Public health not directly tracking tests

The Spectator found several inconsistencies in the way the unit reports and collects its data

A Spectator investigation of COVID-19 testing in Hamilton has revealed gaps in the way public health reports and collects data.

The city’s health unit is not directly keeping track of the number of tests completed at its three assessment centres, and instead relies on the health networks and primary care clinicians who oversee the sites to forward them that information.

It’s a process that, on some occasions, can take days to filter through and reach public health officials, said Dr. Elizabeth Richardson, Hamilton’s chief medical officer of health.

“We get the data in batches,” she said. “We don’t necessarily get it every day and update it every day.”

The city’s east and west assessment centres, which opened March 16, are overseen by St. Joseph’s Healthcare Hamilton and Hamilton Health Sciences. The drivethru test site, which opened April 20, is overseen by a group of primary care clinicians.

Richardson said each assessment centre is responsible for compiling a daily report of how many people visited their sites and the number of tests they’ve completed.

But public health can’t guarantee they receive those reports simultaneously or on the same day.

“The reporting of information out of the assessment centres is one where we don’t necessarily ask them for daily reports,” Richardson said, noting public health’s primary focus amid the crisis has been to understand the number of positive cases and outbreaks, as well as followup on lab results.

“We ask them for reports as soon as they’re able to share them. They may give us a report for one day and then not be able

to send another report until two days later.”

Epidemiologists say the lag in daily testing data that’s available to public health presents transparency issues and limits the public’s capability to understand the extent of an outbreak.

“We should be transparent about our limited data-collection capabilities,” said Colin Furness, an infectious disease expert from the University of Toronto. “The public may not want to hear that … but I think we have to be really transparent about that, because it’s a circus otherwise.

“It means that their (testing) data doesn’t actually mean much day-to-day, because we don’t know if it’s accurate.”

The Spectator has previously reported that public health had “no idea” how many people have been tested in the city.

On Monday, that figure was revealed for the first time — not by public health, but instead by ICES, an independent, nonprofit research institute.

The group stated in a report that 11,182 people have been tested in Hamilton as of April 30, among the highest per region in Ontario. The Ministry of Health said in a statement Thursday that public health units (PHUs) are encouraged to use the report’s findings.

“Given the provincial lab data provides a more complete picture of COVID-19 testing in the province, PHUs may choose to use ICES’ report to gain a more complete understanding of testing,” a ministry spokesperson said.

Public health reports the cumulative number of tests completed as well as the cumulative number of visits per person to its assessment centres on its website daily. The data — which is exclusive to tests completed at the centres and does not include long-term care homes or hospitals — is available to the public for just 24 hours before being updated with new cumulative counts.

The Spectator began taking screenshots of the city’s case status website on April 20 to establish a timeline of how testing numbers have changed over time. It found several inconsistencies in the data and how information is being measured and stored by public health.

In some cases, the number of completed tests spiked drastically before reverting back to a normal rate. In others, the number of completed tests either nearly doubled the number of people who visited assessment centres or matched them exactly.

For instance, cumulative testing totals posted in late April and early May show some large spikes in the numbers. Between April 27 and April 29, Monday to Wednesday of that week, public health reported completing 792 tests from 797 visitors to its three assessment centres. Then from Wednesday to Friday that week, April 29 to May 1, they reported 862 additional completed tests from just 438 visitors.

While that week marks the biggest increase in completed tests since Hamilton opened assessment centres in midMarch, it’s a statistical anomaly.

The spikes were prefaced by two weeks where public health averaged 96.5 completed tests and roughly 103 visits per day. In the 11 days subsequent to May 1, they averaged 62 tests and 62 visits per day.

Public health officials could not explain why the total number of completed tests fluctuated to the extent they did over the last week of April and start of May. They also could not explain why the number of completed tests was nearly twice the number of people who visited assessment centres from April 29 to May 1, nor why those counts were equal in the following 11 days.

Most notable from The Spectator’s analysis was a sudden day-to-day drop of eight per cent in public health’s cumulative count of completed tests.

On May 3, the unit reported having completed a total of 5,450 tests since assessment centre testing began on March 16. On May 4, that number plummeted to 5,021 tests — a drop of 429. It took four more days until the cumulative test count was back to higher than it was on May 3.

Richardson said she wasn’t sure why the decrease occurred.

“The only other thing (it) might have been is the assessment data and the long-term care data being reported together, and we’ve separated those back out again.”

The Spectator presented its collection of data to public health last week and asked for context regarding the data’s inconsistencies. The unit initially said they wouldn’t be able to comment until they knew the source of the numbers. When told the source was the Hamilton Public Health website — and, specifically, screenshots from the daily testing reports — public health began working on a response.

“We don’t keep snapshots of our website,” Richardson said.