This paper utilises data on women from the 1998 Kenya Demographic and Health Survey to examine the conditions under which children under the age of five remain healthy or become ill with three fatal diseases, namely, malaria, diarrhea, and acute respiratory infection. The major objective of the paper was to identify the principal determinants of child morbidity and to pinpoint possible areas where current health programmes could be improved to secure greater child survival. Logistic regression is applied on the data to examine the relationship between women status as indicated by various social and economic factors and the morbidity status of their children. The results show that the marital status, mother’s age, number of children living at home, and region are significant in predicting the morbidity status of children for all the three ailments. The results also show that the location of the household on the wealth scale, occupation status, religious affiliation, literacy status and education were significant in predicting the morbidity status of children for either one or two ailments. On the other hand, media use, and household environmental conditions were not significantly associated with the morbidity status of children for any of the ailments. The factors identified as significant in predicting the morbidity status of children have provided an indication of the direction health programmes could take. There is need for the government to continue efforts to increase availability of health services especially in rural areas. In addition, since the sicknesses considered in this analysis are largely preventable, strengthening various components of the Primary Health Care health provision strategy could substantially improve the health of children.