Do improvements in outreach, clinical, and family and community-based services predict improvements in child survival? An analysis of serial cross-sectional national surveys

BACKGROUND: There are three main service delivery channels: clinical services, outreach, and family and community. To determine which delivery channels are associated with the greatest reductions in under-5 mortality rates (U5MR), we used data from sequential population-based surveys to examine the correlation between changes in coverage of clinical, outreach, and family and community services and in U5MR for 27 high-burden countries. METHODS: Household survey data were abstracted from serial surveys in 27 countries. Average annual changes (AAC) between the most recent and penultimate survey were calculated for under-five mortality rates and for 22 variables in the domains of clinical, outreach, and family- and community-based services. For all 27 countries and a subset of 19 African countries, we conducted principal component analysis to reduce the variables into a few components in each domain and applied linear regression to assess the correlation between changes in the principal components and changes in under-five mortality rates after controlling for multiple potential confounding factors. RESULTS: AAC in under 5-mortality varied from 6.6% in Nepal to -0.9% in Kenya, with six of the 19 African countries all experiencing less than a 1% decline in mortality. The strongest correlation with reductions in U5MR was observed for access to clinical services (all countries: p = 0.02, r(2 )= 0.58; 19 African countries p < 0.001, r(2 )= 0.67). For outreach activities, AAC U5MR was significantly correlated with antenatal care and family planning services, while AAC in immunization services showed no association. In the family- and community services domain, improvements in breastfeeding were associated with significant changes in mortality in the 30 countries but not in the African subset; while in the African countries, nutritional status improvements were associated with a significant decline in mortality. CONCLUSIONS: Our findings support the importance of increasing access to clinical services, certain outreach services and breastfeeding and, in Africa, of improving nutritional status. Integrated programs that emphasize these services may lead to substantial mortality declines.