Objectives: To describe the spatial pattern in under-5 mortality rates in the Basse Health and Demographic Surveillance System (BHDSS) and to test for associations between under-5 deaths and biodemographic and socio-economic risk factors.
Methods: Using data on child survival from 2007 to 2011 in the BHDSS, we mapped under-5 mortality by km(2) . We tested for spatial clustering of high or low death rates using Kulldorff's spatial scan statistic. Associations between child death and a variety of biodemographic and socio-economic factors were assessed with Cox proportional hazards models, and deviance residuals from the best-fitting model were tested for spatial clustering.
Results: The overall death rate among children under 5 was 0.0195 deaths per child-year. We found two spatial clusters of high death rates and one spatial cluster of low death rates; children in the two high clusters died at a rate of 0.0264 and 0.0292 deaths per child-year, while in the low cluster, the rate was 0.0144 deaths per child-year. We also found that children born to Fula mothers experienced, on average, a higher hazard of death, whereas children born in the households in the upper two quintiles of asset ownership experienced, on average, a lower hazard of death. After accounting for the spatial distribution of biodemographic and socio-economic characteristics, we found no residual spatial pattern in child mortality risk.
Conclusion: This study demonstrates that significant inequality in under-5 death rates can occur within a relatively small area (1100 km(2) ). Risks of under-5 mortality were associated with mother's ethnicity and household wealth. If high mortality clusters persist, then equity concerns may require additional public health efforts in those areas.
Keywords: Cox proportional hazards models; Gambia; Gambie; The Gambia; analyse spatiale; análisis espacial; child mortality; demographic surveillance; facteurs de risque; factores de riesgo; modelos de riesgos proporcionales de Cox; modèles de Cox à risques proportionnels; mortalidad infantil; mortalité infantile; risk factors; spatial analysis; surveillance démographique; vigilancia demográfica.
© 2015 John Wiley & Sons Ltd.