Classification of “Near-patient” and “Point-of-Care” SARS-CoV-2 Nucleic Acid Amplification Test Systems and a first approach to evaluate their analytical independence of operator activities

Abstract

Background

European legislation defines as "near-patient testing" (NPT) what is popularly and in other legislations specified as "point-of-care testing” (POCT). Systems intended for NPT/POCT use must be characterized by independence from operator activities during the analytic procedure. However, tools for evaluating this are lacking. We hypothesized that the variability of measurement results obtained from identical samples with a larger number of identical devices by different operators, expressed as the method-specific reproducibility of measurement results reported in External Quality Assessment (EQA) schemes, is an indicator for this characteristic.

Materials and methods

Legal frameworks in the EU, the USA and Australia were evaluated about their requirements for NPT/POCT. EQA reproducibility of seven SARS-CoV-2-NAAT systems, all but one designated as "POCT", was calculated from variabilities in Ct values obtained from the respective device types in three different EQA schemes for virus genome detection.

Results

A matrix for characterizing test systems based on their technical complexity and the required operator competence was derived from requirements of the European In Vitro Diagnostic Regulation (IVDR) 2017/746. Good EQA reproducibility of the measurement results of the test systems investigated implies that different users in different locations have no recognizable influence on their measurement results.

Conclusion

The fundamental suitability of test systems for NPT/POCT use according to IVDR can be easily verified using the evaluation matrix presented. EQA reproducibility is a specific characteristic indicating independence from operator activities of NPT/POCT assays. EQA reproducibility of other systems than those investigated here remains to be determined.

1. Introduction

European legislation defines “near-patient testing (NPT)” as any device that is not intended for self-testing but is intended to perform testing outside a laboratory environment, generally near to, or at the side of, the patient by a health professional [1]. Another widely used and synonymous term is "point-of-care testing (POCT)", created by laboratory medicine to define analytical methods performed by non-laboratory professionals outside of a core laboratory setting [2]. The rapid delivery of test results for both symptomatic patients and asymptomatic screened individuals was essential during SARS-CoV-2 pandemic to allow for a fast isolation of pathogen carriers and thus reduce further spread of infection [3]. Laboratory analysis and diagnostics and the associated sample logistics were extremely challenged and therefore POCT was promoted with the intention of decreasing turnaround times and increasing testing availability [4]. Both succeeded, since the use of POCT systems offers the advantages of no transport times for the samples to the laboratory and the simpler handling of the analysis, so that it can be carried out also by non-laboratory professionals [5,6].

With the entry into force of the European In Vitro Diagnostic Regulation (IVDR) 2017/746 on 26 May 2022, Notified Bodies are responsible for assessing the suitability of NPT systems for which manufacturers apply for a CE-marking to attest compliance with this regulation. The IVD-Regulation is a product legislation setting up a framework mainly applicable to manufacturers that want to place medical devices on the EU market. As such fulfilment of most of the requirements are the responsibility of the manufacturer. This is the case for qualification of the product as a medical device, the subsequent classification that guides the manufacturer to the correct conformity assessment route, and thus the possible involvement of a notified body for all risk classes higher than class A IVDs. It is also the responsibility of the manufacturer to design a device that can fulfil the Annex I General Safety and Performance Requirements and to document this in the technical file. The task of the notified body is mainly to assess and validate the manufacturer's documentation (and their risk management and post market surveillance/vigilance systems) and based on this issue a certificate to the manufacturer for each device. This assessment and certification will not remove the legal responsibility from the manufacturer.

Assays that perform analyses truly autonomous and independent from operator activities should provide results from same samples with only little variance, no matter under whatever conditions and by whom these assays are operated, as long as this is within the manufacturer's specifications. Before the robustness against external influences is examined and the devices are subjected to any relevant stress tests, their inter-laboratory variation must first be known. We revived the idea that the true uncertainty of a quantitative measurement method is reflected by EQA-based variability and assumed that the robustness of operator-independent test systems is expressed objectively by a good reproducibility [7,8]. The EQA reproducibility represents the variability of measurement results over a certain measurement range from a larger number of identical (same type) devices used by different operators at different times at different locations. This characteristic of a device can be an indicator for its independence of operator activities. To examine the usefulness of this indicator, we retrospectively evaluated and compared results obtained from different unit-use cartridge nucleic acid amplification test (NAAT) systems and a laboratory test system for SARS-CoV-2 virus genome detection in external quality assessment (EQA) schemes. In the case of detection of SARS-CoV-2 RNA, those assays indicate the amplification cycle threshold (Ct; syn. Cp for crossing point, Cq for quantification cycle) value at which a positive signal was first detected. Ct values are imperfect theoretical intermediate values in the quantification of DNA molecules using real-time qPCR, which were often misinterpreted as quantitative results in SARS-CoV-2 RNA detection. In order to express the result as viral load or concentration in SI units, they must be converted to copies/mL or international units/mL using instrument-specific factors [9], [10], [11]. However, since this intermediate data provides the basis for a quantitative result, we consider the Ct value to be sufficiently suitable for an initial investigation of the usefulness of EQA reproducibility for evaluating SARS-CoV-2 NAAT systems intended for NPT/POCT use. EQA schemes involve multiple devices of the same type that almost simultaneously analyse identical samples. Since proven homogeneity and stability of samples are prerequisites for their use in such schemes, the basis for all involved laboratories to achieve the same results is given. In addition to their great importance for the harmonization of quantitative results, EQA schemes might therefore also be ideally suited for determining reproducibility [12].

2. Materials and methods

For the legal and terminological part of the study, the IVDR's requirements for test systems intended to be operated by non-laboratory professionals and outside a laboratory environment were evaluated and compared with US and Australian regulations on this subject. From this, criteria for the classification of IVD medical devices according to their properties were derived.

For the technical part of the study, variabilities of Ct values obtained by identical types of test systems for SARS-CoV-2-positve samples in completed rounds of SARS-CoV-2 virus genome detection External Quality Assessment (EQA) schemes were retrospectively evaluated. These schemes were operated by the Austrian Association for Quality Assurance and Standardization of Medical and Diagnostic Tests (ÖQUASTA), Vienna, Austria; INSTAND e.V. Society for Promoting Quality Assurance in Medical Laboratories, Düsseldorf, Germany; and the Royal College of Pathologists of Australasia Quality Assurance Programs (RCPAQAP), St. Leonards, NSW, Australia.

Each EQA organization used their own samples in the respective schemes. If test systems were included in more than one EQA providers’ scheme, the measurement results were considered together. Data were prepared for every data series apart. A data series consists of the data reported for the same sample by laboratories using the same device. Since test systems employ a certain number of specific target genes, the lowest Ct value obtained for all quantified targets by the same laboratory for the same sample was used for evaluation. Data series were first subject to the dip test for bimodality [13]. To accommodate for outliers, only the inner 90% were chosen. To adjust repetitively reported values, a small random value (jitter) was added to the data [14]. If no significant bimodality was found on a significance level of 0.05, the assigned value (mean) and EQA standard deviation (SD) of the data series was calculated according to Huber's H1.5 algorithm, also known as Algorithm A from ISO 13,528 [15,16]. The assigned values and EQA SDs that were obtained with the same methodology were then combined for all samples of which the assigned value had a Ct value ≤36. The relation between the assigned values and EQA SDs was evaluated using a linear regression model. In a first instance, a robust least trimmed squares regression line was calculated, outliers against this regression line were identified and eliminated before calculating a least-squares linear regression line between the assigned values and EQA SDs in order to obtain a mean EQA SD associated with any concentration.

For comparing methods, only those for which measurement results of at least 10 samples were available were taken into account. The difference between the regression lines obtained for each method and the prediction interval around this difference were calculated. As such, the difference between the methods' reproducibilities could be assessed for every investigated concentration. Significance was evaluated at a level of 0.05.

3. Results

The interpretation of the European, American and Australian legislation on requirements regarding test systems for NPT/POCT use is presented in Table 1 . All three legislations describe a class of test systems intended for use outside a laboratory environment. In contrast to the EU-Regulation, the American and the Australian legislation have no restriction to the professional qualification of the intended user, they have to be “well-trained” only. The IVDR requires “health professionals” to perform NPT/POCT, but this job profile and its competencies depend on national legislation. The regulations agree that assays intended for use by non-laboratory staff should be characterized by high robustness to operator and environmental influences. Bases for classifying test systems are established in US and Australian legislation, while the IVDR uses different qualifications of healthcare professionals to perform NPT/POCT.

Table 1.

European, American and Australian legislation on test systems for “near-patient” or “point-of-care” use.

European Union Regulation (EU) 2017/746 of the European Parliament and of the Council of 5 April 2017 on in vitro diagnostic medical devices (IVDR) requires from assays intended for "near-patient" use that (i) the level of training, qualifications and/or experience required by the user is made clear; (ii) the device can be used safely and accurately by the intended user at all stages of the procedure if necessary after appropriate training and/or information; (iii) the risk of error by the intended user in the handling of the device, and, if applicable, the specimen, and also in the interpretation of the results, is reduced as far as possible; (iv) the information supplied in with devices for near-patient testing is provided in paper format and easily understandable. Further IVDR requirements for assays intended for “near patient” use but not directly related to the competences of intended users apply as well (e.g., Section 20.2, Section 20 of Annex I, Section 5.1 of Annex IX).

United States of America US Code of Regulations requires tests to be categorized for complexity (moderate or high) utilizing seven criteria and assigning scores of 1 (low), 2 (medium), or 3 (high) within each criterion [17]. These criteria include demands or efforts for (i) knowledge, (ii) training and experience, (iii) reagents and materials preparation, (iv) characteristics of operational steps, (v) calibration, quality control, and proficiency testing materials, (vi) test system troubleshooting and equipment maintenance, and (vii) interpretation and judgement. Systems with total scores of ≤ 12 are categorized as moderate complexity and those with scores > 12 are categorized as high complexity. Test systems can be classified as “waived complexity” (minimal requirements for testing) if they are (i) cleared by the US Food and Drug Administration (FDA) for home use, (ii) employ methodologies that are so simple and accurate as to render the likelihood of erroneous results negligible; or (iii) pose no reasonable risk of harm to the patient if the test is performed incorrectly. Waived POCT requires no performance validation prior to routine patient testing. There is no formal operator training and competency required. Operators must only follow manufacturer instructions for use. The Clinical and Laboratory Standards Institute (CLSI) published a guideline document which states that operators of (non-waived) POCT devices should be well-trained and pass an appropriate examination [18]. The competence of staff to operate the claimed test system in an organization is compared to the competences required according to the complexity scoring and, if appropriate, permission for an individual to use certain test systems is granted by the laboratory director.

Australia The Therapeutic Goods Administration (TGA) is entrusted with the accreditation of IVD devices into Australia including POCT tests (e.g., for SARS-CoV-2) [19,20]. The TGA uses a risk-based approach and hence the amount of scrutiny before listing on the Australian Register of Therapeutic Goods (ARTG) depends on its risk classification [21]. The following essential principles are verified before a test system can be approved for use and be placed on the ARTG: (i) safety requirements, (ii) chemical, physical and biological properties of the device, (iii) protection from infection and microbial contamination, (iv) appropriate construction and environmental properties, and (v) information to be supplied with the device. The Regulations ensure that these devices may be used outside the laboratory setting by a health practitioner, or trained staff under their supervision, to provide laboratory testing. However, before a laboratory test can be used in Australia and billed to the Australian government, the testing site (GP or laboratory) must be accredited to perform that testing. This accreditation is provided by the National Association of Testing Authorities (NATA) using the standards produced by the National Pathology Accreditation Advisory Council (NPAAC) [22,23]. These standards mandate use of ISO 15,189 for Pathology laboratories as well as a suite of other Technical and Supervision documents. There is a Standard that covers POCT in both GP and Pathology Laboratory settings [24].

Technical complexity and the required competencies of the intended operators can be presented in a scoring matrix. Therefore, SARS-CoV-2 test systems are assigned to one of the classes "low", "medium" or "high" for both their technical complexity and the competences required of operators for safe and error-free operation. This results in a nine-field matrix with one field for each combination of level of technical complexity and the required skills of the operators (Table 2 ). SARS-CoV-2-asscociated tests in the group of low complexity tests are rapid antigen and antibody tests, in the medium complexity group are automated antigen, antibody and NAAT tests, and in the high complexity group are non-automated nucleic amplification and sequencing assays, and virus neutralization tests. Regarding operator competence, only rapid antigen and antibody tests, unit-use cartridge NAAT and antigen, and “Lab on a chip” (LOC) assays are assigned to the group of assays that require only “low” operator competences; those assays are “waived” tests according to US legislation [25]. No SARS-CoV-2 assays could be assigned to the group of “medium” operator competence required to operate test systems. In the group of assays that require “high” operator competences, there are automated and manual antibody, antigen, nucleic acid amplification and sequencing methods. In addition to technical qualifications, methodical training and laboratory experience are required to operate these test systems. Only test systems that require low or medium operator skills, which are specified in detail by the manufacturer, are intended for NPT/POCT use as defined in US and Australian legislation. However, IVDR only refers to competences of professional profiles. Since these are not uniform in the EU, the manufacturer must define the required user skills.

Table 2.

Classification of SARS-CoV-2-associated test systems according to their technical complexity and the required skills of operators.

		Technical complexity
		sample is visually examined by the operator or result is unequivocally visible on test system	test system that automatically carries out reactions, reads the resulting signals and provides results of the analysis	multiple-step non-automated analytic procedures
		low	medium	high
Required operator competence	high	Qualitative results (to visually interpret or detect aspects or properties of a sample; if applicable, after staining or addition of a reagent)	to operate automated test systems including verification, maintenance, quality control, troubleshooting, and calibration procedures, plausibility check of results	same as left, but for non-automated analysis systems; establishing results from raw data or signals, if necessary, using calculations or algorithms; setting cut-offs; handling of large data volumes; if applicable, validation procedures
		(none)	• Automated nucleic acid amplification tests (NAAT) • Automated antibody determination assays • Automated antigen determination assays	• Manual NAAT on thermal cyclers • Virus genome sequencing • Manual antibody determination (microplates) • Virus neutralization tests • Surrogate neutralization tests
		blood smear, stool parasites, identification of irregular blood group antibodies, cytology	common automated clinical chemistry methods, haematology and coagulation analyses	flow cytometry, mass spectrometry
	medium	Semiquantitative results with more than two possible values (to distinguish between colour gradients; if applicable, including semi-quantitative estimation of the approximate concentration of the measurand)	to perform sample preparation procedures; simple maintenance procedures on the device; interpret graphically displayed results; plausibility check of results
		(none)	(none)	(none)
		Urostick, faecal occult blood test, bedside-test	Blood gas analysis, thrombelastography	(Inherently impossible)
	low	Semiquantitative results with two possible values (to distinguish between one or more lines or “+” or “-“ on the reaction field, or interpretation if provided, e.g., “positive”, “negative”, “not detected”, ”high”, “low”, or “failure”) or quantitative results (to read numerical values together with the associated unit from display)
		no manipulation during analysis required	test system requires actions/manipulation by the operator
		• Rapid antigen tests • Rapid antibody tests	• NAAT unit-use cartridge or “lab on a chip (LOC)” assays • Unit-use cartridge antigen assay	(none)
		Lateral-flow based rapid pregnancy tests or tests for various infectious diseases	Non-modifiable test systems for blood glucose, coagulation monitoring, urine analysis, cardiac markers	(Inherently impossible)

Legend: SARS-associated tests in bold larger font and examples of other tests in smaller font and italics.

The investigated seven test systems, number of samples and ranges of Ct values obtained for them, and their mean reproducibility shown in Table 3 .

Table 3.

Investigated SARS-CoV-2 NAAT systems ranked according to their reproducibility.

Method	No. of samples	Ct value		Target gene(s)	Mean reproducibility at Ct (95% CI)
		min	max		21	27	33
“Near-patient use” devices
Cepheid GeneXpert / Xpert Xpress SARS-CoV-2/Flu/RSV plus (CGXplus)	18	23.8	33.5	RdRP, N2, E	0.58 [0.23;0.93]	0.34 [0.22;0.47]	0.11 [0;0.45]
Cepheid GeneXpert / Xpert Xpress SARS-CoV-2/Flu/RSV (CGXflu)	60	20.1	35.5	N2, E	0.2 [0.01;0.4]	0.71 [0.62;0.81]	1.22 [1.07;1.38]
Roche Cobas Liat / LIAT SARS-CoV-2 (Liat)	12	21.1	28.8	orf1a/b, N	0.28 [0.09;0.46]	0.58 [0.45;0.7]	—
Roche Cobas Liat / LIAT SARS-CoV-2 & Influenza A/B (LiatA/B)	20	20.5	35.6	orf1a/b, N	0.71 [0.44;0.98]	0.93 [0.74;1.13]	1.16 [0.81;1.51]
DiaSorin Liaison MDX / Simplexa COVID-19 Direct Reaction Mix (Liaison)	47	18.0	34.6	orf1a/b, S	0.68 [0.61;0.74]	0.78 [0.72;0.84]	0.88 [0.75;1.01]
Cepheid GeneXpert / Xpert Xpress SARS-CoV-2 (CGX)	87	20.5	34.3	N2, E	0.81 [0.62;1]	0.88 [0.8;0.97]	0.96 [0.81;1.11]
Automated laboratory device (non-POCT)
Roche Cobas 6800 / COBAS SARS-CoV-2 Test (RC6800)	89	21.7	35.9	orf1, E	—	1.05 [0.99;1.1]	0.98 [0.9;1.05]

Legend: Cepheid GeneXpert, DiaSorin Liaison MDX and Roche Cobas Liat are designated as for “NPT/POCT” use; Roche Cobas 6800 is a laboratory device to be operated by laboratory professionals. Abbreviations used in text shown in italics. Cepheid GeneXpert / Xpert Xpress SARS-CoV-2/Flu/RSV plus (CGXplus) assay has overall the best EQA reproducibility, except for the lowest Ct values. The Roche Cobas 6800 / COBAS SARS-CoV-2 Test (RC6800) has, except for the highest Ct values, the highest variability. Reproducibilities are only given for Ct values within the range of observed assigned values of the EQA samples. Lower reproducibility, expressed as SD, means better performance.

All test systems have reproducibilities within or just above an intra-assay variability of a magnitude of 1 Ct [27]. The devices designated as POCT (Cepheid GeneXpert, DiaSorin Liaison MDX, Roche Cobas Liat) had better reproducibility with all reagents used (Xpert Xpress SARS-CoV-2/Flu/RSV plus, Xpert Xpress SARS-CoV-2/Flu/RSV, Xpert Xpress SARS-CoV-2, Simplexa COVID-19 Direct Reaction Mix, LIAT SARS-CoV-2, LIAT SARS-CoV-2 & Influenza A/B) than the only as non-POCT designated assay evaluated, the Roche Cobas 6800 with the COBAS SARS-CoV-2 Test. Differences in reproducibility of EQA measurement values between selected combinations of assays are shown graphically in Fig. 1 .

Fig. 1.

Comparison of the difference in EQA reproducibility between each two assays.

Legend: Negative differences indicate a lower EQA reproducibility for the first of the two compared assays. This Figure illustrates the differences in reproducibility of EQA measurement values between selected combinations of assays. As pointed out before, we analyse the data of only those NPT/POCT for which at least 10 measurement results were reported. These NPT/POCT results are mirrored against Roche Cobas 6800 (RC6800), the only automated test and non-POCT system for which enough measurement results were available. Comparisons of further assays can be found in the supplementary data. Abbreviations used are explained in Table 3. The difference between CGXplus and RC6800 is illustrated by Fig. 1a. Over the whole measurement range there is a difference in reproducibility that ranges from 0.5 for the lowest Ct values (18 to 20) to 1 for a Ct value of 36. Differences between the RC6800 assay and other assays are of lesser extent. Compared with the Liat assay for example, the reproducibility of the Liat is inferior, the difference ranges from 1 for low Ct values to near 0 for Ct values of 36 (Fig. 1b). Compared with Liaison and CGX, the reproducibility of the RC6800 assay is 0.5 Ct better for the lowest Ct values and equal for Ct values around 36 (Figs. 1c and 1d). The best overall reproducibility was observed for CGXplus, in particular for the higher Ct values (Figs. 1a, 1e and 1f). The difference between this assay and Liaison is illustrated in Fig. 1e. For Ct values higher than 22, CGXplus has a significantly better reproducibility than Liaison. The fact that the CGXplus has the best overall reproducibility except for the lowest Ct values is illustrated in Fig. 1f. For Ct values below 24, Liat has a significantly better reproducibility than CGXplus, but this situation inverts from Ct values above 25, leading to CGXplus having a significantly better reproducibility than Liat.

4. Discussion

Standard laboratory devices require qualified personnel for routine operations, like supervision of analytics, quality control, calibration, maintenance and, if necessary, their intervention in the analysis process. NPT/POCT systems, however, are intended to be used by personnel with or without such laboratory qualifications, working outside a laboratory environment, and therefore should work independently of operator activities. The proposed classification of SARS-CoV-2 test systems as presented in Table 2 can be helpful in assessing the fundamental suitability of test systems for use by non-laboratory professionals since it combines vividly the operator competence with the technical device complexity. Their reliability can be followed up by means of their EQA reproducibility. When evaluating the suitability of test systems intended for NPT/POCT use, Notified Bodies will verify that (i) the selection of individuals involved in the performance assessments of the device in question was reasonable and balanced, (ii) sampling procedures and, if necessary, sample preparation or storage are clearly described for the intended users so that they can carry out the test properly, and (iii) the results are clearly identifiable for the intended operators, regardless of whether they are qualitative results of classifications (e.g., viral or bacteriological species), semi-quantitative (gradings like +/++/+++/++++, “weakly positive”, “positive” or “negative”, “detectable” or “not detectable”) or quantitative (values with a measurement unit). NPT/POCT systems must reliably detect any error or malfunction in the measurement process and report this clearly instead of a result. Regarding sample preparation, it should be noted that the need to centrifuge sample material does not preclude NPT/POCT use, but pooling by volumetric pipetting of several samples does, unless this is done using a method specified by the manufacturer and certified by the Notified Body. Furthermore, considering the analysis frequency and the risk of patient harm due to incorrect determination can also be helpful when evaluating the manufacturer's information on the frequency of internal quality control (IQC) [26]. Finally, it is of the utmost importance to use both apparatus-based and rapid antigen / antibody tests exactly according to the manufacturer's instructions, and to make both the user and the tested individual aware that the non-detection of the pathogen does not mean its definite absence.

In this study, we compare seven SARS-CoV-2 NAAT assays from which a sufficiently large number of measurement results were available for us to calculate EQA reproducibility. The differences in reproducibilities observed with different reagents on the same device are presumably related to the various composition of the reagents or their different lots. The low inter-laboratory variation implies that different users at different locations have little or no influence on the measurement results of the NPT/POCT assays evaluated here. These results form a good basis for evaluating the robustness of such devices against environmental and operator influences, including any associated stress tests.

A limitation of the usability of the reproducibility derived from EQA results is that assays only appear in such schemes once they are on the market. The reproducibility data derived from EQA results is therefore not available for their initial certification. However, the important role of the reproducibility derived from EQA results for their post-market surveillance in routine use is undisputed. Further limitations are that a certain minimum number of individual devices must be included in EQA schemes and that results from several samples must be compiled in order to derive a reliable value for the reproducibility. For this reason, only seven NPT/POCT systems could be examined here, while a further seven NPT/POCT systems and numerous other automated and manual laboratory test systems were represented in the EQA schemes, but in too low numbers.

5. Conclusion

The EQA reproducibility is a characteristic of NPT/POCT assays providing information on the independence of such test systems of operator activities. Considering the results of this retrospective evaluation, it appears to be a reproducible value that can be easily derived from EQA results and proof is given that methods do have different EQA reproducibilities. To what extent the results for the NPT/POCT systems and the single investigated laboratory test investigated in this study are representative for the variety of NPT/POCT and laboratory tests on the market remains to be investigated in further studies.

CRediT authorship contribution statement

Christoph Buchta: Conceptualization, Investigation, Methodology, Writing – original draft. Heinz Zeichhardt: Resources, Supervision, Writing – review & editing. Tony Badrick: Investigation, Formal analysis, Resources, Writing – review & editing. Wim Coucke: Conceptualization, Data curation, Formal analysis, Visualization, Writing – original draft. Nathalie Wojtalewicz: Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. Andrea Griesmacher: Resources, Supervision, Writing – review & editing. Stephan W. Aberle: Supervision, Writing – review & editing. Ingo Schellenberg: Resources, Supervision, Writing – review & editing. Ellis Jacobs: Investigation, Formal analysis, Writing – review & editing. Gunnar Nordin: Investigation, Methodology, Supervision, Writing – review & editing. Christian Schweiger: Investigation, Writing – review & editing. Karin Schwenoha: Formal analysis, Writing – review & editing. Peter B. Luppa: Investigation, Methodology, Supervision, Writing – review & editing. Ulrich M. Gassner: Formal analysis, Supervision, Writing – review & editing. Thomas Wagner: Investigation, Formal analysis, Writing – review & editing. Martin Kammel: Conceptualization, Investigation, Supervision, Writing – review & editing.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Peter B. Luppa is member of the board of directors of INSTAND e.V., a non-profit organization, named as one of two EQA providers for Germany by the Deutsche Ärztekammer. Heinz Zeichhardt declares that he is co-chairman of the Joint Diagnostic Council of the Deutsche Vereinigung zur Bekaempfung der Viruskrankheiten e.V. (DVV e.V.) and Gesellschaft fuer Virologie (GfV e.V.) and is Advisor for the INSTAND External Quality Assessment (EQA) schemes in virus diagnostics. He is owner and managing director of IQVD GmbH - Institut fuer Qualitaetssicherung in der Virusdiagnostik, Berlin, and was majority owner and managing director of GBD Gesellschaft fuer Biotechnologische Diagnostik mbH, Berlin. He declares that he has no conflicts of interest with regard to the activities mentioned in relation to the publication. The other authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.