False-negative RT-PCR for COVID-19 and a diagnostic risk score: a retrospective cohort study among patients admitted to hospital

Abstract

Objective: To describe the characteristics and outcomes of patients with a clinical diagnosis of COVID-19 and false-negative SARS-CoV-2 reverse transcription-PCR (RT-PCR), and develop and internally validate a diagnostic risk score to predict risk of COVID-19 (including RT-PCR-negative COVID-19) among medical admissions.

Design: Retrospective cohort study.

Setting: Two hospitals within an acute NHS Trust in London, UK.

Participants: All patients admitted to medical wards between 2 March and 3 May 2020.

Outcomes: Main outcomes were diagnosis of COVID-19, SARS-CoV-2 RT-PCR results, sensitivity of SARS-CoV-2 RT-PCR and mortality during hospital admission. For the diagnostic risk score, we report discrimination, calibration and diagnostic accuracy of the model and simplified risk score and internal validation.

Results: 4008 patients were admitted between 2 March and 3 May 2020. 1792 patients (44.8%) were diagnosed with COVID-19, of whom 1391 were SARS-CoV-2 RT-PCR positive and 283 had only negative RT-PCRs. Compared with a clinical reference standard, sensitivity of RT-PCR in hospital patients was 83.1% (95% CI 81.2%-84.8%). Broadly, patients with false-negative RT-PCR COVID-19 and those confirmed by positive PCR had similar demographic and clinical characteristics but lower risk of intensive care unit admission and lower in-hospital mortality (adjusted OR 0.41, 95% CI 0.27-0.61). A simple diagnostic risk score comprising of age, sex, ethnicity, cough, fever or shortness of breath, National Early Warning Score 2, C reactive protein and chest radiograph appearance had moderate discrimination (area under the receiver-operator curve 0.83, 95% CI 0.82 to 0.85), good calibration and was internally validated.

Conclusion: RT-PCR-negative COVID-19 is common and is associated with lower mortality despite similar presentation. Diagnostic risk scores could potentially help triage patients requiring admission but need external validation.

Keywords: COVID-19; epidemiology; molecular diagnostics.

Figure 1

Patient flow diagram by final diagnosis and SARS-CoV-2 RT-PCR status with outcomes. Note ‘presumed COVID-19’ includes patients who were RT-PCR negative (n=293) and those who did not have a valid RT-PCR results (n=109). RT-PCR, reverse transcription-PCR.

Figure 2

(A) ROC for the full diagnostic predictive model. AUC 0.839 (95%CI 0.824 to 0.853), n=2940. (B) Calibration plot showing observed compared with predicted risk of COVID-19 diagnosis as deciles, with 95% CI. The dashed green line shows perfect calibration. (C) Decision curve analysis showing standardised net benefit at different threshold probabilities for diagnosing patients with COVID-19, comparing diagnosing all patients as COVID-19 (blue solid line), diagnosing no patients with COVID-19 (solid red line), and various diagnostic risk models, including the COVID-19 diagnostic score (full model and simplified risk score), CRP over 50, and NEWS of 5 or more. AUC, area under the curve; CRP, C reactive protein; NEWS, National Early Warning Score; ROC, receiver operating characteristic curve.

Figure 3

(A) Overlaid histogram of COVID-19 diagnostic risk score and number of patients with COVID-19 (white) and alternative (not COVID-19) diagnoses. (B) Proportion (%) of patients with COVID-19 (orange) or alternative (not COVID-19, blue) diagnoses by COVID-19 diagnostic risk score. N=2940.