Adolescent non-suicidal self-injury during the COVID-19 pandemic: A prospective longitudinal study of biological predictors of maladaptive emotion regulation.

ABSTRACT

BACKGROUND: Non-suicidal self-injury (NSSI) in adolescence is thought to stem from interactions between vulnerability in developing biological systems and experience of stressors. The current study assesses whether multiple levels of the stress system's response to threat could prospectively predict NSSI engagement during the COVID-19 pandemic, a shared, time-locked stressor. METHODS: Participants were 64 female adolescents (ages 12-16) from community and clinical settings who were oversampled for NSSI histories. Prior to the onset of the COVID-19 pandemic, adolescents completed a protocol that measured hypothalamic-pituitary-adrenal axis response to a social stressor (via salivary cortisol), amygdala volume, amygdala emotion-evoked activation, and frontolimbic resting-state functional connectivity. During early months of the pandemic (Summer 2020), measures of NSSI behavior (Inventory of Statements About Self-Injury), emotion regulation difficulties (Difficulties in Emotion Regulation Scale), perceived stress (Perceived Stress Scale), and pandemic-related stressors (Epidemic Pandemic Impacts Inventory) were collected. Multinomial logistic regression was used to assess if pre-pandemic biomarkers predicted mid-pandemic NSSI engagement persistence of NSSI (Persist; N = 21), cessation of NSSI (Desist; N = 26), and no history of NSSI (Never; N = 17). Linear regressions explored if pre-pandemic biomarkers predicted mid-pandemic difficulties in emotion regulation and perceived stress. RESULTS: Higher pre-pandemic overall cortisol response to stress and amygdala emotion-evoked activation characterized adolescents who persisted in NSSI, compared to those who desisted. These findings remained significant when controlling for pandemic related stressors. Lower prepandemic cortisol reactivity predicted more difficulties in emotion regulation during the pandemic. This finding did not remain significant after controlling for pandemic related stressors. CONCLUSIONS: Findings suggest that patterns in key biological threat response systems may confer vulnerability for risk outcomes including NSSI engagement in adolescent females in the context of a shared, novel, naturally-occurring stressor. The results point to the importance of multi-level, longitudinal approaches for understanding the interface between developing neurobiological systems and experiential stress in at-risk adolescents. Identified patterns give insight into potential risk assessment strategies based on an understanding of the multi-level threat response.