Targeting childhood malaria: insights from multi-country DHS data in Sub-Saharan Africa

Malaria remains a major public health challenge in sub-Saharan Africa, disproportionately affecting children under five and pregnant women. Understanding the socio-demographic and maternal determinants of malaria across multiple countries is critical for designing targeted interventions such as Targeted Drug Administration (TDA), Perennial Malaria Chemoprevention (PMC), Seasonal Malaria Chemoprevention (SMC), Vector Control and Epidemiological Mapping. This study aimed to assess the prevalence and determinants of malaria infection among children under five across seven sub-Saharan African countries and to identify country-specific risk factors.

A cross-sectional analytical study used secondary data from the most recent Demographic and Health Surveys (DHS-8) for Burkina Faso, Côte d'Ivoire, Kenya, Mozambique, Senegal, Tanzania, and for Cameroon, Malaria Indicator Surveys (MIS-8) was used. Two outcome measures were defined: malaria infection based on rapid diagnostic test (RDT) results (five countries: Burkina Faso, Côte d'Ivoire, Mozambique, Tanzania and Cameroon) and maternal self-report (seven countries: Burkina Faso, Côte d'Ivoire, Kenya, Mozambique, Senegal, Tanzania and Cameroon). The data were analyzed using Stata/SE 17.0 and the analytic samples included 20,323 mother-child pairs for RDT analysis and 12,457 for self-reported analysis. The study examined socio-demographic and maternal characteristics, including residence, wealth index, education, parity, body mass index (BMI), and pregnancy status. Analyses incorporated sampling weights and survey design. Descriptive statistics, chi-square tests, and country-specific multivariable logistic regression models were used to identify independent predictors.

Malaria prevalence was higher by self-report (33.5%) than by RDT diagnosis (25.2%), with consistently higher self-reported estimates across Tanzania (31.7% vs. 6.9%), Burkina Faso (58.3% vs. 27.4%), and Côte d'Ivoire (47.1% vs. 36.9%). RDT-confirmed malaria showed stronger and more consistent associations with key risk factors, including rural residence (in Burkina Faso aOR = 3.58, in Côte d'Ivoire aOR = 3.64, in Mozambique aOR = 5.74, in Tanzania aOR = 12.70; all p < 0.001), wealth status has lower malaria risk (richest quintile: Burkina Faso aOR = 0.43, Côte d'Ivoire aOR = 0.26, Mozambique aOR = 0.14, Tanzania aOR = 0.27), and while pregnancy status has significant risk (in Burkina Faso p = 0.008, in Côte d'Ivoire p = 0.005, in Cameroon p = 0.006, and in Mozambique p < 0.001), compared with weaker and sometimes inconsistent self-reported associations. The RDT- malaria diagnosis produced more plausible and statistically robust associations (in Mozambique age 25-34: aOR = 1.32, p = 0.040; in Burkina Faso parity two births: aOR = 1.33, p < 0.001), whereas self-reported malaria showed weaker or contradictory effects suggesting potential misclassification and reporting bias in self-reported diagnosis.

This study demonstrates that important methodological and epidemiological differences exist between self-reported malaria and malaria RDT diagnosis across sub-Saharan African countries. Self-reported malaria showed overestimated malaria prevalence and produced weaker and less consistent associations with established risk factors, indicating limited reliability for identifying true malaria infection patterns. In contrast, malaria RDT diagnosis generated more coherent and epidemiologically plausible associations with rural residence, socioeconomic status, nutritional status, and pregnancy, reflecting more accurately the biological and environmental determinants of malaria transmission.