Hair Cortisol and Psychopathology Among Children with a Chronic Physical Illness

Abstract

Chronic physical illnesses (CPIs), such as asthma, diabetes, and epilepsy, affect approximately 30-50% of children. Living with and managing a CPI in childhood may be a significant source of chronic stress. As a result, children with a CPI are at a greater risk of developing psychopathology (i.e., multimorbidity). Multimorbidity negatively impacts children’s physical and psychosocial health, including child health-related quality of life (HRQoL). Thus, understanding the mechanisms underlying child multimorbidity is a critical public health concern. Chronic stress exposure can lead to dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis, which may be implicated in the development of psychopathology. Over the past two decades, hair cortisol concentration (HCC) has been identified as a viable biomarker for chronic HPA-axis dysregulation in children. However, evidence linking HCC to psychopathology in children is mixed, and very few studies have examined these associations in children with a CPI. Existing studies in children with a CPI are often small, cross-sectional, and illness-specific.

To address these knowledge gaps, this dissertation examined the longitudinal associations between HCC and psychopathology in a large sample of children with diverse CPIs. Specifically, the objectives were to: 1) identify distinct trajectories of HCC among children with a CPI; 2) identify predictors of HCC trajectory group membership; 3) examine associations between HCC trajectories and psychopathology; 4) examine the between- and within-dyad effects of parent distress on child psychopathology; 5) explore child HCC trajectory as a moderator of the associations between parent distress and child psychopathology; 6) examine associations between HCC trajectories and HRQoL; and 7) explore psychopathology as a mediator of the associations between HCC trajectories and HRQoL.

The first study used latent class growth analysis (LCGA) to identify three distinct trajectories of HCC in children with a CPI: Hypersecretion (68%), Hyposecretion (8.6%), and Hyper-to-Hypo (23.4%). Significant predictors of HCC trajectory group membership were child sex, child age, parent age, and time since CPI diagnosis. Linear mixed-effects (LME) models examined associations between HCC trajectories and psychopathology, and found that children in the Hyper-to-Hypo class had fewer internalizing and externalizing psychopathology symptoms compared to the Hypersecretion class. There was no difference in psychopathology symptoms between the Hyposecretion and Hypersecretion classes. These findings suggest that living with a CPI during childhood may be a significant chronic stressor, and HPA-axis dysregulation may be an important mechanism linking childhood CPI to psychopathology.

The second study used LME models to examine the between- and within-dyad effects of parent distress on child psychopathology and explore child HCC trajectory as a moderator of these associations. Higher average parent internalizing symptoms were associated with greater child internalizing and externalizing symptoms. Fluctuations in parent internalizing symptoms only predicted greater child internalizing, not externalizing, symptoms. Higher average levels and fluctuations in parenting stress were associated with greater child externalizing, but not internalizing, symptoms. Child HCC trajectory was identified as a moderating factor, such that the combination of high average parent internalizing symptoms or parenting stress and cortisol hypersecretion predicted greater child internalizing symptoms. These findings suggest that parent and child mental health are strongly linked in families of a child with a CPI, and interventions targeting parent distress may have beneficial downstream effects on child mental health. Moreover, HPA-axis dysregulation in children with a CPI may function as a vulnerability factor that heightens risk for psychopathology when exposed to parent distress.

The third study used latent growth curve models (LGCMs) to examine associations between HCC trajectories and HRQoL over time, and explore psychopathology as a mediator of these associations. Children in the Hyper-to-Hypo class had higher HRQoL in the domains of parents and autonomy, social support and peers, and school environment compared to the Hypersecretion class, and lower psychopathology was identified as a significant mediating mechanism. Children in the Hyposecretion class had higher HRQoL in the physical well-being domain compared to the Hypersecretion class, an effect not mediated by psychopathology. The HCC trajectories were not associated with change in HRQoL over time, but increases in psychopathology were associated with decreases in HRQoL. These findings suggest that HPA- axis dysregulation and psychopathology may be important mechanisms linking childhood CPI to poor HRQoL, and interventions for reducing stress and psychopathology may lead to improvements in HRQoL in this population.

This dissertation fills critical knowledge gaps and contributes to a greater understanding of the mechanisms underlying child multimorbidity. Taken together, these findings can be summarized into three key themes: (1) living with a CPI during childhood may be a significant chronic stressor, and HPA-axis dysregulation may be an important mechanism linking childhood CPI to negative health outcomes; (2) HPA-axis functioning in children with a CPI is heterogenous, and approaches that identify this heterogeneity may be important for understanding health outcomes and informing intervention strategies; and (3) non-categorical, integrated physical-mental, and family-centered care approaches in pediatric healthcare are needed to improve health outcomes for children with a CPI and their families.