Harnessing a real-time location system for contact tracing in a busy emergency department

A H Aung; A L Li; W M Kyaw; R Khanna; W-Y Lim; H Ang; A L P Chow

doi:10.1016/j.jhin.2023.08.015

Harnessing a real-time location system for contact tracing in a busy emergency department

J Hosp Infect. 2023 Nov:141:63-70. doi: 10.1016/j.jhin.2023.08.015. Epub 2023 Sep 1.

Authors

A H Aung¹, A L Li², W M Kyaw², R Khanna², W-Y Lim², H Ang³, A L P Chow⁴

Affiliations

¹ Department of Preventive and Population Medicine, Tan Tock Seng Hospital, Singapore, Singapore. Electronic address: aung_hein_aung@ttsh.com.sg.
² Department of Preventive and Population Medicine, Tan Tock Seng Hospital, Singapore, Singapore.
³ Department of Emergency Medicine, Tan Tock Seng Hospital, Singapore, Singapore.
⁴ Department of Preventive and Population Medicine, Tan Tock Seng Hospital, Singapore, Singapore; Lee Kong Chian School of Medicine, Singapore, Singapore; Infectious Disease Research and Training Office, National Centre for Infectious Disease, Singapore, Singapore; Saw Swee Hock School of Public Health, Singapore, Singapore.

PMID: 37660888
DOI: 10.1016/j.jhin.2023.08.015

Abstract

Background: With the persistent threat of emerging infectious diseases (EIDs), digital contact tracing (CT) tools can augment conventional CT for the prevention of healthcare-associated infectious disease transmission. However, their performance has yet to be evaluated comprehensively in the fast-paced emergency department (ED) setting.

Objective: This study compared the CT performance of a radiofrequency identification (RFID)-based real-time location system (RTLS) with conventional electronic medical record (EMR) review against continuous direct observation of close contacts ('gold standard') in a busy ED during the coronavirus disease 2019 pandemic period.

Methods: This cross-sectional study was conducted at the ED of a large tertiary care hospital in Singapore from December 2020 to April 2021. CT performance [sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and kappa] of the RTLS, EMR review and a combination of the two approaches (hybrid CT) was compared with direct observation. Finally, the mean absolute error (MAE) in the duration of each contact episode found via the RTLS and direct observation was calculated.

Results: In comparison with EMR review, both the RTLS and the hybrid CT approach had higher sensitivity (0.955 vs 0.455 for EMR review) and a higher NPV (0.997 vs 0.968 for EMR review). The RTLS had the highest PPV (0.777 vs 0.714 for EMR review vs 0.712 for hybrid CT). The RTLS had the strongest agreement with direct observation (kappa=0.848). The MAE between contact durations of 80 direct observations and their respective RTLS contact times was 1.81 min.

Conclusion: The RTLS was validated to be a high-performing CT tool, with significantly higher sensitivity than conventional CT via EMR review. The RTLS can be used with confidence in time-strapped EDs for time-sensitive CT for the prevention of healthcare-associated transmission of EIDs.

Keywords: Communicable disease contact tracing; Communicable diseases, emerging; Contact tracing/methods; Epidemiology; Infection control.

MeSH terms

COVID-19* / prevention & control
Computer Systems
Contact Tracing*
Cross-Sectional Studies
Emergency Service, Hospital
Humans