Jac Dinnes and Clare Davenport look at the rollout of the mass testing regime in English schools
The new, voluntary, but “strongly encouraged,” asymptomatic mass testing regime using lateral flow devices (LFDs) in schools in England began last week as children returned to school en masse. We are co-authors of a systematic review of the latest evidence on the accuracy of LFDs and parents of teenagers. We have a particular interest in how the potential benefits and harms from LFD testing are being communicated to students, parents, and teachers. 
To all intents and purposes, asymptomatic mass testing represents the introduction of a screening programme for which principles of informed consent should be followed.  A personal informed choice in regard to testing requires “access to accessible, accurate, evidence-based information covering: the condition being screened for, the testing process, the risks, limitations, benefits and uncertainties, and the potential outcomes and ensuing decisions.”  In contrast to existing screening programmes where testing primarily benefits the individual, the objective of mass testing of asymptomatic individuals in a pandemic shifts the emphasis from individual benefit to one of population benefit. Has this shift away from individual benefit created a disincentive for the provision of clear information about the expected benefits and potential harms from mass testing?
There is a dearth of data on the accuracy of the Innova LFD when used in people with no symptoms of covid-19 and no data on its use for serial testing or in children. [4,5] Innova and similar LFDs that detect viral antigen have been validated by test manufacturers for use in people with symptoms. The development of symptoms is an indicator that viral load is high and therefore virus can be more easily detected on a swab sample. Although people with asymptomatic covid-19 infections may have similar viral load patterns to those with symptoms, they will not be picked up by even the most sensitive antigen detection test unless they are tested within the relatively short window in which they exhibit high enough viral loads. 
Assuming that the Innova test performs at least as well as it did in the Liverpool evaluation of mass testing (sensitivity 40.0% and specificity 99.9% ), at a population prevalence of infection of 0.47% (REACT study 13 to 22 Feb ), one in three positive LFD results will be false positives (positive predictive value 65%) and three out of every five true cases of covid-19 infection will be missed. These figures equate to 288 positive LFD tests per 100,000 people tested with 188 infections correctly detected, 100 people with false positive results and 282 missed cases of infection. Using the Public Health England estimate of Innova specificity of 99.7% reduces positive predictive value to 39% and increases the number with false positive results per 100,000 tested from 100 to 299.  The relative benefits of testing are vulnerable to even small changes in specificity at such low prevalence of infection; similarly, as prevalence falls, positive predictive value falls, increasing the ratio of false positive to true positive test results.  With LFD positivity rates in secondary schools of just 0.05% in the week ending 3 Mar (328 positive out of 663,332 tests), schools and parents may take some convincing regarding potential population gains relative to harms from mass testing. 
In regard to cases missed, there is a commonly circulating argument that any test is better than no test if it detects at least some cases of infection that would otherwise have gone undetected, and furthermore that serial testing with LFDs compensates for their low sensitivity.  However, there are important costs to be quantified, not least the possibility for increased transmission from changing behaviour patterns in those with false negative results.  A recent study of serial daily LFD testing resulted in two outbreaks of covid-19 in asymptomatic intercollegiate athletes, and transmission of infection was not interrupted until serial PCR testing was implemented.  What impact twice weekly Innova testing will have on transmission of infection within schools remains to be seen.
Ten day self-isolation following a positive test result not only affects those testing positive, but extends to their school and household contacts. At the current low prevalence of infection, unnecessary disruption to schooling and any financial burden for families arising from false positive test results become significant. The impact of false positives in population screening programmes are mitigated by the use of a second confirmatory test. For mass testing in schools however, unless a Government U-turn is on the cards, a confirmatory PCR will only be used for tests conducted in the home, a negative test over-ruling the previously positive LFD, and removing the need to self-isolate. Positive LFD results for tests conducted in a school setting will be upheld regardless of any subsequently negative PCR.  Although it is not clear how long on-site school testing will continue beyond testing of those unable or unwilling to use the tests at home, any false positive LFD result has significant implications for students, teachers, and their families. Confirmatory PCR in all those with LFD positive would be a relatively easy and highly recommended solution. 
Government guidance for users of LFDs states that a positive result makes it “extremely likely you are currently infected with covid-19 and risk infecting others” and that “lateral flow tests can detect the vast majority of cases.” [15,16] Others have highlighted even more misleading information about “very accurate” LFD tests that has been provided to schools and forwarded to parents.  A poll taken on 5 March showed schools and parents remain confused about the benefits from mass testing. Only 41% of 200 schools thought the LFDs would keep them safe, more than half reported consent rates below 85%, many even under 50%.  Concerns about potential individual harms from false positive results are likely to contribute to lack of consent. Ongoing engagement with, and confidence in, the covid testing programme more generally has the potential to be undermined by individual experience of false positive test results which come to light following confirmatory PCR testing.
Testing for covid-19 infection is an important public health tool for driving down transmission.  As we’ve seen in the past week, mass asymptomatic testing is problematic and needs to be carefully managed to avoid confusion, increase participation, and maximise potential benefits. As the balance shifts away from individual benefit, it becomes more important to consider the support needs of individuals to engage in activities for which they see no short term gain and at worse, potentially negative consequences.  These concerns are magnified when we consider that both engagement and benefit are likely to be inversely correlated with social disadvantage; individuals who pose the greatest risk of transmission are also those least likely to engage.  Currently, engagement in the school mass testing programme is likely to further widen existing inequalities.
Teachers and students (and their parents) may be willing to consent to testing for the sake of the wider population benefit, if there was explicit consideration and mitigation of the potential harms from false negatives and false positives and a commitment to evaluation of the policy.
Jac Dinnes, Senior Test Evaluation Methodologist, University of Birmingham.
Clare Davenport, Senior Clinical Lecturer, University of Birmingham.
Competing interests: none declared.
- Dinnes J, Deeks JJ, Berhane S, et al., Rapid, point-of-care antigen and molecular-based tests for diagnosis of SARS-CoV-2 infection. Cochrane Database Syst Rev, 2021 (in press).
- Public Health England, Criteria for appraising the viability, effectiveness and appropriateness of a screening programme. 23 Oct. 2015: https://www.gov.uk/government/publications/evidence-review-criteria-national-screening-programmes/criteria-for-appraising-the-viability-effectiveness-and-appropriateness-of-a-screening-programme.
- UK National Screening Committee, Guidance for the development, production and review of information to support UK population screening programmes. 2018: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/730598/UK_NSC_screening_information_development_guidance.pdf.
- Ferguson J, Dunn S, Best A, et al., Validation testing to determine the effectiveness of lateral flow testing for asymptomatic SARS-CoV-2 detection in low prevalence settings. medRxiv, 2020: p. 2020.12.01.20237784.
- University of Liverpool, Liverpool Covid-19 Community Testing Pilot. Interim Evaluation Report. 2020, University of Liverpool: Available from https://www.liverpool.ac.uk/media/livacuk/coronavirus/Liverpool,Community,Testing,Pilot,Interim,Evaluation.pdf.
- Cevik M, Tate M, Lloyd O, et al., SARS-CoV-2, SARS-CoV, and MERS-CoV viral load dynamics, duration of viral shedding, and infectiousness: a systematic review and meta-analysis. The Lancet Microbe, 2021. 2(1): p. e13-e22.
- Riley S, Wang H, Eales O, et al., REACT-1 round 9 final report: Continued but slowing decline of prevalence of SARS-CoV-2 during national lockdown in England in February 2021. medRxiv, 2021.
- Peto T and UK COVID-19 Lateral Flow Oversight Team, COVID-19: Rapid Antigen detection for SARS-CoV-2 by lateral flow assay: a national systematic evaluation for mass-testing. medRxiv, 2021: p. 2021.01.13.21249563.
- Deeks JJ, Why the school testing regime needs to change. Mar 10. https://unherd.com/thepost/why-the-school-testing-regime-needs-to-change/, 2021.
- NHS Test & Trace, Information on Coronavirus (COVID-19) Tests Conducted. 11 Mar. 2021: https://www.gov.uk/government/publications/nhs-test-and-trace-england-statistics-25-february-to-3-March-2021.
- Larremore DB, Wilder B, Lester E, et al., Test sensitivity is secondary to frequency and turnaround time for COVID-19 surveillance. medRxiv, 2020: p. 2020.06.22.20136309.
- West CP, Montori VM, and Sampathkumar P, COVID-19 Testing: The Threat of False-Negative Results. Mayo Clinic Proceedings, 2020. 95(6): p. 1127-1129.
- Moreno, G.K., K.M. Braun, I.W. Pray, et al., SARS-CoV-2 transmission in intercollegiate athletics not fully mitigated with daily antigen testing. medRxiv, 2021: p. 2021.03.03.21252838.
- Carr J, Pupils with positive home Covid test can return after negative lab result, government confirms. 8 March, in Schools Weekly. 2021: https://schoolsweek.co.uk/pupils-with-positive-home-covid-test-can-return-after-negative-lab-result-government-confirms/#:~:text=Pupils%20with%20positive%20home%20Covid,negative%20lab%20result%2C%20government%20confirms&text=Prime%20minister%20Boris%20Johnson’s%20official,used%20by%20pupils%20at%20home.
- Medicines & Healthcare Products Regulatory Agency, For patients, the public and professional users: a guide to COVID-19 tests and testing kits. 3 Mar. 2021: https://www.gov.uk/government/publications/how-tests-and-testing-kits-for-coronavirus-covid-19-work/for-patients-the-public-and-professional-users-a-guide-to-covid-19-tests-and-testing-kits.
- Department of Health and Social Care, Understanding lateral flow antigen testing for people without symptoms. 5 Mar. 2021: https://www.gov.uk/guidance/understanding-lateral-flow-antigen-testing-for-people-without-symptoms.
- Deeks JJ, Gill M, Bird S, Richardson S, and Ashby D. Covid-19 INNOVA testing in schools: don’t just test, evaluate. Jan 12. 2021.
- Carr J, Majority of schools report 90 per cent of pupils taking on-site Covid tests. 8 March, in Schools Weekly. 2021: https://schoolsweek.co.uk/majority-of-schools-report-90-per-cent-of-pupils-taking-on-site-covid-tests/.
- Crozier A, Rajan S, Buchan I, and McKee M, Put to the test: use of rapid testing technologies for covid-19. BMJ, 2021. 372: p. n208.
- Patel J, Fernandes G, and Sridhar D, How can we improve self-isolation and quarantine for covid-19? BMJ, 2021. 372: p. n625.
- Bambra C, Riordan R, Ford J, and Matthews F, The COVID-19 pandemic and health inequalities. Journal of Epidemiology and Community Health, 2020. 74(11): p. 964-968.