4.5 Questions for Alberta Health

One of the ways I’m coping with this pandemic is studying it. Over the span of months I built up a list of questions specific to the situation in Alberta, so I figured I’d fire them off to the PR contact listed in one of the Alberta Government’s press releases.

That was a week ago. I haven’t even received an automated reply. I think it’s time to escalate this to the public sphere, as it might give those who can bend the government’s ear some idea of what they’re reluctant to answer.

1. When will the results of volunteer testing be released? The initial results are already puzzling; even if we assume all the provinces’ COVID-19 cases were in Calgary, 50 positive tests out of 1,500 performed combined with a case fatality ratio of 1% implies that 445 Albertans should be dead, nearly three times higher than the current total of 128. Either the typical estimates for the case fatality ratio are too high, in which case Alberta Health Services has invaluable information on the pandemic, or the false positive rate of their tests is roughly 2%, which is quite reasonable but has rather large consequences for the true number of COVID-19 cases in the province.

2. What sorts of tests are being performed? If those 1,500 were detecting viral RNA, that implies 44,533 active COVID-19 cases in Calgary, 56 times the official count. For that reason alone I’m fairly confident they were antibody tests, which combined with a false positive rate in the ballpark of 2.6% would mean the true number of total ever-infected is three times higher than official reports, in line with other research. If the official number of tests includes both viral and antibody tests, though, this makes it much tougher to determine the true extent of COVID-19 in this province. This conflation has led to artificially-inflated test numbers in the US and become a minor scandal, so some clarity on which tests are being performed in what context would be appreciated.

3. Are there plans to release data on excess mortality? There’s substantial evidence for uncounted or indirect deaths related to COVID-19 in other jurisdictions. Canada in general does not do a good job gathering the relevant information, but my impression is that Alberta does better than other provinces. By releasing a report or even just data on the topic, you could build public trust that’s invaluable during a pandemic, while both earning a PR win over other provinces and shaming them into doing a better job.

4. I’m very grateful for the statistics AHS has released to the public, they’re world-class and an invaluable resource. This openness can also reveal flaws in the process or oddities, unfortunately. I’m well aware that case statistics can change a week or two after first being reported, and I tend to ignore any discrepancies within that window. There is something odd going on with old test counts, though. Here’s a table of the number of tests performed on a specific date, with the second column representing the numbers reported on April 22nd and the third being those same numbers from yesterday.

Date On April 22nd On May 20th
2020-03-04 71 71
2020-03-05 55 55
2020-03-06 62 62
2020-03-07 276 274
2020-03-08 443 442
2020-03-09 723 722
2020-03-10 971 971
2020-03-11 966 966
2020-03-12 1144 1141
2020-03-13 1421 1420
2020-03-14 1608 1605
2020-03-15 1788 1785
2020-03-16 1324 1321
2020-03-17 2141 2137
2020-03-18 2884 2881
2020-03-19 2802 2802
2020-03-20 3418 3418
2020-03-21 3563 3563

The decreases seem too small to be malicious, but I can think of no explanation for why they’re occurring at all. Can you shed some light on this?


…. Apparently not, at least to me.

As a post-script, I’ve noticed something else odd in the data since I punched “send.” There’s two ways to learn the numbers in Alberta, the official stat website and official press releases. The latter are somewhat sloppy, one time announcing a death that didn’t happen, but in theory both draw from the exact same source.

The Provincial Surveillance Information system (PSI) is a laboratory surveillance system which receives positive results for all Notifiable Diseases and diseases under laboratory surveillance from Alberta Precision Labs (APL). The system also receives negative results for a subset of organisms such as COVID-19.

And yet, when you chart the numbers from the website against those from the press releases, you get this:
Deaths in Alberta due to COVID-19, based on where the data is coming from.To be fair, there’s a reasonable explanation. Since the website pulls directly from PSI, it knows when each death occurred and can properly back-date them to coincide with the time of death. Press releases should not be tweaked after release, so they can only announce deaths on the day the case was confirmed to be COVID-19.

That’s not my question, though. This is: why the heck is it taking a week and a half to confirm someone died of COVID-19? That seems inordinately long, given that most of these people should have had a positive COVID-19 diagnosis before they died. If they aren’t, then given that about three-quarters of all COVID-19 deaths in this province have been at continuing care facilities, that implies Alberta Health isn’t doing systematic testing of seniors despite knowing they were a potential hot spot for two months and vowing to expand testing a month ago.