Mechanistic effects of a Family Strengthening Intervention on child outcomes in Rwanda - the roles of caregiver mental health

• In Event: Factors affecting child development and executive functioning in three low income counties

Abstract

Background: Rich evidence suggests that families living in extreme poverty benefit from interventions to support early childhood development. Poverty-related risks that are known to affect child development and that can be targeted in intervention include low stimulation, poor hygiene and nutrition, violence, and neglect. We present data from a longitudinal cluster-randomized trial examining the effectiveness of Sugira Muryango (Strengthen the Family) in Rwanda. Families were eligible to participate in Sugira Muryango if they belonged to the poorest categorization of poverty as defined by the Rwandan government and had at least one child aged 6-36 months in Rwanda. Sugira Muryango was delivered by trained lay workers via active coaching in the family’s home. The curriculum covers responsive parenting as well as active education and problem solving to promote family functioning, increase father engagement, and improve family health (Figure 1). Result from the pre- to post-intervention assessment suggested that participating in Sugira Muryango was associated with increased stimulation and positive parenting, improved caregiver mental health, and reduced family violence including harsh discipline (redacted, 2020). Here we present findings from the 12 months follow up and explore the extent to which improved caregiver mental health mediate positive effects of Sugira Muryango on children’s developmental outcomes. Mental health problems are particularly prevalent among people living in poverty and can impact parents’ ability to attend to children’s developmental needs. Parental mental health is likely to both mediate and moderate effects of parent-focused interventions to promote early childhood development.

Methods: The cluster randomized trial enrolled 1049 families. Data were collected at three timepoints: baseline, immediately after the end of the intervention, and 12 months after the intervention ended. Developmental outcomes were assessed using the Ages and Stages Questionnaire (ASQ-3). Caregiver mental health was assessed using Hopkins Symptoms Checklist (HSCL-25) to assess anxiety and depression. Latent growth modeling in Mplus was used to examine indirect effects of caregiver mental health on child outcomes.

Results: We include 541 intervention families and 508 control families in the analyses. Attrition was very low (2.0% children; 9.6% caregivers) and addressed using hot deck imputation. Children in families receiving Sugira Muryango improved more on gross motor (d=0.162, 95%CI: 0.065–0.260), communication (d=0.081, 95% CI: 0.005–0.156), problem solving (d=0.101, 95% CI: 0.002–0.179), and personal-social development (d=0.096, 95% CI: -0.015–0.177) on ASQ-3. Latent growth models showed that changes in caregiver depression explained some of the change in children’s problem solving such that higher rates symptoms were associated with a more negative slope (= -0.067, p=0.010). Caregiver mental health did not mediate effects on gross motor, communication or personal-social development.

Conclusions: Understanding mechanisms by which early parenting interventions improve child development is important for optimization and scale up of programs to reach vulnerable families. Findings suggest that parental mental health is one mechanism through which parenting intervention generates positive outcomes, but that more mechanisms must be at play. Other potential mediators of intervention effects include stimulation in the home, father engagement and harsh discipline will be discussed.

Authors

• Sarah Jensen, Boston College
  Presenting Author

• Matias L. Placencio-Castro, Boston College
  Non-Presenting Author

• Jordan Farrar, Boston College
  Non-Presenting Author

• Robert T Brennan, Research Program for Children and Adversity at Boston College
  Non-Presenting Author

• Theresa Stichick Betancourt, Boston College
  Non-Presenting Author