Prevalence and risk factors associated with malaria infection in children under the age of fourteen years in Kenya.
Ongoma, Dona Namukowa.
MetadataShow full item record
Despite various efforts by multilateral agencies and governments to prevent, control and eliminate malaria, it continues to be a major plague with more than 300 million at risk of infection worldwide. In Kenya, it is still a leading cause of morbidity and mortality, affecting more than 70% of the population. Malaria is endemic in most parts of the country with either high to moderate transmission patterns or seasonal epidemic patterns. Statistics from the ministry of health records show that it accounts for about 30% of outpatient care and 20% of the admissions in hospitals nationwide. Therefore, it is important to constantly review and understand the epidemiology, and the risk factors associated with malaria infection. Such efforts will help the government and the multilateral agencies in their planning, monitoring and evaluation efforts to control and eventually eradicate malaria. The main objective of the study was to identify the risk factors associated with malaria infection in children under the age of fourteen years. To achieve this, three different statistical methods for analysing complex survey data with a binary outcome, with both linear and non-linear covariates were used. The data used was obtained from a household survey conducted by the government of Kenya in the year 2010 during the peak malaria transmission period. A total of 240 clusters with 30 households in each cluster was sampled from highland epidemic, lake endemic, coastal endemic, seasonal risk and low risk epidemiological regions of Kenya. Probability weights were assigned at each stage of sampling to provide accurate estimates. A total of 11; 310 children between 3 months and 14 years were the identified study subjects. To account for the complexity in the sampling design, survey logistic regression (SLR), a special model under the generalized linear models (GLM) framework was used to identify the risk factors associated with childhood malaria infections. However, the SLR model fails to account for variability arising from correlation between subjects from the same household and clusters. Therefore, the generalized linear mixed effects model (GLMM) was also applied to the data. To relax the assumptions of normality and linearity in the two parametric models, the semi-parametric generalized additive mixed effects model (GAMM), was finally applied to the data. The findings of the study showed that age of the child, cluster altitude in metres, region, place of residence, type of housing structure, availability of toilet facilities, use of insecticide treated bed nets and mother's level of education were the key determinants of the risk malaria. In the fght to control and eliminate malaria, the results of the study can aid in policy formulation.