Effectiveness of SARS-CoV-2 testing strategies inreducing COVID-19 cases, hospitalisations, and deaths.

Abstract

Rationale: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has substantially affected daily life. Sustainable testing practices are essential to balance the resource demands of widespread testing with the need to reduce the health impacts of COVID-19. However, the effectiveness of specific testing strategies for symptomatic and asymptomatic individuals in reducing COVID-19 cases, hospitalisations, and deaths remains uncertain.

Objectives: To evaluate the effectiveness of different SARS-CoV-2 testing strategies in reducing COVID-19 cases, hospitalisations, and deaths amongst suspected cases and asymptomatic individuals.

Search methods: We searched CENTRAL, MEDLINE (Ovid), Embase (Elsevier), Europe PMC, ClinicalTrials.gov, and the World Health Organization (WHO) International Clinical Trials Registry Platform. We also conducted reference checks, citation searches, and contacted study authors to identify eligible studies. The most recent search was conducted on 07 October 2024.

Eligibility criteria: We included randomised controlled trials (RCTs), non-randomised studies of interventions (NRSIs), controlled before-and-after studies (CBA), matched cohort studies, and observational studies with a comparison group involving suspected or asymptomatic individuals. Eligible studies compared testing strategy versus no testing or standard care or usual practice; one testing strategy with another, such as antigen-detecting rapid diagnostic tests (RDTs) versus nucleic acid amplification testing (NAAT), including reverse transcription polymerase chain reaction (RT-PCR); home-based versus provider-administered testing; one-time testing versus repeated testing at different frequencies; and targeted testing versus widespread testing. Combinations of these components were also considered. In this review, we define 'SARS-CoV-2 testing strategy' as a complex intervention comprising multiple varying components, including test type (e.g. NAAT, antigen-detecting RDT), sample type (e.g. nasopharyngeal swab, saliva), target population (e.g. symptomatic, contacts), setting (e.g. home, clinic, congregate), frequency of testing (e.g. one-time, weekly, daily), and response protocol (e.g. isolation, confirmatory testing, treatment). We excluded single-arm studies, reviews, theses, editorials, letters, commentaries, studies reported solely in abstract form, laboratory or animal studies, mathematical modelling studies, and diagnostic test accuracy studies.

Outcomes: Our critical outcomes were: COVID-19 cases avoided (reduction in new cases); COVID-19-related hospitalisations avoided (reduction in hospital admissions); COVID-19-related deaths avoided (reduction in mortality); and serious adverse events related to testing, including unnecessary interventions, employment impacts, isolation effects, and psychological harms.

Risk of bias: We used the Risk of bias 2 (RoB 2) tool to assess the risk of bias in RCTs and the ROBINS-I tool to assess the risk of bias in NRSIs, CBA studies, and matched cohort studies.

Synthesis methods: As a meta-analysis was not feasible due to the nature of the data, we applied Synthesis Without Meta-analysis (SWiM) methods. We assessed the certainty of the evidence for each outcome using the GRADE approach.

Included studies: We included 21 studies (10 RCTs and 11 NRSIs) with 13,312,327 participants. Among these, 13 studies-comprising eight RCTs and five NRSIs-either reported one or more prespecified outcomes (four studies), provided relevant information through proxy measurements (five studies), or supplied information following author correspondence (four studies).

Synthesis of results: We present the prioritised comparisons and critical outcomes. For the comparison testing strategy versus no testing or standard care or usual practice, one included study measured two critical outcomes. The study did not measure the other critical outcomes: COVID-19 cases avoided, and serious adverse events related to testing. No studies measured any critical outcomes for the other prioritised comparison: antigen-detecting RDT versus NAAT testing. Benefits and harms of testing strategy versus no testing or standard care or usual practice One observational study with a comparison group, conducted in a long-term care facility in Israel, compared weekly SARS-CoV-2 RT-PCR testing with no testing and measured two of our critical outcomes. Based on the analysis, the evidence is very uncertain about the effect of SARS-CoV-2 RT-PCR testing on reducing hospitalisation (decrease in the hospitalisation rate from 13.59% to 11.41%; 1 study, 162,205 participants, very low-certainty evidence) and mortality (33.8% decrease in expected mortality; 1 study, 162,205 participants, very low-certainty evidence) compared to no testing. We downgraded the certainty of the evidence because of methodological limitations, indirectness, and imprecision.

Authors' conclusions: The available data are of very low-certainty. Only one of the 21 included studies reported hospitalisations or deaths; therefore, we cannot draw conclusions about the effects of testing strategy versus no testing on reducing hospitalisation and mortality. No studies evaluated other critical outcomes i.e. COVID-19 cases avoided, and serious adverse events related to testing. Future research should aim for consistency and relevance by using clearly defined outcomes, preferably based on a standardised core outcome set. A qualitative evidence synthesis (QES) would help identify barriers and facilitators to routine SARS-CoV-2 testing in healthcare settings, which could help inform intervention development. The QES would explore factors affecting the implementation of routine testing, drawing on the perspectives of healthcare providers, patients, and other interest holders.

Funding: This Cochrane review was partially funded by the World Health Organization (WHO) and the Health Research Board of Ireland.

Registration: Protocol (2025) DOI: 10.1002/14651858.CD016192.