Biopsychosocial approach to the examination of potential biomarkers of depression, anxiety and wellbeing in adolescence

Abstract

Adolescence constitutes a period of vulnerability for the onset of psychological distress. The onset of psychological illness during this period is associated with a more severe and enduring course. Considerable research investigates risk factors and precursors of illness but is largely dependent on research involving adult samples. Additionally, the majority of research focuses on symptoms of distress without examining measures of wellbeing. This is a significant limitation as it may only partly capture experiences. This thesis will examine symptoms of depression and anxiety related distress alongside measures of wellbeing in adolescence. Cognitive biases, attachment, emotional reactivity, emotional regulation, the stress-response system and sleep are all components that have been implicated by previous research as related to the onset of disorders; but, lack robust findings within adolescent samples. This thesis comprises of four studies which discuss empirical research examining these factor and the extent to which they are able to predict symptoms of depression, anxiety and wellbeing. Two different samples were involved in this research. Data informing chapters Two and Three are based on the same sample, while data informing chapters Five and Six are from a second sample. Both samples comprise of adolescents aged 13-18, who were recruited from Scottish secondary schools and participated in self-report measures, experimental paradigms and provided biological samples for analysis. Overall, results suggest that tasks assessing self-referential memory and interpretation of ambiguous scenarios, alongside measures of rumination and dysfunctional attitudes, were able to predict symptoms of depression, anxiety and wellbeing. Regression models were able to predict approximately 60% of the variance in depression, anxiety and wellbeing. Activation of the stress-response system was approximated via saliva and hair samples analysed for cortisol exposure. Results indicate that physiological measures were shown to have less predictive utility than cognitive mechanisms. Analyses indicate that lowered morning and higher evening cortisol was associated with increased depression and lower wellbeing. However, although this relationship was significant, it was small and may have limited clinical utility. Three-month retrospective measurement of cortisol was not significantly related to measures of depression, anxiety or wellbeing. The examination of sleep architecture, using measured through Consensus Sleep Diaries and Philips Actiwatch 2 actigraphic wrist monitors partially supported relationships between depression and sleep. Self-report measures were significantly predictive of depression, anxiety and wellbeing. Specifically, increased wakefulness while in bed is implicated in increased depression and anxiety and lower wellbeing. Overall results indicate that psychological factors are more salient markers of mental health than physiological markers in this sample. Overall results emphasise the importance of psychological factors suggesting that these may be ideal targets for developmentally and individually appropriate interventions. These results have implications for the identification of individuals at risk for onset of mental health conditions as well as psychological components of importance for wellbeing.