The accuracy of recording malaria rapid diagnostic test (RDT) results in public health facilities in Benin; results from the MaCRA project

Abstract Background: Accurate interpretation and recording of malaria rapid diagnostic tests (RDTs) are critical for case management and surveillance in malaria-endemic settings. In Benin, where over 90% of malaria diagnoses rely on RDTs, concerns remain about the accuracy of the reporting and recording of RDT results. This study assessed the fidelity of RDT recording by healthcare workers (HCWs) in public health facilities and explored associated factors. Methods: A six-month mixed-methods, prospective observational study was conducted in 16 public health facilities across two departments in Benin. For each RDT performed, an image was captured using a digital RDT reader (HealthPulse, Audere, Seattle, WA USA) and independently interpreted by an external trained panel. HCW-recorded results were compared to panel interpretations. A knowledge, attitudes, practices, and beliefs (KAPB) survey and structured observations of RDT performance were conducted, alongside in-depth interviews with selected HCWs. Results: Of 35,706 RDTs assessed, overall agreement between HCW and reference panel interpretations was 94.3% (Cohen’s kappa = 0.88). Results misrecorded as positive (5.0%) were more frequent than results misrecorded as negative (0.7%). Agreement varied by patient age, HCW experience, and facility characteristics. Accuracy was highest with children under 5 years (96.7%) and lowest with patients over 15 years (91.6%). HCWs with ≥ 10 years of experience, and access to electricity and internet performed better. From 226 HCWs surveyed, 89.4% believed a patient with malaria could have a negative RDT, though only 19.5% supported treating such cases with antimalarials. While most HCWs were proficient in performing RDTs, only 40.5% waited the recommended time before reading results, and glove use was low (15.6%) highlighting safety gaps. RDT use was primarily motivated by adherence to guidelines (60.2%), rather than patient or supervisor expectations. Qualitative interviews highlighted contextual challenges including workload, lighting conditions in health facilities, and resource constraints. Conclusion: HCWs in Benin showed high accuracy in interpreting and reporting malaria RDT results, likely supported by recent nationwide RDT cassette validations. Performance was strongest among those with more experience, training, and adequate infrastructure. However, negative results misrecorded as positive, especially in adult patients, remains a concern. Targeted training and supportive supervision may help strengthen confidence in negative results and improve overall diagnostic accuracy.