BackHome
Refresh: OffView Personal Schedule
Individual Submission Summary
Share...

Direct link:

Integrating early stimulation and play at scale: Evidence from the MAHAY Mikolo Trial in Madagascar

Thu, March 23, 11:45am to 1:15pm, Salt Palace Convention Center, Floor: 1, Grand Ballroom E

Abstract

Background: A recent study in Madagascar (MAHAY) investigated the impact of a scaled-up adaptation of a home-visiting program. The program included two years of fortnightly home visits by a community health worker (CHW), focusing on early childhood development (ECD). Although MAHAY was carefully implemented and rigorously supervised, it failed to demonstrate significant effects on child outcomes. The small effect sizes were likely due to long travel times and threats to personal safety for CHWs to conduct home visits, and lack of resources such as toys that prevented caregivers from engaging in CHWs’ suggestions regarding play.

A group-based adaptation, MAHAY Mikolo (“know how to take care”) was designed to improve effectiveness of the intervention, while also increasing coverage of early childhood development support while focusing on scalability. The current program was integrated into the existing community-based maternal and child health and nutrition program in Madagascar.
The specific goals of this analysis were to:
1) Measure how the integration of ECD activities affected time and task allocation of CHWs and CHW psychosocial well-being;
2) Test whether the integration of ECD activities affected take-up of standard health and nutrition activities;
3) Test whether the availability of age-appropriate play materials at home enhanced caregiver-child participation in ECD group activities.

Methods: The study population consisted of (i) CHWs serving 75 community-based health and nutrition sites (of about 225,000 children) in two regions of Madagascar and (ii) eligible caregivers with 6- 30 month-old children.

Communities were randomly assigned to one of three study arms:
• Treatment [T], which is [C] + ECD group sessions;
• Treatment Plus [T+], which is [T] + an enhanced package of play materials for home use;
• Comparison group [C] (standard of care community-based health and nutrition program);
The intervention began in February 2021 and ended in March 2022 and was comprised of fortnightly group sessions, organized by age groups. CHWs provided caregiver-child dyads with information relating to ECD; sessions also included structured time for caregivers to practice the play and child stimulation activities they learned. Monthly surveys were administered to measure CHW time use and task allocation, with additional measures of CHW psychosocial well-being administered at select survey waves. Monthly caregiver-child participation in sessions was recorded in administrative data. Intention-to-treat analysis was estimated with strata and time fixed effects.

Results: CHWs in the treatment group [T] reported increased time spent on CHW work by approximately 15 percent (p < .10), and [T+] reported increased time spent on CHW work by approximately 30 percent (p < .01) relative to the control group mean. These increases were compensated by a decrease in time spent on outside work and household chores. There was little evidence of crowding-out of health and nutrition activities. There were no significant effects of treatment on self-reported depressive symptoms or perceived stress, nor across a variety of motivation-related measures such as autonomy, perseverance, and growth mindset. Ongoing analyses will expand the analysis to the participation to health, nutrition and ECD activities, and model treatment effect heterogeneity by CHW-, household-, and community-level characteristics.

Authors