SARS-CoV-2 Rapid Antigen Tests for Health Care Workers

Assessment of benefit by means of a questionnaire

One instrument to contain the SARS-CoV-2 pandemic, which has been in use since November 2021, is the requirement for health care workers to perform regular testing using a rapid antigen test (RAT). This measure is intended to detect infections in health care workers who are already infectious in the pre- or asymptomatic state in order to prevent further transmission of the virus to patients and other health care workers. The aim of this study was to determine how many SARS-CoV-2 infections were detected through symptom-free rapid antigen testing. Since a key reason for symptom-free testing is the potential for clinically non-apparent SARS-CoV-2 infection, an additional aim was to establish the actual number of cases with such an infection.

Methods

From March 2020 to June 2022, health care workers at the Medical Center – University of Freiburg with suspected SARS-CoV-2 infection were offered easily accessible polymerase chain reaction (PCR) testing. In addition, positive RAT results had to be confirmed by PCR testing. As part of our study (file number 22–1163, Ethics Committee of the Albert Ludwigs University of Freiburg), a questionnaire was sent to all members of the staff with a positive PCR result during the period September 2021 to April 2022. The questionnaire included: the reason for the requested PCR test (symptoms, contact with an infected person, positive RAT result, other); the information whether a RAT had been performed and its result; presence and nature of symptoms at baseline and subsequent follow-ups; age and sex. The response options could be ticked in combination and free text could be added to some of the responses. During the survey period, 815 443 rapid antigen test kits (unit price of 1.60 Euro) were handed out by the employer to members of the staff.

Results

A total of 2072 questionnaires were sent out (to women: 64.9%; to men: 35.1%). Of these, 750 were returned and 738 of the returned questionnaires could be included in the analysis (from women: 67.5%, from men: 32.5%; response rate of 35.6%). Of the 738 members of the staff, 716 underwent rapid antigen testing on the day of PCR testing (Figure 1). In 585 of the 716 cases (81.7%), the RAT was positive; of these, 550 persons (94%) experienced symptoms on the day of the positive test. A total of 35 persons with a positive RAT had no symptoms. In 24 persons of this group, the positive RAT was the only reason for PCR testing; 11 persons additionally had contact to a person who tested positive. Nine of the 35 asymptomatic persons with positive RAT also did not develop symptoms later on; of these, two had contact to an infected person. This leaves seven infected staff members who were only detected because of rapid antigen testing (1%). At the onset of symptoms, 173 (24.2%) persons still had a negative RAT result. Of these, 67 had only initiated PCR testing on the day the RAT returned a positive result. In most cases, the RAT became positive 1–4 days after symptom onset. 15 persons stated that they never tested positive on RAT.

Figure 1

Flowchart for the 716 persons who had a rapid antigen test (RAT) performed on the day of PCR testing. * In 67 persons, PCR testing was only performed when rapid antigen testing became positive, despite the fact that they had already experienced symptoms days before. PCR, polymerase chain reaction

In 40 (5.6%) persons with symptoms, multiple rapid antigen tests were performed with varying results. Of these, 21 stated that the RAT was negative with a nasal swab, but positive with a throat swap. Of all 738 respondents, twelve stated that even over time they did not develop any symptoms (1.6%).

Discussion

Symptom-free rapid antigen testing is intended to identify health care workers with asymptomatic infection in order to prevent transmission of the virus within health care facilities. However, it has been found early on that rapid antigen tests only become reliably positive in the symptomatic phase of the disease and at Ct values ≤ 25 (1, 2). A study conducted by the German Society of Hospital Hygiene (DGKH, Deutsche Gesellschaft für Krankenhaushygiene) found that the use of RAT in care homes did not result in a reduction of outbreaks (3). Many staff members reported having continued to work despite symptoms because the RAT was negative. Using this approach, there is an increased risk for possible transmission of SARS-CoV-2 to patients or co-workers. The number of cases with nosocomial SARS-CoV-2 infection increased during the survey period.

Of the 716 infections confirmed by PCR testing, seven were only detected because of a positive RAT. Assuming that the persons who returned a questionnaire constituted a representative section of the respondents, an extrapolated number of 20 (7×2072/738 ≈ 19.7) infections results. In addition, 17 persons were identified early on by rapid antigen testing, extrapolating to approximately 47 persons. This compares to 173 persons who were not identified by rapid antigen testing despite symptoms, extrapolating to 486 persons. It is possible that health care workers with symptoms and negative RAT did not seek further testing. Therefore, our results are in support of preexisting data showing that rapid antigen testing without subsequent PCR confirmation is to be approached critically (4) and RAT strategies are problematic in general (5).

The Paul Ehrlich Institute (PEI) specified the sensitivity of the rapid antigen tests used as 100% with Ct values ≤ 25. In our cohort, some of the Ct values were considerably lower (Figure 2). This may be explained by the fact that rapid antigen testing is performed using nasal swabs, while PCR testing is performed using throat swabs. Another possible explanation lies in the replication dynamics of the Omicron variant.

Figure 2

Ct values in relation to a positive or negative RAT on the day of PCR testing. RAT, rapid antigen test; PCR, polymerase chain reaction

Conclusion

Mandatory testing of health care workers is a resource-intensive measure with little benefit and clear risks.

Erik Huzly, Daniel Steinmann, Stefanie Kramme, Marcus Panning, Daniela Huzly

Conflict of interest statement

Daniela Huzly gives scientific lectures for Abbott.

The remaining authors declare no conflict of interest.

Manuscript received on 9 November 2022, revised version accepted on 29 December 2022.

Translated from the original German by Ralf Thoene, MD

Cite this as:
Huzly E, Steinmann D, Kramme S, Panning M, Huzly D: SARS-CoV-2 rapid antigen tests for health care workers—assessment of benefit by means of a questionnaire. Dtsch Arztebl Int 2023; 120: 170–1. DOI: 10.3238/arztebl.m2022.0412

References

1. Bornemann L, Kaup O, Kleideiter J, Panning M, Ruprecht B, Wehmeier M: Real-life evaluation of the Sofia SARS-CoV-2 antigen assay in a large tertiary care hospital. J Clin Virol 2021; 140: 104854

2. Dinnes J, Sharma P, Berhane S, et al.: Rapid, point-of-care antigen tests for diagnosis of SARS-CoV-2 infection. Cochrane Database Syst Rev 2022; 7: CD013705

3. Gleich S, Walger P, Popp W, Lemm F, Exner M: Nosokomiale COVID-19-Ausbrüche in vollstationären Pflegeeinrichtungen Ursachen und Forderungen. Hygiene und Medizin 2021; 46: 1–8.

4. Freund T, Friesen J, Stolle JM, et al.: Rapid tests detecting SARS-CoV-2-specific antigens: variable value in different periods of the pandemic in Berlin, Germany. Dtsch Arztebl Int 2022; 119: 647–8

5. Hirsch O, Bergholz W, Kisielinski K, Giboni P, Sönnichsen A: Methodological problems of SARS-CoV-2 rapid point-of-care tests when used in mass testing. AIMS Public Health 2022; 9: 73–93