IntroductionMalaria is a public health concern in Zambia. In 2018, Zambia reported a 9% malaria prevalence in under-five children, with some provinces reporting above 20%. Factors still driving malaria infection among the country’s under-five children must be investigated. Factors associated with malaria infection among under-five children were investigated.MethodsThis was a secondary analysis of the cross-sectional study for the Zambia Malaria Indicator Survey 2018 (MIS). Multistage sampling was used in the malaria indicator survey. All children aged 6 to 59 months who received malaria rapid diagnostic tests (RDT) in the data set were considered for the study. Malaria infection: Tested positive or negative for malaria RDT. Sample weights and multivariable logistic regression were used. Backward stepwise regression was used to determine the best model, and the Akaike information criterion was used to select the best model that best fits the data. The Odds ratio measured the association at a 95% confidence level.ResultsA total of 2400 children were analysed. 24.3% (583) tested positive. The median age was 32 (interquartile range (IQR:8-46)) months. Males were 52% (1,249). The odds of malaria infection increased as the child’s age in months increased (aOR=1.004, 95% CI: 1.003, 1.005). Children in the rural had higher odds of malaria than urban children (aOR=1.11,95% CI: 1.05, 1.17). The odds of malaria in children in Copperbelt, Luapula, Muchinga and North-Western Provinces were higher than in children in Central Province (p-value <0.05). Children whose houses did not receive IRS had increased odds of malaria compared to children whose houses received IRS (aOR=1.05, 95%CI: 1.01, 1.09). Sex was not statistically significant.ConclusionAn increase in age, living in the rural and northern parts of Zambia was associated with malaria infection. Increasing malaria prevention and control measures for older children, rural communities, and the northern parts of Zambia may help reduce malaria prevalence.Strengths and LimitationsThe study used a large dataset of the Malaria Indicator Survey 2018 that was powered to be nationally representative of Zambia, urban and rural strata, and provincial levels.The study used a complex data analysis considering household weights, census enumeration areas and provinces, addressing intra-cluster correlations.The study addressed confounding using multivariable regressing and determined the best-fit model using Akaike Information Criteria (AIC) in a backward stepwise regressing.The study’s main limitation is that it is a survey that simultaneously assessed the outcome and exposure variables, eliciting associations only and not causality.