Towards new forms of communication and surveillance: a mixed methods study of rapid respiratory virus assessment in general practice during the SARS-CoV-2 pandemic

Introduction

Accurate diagnosis and assessment of influenza and other respiratory illnesses in primary care are important for surveillance, potential treatment advances¹ and patient communication. Point-of-care testing (POCT) has long been integral in health care,² initially in hospital care then extended to primary care,³ with tests such as blood glucose monitoring and pregnancy tests.

Rapid reporting in respiratory infection is a relatively recent development, initially using C-reactive protein (CRP),⁴ with much of POCT use being focused on antibiotic and more recently, anti-viral prescribing.^5,6 Polymerase chain reaction (PCR) testing is recognised as a rapid and sensitive method for respiratory virus surveillance^7,8 with studies demonstrating POCT being both feasible and having an antimicrobial prescribing impact.^9,10

While PCR testing is the ‘gold standard’ test for virus identification, it requires centralised laboratory-based testing, and creates delays in clinical decision making. Multi-viral rapid antigen testing (RAT) has enabled testing for multiple respiratory viruses on one sample, but early platforms had low sensitivity and specificity.¹¹ Despite primary care internationally showing enthusiasm for POCT, reservations remained about test accuracy and results interpretation.^12,13

The Aotearoa New Zealand (NZ) Ministry of Health funds the Institute of Environmental Science and Research (ESR) to undertake winter surveillance for influenza like illness (ILI) using weekly nasopharyngeal swabs for PCR testing from sentinel general practices using the World Health Organization international standard for ILI.¹⁴

The arrival of COVID-19 highlighted the critical need for rapid diagnosis and genetic sample collection for a novel respiratory virus.^15,16 To effectively identify SARS-CoV-2, primary care needed to adapt traditional acute respiratory illness (ARI) screening¹⁷ to recognise a broader list of respiratory symptoms. Therefore, a pragmatic solution was needed to combine screening for SARS-CoV-2 alongside influenza viruses without overburdening an already strained workforce.¹⁸

With improvements in the diagnostic accuracy of tests across multiple pathogens,¹⁹ we determined it appropriate to pilot using a RAT enabling an immediate test for SARS-CoV-2 alongside PCR-based testing for multiple respiratory viruses. We also aimed to explore facilitators and barriers to its use,²⁰ and to understand the type and incidence of respiratory viruses among consultation-seeking patients with an ARI.

Methods

We recruited a purposive sample of six general practices with a range of geographical and socio-demographic characteristics. Nasopharyngeal viral swab collection commenced between May and June 2022 and ended on 11 December 2022. An electronic collection template for virology and clinical data entry was developed for recording viral swabs, clinical details and swab results.

Practices were informed that the ARI surveillance would use a routine nasopharyngeal swab (at that time current practice indicated for COVID-19 testing – ie no change in practice) to test for influenza and non-influenza respiratory viruses and SARS-CoV-2 with a single PCR swab. A RAT test for immediate COVID-19 results was undertaken by using the same PCR swab to allow for immediate feedback for the SARS-CoV-2 results.

The case definition for ARI consisted of any acute respiratory symptoms such as new or worsening cough, fever (at least 38°C), shortness of breath, sore throat, coryza (runny nose), anosmia and dysgeusia (altered sense of taste). Patients presenting with less typical symptoms (eg headache, myalgia) were also recommended for testing, as were patients with any other symptoms consistent with COVID-19.

A nasopharyngeal/mid-turbinate swab was taken from the patient and inserted into liquid transport media. The liquid media was sampled using a RAT swab, processed on site for SARS-CoV-2 and the remaining viral transport media was sent to a central laboratory for multiplex PCR testing. The samples were tested there for SARS-CoV-2, influenza A and B and non-influenza respiratory viruses (respiratory syncytial virus, parainfluenza virus types 1, 2, 3 and 4, rhinovirus, adenovirus and human metapneumovirus (hMPV). PCR test results for SARS-CoV-2 were available on the practice management system (PMS) within 48 h and for the other viral tests within 7 days.

A proof of concept validation for using a single viral liquid media for both RAT and laboratory-based PCR testing was undertaken using 60 nasopharyngeal/mid-turbinate swabs, comparing RAT results with PCR results. The good correlation of results in samples where the cycle threshold (C_t) value was <25 confirmed that the same sample could be used for both tests.

A questionnaire with 27 questions exploring the introduction of the viral testing programme and facilitators and barriers to its use was completed by the practice manager or a lead clinician in each practice at the end of the study period (questionnaire outline is included in Appendix 1).

Ethics approval was obtained from the Health and Disability Ethics Committees NTX/11/11/102/AM66.

Results

Overall 4135 patients met ARI eligibility criteria and 2233 (54%) tested positive for at least one of the viruses tested. Table 1 shows the results of testing.

Rhinovirus was the most common (835/4135, 20.2%) followed by hMPV (382/4135, 9.2%).

The results were used as part of the weekly reporting from ESR during this period providing a valuable addition to other sources of virology surveillance (Fig. 1).

Fig. 1.

Virology reporting from participating general practices.

Discussion

This project demonstrated the feasibility of using standard primary care practice to collect national surveillance data during a pandemic. The use of a single sampling event was acceptable and advantageous for practices and patients. The data assisted national surveillance in identifying community transmission of ILIs and documenting the relative prevalence of circulating respiratory viruses, including influenza and COVID-19, at a time of uncertain and changing viral patterns.

In keeping with recent overseas studies evaluating the use of POCT molecular testing in general practices, our practices perceived benefits to having increased diagnostic information regarding viral aetiology leading to potential changes in prescribing and communication strategies with patients.^13,21–23

The strength of our study is the incorporation of the process and technology into routine practice at a time of significant workload pressure from respiratory illness. Enthusiastic participation was achieved in a range of different types and settings, including a Māori health provider serving a high needs population.

Limitations included questions about the feasibility of swabbing infants and young children within routine clinical practice.

We have demonstrated the feasibility of introducing routine POCT RAT for respiratory viral infections in general practice with reported satisfaction from both clinicians and patients with the information gained.

With the introduction of multi-viral RAT technology, there is the potential to have a single swab provide immediate diagnostic information on a range of different viruses with utility both for individual patient management and also for population based surveillance of influenza, SARS-CoV-2 and other viral pathogens.²⁴