Send to

Choose Destination
BMJ Glob Health. 2017 Oct 25;2(4):e000209. doi: 10.1136/bmjgh-2016-000209. eCollection 2017.

Variation in neonatal mortality and its relation to country characteristics in sub-Saharan Africa: an ecological study.

Author information

Julius Global Health, Julius Center for Health Sciences and Primary Care, University Medical Centre Utrecht, Utrecht, The Netherlands.
International Research Centre of Excellence, Institute of Human Virology, Abuja, Nigeria.
Global Geo and Health Data Center, Utrecht University, Utrecht, The Netherlands.
Postdoctoral Unit, Noguchi Memorial Institute for Medical Research, University of Ghana, Accra, Ghana.
Department of Epidemiology and Disease Control, School of Public Health, University of Ghana, Accra, Ghana.
Research and Development Division, Ghana Health Service, Accra, Ghana.
Warwick-Centre for Applied Health Research and Delivery, Division of Health Sciences, Warwick Medical School, The University of Warwick, Coventry, UK.
International Health Group, Liverpool School of Tropical Medicine, Liverpool, UK.
Division of Epidemiology and Biostatistics, Faculty of Health Science, School of Public Health, the University of Witwatersrand, Johannesburg, South Africa.



A substantial reduction in neonatal mortality is the main priority to reduce under-five mortality. A clear understanding of the variation in neonatal mortality and the underlying causes is important for targeted intervention. We aimed to explore variation in neonatal mortality and identify underlying causes of variation in neonatal mortality in sub-Saharan Africa (SSA).


This ecological study used 2012 publicly available data from WHO, the US Agency for International Development and the World Bank. Variation in neonatal mortality across 49 SSA countries was examined using control chart and explanatory spatial data analysis. Associations between country-level characteristics and neonatal mortality were examined using linear regression analysis.


The control chart showed that 28 (57%) SSA countries exhibited special-cause variation, 14 countries were below and 14 above the 99.8% control-limits. The remaining 21 (43%) SSA countries showed common-cause variation. No spatial clustering was observed for neonatal mortality (Global Moran's I statistic -0.10; p=0.74). Linear regression analysis showed HIV/AIDS prevalence among the population of reproductive age to be positively associated with neonatal mortality (β=0.463; 95% CI 0.135 to 0.790; p<0.01). Declining socioeconomic deprivation (β=-0.234; 95% CI -0.424 to -0.044; p<0.05) and high quality of healthcare governance (β=-1.327, 95% CI -2.073 to -0.580; p<0.01) were inversely associated with neonatal mortality.


This study shows a wide variation in neonatal mortality in SSA. A substantial part of this variation can be explained by differences in the quality of healthcare governance, prevalence of HIV and socioeconomic deprivation. Future studies should validate our findings using more rigorous epidemiological study designs.


child health; health systems

Supplemental Content

Full text links

Icon for BMJ Publishing Group Icon for PubMed Central
Loading ...
Support Center