Identifying COVID-19 cases in outpatient settings

Case identification is an ongoing issue for the COVID-19 epidemic, in particular for outpatient care where physicians must decide which patients to prioritise for further testing. This paper reports tools to classify patients based on symptom profiles based on 236 severe acute respiratory syndrome coronavirus 2 positive cases and 564 controls, accounting for the time course of illness using generalised multivariate logistic regression. Significant symptoms included abdominal pain, cough, diarrhoea, fever, headache, muscle ache, runny nose, sore throat, temperature between 37.5 and 37.9 °C and temperature above 38 °C, but their importance varied by day of illness at assessment. With a high percentile threshold for specificity at 0.95, the baseline model had reasonable sensitivity at 0.67. To further evaluate accuracy of model predictions, leave-one-out cross-validation confirmed high classification accuracy with an area under the receiver operating characteristic curve of 0.92. For the baseline model, sensitivity decreased to 0.56. External validation datasets reported similar result. Our study provides a tool to discern COVID-19 patients from controls using symptoms and day from illness onset with good predictive performance. It could be considered as a framework to complement laboratory testing in order to differentiate COVID-19 from other patients presenting with acute symptoms in outpatient care.

Comparison of the novel ARCHITECT procalcitonin assay with established procalcitonin assay systems

Aims

This study assessed the performance of a new fully automated immunoassay, ARCHITECT B.R.A.H.M.S procalcitonin (PCT), comparing the results with other commercial assays on routine clinical specimens.

Methods

At nine sites from eight countries, precision analysis was carried out on controls by ANOVA. Threshold and linearity were verified according to standard procedures. Comparison of ARCHITECT B.R.A.H.M.S PCT with the Cobas^®, LIAISON^®, VIDAS^® and Kryptor^® PCT assays was evaluated using Passing-Bablok and Deming regression analyses.

Results

The within-laboratory standard deviation and %CV across all sites ranged from 0.005 to 0.008 and 2.7 to 4.1; 0.040 to 0.212 and 2.1 to 11.7; 1.628 to 4.191 and 2.5–6.3 for the three control levels, respectively. The mean slope (linearity analysis) across all sites ranged from 0.85 to 1.03, with a mean y-intercept ranging from –6.15 to + 1.71 and a correlation coefficient ranging from 0.94 to 1.00. The LoB, LoD, and LoQ claims were verified. Deming regression analysis of 1116 plasma or serum samples with PCT results detected across a dynamic assay range of 0.02–100 μg/l using the ARCHITECT B.R.A.H.M.S PCT assay yielded results of r = 0.989 vs. Roche Cobas^®, r = 0.986 vs Kryptor^® B.R.A.H.M.S, r = 0.987 vs BioMèrieux VIDAS^® and r = 0.972 vs. Diasorin LIAISON^®, respectively. Concordance at cut-offs of 0.25 μg/l and 0.50 μg/l were 96.9% and 98.1% with Roche Cobas^®, 95.4% and 96.1% with B.R.A.H.M.S Kryptor^®, 93.8% and 98.4% with BioMèrieux VIDAS^®, and 92.7% and 93.9% with Diasorin LIAISON^®.

Conclusions

Compared with other assays, ARCHITECT B.R.A.H.M.S PCT offers excellent precision and low-end sensitivity.