Modeling urban malaria infection in Anopheles stephensi hotspot area in Eastern Ethiopia: application of Structural Equation Modeling

In Ethiopia, the fight against malaria faces significant challenges, including the emergence of insecticide resistance, vector behavioral change, population movement, climate change, civil unrest, emergence of COVID-19 pandemic, unplanned urbanization, invasion and spread of urban malaria vector Anopheles stephensi. Modeling the complex relationship and contribution of these factors to malaria infection is essential for ultimate malaria elimination. Hence, the aim of this study is to model the direct and indirect effect of factors affecting the risk of urban malaria infection in eastern Ethiopia where an invasive malaria vector has been recently detected. A facility based cross-sectional study was conducted among 329 febrile urban resident patients visiting public health facilities of Dire Dawa city using an interviewer administered questionnaire. Structural Equation Modeling (SEM) was done to identify the direct and indirect effects of factors for malaria infection. Lavaan (Latent variable analysis) package was used in R and diagonally weighted least square (DWLS) estimation method was employed. The confirmatory factor analysis indicated that all selected factors were significantly loaded on their respective latent variables. The direct effect of the final model indicated that wealth index had a negative statistically significant effect on insecticide treated nets (ITN) utilization (-0.66; p < 0.001) and knowledge on malaria and its prevention (-0.63; p < 0.001). Attitude had positive effect on ITN utilization (0.16; p = 0.049) and having history of travel outside the city had significant positive effect on malaria infection (0.969; p = 0.01). The indirect effect analysis revealed two pathways in which attitude and utilization as the mediating factor significantly influenced the risk of malaria infection (indirect path coefficient=-0.091; p = 0.038) and (indirect path coefficient = 0.029; p = 0.048) respectively. SEM is an effective technique that identified the direct and indirect effects of wealth index, ITN utilization, knowledge, attitude and history of travel on risk of urban malaria infection. Hence, strengthening holistic approach and urban-targeted malaria interventions should be enhanced to prevent and control malaria infection in urban settings.