677. Compliance and Performance Characteristics of Subject Collected Versus Health-care Worker Collected Nasal Swabs for Respiratory Viral Surveillance

Abstract

Background Self-collection of mid-nasal swabs (SCNS) at home is a convenient alternative to health-care worker-collected nasal swabs (HCWC) for determining the pathogen-specific epidemiology of influenza-like illness (ILI). We evaluated the compliance and performance characteristics of SCNS vs. HCWC for respiratory pathogens during 2019-2020 flu season. Methods Adult Military Health System (MHS) beneficiaries were enrolled in an influenza vaccine effectiveness trial (PAIVED). Following vaccination, subjects were instructed on SCNS and completion of a symptom diary and were contacted weekly to ascertain ILI symptoms (fever, sore throat, and/or cough). In the event of an ILI, subjects completed the symptom diary and SCNS and were scheduled a clinic visit for HCWC. Swabs were tested with the Luminex NxTAG® Respiratory Pathogen Panel. We evaluated compliance with swab collection, positive percent agreement (PPA) of SCNS using PCR detection from either HCWC or SCNS as the reference standard, and agreement between paired swabs using the Cohen Kappa coefficient (Κ). Results 1808 ILI were reported by 972 participants enrolled during the study period. Compliance with HCWC was higher than SCNS (58% [1042] vs. 42% [766]; p< 0.001). SCNS were associated with a shorter interval from symptom onset (median: 4 days [IQR:2-6 days] vs. clinic collect: 7 days [IQR:4-9 days]; p < 0.001). 663 paired swabs were available for 609 participants (Table 1). The overall detection rate was higher in SCNS (36%) than HCWC (26%; p< 0.001) (Figure 1). The overall PPA was 85.7% and a PPA of approximately 80% of greater was observed for influenza, rhino/enterovirus, parainfluenza and respiratory syncytial virus. Agreement between paired swabs was poor due to the lower detection rates in HCWC. Table 1. Demographics and swab collection data for 609 participants who provided 663 paired swabs Figure 1. Detection by pathogen in 663 paired swabs Conclusion SCNS were associated with higher detection rates compared to HCWC, likely due to the shorter interval between symptom onset and swab collection. Strategies to improve compliance with SCNS and minimize the interval between symptom onset and swab collection are needed to optimize detection of respiratory pathogens in this MHS cohort. Disclosures Ryan C. Maves, MD, EMD Serono (Advisor or Review Panel member)Heron Therapeutics (Advisor or Review Panel member) Jitu Modi, MD, GSK (Speaker's Bureau)