Implementation Research on an Effective Parenting Program in Kisumu, Kenya: Comparing Group and Home Delivery

• In Event: Global Parenting Interventions: Advances in Measurement and Scale Up

Abstract

Study Objective: To describe inputs and outputs of the parenting program and relate them to outcomes. Specific research questions were:
1. What program resources were put in place ahead of implementation to fit the needs of parents and delivery agents? Was the program content and delivery acceptable to them?
2. How did Outputs such as delivery fidelity and parental attendance and enactment change over time, and what determined these Outputs?
3. What features of Inputs and Outputs were in turn associated with child and parent Outcomes, and how did this differ by delivery mode?

Background:
Based on the Nurturing Care Framework (World Health Organization, 2018), the Msingi Bora parenting program trained community health volunteers (CHVs) to deliver messages to parents of children 6 to 24 months on topics of responsive care and early learning, along with nutrition, hygiene and love and respect in the family. In order to provide operational guidance on government implementation, we collected data advocated by the CARE reporting guidelines (Consolidated Advice for Reporting of ECD implementation; Yousafzai et al., 2018). We compared delivering all 16 sessions to large groups of 18 mother-child dyads with a mixed-delivery in which four of the sessions were delivered during a home visit.

Methods and Analyses:
Quantitative assessments of fidelity/quality of session delivery, and attendance at group/home meetings were analyzed statistically. Qualitative semi-structured interviews with mothers, CHVs, and local supervisors regarding their acceptance of the program, motivations for attendance, barriers and enablers of practice enactment, and views on group versus home delivery formats were content analyzed by two coders to ensure reliability.

Results:
1. The interviews revealed that the five practices were acceptable to parents, CHVs and supervisors, and the use of a structured manual with practical activities was appreciated.
2. Monitoring forms showed gradual and statistically significant improvement across sessions in delivery quality (Figure). Following session content, expressing acceptance, being prepared, and using visual aids were rated better than facilitating a discussion on barriers to practices. Determinants of fidelity raised during interviews showed how training and supervision influenced CHVs' perceived delivery preparedness. Only CHVs' knowledge test score after training was significantly correlated with delivery quality, while delivery quality was itself a significant determinant of mothers' attendance. Common barriers to enactment included resources and poor understanding of responsive stimulation, whereas common enablers were the program teachings and child benefits.

3. CHV fidelity scores, CHV older age, and Knowledge Test score predicted outcome HOME scores, age and knowledge predicted child Bayley scores. Children in the all-group delivery villages performed better on the Bayley than those in the mixed-delivery villages. Mothers had more positive views on groups, citing their children's joy and benefit interacting with others; whereas CHVs and supervisors had relatively more positive views on home visits.

Discussion:
These findings inform how the program can be scaled up to more districts and to the government's health system, as well as how inputs and outputs are associated with effective outcomes for children and parents.

Group Authors

"As per Anne Perdue please consider these flash talks as one session. They should either be accepted or rejected as a whole.

Flash talks should be ordered as follows: 1707093, 1705221, 1705962, 1707008, 1705220, 1705994, 1706107, 1706048"

Authors

• Jill Luoto, RAND Corporation
  Presenting Author

• Ronald Otieno, Safe Water and AIDS Project (SWAP)
  Non-Presenting Author

• Italo Lopez Garcia, RAND Corporation
  Non-Presenting Author

• Edith Alu, SWAP Kenya
  Non-Presenting Author

• Rebecca Zhu, University of California - Berkeley
  Non-Presenting Author

• Frances E Aboud, Centre universitaire de santé McGill
  Non-Presenting Author