Adverse childhood experiences and biological embedding: A systematic review and meta-analysis of developmental trajectories

• In Event: Poster Session 04
  In Poster Session: PS 04 Topic - Biological Processes: Psychophysiology

Abstract

Nearly two-thirds of U.S. adults have had adverse childhood experiences (ACEs), which have been linked to chronic illnesses such as cancer, diabetes, and heart disease (Merrick et al., 2019). Yet, it is unclear which biological processes are most affected by ACEs and how the biological impact changes over time. This review aims to (a) elucidate mechanistic pathways of how ACEs may influence biomarkers across the lifespan, (b) identify stages of development during which it may be more harmful to experience ACEs, and (c) suggest potential prevention and intervention strategies to avoid the progression of harmful health effects. The review protocol was pre-registered on PROSPERO. We searched three electronic databases (PubMed, APA PsycINFO, CINAHL) for eligible studies, which were screened and coded independently and in duplicate. Included studies assessed at least one of 10 ACEs in the domains of child maltreatment and household dysfunction identified by Felitti et al. (1998) and measured any biomarker (e.g., cortisol, C-reactive protein) at more than one time point. The search returned 4,096 reports, and 514 were screened at the full-text level; 83 studies were retained for the current review (see Figure 1). Extracted data included demographics, adversity experienced, biomarkers, numeric data for effect size calculations, and a formal study quality assessment from a National Institutes of Health tool. We will calculate effect sizes as correlation coefficients, transformed to Fisher’s z for analyses; the sign of each will be coded as positive for findings that imply a greater biological risk for those who experienced more ACEs. We will fit models with robust variance estimators, which enables the pooling of dependent observations (e.g., multiple effect sizes per study), rather than selecting only a single observation from each study (Tanner-Smith & Tipton, 2014); this way, we can examine linkages of ACEs to multiple outcome variables simultaneously. We will also separate biomarkers into areas related to physiological functioning (e.g., neuroendocrine, metabolic) for independent analyses. We will evaluate the extent to which results are heterogeneous with I², Q, and τ² values (Borenstein et al., 2011) and use univariate and multiple-moderator meta-regression to examine multiple moderators simultaneously. We will examine included studies for effect size and study sample outliers and conduct sensitivity analysis as appropriate. Publication bias across studies will be assessed using funnel plots and formal tests (e.g., Egger’s, PET-PEESE), and we will interpret results in light of the moderators if significant bias is present. Presently, our team has retrieved and coded 83 studies, and the database is nearly ready for meta-analytic modeling. These findings will provide insight into how ACEs influence developmental trajectories through biological embedding.

Authors

• Julie M Brisson, University of Connecticut
  Presenting Author

• Blair T. Johnson, University of Connecticut
  Non-Presenting Author

• Preston A Britner, University of Connecticut
  Non-Presenting Author

• Justin M Le, University of Connecticut
  Non-Presenting Author

• Emily A Hennessy, University of Connecticut
  Non-Presenting Author

• Rebecca L Acabchuk, University of Connecticut
  Non-Presenting Author