Are at-home, parent-collected, Royal Mail-transported upper respiratory tract dry specimens equivalent to nurse-collected, same-day, on-ice, viral-medium transported specimens for the detection of respiratory microbes in young children?

Talk Code: 
Claire A Woodall
Claire A. Woodall1, Hannah V. Thornton1, Emma C. Anderson2, Suzanne M. Ingle3, Peter Muir4, Barry Vipond4, Denise Longhurst4, John P Leeming5, Charles R. Beck3 6, Adam Finn7 and Alastair D. Hay1
Author institutions: 
1Centre for Academic Primary Care, Population Health Sciences, Bristol Medical School, University of Bristol, Canynge Hall, 39 Whatley Road, Bristol BS8 2PS, UK. 2Centre for Academic Child Health, Population Health Sciences, Bristol Medical School, University of Bristol, 1-5 Whiteladies Road, Bristol BS8 1NU, UK. 3National Institute Health Research, Health Protection Research Unit in Evaluation of


Individuals are increasingly self-collecting upper respiratory swabs for the detection of Covid-19. Community self-collected swabs could reduce costs and infection risk, but there is a paucity of evidence comparing parent-collected (PC) and nurse-collected (NC) upper respiratory specimens. We aimed to assess PC vs. NC saliva and nasal sample quality and performance.


Children were recruited into a community inception respiratory infection cohort study between 26 February and 1 July in 2016. PC nasal and saliva swabs were completed at-home and posted to a single research laboratory. NC nasal and saliva swabs independently, and then transported them same-day in viral transport medium on ice. Reverse transcriptase (RT)-PCR was used to detect the human gene, 18S rRNA, to measure swab quality; and a 42-pathogen TaqMan array card to detect respiratory microbes, with a cycle threshold ≤38 reported as positive. Agreement (Cohen’s kappa, κ), sensitivity and specificity were calculated for the detection of human control genes and pathogens with PC and NC microbe results considered the ‘test’ and ‘reference standard’ respectively.


Pairs of PC and NC nasal and saliva samples were available in 91 and 92 children respectively, median age 4 years (IQR 2 – 8). Swabs were taken at a mean of 6.4 days (95% Cl, 5.8 – 6.9) after symptom onset. PC swabs arrived at the laboratory a mean of 2.7 days (95% Cl, 2.33 – 3.06) post-collection whereas NC swabs arrived at a mean of 0.05 days (95% Cl, 0 – 0.1). 18S rRNA detection was 100% sensitive and specific (κ = 1). Combining all viruses, nasal swabs had a sensitivity, specificity and kappa of 91.6%, 98.8% κ = 0.84 respectively. For all bacteria, the same parameters were 91.5%, 96.2%, κ = 0.84. Saliva specificity was similar to nasal swabs for virus and bacteria. However, for viruses and bacteria respectively, sensitivities were lower at 69.4% and 78.1%, and was in agreement, κ = 0.64 and κ = 0.73.


Our results suggest that PC nasal swabs at home, posted dry could be used instead of healthcare professional swabbing with same-day transport, on ice, in transport medium. Our results also suggest PC saliva samples should not replace NC saliva samples. PC home taken nasal samples could be used in clinical and research settings.

Submitted by: 
Claire Woodall
Funding acknowledgement: 
This work was supported by in part by the Medical Research Council and Wellcome Trust Institutional Strategic Support Fund (WT ISSF), awarded to CAW on a Daphne Jackson Trust Development Fellow in collaboration with the Elizabeth Blackwell Institute at the University of Bristol. The WT ISSF3 grant number 204813/Z/16/Z.