A Meta-Analysis of Maternal Postnatal Depressive Symptoms and Children’s Cognitive and Developmental Outcomes

• In Event: Moderators and Protective Factors for Pre- and Post-natal Maternal Distress Associations with Child Neurodevelopmental Outcomes

Abstract

Introduction: It is well established that maternal depressive symptoms, in both the prenatal and postnatal periods, have negative implications for maternal health and well-being, as well as the health and development of her offspring (Bussieres et al., 2015; Field, 2010; Goodman et al., 2011; Madigan et al., 2018; Wachs, Black, & Engle, 2009). One domain of child development that has been shown to be particularly impacted by maternal depressive symptoms is child cognitive development (Cortes Hidalgo et al., 2020; Lupien, McEwen, Gunnar, & Heim, 2009; Nisbett et al., 2012; Tarabulsy et al., 2014). However, meta-analyses examining the association between maternal postnatal depression and child cognitive outcomes have excluded screening measures of early learning and development (i.e., achievement of developmental milestones) and moderators of these associations have yet to be examined. The current meta-analysis evaluates the associations between maternal postnatal depressive symptoms and two related child outcomes, achievement of developmental milestones and cognitive skills, and examines moderators of these associations.

Hypotheses: Based on previous work (e.g., (Beck, 1998; Liu et al., 2017)), we anticipated small but significant associations between maternal postnatal depression and child cognitive outcomes.
Study Population: A total of 55 studies were included.

Methods: Searches were conducted in MEDLINE, Embase, and PsycINFO up to November 2018 by a health sciences librarian. Inclusion criteria for the current meta-analysis were the following: (1) maternal depressive symptoms was measured following the child’s birth was collected; (2) child developmental or cognitive outcomes were collected prior to 18 years of age; (3) the study reported a statistic that could be transformed into an effect size; and (4) the full-text article was available and written in English. Potential moderators were also extracted and included: study design (cross-sectional versus longitudinal), socioeconomic status (low versus mid to high/mixed), percentage of male participants, child age at the postpartum depressive symptoms exposure, maternal age at the time of depressive symptoms measurement, time between the depressive symptoms and cognitive outcome measurement, syndromal level of maternal depressive symptoms (i.e., clinically elevated versus not), and study quality. Data analysis was conducted using Comprehensive Meta-Analysis (CMA, Version 3.0) (Borenstein, Hedges, Higgins, & Rothstein, 2013).

Results: The initial electronic search from three databases yielded 21,217 articles after duplicates were removed and hand-searched articles were added. Following the review of titles and abstracts, 403 studies were identified as potentially meeting inclusion criteria and the full-test article was reviewed. Subsequent to full article review, n=55 studies met full inclusion criteria (14 with child development outcomes, 41 with child cognitive outcomes; See Table 1). The pooled effect size for the association between postnatal depressive symptoms and child developmental (N = 14; r =-.08; 95% CI= -.13, -.04) and cognitive outcomes (N =41; r = -.16; CI = -.21 to -.11) were statically significantly, and small in magnitude (See Figure 1). Moderator analyses indicated that effect sizes for child developmental outcomes were more robust when the duration between the assessment of depressive symptoms and developmental outcomes was greater and when study quality was stronger.

Authors

• Nicole Racine, University of Ottawa
  Presenting Author

• Pauline Wu, University of Calgary
  Non-Presenting Author

• Rachel Pagaling, University of Calgary
  Non-Presenting Author

• Katarina Zumwalt, University of Calgary
  Non-Presenting Author

• Sheri Madigan, University of Calgary (U of C)
  Non-Presenting Author