My children, and future generations, will never know the strangely hypnotic allure of "noise" — the random static flickerings — on an old television screen that has lost its signal. (I mean, unless they go looking for it on YouTube, but no rational person would do that.) Modern TVs just cut discreetly to a blue or black screen or to a screensaver of some kind. I frankly preferred the noise. You never had to stare at it long before your brain — which is, after all, a pattern-recognition computer — started to see images in the random flashes.
Mind you, then again, you can get a pretty good recreation of that experience by trying to read too much meaning into each day's updated count of new COVID-19 cases in Ontario.
The data matters, of course — and no disrespect is intended to my colleagues who put together the charts each day. My point is simply that any single day's number is just a bit of noise, and you can read too much into it unless you have a fuller picture. You see this in headlines a lot, as overworked web editors rush stories online. Numbers "surge," or they have a "big drop" from previous highs. Then everyone with a preconceived notion and an axe to grind seizes the numbers and runs with them to prove their point. On the weekend, Ontario broke through 1,000 daily new cases; on Monday, it was back down into the mid-800s. That seems like a big improvement, but it'll be days before we know whether either number is just a blip. Each day’s numbers are really only truly valuable as part of a pattern, and that pattern isn’t clear for weeks.
Are you appreciating this article?
Donate today to support TVO's quality journalism. As a registered charity, TVO depends on people like you to support original, in-depth reporting that matters.
There's really only one set of daily numbers that matters: hospitalizations. Case counts may give us a sense of our future, but the real number is how many Ontarians are in hospitals and, especially, how many are in ICUs.
This isn't just my opinion. Ontario public-health authorities have explicitly identified hospital capacity, particularly ICU capacity, as the province's metric for large-scale restrictions. They did that almost a month ago and set a lower limit of 150 Ontarians in intensive-care beds. So long as we're below that level, the medical system can function at something close to normal levels and even continue clearing the backlog of tests and procedures created by the first wave. If the number goes above 350, to use the technical term, we're in deep [expletive deleted].
And how are we doing? Thus far, actually fairly well. As of Monday, Ontario had 78 patients in ICUs with COVID-19. (More than 200 others are in the hospital but not in the ICU, which would include some patients who were previously in the ICU but are now on the road to recovery.) This is genuinely good news. Five or so weeks into a second wave, our hospital system remains intact and functioning, with room to spare. Hospitalization is a lagging indicator, so this doesn't mean we're home free and out of the woods. And we shouldn’t lose sight of the fact that some hospitals may well be struggling even if the overall system remains in solid shape — as in so many other ways, COVID-19 does not land on everyone equally.
But still. These numbers are a lot better than might have been reasonably feared even a month ago.
The bad news, of course, is that, a month ago, we had way fewer people in ICUs than we do today. On September 26, instead of 78 patients in the ICU, there were 28. We've almost tripled that number since. There's still a lot we don't know about COVID-19, but anyone who was hoping that this would be a "casedemic" where expanded testing would catch a bunch of cases from people who'd never get sick will be disappointed. When COVID-19 cases rise, hospitalizations rise. No reasonable person can deny it at this point.
(Please note that my counts above are taken from Jennifer Kwan, an Ontario family physician whose Twitter account, with daily summaries of numerous COVID-19-related metrics, has been an invaluable resource.)
The hospitalization numbers are the whole ball game. Relying on case counts was always going to be an inexact metric, because some proportion of those cases was always going to be taken up by relatively young people with very low risk of major COVID-19-related illness. Higher numbers are always worse than lower, but cases are not equal — 100 cases among young Ontarians will, statistically speaking, result in no deaths, but 100 in a long-term-care home could kill dozens. Making sense of Ontario’s numbers was also complicated by the government’s decision to change the testing criteria. Today’s numbers are apples to last month’s oranges.
Again, this isn’t a knock on the raw data or a suggestion that we shouldn’t collect it. Case counts are a useful measure, but only when viewed as part of a larger data set that gives you a rolling average over days or even a week. (Come to think of it, I guess that’s another number that really matters, too.) Analyzing case counts is even easier when they’re broken down by location and age group.
These figures are all available. A small group of dedicated reporters and citizens track all this and publicize the data daily. And God bless ’em for it. But each day, when the numbers come out, ignore the top-line number, and make a beeline to the two that really matter — the rolling average and the ICU figures. If, in the weeks to come, we see that rolling average coming down even as ICU usage creeps up, that will be a good sign: we will be on the right path, and we can be reasonably confident that the hospitalization numbers will follow the case counts down in a few weeks, remaining below the 150 lower-end figure.
On the other hand, if that ICU figure keeps edging up, and the rolling average is still moving in an upward direction, too, well, get your haircuts now, people. We might be in for a long, quiet winter.