Introduction: the World Health Organization (WHO) recommends testing of suspected malaria cases before treatment. Malaria rapid diagnostic test (mRDT) has been recommended for this purpose in endemic countries where microscopy is not accessible. However, its diagnostic performance remains a concern in clinical settings. We assessed diagnostic performance of RDT among febrile under-five children (U5) presenting at Oni Memorial Children's Hospital, Ibadan (OMCH).

Methods: a cross-sectional study was conducted among 370 febrile U5 attending OMCH February to May, 2016. We examined their finger prick blood samples for malaria parasitaemia using CareStartTM histidine rich protein II (HRP-2) RDT and microscopy. The sensitivity, specificity, positive and negative predictive values (PPV, NPV), false positive (FP), invalid rates (IR), likelihood ratio of positive and negative tests (LRP and LRN), were calculated.

Results: mean age of the children was 28.17 ± 15.59 months. Malaria prevalence was 21.6% and 15.1% by mRDT and microscopy, respectively. Sensitivity of CareStartTM HRP-2 RDT was 94.6% (95% confidence interval (CI): 84.2-98.6), specificity: 91.4% (CI: 87.6-94.2), PPV: 66.3% (CI: 54.7-76.2), NPV: 98.9% (CI: 96.8-99.7), FPR 6.5%, IR 8.1%, LRP:10.6 and LRN:0.1.

Conclusion: diagnostic performance of CareStart^TM used in the study met the ≥ 95% sensitivity at 100 parasites/µL recommended by WHO. This finding provides clinical evidence that testing before anti-malarial treatment as recommended by WHO will identify cases of malaria infection and reduce unnecessary use of drugs. Healthcare workers should be educated on diagnostic accuracy of mRDT and adhere to the WHO's test-treat strategy for anti-malaria therapy.