A study published today in Nature shows that the NHS COVID-19 contact tracing app successfully predicted people’s risk of catching the virus following exposure to an infected person.
Researchers at the University of Oxford's Pandemic Sciences Institute (PSI) analysed the app’s exposure measurements for 7 million people who were notified by the app that they had come into contact with the virus. The researchers found that the app correctly predicted people’s risk of infection based on these measurements.
An estimated 40 per cent of people analysed caught COVID-19 from a household contact, despite these accounting for only 6 per cent of contacts logged in the app. Duration of exposure to infected individuals was found to be the most significant determinant of transmission risk, with fleeting contacts (less than 30 minutes) representing half of reported contacts but very few transmissions.
The data provides new insights into how COVID-19 is spread. Getting closer to those with COVID-19, and spending more time close to them, has been widely known to increase the risk of someone catching the virus – but this is the first large-scale study to reveal how much, and how these two factors act together.
Lead analyst Dr Luca Ferretti, Research Fellow at the PSI, said: “This study found that how long you are exposed to someone infected with COVID-19 is the biggest factor affecting whether or not you’ll become infected yourself.
“Many infections resulted from long exposures. The risk of infections keeps increasing for every hour spent in close proximity. Infection is not inevitable after a short exposure, rather the risk keeps building up as the hours go by.
“18 million people in England and Wales used the NHS COVID-19 app, and thanks to them, millions of cases and over 10,000 deaths were averted.”
These latest findings have implications for other respiratory infections, including the design of measures to prepare for a future pandemic.
Professor Christophe Fraser, the study’s Principal Investigator and Moh Family Foundation Professor of Infectious Disease Epidemiology at the PSI, said: “These findings show how good data could be used in a future outbreak to inform social distancing precautions. In the case of COVID-19, this might have led to more emphasis on duration of contact as a risk for transmission.
“As governments and the global health community consider measures to address how to manage future epidemics and pandemics, improving our tools to understand how viruses spread is crucial. Risk management decisions need to be firmly based on quantitative evidence such as this.”
Professor Steven Riley, Director General for Data and Surveillance at UK Health Security Agency (UKHSA) said: “It's wonderful to see more great insight from this collaboration between UKHSA and the University of Oxford. Thanks to the many people in the UK who used the app, we now know how risk of COVID-19 transmission was affected by the proximity and duration of a contact.
“In the future, it’s possible that the methods and technology used in this study could give us much better evidence than we have had until now to determine the best response in the early stages of the next pandemic.”
Key findings:
- The duration of exposure plays a major role in whether transmission occurs.
- Most exposures were very short (less than one hour) but exposures resulting in transmission typically lasted hours to days.
- Individuals exposed within a household setting made up only 6 per cent of all ‘pinged’ individuals, but 40 per cent of transmissions detected by the app.
- Transmission risk continued increasing after several days of exposure, implying that exchange of the virus between household members is not a foregone conclusion, and could be prevented with precautions within the home.
Previous studies from the same University of Oxford and UKHSA team estimated the impact of the NHS COVID-19 contact tracing app on the pandemic in England and Wales. They concluded that the app saved 10,000 lives in its first year.
The research was led by the University of Oxford in partnership with the UKHSA and the University of Warwick.
Read the paper: Digital measurement of SARS-CoV-2 transmission risk from 7 million contacts