Community interventions to reduce child mortality in Dhanusha, Nepal: study protocol for a cluster randomized controlled trial

Abstract

Background: Neonatal mortality remains high in rural Nepal. Previous work suggests that local women's groups can effect significant improvement through community mobilisation. The possibility of identification and management of newborn infections by community-based workers has also arisen.

Methods/design: The objective of this trial is to evaluate the effects on newborn health of two community-based interventions involving Female Community Health Volunteers. MIRA Dhanusha community groups: a participatory intervention with women's groups. MIRA Dhanusha sepsis management: training of community volunteers in the recognition and management of neonatal sepsis.The study design is a cluster randomized controlled trial involving 60 village development committee clusters allocated 1:1 to two interventions in a factorial design. MIRA Dhanusha community groups: Female Community Health Volunteers (FCHVs) are supported in convening monthly women's groups. Nine groups per cluster (270 in total) work through two action research cycles in which they (i) identify local issues around maternity, newborn health and nutrition, (ii) prioritize key problems, (iii) develop strategies to address them, (iv) implement the strategies, and (v) evaluate their success. Cycle 1 focuses on maternal and newborn health and cycle 2 on nutrition in pregnancy and infancy and associated postpartum care practices. MIRA Dhanusha sepsis management: FCHVs are trained to care for vulnerable newborn infants. They (i) identify local births, (ii) identify low birth weight infants, (iii) identify possible newborn infection, (iv) manage the process of treatment with oral antibiotics and referral to a health facility to receive parenteral gentamicin, and (v) follow up infants and support families.

Primary outcome: neonatal mortality rates.

Secondary outcomes: MIRA Dhanusha community group: stillbirth, infant and under-two mortality rates, care practices and health care seeking behaviour, maternal diet, breastfeeding and complementary feeding practices, maternal and under-2 anthropometric status. MIRA Dhanusha sepsis management: identification and treatment of neonatal sepsis by community health volunteers, infection-specific neonatal mortality.

Figure 1

Design of the study. Essential Newborn Care training was provided throughout the district, followed by exclusion of small and conflict-affected clusters and stratified block randomisation on the basis of cluster size to women's groups intervention and control. A second block randomisation allocated equal numbers of clusters to sepsis management intervention within each women's group study arm.

Figure 2

The sequence of women's group activities. Meetings conducted on a monthly basis between May 2007 and April 2011 by 270 groups in 30 study clusters. Strategies initiated after meeting 10 in both cycles may be implemented continually while meetings on other topics are ongoing and also after the end of the intervention period.

Figure 3

Flowchart of community-based sepsis management of babies 0-2 months by Female Community Health Volunteers in Dhanusha. Actions and decisions undertaken by FCHVs when checking birth weight and for signs of possible local bacterial infection or possible severe bacterial infection in babies born in their working areas.