Regional variation in infant mortality in Kenya: A search for explanations

Abstract

This paper seeks to identify some the factors that underlie regional variation in infant mortality in Kenya. The data drawn from the 1988/89 Kenya Demographic and Health Survey were used for this purpose. Logistic regression is used to analyse the data. On the basis of infant mortality estimates obtained, provinces were grouped into two groups: High (HMP) and low (LMP). The results obtained show that the values of explanatory variables in LMP region than in the high mortality region. However, their differences did not explain much of the variation in infant mortality between the two mortality regions. Decomposing the results revealed that the differences were largely due to the differences in the nature or structure of relationships, as represented by logit coefficients, between mortality and explanatory variables. The results indicate that the lower average level of maternal education, higher proportion of preceding child loss, higher proportion mothers belonging to low economic status households and a lower proportion of mothers belonging to households possessing livestock and lower use of modern contraception modestly contributed to high infant mortality in the high mortality region