Anaemia severity as a predictor of false-negative and false-positive malaria RDT results among children under five in sub-Saharan Africa

Malaria rapid diagnostic tests (RDTs) are the primary point-of-care diagnostic tool for malaria across sub-Saharan Africa, yet their accuracy varies with patient characteristics. This study examines how RDT performance changes across the full spectrum of anaemia severity in children under five, with particular attention to both false-negative and false-positive errors. Cross-sectional analysis of nationally representative DHS data from 96,611 children aged 6 to 59 months across 20 sub-Saharan African countries. Microscopy was the reference standard. RDT sensitivity, specificity, false-negative rates (FNR), and false-positive rates (FPR) were estimated within four WHO anaemia categories. Survey-weighted logistic regression modelled the associations between anaemia severity and each error type, adjusting for age, sex, residence, wealth, ITN use, and maternal education; modified Poisson regression provided risk ratios on the same scale. In the eight surveys that recorded antimalarial treatment history for RDT-positive children, we tested whether recent treatment was associated with a microscopy-negative result. Pooled sensitivity was 65.1% in non-anaemic children and rose to 94.5% in those with severe anaemia. The false-negative rate fell from 34.9% to 5.5% across the same range. Specificity moved in the opposite direction: 94.0% in non-anaemic children falling to 61.3% in severe anaemia, with the FPR reaching 38.7% in that group. Adjusted odds of a false-negative result were 0.09 (95% CI 0.07–0.12, p <0.0001) for severe anaemia compared to non-anaemic children. Adjusted odds of a false-positive were 10.79 (95% CI 9.03–12.88, p <0.0001) for severe anaemia. Of all false-negative results, 41.9% occurred in children who also had moderate or severe anaemia. Among RDT-positive children in the six surveys recording recent artemisinin-based combination therapy, 66.0% of those treated in the previous two weeks were microscopy-negative compared with 38.2% of those not treated (adjusted OR 3.29, 95% CI 2.75–3.93). RDT performance varies systematically with anaemia severity in opposite directions for sensitivity and specificity. A negative RDT in a severely anaemic child still carries a clinically meaningful miss rate, while a positive RDT in the same child was strongly associated with recent antimalarial treatment, consistent with persisting antigen rather than active parasitaemia. Haemoglobin status should inform how RDT results are interpreted in clinical practice.