Does Maternal Depression During Early Childhood Cause Child Externalizing and Internalizing Problems? A Propensity Score Study.

• In Event: 3-233 - Parental Depression and Youths’ Outcomes: Integrating Advanced Methods to Gain Novel Insights

Integrative Statement

Introduction. Hundreds of studies have linked maternal depression to negative child outcomes. However, because the vast majority of these studies have been correlational, they cannot rule out alternative causal explanations, such as that child characteristics evoke maternal depression or that confounding variables are causes of both phenomena. We aim to make use of advances in the field of causal inference to address the limitations of previous research and better estimate the causal effect of maternal depression on child outcomes. We hypothesize that the effect will be substantially reduced in magnitude after accounting for confounding variables.

Methods. We implemented a propensity-score-based design with data from the Early Steps Multisite Trial, a sample of 731 low-income families assessed on 8 occasions from ages 2 through 10. Mothers with a total score above 15 on the Center for Epidemiological Studies Depression (CES-D) scale at age 2 were classified as depressed. Families with depressed and non-depressed mothers were equated on propensity scores based on a large set of baseline characteristics, producing two groups that were similar across all measured characteristics except for the presence of clinically significant maternal depression during at baseline (i.e., child age 2). Children’s behavioral outcomes were compared across the equated groups to estimate the causal effects of maternal depression. Mothers completed the Child Behavior Checklist (CBCL) at ages 3, 4, 5, 7, 8, 9, and 10. Teachers completed the Teacher Report Form (TRF) at ages 7, 8, 9, and 10. The externalizing and internalizing subscales of both measures were analyzed as dependent variables.

Results. Propensity score matching was successful in finding a subsample of 404 families in which those with depressed and non-depressed mothers were similar (i.e., standardized mean differences [SMDs] < 0.10) across more than 80 different confounding variables (e.g., income, living situation, child’s behavior problems, mother’s relationship satisfaction, genetic risk score). Child outcomes observed for each group were compared to calculate adjusted effects of maternal depression. Prior to any adjustment, the negative effect of maternal depression at child age 2 on mothers’ later ratings of child externalizing and internalizing behavior (i.e., CBCL) was in the medium range (SMDs~ 0.40−0.60). The magnitude of these effects was similar on externalizing and internalizing behavior, and across measurement occasions ranging from age 3 to age 10. After matching (i.e., equating of the depressed non-depressed groups), effects on mothers’ ratings of child externalizing and internalizing behavior (i.e., CBCL) were reduced to the small range (SMDs~ 0.15−0.25). The unadjusted effects of maternal depression at child age 2 on later teacher ratings of child externalizing and internalizing behavior (i.e., TRF) were smaller than those observed for mothers’ ratings (SMDs~ 0.10−0.30), and were less affected by adjustments for confounding variables (SMDs reduced by ~ 0.05).

Conclusions. Findings are among the strongest evidence to date that maternal depression during toddlerhood has a causal, negative impact on children’s behavioral outcomes in early and middle childhood. However, this impact is smaller than previously documented after accounting for other covarying risk factors. Implications for prevention and treatment are discussed.

Authors

• William E Pelham III, Arizona State University
  Presenting Author

• Stephen West, Arizona State University
  Non-Presenting Author

• Kathryn Lemery-Chalfant, Arizona State University
  Non-Presenting Author

• Sherryl Goodman, Emory University
  Non-Presenting Author

• Daniel S. Shaw, University of Pittsburgh
  Non-Presenting Author

• Melvin N. Wilson, University of Virginia
  Non-Presenting Author

• Thomas Dishion, Arizona State University
  Non-Presenting Author