We first set out to establish whether childhood adversities increase the risk of psychological distress, obesity, and their comorbidity in early adulthood. We then examined several lifestyle factors as moderator variables, to determine whether adopting a healthy lifestyle can provide some protection against the mental and physical health consequences of childhood adversities.
Childhood adversities increased the risk of comorbidity to a greater extent than psychological distress or obesity alone.
We demonstrated a robust association between childhood adversities and the comorbidity of psychological distress and obesity, i.e., mental-physical comorbidity. Each reported exposure to an adversity, corresponding to a one-item increase in a cumulative measure of childhood adversities (0-33), increased the odds of developing comorbidity by 16%. After assigning comorbidity as the reference category, childhood adversities were further shown to be stronger risk factors for comorbidity than psychological distress or obesity alone. These findings align with a mega-analysis of data from middle- and older-aged adults (n = 156,511, 40-69 years), which found that childhood maltreatment increased the risk for the comorbidity of depression and cardiometabolic diseases to a greater extent than either condition alone (7).
Whilst the same study showed childhood maltreatment to increase the risk of comorbidity in one cohort of young adults (ALSPAC mothers; n = 3,927, mean age = 29 years), as far as we are aware, the current study is first in this age group to show that childhood adversities are more strongly related to the comorbidity of psychological distress and obesity than each health problem alone. Given the known associations between depression, obesity and cardiometabolic diseases (3,57,58), these results could indicate that psychological distress-obesity comorbidity in early adulthood could lie on the risk pathway from childhood adversities to depression-cardiometabolic disease comorbidity. These hypotheses could be tested by mapping multimorbidity trajectories of depression, obesity and cardiometabolic disease (59), and exploring childhood adversities as predictors of these trajectory classes, where advice on predicting class trajectories is readily available (60).
The findings build upon previous analyses in BCS70 and other UK birth cohorts, demonstrating that socioeconomic disadvantage in childhood increases the risk of psychological distress-obesity comorbidity in early adulthood (14,15). A strong relationship between childhood maltreatment (i.e. neglect, abuse, etc.) and childhood socioeconomic status has been documented elsewhere (16), and accounting for these risks within a single scale may help capture combined effects in relation to health outcomes (61). In examining this, this study developed a cumulative measure of childhood adversities as the main exposure, which can be utilised in future BCS70 studies. In doing so, we also show that this broader definition of adversity in childhood is a strong predictor of mental-physical comorbidity, indicating that a wider range of childhood adversities, beyond low socioeconomic status, contribute to comorbidity risk.
Despite childhood adversities being a stronger predictor of comorbidity, a 1-item increase on the adversity scale elevated the risk of psychological distress by 11% and obesity by 5% as individual outcomes. This finding conforms with other studies which found that childhood adversities increase the risk of mental and physical health problems in isolation, and in regard to these associations, the increased risk due to childhood adversities was larger for mental vs. physical health problems (6,7). Given that the established associations between childhood maltreatment and obesity became non-significant when adjusting for current depression in a separate study (62), there is support for a hypothesis suggesting depression makes a greater contribution than obesity to the mediating pathways between childhood adversities and comorbidity of depression and obesity (62–64).
Lifestyle factors did not moderate the association between childhood adversities and comorbidity or individual health outcomes.
In addressing the study’s main aim, a series of moderation analyses found that contrary to our hypotheses, adhering to a healthy lifestyle did not weaken the association between childhood adversities and comorbidity. Dysfunction of biological systems have already been highlighted as mechanisms which link childhood adversities and health problems across the life course, including under and over-activation of stress and immune systems (inflammation, hypothalamic pituitary adrenal axis, etc.), metabolic alterations and disruptions to the microbiome (65–67). Biological pathways to comorbidity could also be partially explained by overlapping genetic risk variants for depression and obesity (68). Due to how the examined lifestyle factors were not significant moderators, this study emphasises the need to identify other factors which could help weaken the association between childhood adversities and comorbidity.
Factors relating to study design may have also contributed to our findings, including the absence of moderating effects. Whilst the measurement timepoints for lifestyle factors (16 years) and comorbidity (30 years) helped control for reverse causality between outcome and moderators, the developmental period from 16 to 30 years represents a transitional stage of life characterised by greater autonomy (leaving home, leaving education, entering employment, etc.). This may have led to changes in lifestyle behaviours in BCS70 participants, as documented in other cohorts (69–72), although some behaviours such as sleep duration and smoking, may be more stable from adolescence to adulthood (73,74). This point is highlighted by the fact that school-time physical activity contributed to a greater extent than extra-curricular sports to our measure of total physical activity, which could change after leaving school.
Sleep duration was also self-reported based on a single night prior to arrival at the assessment centre, which may not be representative of typical sleep patterns (75). There is also the potential for social desirability bias in self-reported health behaviours, in particular alcohol use (16 years) was recorded before the legal age of consumption (18 years), and smoking was measured shortly after the legal age of use in the UK (16 years). These health-risk behaviours have been shown to be under-estimated when self-reported (76–78), highlighting the need to examine both objective and/or repeated measures of lifestyle factors from adolescence to young adulthood.
As well as examining total levels of physical activity, we also examined different types of physical activity, because one study found consistent participation in team sports, but not individual sports, to reduce depressive symptoms in early adulthood (56). This indicates that engaging in sports which rely on social cohesion as a core element, may be more beneficial at protecting against mental health problems. This distinction was investigated within our post-hoc analyses, which found that considering team-based sports, individual sports, or skill-based sports separately did not change the main conclusions of the study (i.e., no moderation of the association between childhood adversities and the health outcome categories). Further examination of specific subtypes of physical activity is required to determine beneficial components of physical activity on mental health.
Limitations
Our findings need to be considered within the context of the following limitations. Firstly, despite a comprehensive imputation model, high levels of missing data means that these results should be classified as hypothesis-generating, and the analyses should be replicated in alternative samples to confirm inferences (79). This is particularly important given the focus of this study, as childhood adversities, poor mental health and risk behaviours have been shown to be associated with higher rates of attrition in longitudinal studies (45,80–82). Secondly, each adversity in childhood was dichotomised to denote risk or no risk, which whilst designed to give equal weight to each type of adversity, represents a simplistic view of stressful events. Although an attempt was made to use evidence-based cut-offs, due to the response options given to participants some decisions required subjective deliberation and were based on researcher judgement. To mitigate this, we have been transparent about our dichotomisation criteria (see Supplementary Table 1) to allow other researchers to improve or modify the measure with improved knowledge of what constitutes an adversity. Finally, this study also did not account for any directionality between the cooccurrence of psychological distress and obesity, with some studies suggesting that psychological distress more commonly precedes obesity (83), or vice versa (84).