COVID-19 cluster surveillance using exposure data collected from routine contact tracing: The genomic validation of a novel informatics-based approach to outbreak detection in England

PLOS Digit Health. 2024 Apr 25;3(4):e0000485. doi: 10.1371/journal.pdig.0000485. eCollection 2024 Apr.

Abstract

Contact tracing was used globally to prevent onwards transmission of COVID-19. Tracing contacts alone is unlikely to be sufficient in controlling community transmission, due to the pre-symptomatic, overdispersed and airborne nature of COVID-19 transmission. We describe and demonstrate the validity of a national enhanced contact tracing programme for COVID-19 cluster surveillance in England. Data on cases occurring between October 2020 and September 2021 were extracted from the national contact tracing system. Exposure clusters were identified algorithmically by matching ≥2 cases attending the same event, identified by matching postcode and event category within a 7-day rolling window. Genetic validity was defined as exposure clusters with ≥2 cases from different households with identical viral sequences. Exposure clusters were fuzzy matched to the national incident management system (HPZone) by postcode and setting description. Multivariable logistic regression modelling was used to determine cluster characteristics associated with genetic validity. Over a quarter of a million (269,470) exposure clusters were identified. Of the eligible clusters, 25% (3,306/13,008) were genetically valid. 81% (2684/3306) of these were not recorded on HPZone and were identified on average of one day earlier than incidents recorded on HPZone. Multivariable analysis demonstrated that exposure clusters occurring in workplaces (aOR = 5·10, 95% CI 4·23-6·17) and education (aOR = 3·72, 95% CI 3·08-4·49) settings were those most strongly associated with genetic validity. Cluster surveillance using enhanced contact tracing in England was a timely, comprehensive and systematic approach to the detection of transmission events occurring in community settings. Cluster surveillance can provide intelligence to stakeholders to support the assessment and management of clusters of COVID-19 at a local, regional, and national level. Future systems should include predictive modelling and network analysis to support risk assessment of exposure clusters to improve the effectiveness of enhanced contract tracing for outbreak detection.

Grants and funding

All work was funded by the UK Health Security Agency (previously Public Health England) as part of the public health response to the COVID-19 pandemic in England. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.