Error rates in SARS-CoV-2 testing examined with Bayes' theorem

Cost and performance analysis of efficiency, efficacy, and effectiveness of viral RNA isolation with commercial kits and Heat Shock as an alternative method to detect SARS-CoV-2 by RT-PCR

In late 2019 a new virus reported in Wuhan, China, identified as SARS-CoV-2 spread rapidly challenging the healthcare system around the world. The need for rapid, timely and accurate detection was critical to the prevention of community outbreaks of the virus. However, the high global demand for reagents during the years 2020 and 2021 generated a bottleneck in kits used for detection, greatly affecting developing countries, lagging their ability to diagnose and control the virus in the population. The difficulty in importing reagents, high costs and limited public access to the SARS-CoV-2 detection test led to the search for alternative methods. In this framework, different commercial nucleic acid extraction methodologies were evaluated and compared against heat shock as an alternative method for SARS-CoV-2 detection by RT-PCR, in order to determine the diagnostic yield and its possible low-cost compared to other methodologies. Nasopharyngeal samples were used where the diagnostic efficiency of the alternative method was 70 to 73%. The evaluation of the discriminatory efficacy of the method took the sensitivity and specificity to establish its cut-off point, being 0.73 to 0.817, which allows discriminating between COVID-19 positives and negatives. As for the diagnostic effectiveness expressed as the proportion of subjects correctly classified, it is between 80 and 84%. On the other hand, in terms of the costs necessary to carry out the detection, the alternative method is more economical and accessible in terms of direct cost close to 47 and 49 USD, and indirect cost around 35 and 50 USD compared to the commercial methods available in this comparison and evaluation, being possible its implementation in developing countries with high infection rates, allowing access to the diagnostic test with a reliable and low-cost method. Keywords: COVID-19, RT-PCR, Viral RNA.

On some fundamental challenges in monitoring epidemics

Epidemic models often reflect characteristic features of infectious spreading processes by coupled nonlinear differential equations considering different states of health (such as susceptible, infectious or recovered). This compartmental modelling approach, however, delivers an incomplete picture of the dynamics of epidemics, as it neglects stochastic and network effects, and the role of the measurement process, on which the estimation of epidemiological parameters and incidence values relies. In order to study the related issues, we combine established epidemiological spreading models with a measurement model of the testing process, considering the problems of false positives and false negatives as well as biased sampling. Studying a model-generated ground truth in conjunction with simulated observation processes (virtual measurements) allows one to gain insights into the fundamental limitations of purely data-driven methods when assessing the epidemic situation. We conclude that epidemic monitoring, simulation, and forecasting are wicked problems, as applying a conventional data-driven approach to a complex system with nonlinear dynamics, network effects and uncertainty can be misleading. Nevertheless, some of the errors can be corrected for, using scientific knowledge of the spreading dynamics and the measurement process. We conclude that such corrections should generally be part of epidemic monitoring, modelling and forecasting efforts. This article is part of the theme issue 'Data science approaches to infectious disease surveillance'.

Comparing the diagnostic accuracy of point-of-care lateral flow antigen testing for SARS-CoV-2 with RT-PCR in primary care (REAP-2)

BACKGROUND

Testing for COVID-19 with quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) may result in delayed detection of disease. Antigen detection via lateral flow testing (LFT) is faster and amenable to population-wide testing strategies. Our study assesses the diagnostic accuracy of LFT compared to RT-PCR on the same primarycare patients in Austria.

METHODS

Patients with mild to moderate flu-like symptoms attending a general practice network in an Austrian district (October 22 to November 30, 2020) received clinical assessment including LFT. All suspected COVID-19 cases obtained additional RT-PCR and were divided into two groups: Group 1 (true reactive): suspected cases with reactive LFT and positive RT-PCR; and Group 2 (false non-reactive): suspected cases with a non-reactive LFT but positive RT-PCR.

FINDINGS

Of the 2,562 symptomatic patients, 1,037 were suspected of COVID-19 and 826 (79.7%) patients tested RT-PCR positive. Among patients with positive RT-PCR, 788/826 tested LFT reactive (Group 1) and 38 (4.6%) non-reactive (Group 2). Overall sensitivity was 95.4% (95%CI: [94%,96.8%]), specificity 89.1% (95%CI: [86.3%, 91.9%]), positive predictive value 97.3% (95%CI:[95.9%, 98.7%]) and negative predictive value 82.5% (95%CI:[79.8%, 85.2%]). Reactive LFT and positive RT-PCR were positively correlated (r = 0.968,95CI=[0.952,0.985] and κ = 0.823 , 95%CI=[0.773,0.866]). Reactive LFT was negatively correlated with Ct-value ( r  = -0.2999, p  < 0.001) and pre-test symptom duration (r = -0.1299,p = 0.0043) while Ct-value was positively correlated with pre-test symptom duration (r = 0.3733),p < 0.001).

INTERPRETATION

We show that LFT is an accurate alternative to RT-PCR testing in primary care. We note the importance of administering LFT properly, here combined with clinical assessment in symptomatic patients.

FUNDING

Thomas Czypionka received funding from the European Union's Horizon 2020 Research and Innovation Programe under the grant agreement No 101016233 (PERISCOPE). No further funding was available for this study.