Estimation of Childhood Mortality in Kenya Using the 2015/2016 Kenya Integrated Household and Budget Survey

Abstract

Childhood mortality is one of the key indicators used in determining both the general and national socio-economic development, such as resource allocation and policymaking, among other uses. This can only be achieved by ensuring that data are accurate at both the national and county level. This study aimed to estimate childhood mortality in Kenya using the 2015/2016 Kenya integrated household and budget survey (KIHBS). The study specifically sought to establish the quality of KIHBS data, levels of under-five mortalities at the county level, and whether the KIHBS estimates on childhood mortality at county levels are plausible. A survey research design was adopted in this study. Secondary data from the 2015/2016 Kenya Integrated Household Budget Survey (KIHBS) and statistics on child deaths and childbirths between September 2015 and August 2016 was used. The survey design deliberately captured a family range of demographic data such as age, child deaths and childbirths, and women with the highest educational level. Data quality was checked using the age ratio method, Whipples, and United Nations accuracy score method. The differential proportion of children dead by the women's age was also assessed to establish the pattern across age groups. Coale–Demeny West model was used to calculate the under-five mortality. Regression was used to check the plausibility of U5MR from the 2015/2016 Kenya Integrated Household Budget Survey (KIHBS). The study found that the 2015/2016 Kenya Integrated Household Budget Survey (KIHBS) data was inaccurately reported with large fluctuations in age ratios for males and females, indicating persons of various ages being carried across age group boundaries or persons misreporting their ages for various reasons. The study concludes that quality issues exist in the 2015/16 KIHBS data. It is not suitable to estimate the under-five mortality rate at the county level because it did not have recent fertility (births in the last 12 months) and had inconsistent results for all the counties. The study recommends that agencies involved in data collection should engage the public in understanding the importance of accurate reporting by creating policies aligned to supporting mass awareness and educating people on how false reporting during surveys distorts information derived from data and how their lives are likely to be affected. DocuSign Envelope ID: DB87F651-ABCC-49E1-B078-163DA6F1D3B4