxiety and depression.

Several limitations should also be considered in the context of our findings. Although weight and height data were prospectively collected, we relied on self-reported measures at ages 30 and 43. Even though a strong correlation has been found between self-reported and measured weight and height, self-report could be prone to underestimation, particularly in individuals with overweight or obesity [27]. This would have tended to bias our results towards the null. Whilst we used validated, self-report questionnaires to assess anxiety and depression, the GAD-7 and PHQ-9 are not diagnostic tools. Nevertheless, these instruments are widely used in epidemiological studies and scores greater than ten have previously demonstrated high sensitivity and specificity in detecting probable cases [11, 12]. Furthermore, it is important to note that the GAD-7 and PHQ-9 were only administered at one time point, so we did not have prior data in adolescents/young adulthood to determine the age of depression or anxiety onset, nor could we exclude the possibility of reverse causation.

Despite using a DAG to inform confounder adjustment, our simple causal diagram could not depict the complex interrelationships between a time-varying exposure and time-varying confounders. Additionally, several known risk factors identified by our DAG were not measured in the TAHS cohort and could not be adjusted for in our analyses. These included family history of depression and obesity, medication use, genetics, history of trauma or chronic stress and substance abuse. Consequently, our results might have some residual confounding. Attrition bias could be a possibility as men and those with obesity at age 43 were more likely to be lost to follow-up. However, no differences in the proportion of participants lost to follow up between BMI trajectories were seen. Moreover, the persistently high and child average-increasing trajectories had the smallest sample sizes and may have lacked precision, particularly in stratified analyses. Although the shape and size of our trajectories are comparable to prior research in the field [8], TAHS participants were born in 1961, more than a decade before the prevalence of childhood obesity began to notably increase in the community. Based on current trends, we would expect that present-day trajectories may differ in shape and size, with a greater proportion of participants belonging to the highest trajectories. Therefore, further research is necessary to quantify the effects of more recent trends in BMI trajectories from childhood. Lastly, multiple comparisons might have increased the risk of type 1 error, especially given that we considered p-values of the interaction terms to be significant at a level of 0.1. Therefore, the results of effect modification analyses should be interpreted with some caution.

留言 (0)

沒有登入
gif