Diagnostic Accuracy of Malaria Rapid Tests in Routine Field Settings in Kassena-Nankana District of Northern Ghana: Implications for the Test and Treat Policy

Abstract

Rapid diagnostic tests for malaria (RDTs) carry the hope that malaria treatment can be prompt, accurate and cost-effective. However, malaria rapid tests can achieve this goal only if they give accurate results comparable to those recommended by World Health Organization, and if health workers apply treatment according to test results. The study evaluated the diagnostic accuracy of malaria rapid tests in routine field settings and the implications for the test-based management of malaria. Out patients presenting for a malaria test requested by their doctor were serially enrolled into the study. Their folder/OPD numbers were noted. A finger prick blood sample for thick and thin film slides was collected in a micro-EDTA tube soon after a blood sample for a malaria rapid test had been collected by the heath staff performing the test from the same finger prick site. Malaria rapid test results, as read and interpreted by the person who did the test, were recorded and later compared to expert microscopy which was taken as the gold standard. A structured case record form was used to collect primary data on factors with the potential to affect the diagnostic performance of RDTs in those settings. Patients’ folders were later retrieved from the dispensary or records department. Diagnosis, prescription and medicines dispensed were extracted from the folders. Prevalence of parasitaemia was 5%, with P. falciparum as the only species transmitting and infecting people at the beginning of the high transmission season. Sensitivity, specificity, positive predictive value, negative predictive value, false positive rate and false negative rate were 69.6%, 92.4%, 32.7%, 98.3%, 7.6%, 30.4%, respectively. Parasitaemia of ≤100/μL was the single lone predictor of low sensitivity. Body temperature of ≥37.5˚C was found to have a rather bizarre association with low specificity which has no plausible biological explanation. Prescriber adherence to both positive and negative RDT results was 100%. RDTs have substantially improve the targeting of ACTs to malaria patients. However, their diagnostic performance is insufficient to be a standalone method upon which to wholly base clinical decision. They must be complemented with other better performing diagnostic technologies and other setting-specific malaria management guidelines to sustainably support the Test and Treat strategy.