Skip to main content

The Public Health Agency of Canada says “discrepancies” with how it interpreted Alberta COVID-19 testing data artificially inflated the national test positivity rate.

Public Health said in an advisory that it had temporarily removed Alberta’s COVID-19 testing data from its national statistics. It said unspecified issues with provincial testing figures overestimated the percentage of tests that came back positive, although it did not elaborate.

The agency later updated its advisory acknowledging that it had made the error, not the Alberta government.

“Following dialogue with the province of Alberta, it was determined that a field of submitted data had been interpreted incorrectly by PHAC,” the corrected advisory said. “This issue has since been resolved. Previous data reports were not affected by this issue.”

The agency did not specify how significantly this error affected the positivity rate. It said the national positivity rate for the week of Nov. 29 to Dec. 5 was 6.5 per cent after the Alberta data was removed. For the previous week, the rate was 7.4 per cent with the Alberta data included.

Alberta Health spokesman Tom McMillan said in an e-mail that the issue was simply a misunderstanding. He said the province has not made any changes to how it reports data.

He said the issue appears to be related to the province reporting two different statistics – the total number of tests completed every day as well as the number of people tested for the first time, which is much lower.

“It is our assumption that, in the recent reports, PHAC may have incorrectly calculated the positive percentage using the number of new people tested rather than total tests completed,” Mr. McMillan wrote.

“This would be incorrect and lead to errors, but those would not be due to the information provided by Alberta Health.”

Mr. McMillan said the province will discuss the issue with Public Health and will change how it reports testing data to ensure the issues are fixed.

Publicly available data from both Alberta Health and the agency align with this explanation.

The Alberta government has been publishing two sets of testing statistics on its COVID-19 data tool throughout the pandemic. The province calculates its own positivity rate, which is currently at 9.5 per cent, using the total number of all tests conducted. On Dec. 9, for example, the province conducted 16,793 tests.

It also publishes a much smaller number that is described as “people tested.” While it is not explained on the website, this figure refers only to the number of people tested in a given day for whom it was their first-ever test. On Dec. 9, that number was 6,841.

Provincial-level data released by the public-health agency has been close to the lower number for Alberta. On Dec. 9, the agency reported 6,551 tests in the province. That would produce an incorrect positivity rate for the province of about 23 per cent.

The positivity rate is one of several metrics used to track both the spread of COVID-19 and the adequacy of the testing system.

A high positivity rate can either mean that infection rates are high or that there are too few tests being conducted. A high rate suggests there could be a significant proportion of infections that are not being detected.

The World Health Organization recommends keeping the positivity rate below 5 per cent.

Alberta has the highest per capita rates of new infections and active cases in the country and among the highest rates of hospitalizations. The province added 1,566 confirmed infections on Monday and 13 deaths.

There were 682 people in hospital, including 124 in intensive care.

We have a weekly Western Canada newsletter written by our B.C. and Alberta bureau chiefs, providing a comprehensive package of the news you need to know about the region and its place in the issues facing Canada. Sign up today.

Editor’s note: A previous version of this story cited the Public Health Agency of Canada describing "discrepancies" in testing data it received from the Alberta government. The agency has since issued a correction, acknowledging that it had made the error, not the province.