Abstract
INTRODUCTION: RT-PCR has suboptimal sensitivity for the diagnosis of COVID-19. A composite reference standard comprising RT-PCR plus radiological and clinical features has been recommended for diagnostic accuracy studies. The FebriDx finger prick point-of-care test detects an antiviral host response protein (MxA) in 10 min. We evaluated the diagnostic accuracy of FebriDx and RT-PCR compared to a composite reference standard. METHODS: Adults presenting to hospital with suspected COVID-19 were tested by FebriDx and RT-PCR. A composite reference standard was used to classify patients as having COVID-19 based on RT-PCR positivity, or RT-PCR negativity with COVID-19 radiological findings or other clinical criteria. Measures of accuracy were calculated for MxA alone, RT-PCR alone, and both combined. This study is registered with the ISRCTN (ISRCTN14966673) and has completed. RESULTS: A total of 478 patients were tested, with valid results in 475. Of these 475 patients, 222 (46.7%) were classified as having COVID-19; 192 (40.4%) were RT-PCR positive, and 30 (6.3%) were RT-PCR negative and diagnosed on radiological/clinical criteria. Sensitivity of FebriDx MxA vs the composite reference standard was 186/222 (83.8%, 95% CI 78.3-88.4) and was similar to the sensitivity of RT-PCR (192/222 (86.5%, 95% CI 81.3-90.7), (difference of 2.7%, 95% CI - 3.9 to 9.3, p = 0.42). The sensitivity of combined FebriDx and RT-PCR was 208/222 (93.7%) which was superior to both RT-PCR alone (difference of 9.9, 95% CI 4.1-15.9; p = 0.001) and FebriDx MxA alone (difference of 7.2, 95% CI 1.6-12.9; p = 0.011). CONCLUSION: Sensitivity of combined FebriDx and RT-PCR testing was superior to each alone for the detection of COVID-19 in hospital and may improve infection control and treatment decisions.