Seasonal variations in under-five mortality, stratified by neonatal, post-neonatal and child mortality, in Korogocho and Viwandani urban slums in Nairobi, Kenya : a time-series analysis on secondary data from the Nairobi Urban Health and Demographic Surve

Abstract: Abstract Background  Globally, the amount of people living in urban poor settings is increasing rapidly. In order to effectively direct resources and achieve equity in health, an understanding of how contextual factors are affecting mortality in marginalised groups is first needed. Aim  This study aims at clarifying if seasonality can be associated with mortality in different child age spans i.e. neonatal, post-neonatal and children under five in two urban slum settlements in Nairobi, Kenya. Methodology  Secondary seasonal mortality data from the Nairobi Urban Health and Demographic Surveillance System (NUHDSS), collected in the years 2008–2010 in Viwandani and Korogocho slums in Nairobi, Kenya was decomposed into seasonal components with a moving average comparison. Seasonal indices were created on disaggregated data for the different mortality groups and gender strata. Results  The annual mean prevalence of under-five mortality [U5M] and neonatal mortality [NM] was 86 and 20 respectively. Overall U5M increased from April–August. Post-neonatal mortality [PM] had the earliest onset (April–May) followed by NM (May–August). Child mortality [CM] had two peaks, June–August and November. Conclusion  The displayed seasonal pattern in NM, PM, CM and U5M seemingly corresponded to the yearly fluctuation of temperature and precipitation, with a predominant intensification in the wettest and coldest months. CM seemed to be the only group with increased mortality also in the second rain period, not reflected in the overall mortality index. These findings could be used to set the agenda for preventative interventions aimed at reducing childhood mortality and morbidity.