Modeling the associations among diurnal alpha-amylase, diurnal cortisol, and facets of emotion dysregulation in children

Thumbnail Image
McCurdy, Bethany
Major Professor
Weems, Carl F.
Gilligan, Megan
Lee, Daeyong
Melby, Janet
Neppl, Tricia
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Is Version Of
Human Development and Family Studies
Associations between salivary alpha-amylase (sAA), cortisol, and mental health symptoms are complex in nature, and research and theory suggest an important role of sAA and cortisol in mental health. This dissertation aimed to clarify the associations between sAA and cortisol, explore the associations of sAA and cortisol with symptoms of anxiety, depression, and post-traumatic stress, and emphasize the importance of using various statistical models for analyzing time-series data. This study examined diurnal patterns of sAA and cortisol in relation to symptoms of anxiety, depression, and post-traumatic stress in a sample of underserved youth (n = 100; mean age = 10, SD = .64; 79% Latino; 67% received free or reduced lunch). The findings revealed that diurnal trajectories of sAA and cortisol were consistent with previous literature, showing cubic patterns. Findings also revealed direct cubic associations between diurnal levels of sAA and cortisol, highlighting the nonlinear nature of these biomarkers. Notably, age significantly predicted sAA levels, with 8-year-olds exhibiting distinct patterns compared to older youth. For mental health symptoms, anxiety was negatively associated with sAA, contrary to expectations, suggesting non-normative patterns in highly anxious individuals. Depression predicted lower levels of sAA, and when combined with post-traumatic stress symptoms, it significantly predicted diurnal trends. Anxiety was associated with positive slopes and higher evening levels of cortisol, while post-traumatic stress symptoms showed no direct associations. Depression predicted lower awakening and midday cortisol and higher evening cortisol. Overall, this dissertation highlights the importance of considering psychological symptoms in tandem and using different statistical approaches to interpret findings. Further investigation is needed to explore the developmental changes of these biomarkers and replicate the inconsistent associations between anxiety, depression, and cortisol levels.