Testing the Effectiveness of a Modified Community Model in Improving Child Health Outcomes in Mashonaland East, Zimbabwe

Overview

This study modified and contextualized a community mobilization approach in a bid to find a solution to reduce the high incidence and prevalence of child morbidity and mortality in Zimbabwe.The developed model will be tested for its effectiveness in reducing child morbidity and mortality at community level by comparing the effect of the intervention to that of the conventional community interventions.

In Zimbabwe the Neonatal and infant mortality rate are 31/1000 and 57/1000 respectively (ZDHS, 2011). Seventy -four percent of these death occur at community level due to the delay in deciding to seek care . The government of Zimbabwe has intensified facility based health systems though very little has been done in community health system strengthening yet there is glaring evidence from recent studies that the burden of infant and child morbidity and mortality is largely at community level. The effect of social mobilization approaches in reducing socio-cultural causes of morbidity and mortality is not known. Evidence base for participatory models and their effectiveness remains scanty and in-country, there is no approved , integrated community mobilization model that achieves saturation coverage in addressing all facets of community maternal newborn and child health along the continuum of care. Community immersion with health services is in line with the Alma Ata Declaration of the Primary Health Care approach that seeks to decentralize health services to the marginalized people. The study will test the effectiveness of the contextualized model in improving child health outcomes. A cluster randomized controlled design will be used in 2 randomly selected districts of Mashonaland East , Zimbabwe. Women 18- 49 years with children 0-48 months in the selected districts will be recruited and followed up for 9-12 months . Stratified randomization will be used to select 11 pairs of participating villages . Central randomization will be employed for allocation concealment . The sample size was calculated in STATA 13. assuming a morbidity prevalence of 50% with a margin error of 5%. The assumed intra-cluster correlation coefficient (ICC) of 0.05 with a level of significance of 5% , design effect of 2.45 and power of 0.9. The sample size per arm will be 330 mother-child pairs and each arm will have 11 clusters. Data will be collected monthly using monthly data collection tools. The study will require participants to commit at most 2 hours once in every two weeks. The main intervention is provision of critical health education and targeted promotion services to the participants in their locality. Participatory approaches will be employed in the intervention arm for learning and sharing of good community IMNCI practices for the prevention and management of childhood illnesses. Participants will be motivated through learning exchange visits and provision of promotional materials as incentives for successful completion and practice of recommended behaviors and child care practices.