Assessing the accuracy of the recording and reporting of malaria rapid diagnostic test results in four African countries: methods and key results
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Rapid diagnostic test (RDT) results are the foundation of both case management and malaria surveillance across sub-Saharan Africa. However, RDT results may be misrecorded in health facility registers by healthcare workers (HCWs), either unintentionally or deliberately, for example, to justify treatment based on clinical judgment. A multi-country evaluation was conducted to quantify the extent of RDT misrecording and identify factors associated with recording and reporting accuracy. This report summarizes the study’s methods, key findings, and implications for improving malaria RDT data quality.
Methods
A mixed-methods observational study was implemented in Benin, Côte d'Ivoire, Nigeria, and Uganda in 2023. At participating health facilities, images of RDTs were captured at the point of care and later reviewed by a trained, external panel. Agreement between the panel’s interpretation and the result recorded in the register was assessed using weighted mean Cohen’s kappa (κ). Secondary objectives included identifying factors associated with inaccurate recording, assessing the durability of RDT results after 1 month of storage, evaluating accuracy of data obtained from the District Health Information System 2 (DHIS2), and measuring the observer effect of the start of the study on test positivity rates (TPR) reported to DHIS2.
Results
A total of 102,337 RDT results was observed. Agreement between register-recorded results and the external panel was high, ranging from κ = 0.80 (95% confidence interval [CI] 0.75, 0.85) in Nigeria to κ = 0.88 (95% CI 0.84, 0.92) in Benin. HCWs were more likely to misrecord results as positive (range: 5.1–7.3%) than negative (range: 0.7–3.7%), and patient age was associated with misrecording results in all countries except Nigeria. After the study began in Côte d’Ivoire, TPRs reported from the DHIS2 declined significantly more in study than control facilities (rate ratio: 0.80, 95% CI 0.76, 0.84).
Conclusions
Although HCWs generally recorded RDT results accurately, the disproportionately higher proportion of results misrecorded as positive raises concern about possible intentional misreporting. The observed TPR decrease after study onset in Côte d’Ivoire suggests some HCWs can be motivated to reduce misrecording and improve the quality of malaria case management and surveillance data.
