Internal Publication
Estimating causal effects of malaria infection on anaemia among children under five years in Uganda: evidence from a cross-sectional malaria indicator survey
Background and Aims: Malaria and anemia remain major public health challenges contributing substantially to morbidity and mortality among children under 5 years in sub‐Saharan Africa (SSA). Although conventional regression analyses have reported a significant association between malaria infection and anemia, these approaches often inadequately account for socioeconomic and demographic confounders, thereby limiting causal interpretation. This study aimed to estimate the causal effect of malaria infection on anemia among children under five in Uganda using a robust quasi‐experimental approach.
Methods: We conducted a Propensity Score Matched (PSM) analysis using data from the 2018–2019 Uganda Malaria Indicator (Cross‐Sectional) Survey (UMIS). The study population comprised children under 5 years of age with complete data on malaria
status (exposure) and anemia status (outcome). A logistic regression model was used to estimate the propensity scores, incorporating a comprehensive set of child‐level, maternal, and contextual covariates. The Average Treatment Effect on the Treated (ATT) was estimated in Stata v14.0 using survey weights to account for the UMIS complex sampling design.
Result: The prevalence of malaria infection among children under five was 23.6% [95% CI: 19.8, 27.8], while the prevalence of anaemia (mild, moderate, or severe) was 51.2% [95% CI: 48.3, 54.1]. The effect of exposure to malaria infection on anaemia was
positive and significant [Average Treatment Effect of the Treated (ATT) = 25.6%, 95% CI: 17.6%, 33.5%, p < 0.001]. These results indicate that malaria infection increases the likelihood of anaemia by 25.5 percentage points among infected children under five in Uganda, after adjusting for key confounders.
Conclusion: Malaria infection is a major driver of anaemia among Ugandan children under five. Strengthening malaria prevention and prompt treatment could substantially reduce childhood anemia and its associated health burden.
Author: Mwebesa, Edson | ORCID: https://orcid.org/0000-0003-4421-5356
Co-author: Agasa, Lameck Ondieki | ORCID: https://orcid.org/0000-0002-3740-8865
Co-author: Awor, Susan | ORCID: https://orcid.org/0009-0002-4106-3553
Co-author: Kazonda, Herbert Tafadzwa | ORCID: https://orcid.org/0009-0009-1788-2285
Co-author: Jimmy, Abraham Isiaka | ORCID: https://orcid.org/0009-0005-3885-1761
Co-author: Essetcheou, Scholastique M. Merveille | ORCID: https://orcid.org/0009-0004-7306-9962
Co-author: Tum, Isaac Kipkosgei | ORCID: https://orcid.org/0009-0000-1552-2328
Co-author: Kerich, Gregory Kibet
Co-author: Mwangi, Ann | ORCID: https://orcid.org/0000-0001-5279-8622
Co-author: Agbi, Delight Mawufemor | ORCID: https://orcid.org/0000-0002-1876-8419
Institution: Moi University | Centre: DAAD
Type: Journal article | English | Peer Reviewed
Subjects: Health and Pandemic Prevention
Published in: Health Science Reports, ISSN 2398-8835, EISSN 2398-8835 | Volume 9, issue 8
Publisher of document: Hoboken, NJ: Wiley Periodicals LLC
Date: 5 August 2026 | Pages: 9
https://doi.org/10.1002/hsr2.73001
Copyright: © 2026 The Author(s) | License: Open Access: Creative Commons (CC BY Attribution 4.0 International)
