Biobehavioral markers of prenatal resilience associated with positive postpartum maternal mental health and well-being

PROJECT SUMMARY/ABSTRACT Exposure to chronic life stress is associated with increased risk for postpartum depression and anxiety. However, many individuals fare well in the postpartum period despite these exposures. This may, in part, be due to positive experiences early in life and/or prenatal biobehavioral resilience (e.g., greater heart rate variability [HRV], self- reported tenacity), although longitudinal studies examining mechanisms that promote postpartum well-being have been limited. Thus, this K01 application is informed by a strengths-based framework to test the central hypothesis that, in the context of chronic stress, experiences of positive parenting and supportive peer relations during childhood will be associated with physiological and self-reported psychological resilience in pregnancy (Aim 1), which, in turn, will predict early positive postpartum mental health and well-being (Aim 2). In Aim 3, novel machine learning (ML) algorithms will extract multi-level data to test the mediating role of prenatal biobehavioral resilience in the association between positive childhood experiences and postpartum mental health and well- being. ML uses decision trees to find and extract cross-validated, generalizable features in rich, multimodal, time- series data to maximize the likelihood of traditional mediation models. Participants who have experienced chronic stress from ages 8-17 years (N=116) will be recruited from a population-based sample of pregnant people (67% Black, 28% white, 5% mixed-race; 99% cis-female) enrolled in the longitudinal Pittsburgh Girls Study (PGS) that began in childhood. At 20-30 weeks gestation, HRV will be measured using a wearable, passive sensor for 7 days and psychological resilience will be assessed by self-report. At 3-months postpartum, mental health and well-being will be measured. These new data will be linked to extant PGS data on positive childhood experiences that were measured prospectively from 8-17 years. The award candidate, Dr. Magee, is a licensed clinical psychologist experienced in studying mood and behavior around key life transitions. Dr. Magee will gain specialized conceptual and methodological training in: (1) developmental models of stress, resilience, and peripartum mental health and well-being; (2) acquiring, processing, analyzing, and interpreting ambulatory cardiovascular psychophysiological data; and (3) ML and deep learning techniques to extract multi-modal data. Dr. Magee’s training and research will be supported by a strong team of mentors (Drs. Hipwell, Levine, Bylsma, and Akcakaya) and a consultant (Dr. Cheng) with expertise that spans domains of preconception stress and resilience, postpartum mental health, ambulatory assessment of prenatal psychophysiology, multilevel modeling, and ML techniques. The University of Pittsburgh is a leading research institution that will provide an exceptional training environment for Dr. Magee to meet her research and career development goals. The proposed K01 research aligns with and advances the NIMH’s mission and scientific priorities and is the first step in Dr. Magee’s goal to create a larger program of independent research focused on identifying promotive factors that inform the development of peripartum mental health interventions.