Abstract

Early diagnosis and prompt treatment of malaria in young children are crucial for prevention of serious stages of malaria. In areas where general delayed treatment-seeking habits are observed, targeted campaigns and interventions could be implemented. In this chapter, we applied multivariate binary logistic regression diagnostics and geospatial logistic model to identify traditional authorities having caregivers with outlying health-seeking behaviour for childhood malaria. The 2021 Malawi Malaria Indicator Survey data was adopted for illustration. We used R software version 4.3.0 for regressions and STATA version 14.0 for data cleaning. Both models reported significant between-village variability of treatment-seeking habits of caregivers. The results of the findings of the mixed-effects logit model’s diagnostic assessment indicated that the traditional authorities Vuso Jere, Kampingo Sibande, Ngabu, and Dzoole were outliers in the model. The majority of the caregivers in these traditional authorities sought treatment within twenty-four hours of the onset of malaria symptoms in the children, although these areas had characteristics that promoted late reporting of malaria at clinics. The geospatial logit model estimates showed that the predicted covariate-adjusted prevalence of late seeking of malaria treatment was high in most locations of the country, except few areas such as the traditional authorities of Mwakaboko, Mwenemisuku, Mwabulambya, Mmbelwa, Mwadzama, Zulu, Amidu, Kasisi, and Mabuka. These findings suggest that multivariate regression model residuals and geospatial statistics can complement each other in determining communities with unique treatment-seeking behaviour for childhood malaria in a population.

Original language	en
Pages (from-to)	499-524
Publication status	Published - 2025