Sensitive Periods Across Development: Prospectively Assessed Adversity Exposures and Depressive Symptoms in Late Adolescence

• In Event: 1-096 - Sensitive Periods in Development Shaping Risk for Psychopathology Across the Life Course

Integrative Statement

Introduction: Although exposure to environmental stressors are known determinants of poor mental health, a major challenge in psychiatric studies relates to modeling the effect of environmental exposures on disease risk, particularly when these exposures are assessed repeatedly across time. Innovative life course modeling approaches now allow researchers to test the effects of a given repeated exposure––as well as the effect of the timing and duration of that exposure––on mental health outcomes. However, few practical applications of these approaches exist, and among these even fewer have studied repeated exposures across a long developmental time frame, such as from birth through adolescence.

Hypotheses: The current study aimed to address these gaps by focusing on the role of two commonly-assessed environmental exposures on subsequent risk for depression in late adolescence. We were particularly interested in understanding whether there were sensitive periods in childhood and adolescence when exposure to these adversities conferred a particularly high risk for depression.

Study Population: Prospective data came from the Avon Longitudinal Study of Parents and Children (ALSPAC), a two-decade-long birth cohort; these analyses were based on an analytic sample comprising n=3,263 children.

Methods: Two common risk factors for depression––physical or emotional abuse by a caregiver (characterized by physical or verbal cruelty) and financial stress (characterized by the inability to afford basic goods or services)––were each measured via parental report on at least seven occasions between birth and age 18. Using a two-stage life course modeling approach, we first evaluated if there was a time-period of exposure that explained the most variability (r2) in depressive symptoms measured at age 18.5 using the Short Moods and Feelings Questionnaire, and determined the corresponding magnitude of those associations. We then explored whether a competing life course theoretical model, specifically an accumulation of risk model, explained more of the observed variation in depressive symptoms than any single sensitive period.

Results: We identified three sensitive periods (Table 1): the first was in very early childhood (spanning 8 months to 2.75 years of age) for the effect of financial stress in girls (r2=0.45%), the second was in middle childhood (spanning 7 to 9 years of age) for the effect of caregiver physical or emotional abuse in boys (r2=0.81%), and the third was in early adolescence (11 years of age) for the effect of caregiver physical or emotional abuse in girls (r2=0.20%). Substantial effect sizes were observed (standardized mean differences: 1.62 for financial stress in girls; 2.55 for caregiver physical/emotional abuse in boys; 2.14 for caregiver physical/emotional abuse in girls). For boys, the effects of this sensitive period persisted even after accounting for the accumulation of exposure (r2=0.71%; Table 2). These findings suggest that to better understand the link between environmental exposures and psychiatric disease risk, we may need to move beyond comparisons of those “exposed” vs. “unexposed” and begin to account for the timing of adversity exposure. Novel life course modeling approaches can help identify sensitive periods in development, which could guide intervention efforts in the prevention and treatment of psychiatric illness.

Authors

• Kathryn Anne Davis, Massachusetts General Hospital
  Presenting Author

• Andrew D.A.C. Smith, University of the West of England
  Non-Presenting Author

• Yiwen Zhu, Massachusetts General Hospital
  Non-Presenting Author

• Henning Tiemeier, Harvard TH Chan School of Public Health, Boston
  Non-Presenting Author

• Ezra S. Susser, Columbia University
  Non-Presenting Author

• Erin C Dunn, Massachusetts General Hospital
  Non-Presenting Author