Influence of household environment on child morbidity and mortality in slums: a case study of Mathare and Soweto, Nairobi
Child mortality and morbidity disparities continue to be observed in this era of improved expansion of the provision of health care services. Some areas have low rates while others have high. The observed patterns are attributed to uneven distribution of health facilities, which guided the choice of study sites, namely Mathare and Soweto slum areas in Nairobi. The study objective of the study was to assess the perception and awareness of slum dwellers about the influence of household environment and behaviour patterns on child morbidity and mortality, given the available health, information and education service inputs. The study hypothesis was ''Households whose women believed that their children were more susceptible to child sickness in the environment they lived in and who had a positive perception towards a healthy behaviour were likely to be negatively correlated to child morbidity and mortality''. The methods that were used to collect data were: structured questionnaires, interviews, focused group discussion (FGDs), documentary information and observation methods. The study targeted women aged between 15-49 year who had children under 5 years of age. A total of 599 women from the study area were included in the sample. The Cluster sampling technique was applied. The main methods of data analysis for estimating child mortality were the Coale and Trussel techniques, ordinary least squares regression, the binomial distribution and descriptive techniques. A key finding of the study is that the probability of a child dying at exact age 2 (q2) values for Soweto and Mathare slums were estimated as 69.9 and 64.9 deaths per thousand live births respectively, while the estimates of the probability of dying at exactly age 5 (q5) values were 105.3 and 146.7 deaths per thousand live births, for Soweto and Mathare slums respectively. These values are high compared to the national values that have a q2 and a q5 of 60 and 96, respectively. The behavioural variables that were significant and positively correlated with child loss were: self-administration of medicine, incomplete immunization, type of medical treatment sought and level of toilet cleanliness. The psycho-social variable that was significant and negatively correlated to familial risk to child deaths was the perception of household residential arrangement as consisting of wrong people and being over-crowded. Those that were consisting of wrong people and being over-crowded. Those that were significant and positively correlated to familial risk to child death include; pre-natal clinical attendance and water quality status. The household variable that was significant and positively correlated to familial risk of child deaths was household size (7+). The individual variables that were significant and positively correlated to familial risk to child loss were; households whose women were aged between 35 and 49 and did not belong to any denomination. The study findings show that those households that had higher familial risk to child loss were mainly poor and had negative attitude towards modern medicine, and generally had low hygienic standards. The study recommends that public awareness on proper hygienic standards should be advocated. Also, the government should undertake training of the required health workers who will later be deployed into these areas (slums) in order to improve the living conditions. The Nairobi City Council should mobilize social and health workers to educate slum dwellers on proper hygiene and sanitation (outreach services), while there is need for a strong legislation guiding house construction and other social amenities.