Accuracy of recording of malaria rapid diagnostic tests in Côte d’Ivoire

Abstract Background Accurate malaria diagnosis and reporting are critical for effective case management and surveillance. In sub-Saharan Africa, rapid diagnostic tests (RDTs) are widely used to support clinical decision-making. However, limited data exist on the accuracy of recorded RDT results reported to health information systems. This study assessed the accuracy of recording of malaria RDT results documented in health facility registers in Côte d’Ivoire, as part of a multi-country evaluation. Methods A mixed-methods, observational study was conducted across 16 primary health care facilities in two regions of Côte d’Ivoire between August and December 2023. For each patient tested for malaria, a digital image of the RDT cassette was captured and linked to the corresponding register entry. An independent panel of trained reviewers interpreted the RDT images, which served as the reference standard. Agreement between panel and register results was assessed using percent agreement and Cohen’s kappa (κ). The positive predictive value (PPV) and negative predictive value (NPV) were calculated for the register result. Meta-regression was used to identify facility, healthcare worker (HCW), and patient characteristics associated with agreement. Results Of 11,161 matched RDT images and register entries, 57.9% were interpreted as positive by the external panel. Overall agreement between panel and register results was strong (κ = 0.83, 95% CI 0.77, 0.88), with a PPV of 90.7% and NPV of 94.8%. However, negative or invalid results were more frequently incorrectly recorded as positive (5.9%) than the reverse (2.0%). Recording errors occurred more often among patients noted in the register with a diagnosis of malaria or antimalarial prescription, suggesting potential systematic bias. District, patient volume, HCW cadre, and education level of the health worker were associated with agreement. Notably, HCWs who frequently performed or recorded RDTs had lower agreement levels. Most HCWs believed that a negative RDT could miss malaria and that treatment could still be warranted. Conclusion The results from the study demonstrate that RDT results are recorded fairly accurately in Côte d’Ivoire. However, the disproportionate misclassification of negative results as positive may distort malaria surveillance data and test positivity rates. Strategies such as regular comparison of RDT cassettes with register entries, enhanced HCW training, and reinforcement of adherence to diagnostic guidelines may improve data quality and support evidence-based decision-making.