The overall objective of the study is to examine the effects of integrating early child development group sessions into the existing, at-scale, community health and nutrition programs administered by the government in Madagascar.

This research is a cluster-randomized trial that will assess the effects of integrating early child development activities on community health worker time use and caregiver time use and participation in the program. The evaluation will compare the addition of early child development sessions to the status quo health and nutrition program, with an additional intervention arm looking at the addition of toys to the program.

This study has four objectives, which will be measured at the community health worker level, and at the caregiver level for caregiver-child dyads (where children are between 6-30 months old at time of intervention launch):

1. Measure how the integration of ECD activities with standard nutrition programming affects the time and task allocation of the community health workers
2. Measure the extent, if any, that the addition of ECD activities crowds out community health worker health/nutrition tasks that are part of the standard community program
3. Measure how the integration of ECD activities affects rates of caregiver participation in nutrition and health activities over time (e.g., during changes in seasonal activities).
4. Measure the impact of enhanced availability and maintenance of age-appropriate play materials/activities on sustained caregiver participation rates in the program.

Inclusion Criteria for communities:

1. No other ECD programs present in the community or neighboring communities;
2. At least 40 children monitored at the nutrition center in the target age range of 6-30 months (monitored is defined as attending at least one growth monitoring session between July 2019 to August 2020);
3. Stable supervisory presence. Local NGOs are contracted to monitor and supervise all community health workers and help to strengthen linkages with local structures such as the community health committees and primary care facilities. Each NGO supervisor is in charge to up to 9 communities, which they visit twice a month. The investigators defined stable presence as not having had NGO turnover within the last year.

Among the communities included, children 6-30 months were eligible to participate in the intervention. There was no inclusion criteria for community health workers.

In accordance with the World Bank's Open Data and Knowledge Initiative, the deidentified data collected in the study will be made publicly available at the data repository at the World Bank, which is expected to be within 24 months of the final data collection date. Metadata and critical documents (i.e. protocols and questionnaires) will conform to the standards of the Data Documentation Initiative and will be made available within one year of the end of data collection.

The investigators that are analyzing the data are blinded to treatment assignment during analysis.