5 - (IOP 7) Patterns of Trauma Exposure Interact with Stress Reactivity to Predict Change in Anxiety and Externalizing Symptoms in Youth

Potentially traumatic events (PTEs) are associated with increased risk of both anxiety and externalizing problems in youth (Chapman et al., 2007). Several theories suggest PTEs may be particularly detrimental to those who are physiologically either hyper- or hypo-reactive to stress (e.g., Del Giudice et al., 2011). Prior work examining interactions between PTE exposure and stress reactivity has often conceptualized PTEs as a binary (present vs. absent), a count (total PTEs), or compared specific PTEs (e.g., physical abuse vs. neglect). However, exposure to one PTE is associated with increased risk for additional PTEs, and some types may be more or less likely to co-occur (Jenness & McLaughlin, 2015). Therefore, person-centered approaches to studying interactions between patterns of PTE exposure and stress reactivity on risk for psychopathology are needed to better capture the experiences of individuals seen in clinical practice. This study addresses this gap by considering how classes of PTE exposure differed in associations between hypothalamic-pituitary-adrenal (HPA) axis reactivity (a component of the mammalian stress response), anxiety, and externalizing symptoms.

A community sample of youth (N = 326, Mage = 11.52 years, range = 8 – 15, 50% female; 54% identifying as racial/ethnic minority) were recruited. Data were collected at baseline (T1) and 3 years later (T2). T1 PTEs were assessed through child and caregiver interview. Prior work with this dataset identified three classes of PTE exposure - C1 (7.2% of sample) is characterized by family violence (e.g., physical and emotional abuse, domestic violence, neglect); C2 (33.1% of sample) is characterized by exposure to community violence; C3 (59.6% of sample) is characterized by low levels of PTEs. Anxiety and externalizing problems were assessed using self-report questionnaires at T1 and T2. T1 HPA axis reactivity to a social stress task was calculated using two methods – area under the curve with respect to ground (AUCg; total task cortisol output), and with respect to increase (AUCi; task change in cortisol levels). The Bolck-Croon Hagenaars method was used within Mplus to assess PTE class differences in associations between HPA axis reactivity and anxiety and externalizing problems. Gender, pubertal status, and time between youth awakening and cortisol collection were included as covariates.

Associations between did not differ by PTE class at T1. However, differences emerged in longitudinal analyses. Low AUCg and AUCi predicted increases in both anxiety and externalizing from T1 to T3 for C1 (ps < .001 - .005), whereas high AUCi predicted increases in anxiety for C2 (p = .02). In short, for those with PTE experiences characterized by family violence, hypo-reactivity predicted increased anxiety and externalizing symptoms, whereas hyper-reactivity predicted increased anxiety for those whose experiences were characterized by community violence.
Strengths of the study include a diverse sample and longitudinal design. Limitations include the reliance on self-report and the small size of C1. Findings are consistent with theory and highlight the importance of person-centered approaches understand associations between trauma, stress reactivity, and psychopathology.