September 23, 2021

How many lives did Canada’s COVID-19 Alert app save? This new study did the math


An independent analysis of the federal government’s COVID Alert smartphone app concludes that over five months it reduced the number of cases in Canada by between 6,284 and 10,894 — and saved between 57 and 101 lives — but that few Canadians actually used it.

“With even relatively modest adoption, as in Newfoundland and Labrador, our modelling suggests that the COVID Alert app was able to avert a substantial proportion of cases,” said Erica Moodie, a biostatistics professor at McGill University and one of the researchers involved in the study, in an email to the Star.

“Wider adoption could have averted many cases and subsequent deaths.”

The independent study was released Tuesday and prepared by four researchers from McGill who were able to access raw data from the application, which was rolled out by the federal government in July 2020 to help with exposure notifications and contact tracing. The analysis looked at data gathered between March and July of 2021.

The researchers said they used statistical modelling and a comparative analysis with other regions in the world to arrive at their conclusions.

Between July 2020 and July 2021, there were 6.6 million downloads of the app, said the report. However, in British Columbia, Alberta, Yukon and Nunavut, users could not report a COVID-19 diagnosis on it.

The researchers focused on jurisdictions where more than 200 notifications were sent in the five-month period: Manitoba, Newfoundland and Labrador, Nova Scotia, Ontario, Quebec and Saskatchewan.

Challenges persisted for the team as it looked at the data, including restrictions on what was captured by the app due to privacy concerns. Comparative analysis to other countries where smartphone apps were used was also “not really a fair assessment of the effectiveness,” added Moodie, since the period of reporting of data elsewhere and the timing of overall case surges differed.

Still, while the report said that comparisons to other countries where exposure apps were used was difficult due to unreliable data that “among countries that reported adoption rates for their apps, that of COVID Alert was comparatively low.”

Adoption of the app was hindered by provinces — such as B.C. and Alberta — opting to use their own apps instead of the COVID Alert app. Also, some people whose first language is not English or French, as well as older residents without smartphones, could have not used it, Moodie said.

The report noted that where there was a high level of uptake, such as in Newfoundland and Labrador as well as Nova Scotia, there were “dramatically higher ratios of averted cases and averted deaths.”

“In Newfoundland and Labrador, for example, the upper bound of the number of cases averted was greater than 60% of total recorded cases in the period of analysis,” said the report.

Meanwhile, in Alberta, the ABTraceTogether app has been criticized for its low levels of uptake. In January, Global News reported that the app had been used to track 32 positive cases and flag 141 close contacts as of that month.

The report looked at the number of downloads in different countries. In Canada, between July 31, 2020, and July 15, 2021, there were 6.6 million downloads, representing just over 17 per cent of the population. This was lowest among the nine jurisdictions included.

France saw 38 per cent of its population download its app, and New Zealand had 60 per cent of its population download an app over the year.

The United Kingdom saw 21 million downloads of the app, representing just over 30 per cent of its overall population. For the month of March 2021, the U.K. had 16.5 million active users, representing 78 per cent of the total downloads. Meanwhile, in the same month, Canada had just over three million active users, representing 46 per cent of the total number of downloads.

The report estimated that the U.K. averted between 284,000 and 594,000 COVID-19 cases through the use of its app. This would account for between five and 10 per cent of the total number of cases in the U.K., the report said.

In Canada, the report did note, usage of the app was likely higher before the time period that the researchers were able to assess data for. However, “we do not believe that these higher values would markedly change the trends and relative impact reported.”

Moodie said that waves of cases happening at different times in different places, including various provinces, means that apps can show varying degrees of effectiveness depending on when they’re analyzed.

“This is precisely what we see when comparing across provinces: Although the calendar time of the assessments are the same, the ‘pandemic time’ in terms of the magnitude of the waves is not and so the number of cases of averted can be substantially different,” noted Moodie.

“That said, the U.K. app appears to have averted many cases and preserved user privacy while still collecting very useful data for effectiveness analyses.”





Source link