Individual level analysis of digital proximity tracing for COVID-19 in Belgium highlights major bottlenecks
Abstract To complement labour-intensive conventional contact tracing, digital proximity tracing was implemented widely during the COVID-19 pandemic. However, the privacy-centred design of the dominant Google-Apple exposure notification framework has hindered assessment of its effectiveness. Between October 2021 and January 2022, we systematically collected app use and notification receipt data within a test and trace programme targeting around 50,000 university students in Leuven, Belgium. Due to low success rates in each studied step of the digital notification cascade, only 4.3% of exposed c... Mehr ...
Verfasser: | |
---|---|
Dokumenttyp: | Artikel |
Erscheinungsdatum: | 2023 |
Reihe/Periodikum: | Nature Communications, Vol 14, Iss 1, Pp 1-12 (2023) |
Verlag/Hrsg.: |
Nature Portfolio
|
Schlagwörter: | Science / Q |
Sprache: | Englisch |
Permalink: | https://search.fid-benelux.de/Record/base-27391909 |
Datenquelle: | BASE; Originalkatalog |
Powered By: | BASE |
Link(s) : | https://doi.org/10.1038/s41467-023-42518-6 |
Abstract To complement labour-intensive conventional contact tracing, digital proximity tracing was implemented widely during the COVID-19 pandemic. However, the privacy-centred design of the dominant Google-Apple exposure notification framework has hindered assessment of its effectiveness. Between October 2021 and January 2022, we systematically collected app use and notification receipt data within a test and trace programme targeting around 50,000 university students in Leuven, Belgium. Due to low success rates in each studied step of the digital notification cascade, only 4.3% of exposed contacts (CI: 2.8-6.1%) received such notifications, resulting in 10 times more cases detected through conventional contact tracing. Moreover, the infection risk of digitally traced contacts (5.0%; CI: 3.0–7.7%) was lower than that of conventionally traced non-app users (9.8%; CI: 8.8-10.7%; p = 0.002). Contrary to common perception as near instantaneous, there was a 1.2-day delay (CI: 0.6–2.2) between case PCR result and digital contact notification. These results highlight major limitations of a digital proximity tracing system based on the dominant framework.