‘Looking in the rear-view mirror’: A scientist on the challenges with tracking Omicron

TVO.org speaks with physician Jeff Kwong about data gaps — and which indicators we should be watching
By Justin Chandler - Published on Jan 07, 2022
Jeff Kwong serves as a scientist with Public Health Ontario. (Courtesy of Jeff Kwong)

Comments

X

HAMILTON — COVID-19 testing is now severely limited in Ontario, and people with symptoms are being told just to assume they have the virus. With that, many experts say daily case counts are undercounting the real reach of COVID-19.

“Our COVID-19 case numbers are very much underestimated with the rapid spread of Omicron and reduced access to testing. We are just reporting on the tip of the iceberg when it comes to the number of cases that are here in Hamilton,” Elizabeth Richardson, Hamilton’s medical officer of health, said in a news conference Tuesday. “As we go forward, we’re going to start predominantly focusing on the key health indicators that we’re tracking at this point in the pandemic, which are our hospitalizations and ICU admissions, as well as deaths.”

When asked by TVO.org whether public health will change how it communicates data to reflect these changing priorities, Richardson said her team will “continue to look at how we can make those numbers available.” She added that outbreak numbers and the percentage of positive tests will be important indicators to follow.

A man filming in The Agenda studio

Our journalism depends on you.

You can count on TVO to cover the stories others don’t—to fill the gaps in the ever-changing media landscape. But we can’t do this without you.

Agenda segment, Jan. 14, 2021: How tracking Omicron is different

Jeff Kwong,  the program leader of the Populations and Public Health Research Program at ICES — the independent Ontario health-research organization — and a family physician, says that our current ability to understand spread is like “looking in the rear-view mirror.” TVO.org speaks with Kwong about gaps in our current COVID-19 data, what we know about Omicron, and which indicators to watch going forward.

TVO.org: As things stand, how well are our traditional analytics tracking the pandemic?

Kwong: Some of the numbers are going to be harder to interpret moving forward. The case counts are not going to be interpretable. So I think switching over to looking at the per cent positivity is probably the best thing that we can do now, besides looking at the hospitalizations and ICU numbers. But the issue with those is that those are lagged, so what’s happening today will not show up in the hospitals for a few weeks.

TVO.org: Let’s start with the per cent positivity, then. Some experts say that number may also be skewed because only the highest-risk people are now being tested. How should we interpret that?

Kwong: Yeah, there might be some of that. You have to look at who’s eligible to get tested. One of the big groups is health-care workers, and a lot of them are getting infected through the community and through their household members and other contacts, not necessarily at work. I think that still is a good representation of how much Omicron is out there. Right now, it’s about 30 per cent. Whether it’s 25 per cent or 30 per cent — that’s not such a big difference, as opposed to seven or eight weeks ago, when it was around 1 per cent. The difference between 1 per cent and 30 per cent is not subtle. It’s a massive difference, and whether it’s truly actually only 25 because it’s a more restricted population is not a big deal.

Agenda segment, January 4, 2022: The latest on COVID-19 in Ontario

TVO.org: The scale is so massive that it tells you what you need to know, even if it’s not quite as precise as it might be if everyone were getting tests.

Kwong: Exactly.

TVO.org: Public-health units have been saying they’ll be paying close attention to hospitalizations. Can you tell me a little bit more about why that’ll be helpful?

Kwong: One of the big goals throughout the pandemic has been to ensure that our health systems don’t get overwhelmed. And one of the tricky things right now is making the distinction between people who are hospitalized because of COVID versus people who are hospitalized for something else but happen to test positive for COVID. And that’s something that the ministry has asked the hospitals to report separately.

But sometimes it’s hard to distinguish. Let’s say somebody has an underlying medical condition, and it was COVID that caused an exacerbation of that. So it’s hard to necessarily say this one is definitely because of COVID and that one’s definitely not because of COVID. But we’ll see what the data show in the coming weeks now that the hospitals are expected to report it split up.

TVO.org: And you mentioned the difficulty in terms of lag. How does that make things tougher for decision-makers or analysts?

Kwong: It’s kind of like driving by looking only in the rear-view mirror. Anything that you do, you’re not going to see a change in hospitalization for a couple of weeks. It’s hard to assess the impact interventions are making. By the time hospitalization is going up, you’re already kind of too late. People are going to get infected today, and there are going to be even more hospitalizations in two weeks. It’s almost a guarantee. By the time you see hospitalizations going down, the per cent positivity should already have gone down and the case counts have gone down.

Agenda segment, January 4, 2022: Battling COVID-19 frustration and burnout

TVO.org: When our testing capacity better reflected the virus’s spread, would we generally use the number of positive tests and the per cent positivity to predict hospitalizations? And is doing that more difficult now?

Kwong: Yeah, exactly. We always knew there was some under-diagnosis. There would be some cases out there that were not getting picked up. But assuming that everyone who wanted to get a test could get a test, we would pick up most of the cases. From there, the modellers were able to project how many hospitalizations there would be.

This latest round of modelling was challenging because we didn’t know how much less severe Omicron was going to be compared to previous variants. It made it really tough for the modellers to project how many ICU beds would be needed and that sort of thing. They were really accurate at modelling the projected numbers of cases. We knew this back in mid-December, the week that Steini [Brown, co-chair of the  Ontario COVID-19 Science Advisory Table] released the modelling and science-table recommendations calling for the circuit breaker.

“Flying blind” is the expression that a lot of people are using, because we just don’t really know what’s going on out there. We just know that there’s a lot out there, and to quantify exactly how much is really hard.

TVO.org: You mentioned ICU admissions specifically. Is it useful to drill more into the difference between general hospitalization and ICU hospitalization — in the same way that it might be useful to know hospitalized with COVID as opposed to hospitalized because of COVID?

Kwong: I think it is useful to know that distinction. It gives a sense of how many are severely ill. Basically, if you need a ventilator, that means you need to be in ICU. There are many reasons why someone needs to end up in an ICU, but it means that we have to cancel surgeries. Some people need to be in an ICU right after surgery for the first day or so. If there’s no ICU bed, then you can’t do that surgery for that person. At the same time, it’s hard to know what the limit of ICU beds is because you can set up those field hospitals [for example]. Buying ventilators is the easy part, but the problem is that you’re limited by your health human resources. If you have no one to look after the people in the beds, then you’re still stuck.

There are a lot of people being admitted to hospital who don’t require ICU. They don’t need to be ventilated. So that’s good news. But it is still a strain on the health system, because if you have a lot of patients in regular hospital beds, it means that other people who have other conditions can’t be admitted to the hospital. It takes away from the other aspects of the health system in terms of the human resources, the bed capacity — all those things.

Agenda segment, December 10, 2021: Navigating Omicron

TVO.org: The heads of Hamilton Health Sciences and St. Joseph’s Health Care spoke Tuesday about staffing concerns. It’s not just how many hospitalizations there are or how many beds there are but how many staff are at home isolating or at home sick and can’t be at those beds.

Kwong: Yeah, in addition to needing more staff, you’re at the same time having staff ill because of COVID or exposures. It’s a double whammy.

TVO.org:  Do you think that makes it harder to parse the data when we think about hospitalizations?

Kwong: That’s a good question. How many people who need admission are not getting admitted because of concerns about, like, the condition becoming unsafe, and there are just no staff to look after them? I don’t know the answer to that. I would hope we don’t get to that point. But I can’t say for sure. There could be collateral damage.

Let’s say someone comes in for something else, and it’s borderline: Do they need to be admitted to hospital or not? And then the person in emergency might say, “Well, the hospital is full. We don’t have any more staff to look after more people; we’re going to send that person home.”

Something could happen to them while they’re at home. If they had been admitted to hospital, then maybe they wouldn’t have had a bad outcome. And then, in terms of the numbers, there may be some people who in the prior waves would have been admitted to hospitals, but now we’re not admitting them.

TVO.org: Are there any other indicators that you think we should be watching more closely?

Kwong: Well, another one is absenteeism. I think there are some data sources that speak to workplace absenteeism, and when the kids go back to school, school absenteeism. Sometimes these sorts of data are being collected for other purposes. This is what we call syndromic surveillance.

TVO.org: What does that mean?

Kwong: It’s using non-traditional data sources to track for a disease. Examples would be pharmacy sales of medications like anti-diarrheals or something like Imodium. That would track, is there a diarrheal disease going around? And school absenteeism during influenza season, or now, during COVID.

TVO.org: So if we knew so many of such and such a working population are absent, we could then infer that maybe they’ve got COVID?

Kwong: It’d be useful in a broad sense for areas — like, let’s say, “Oh, there’s way more absenteeism in Hamilton than in Toronto.”

TVO.org: There might be people who would read our discussion and wonder, why does it matter to have all these specific numbers if we know that all sorts of people are going to get COVID-19, and we can just assume people are sick and are going to get it? What’s the value in making more precise determinations?

Kwong: Yeah, I think that’s a very good question. If we liken this to just a regular cold, then maybe there’s no need to do any of this stuff. I think what’s tricky is that there are still some people — more than the usual, I think — who would still get very sick, and so having a rough sense of how many people are going to need hospital care is going to be useful. But it’s just at this point really hard to know, based on the lack of data.

Also, I think what’s hard, and this happens every winter, is that we have other respiratory viruses also circulating. Just because somebody’s sick does not 100 per cent mean they have COVID, but we’re just going to say, for now, treat it as if it is COVID, because we have no way to distinguish. I mean, hopefully, COVID does become a milder illness, and most people who get it will not get very sick and won’t need hospital care. That’s the kind of the best outcome, and then they develop immunity, we carry on with life, and they don’t get long COVID. But we’ll have to see how things go in the coming weeks and months.

TVO.org: Long COVID is another thing I wanted to ask you about. Are you aware of what metrics we could be tracking for that and how we can collect information on long COVID?

Kwong: Yeah, that’s really, really tricky. Right now, we don’t have a good handle on that at all. We don’t know what per cent of people are developing long COVID. We don’t have a good definition of what long COVID is. I think there are a lot of things that are being called long COVID that may or may not be long COVID, and then there’s a huge range of variation of what long COVID is, and I don’t know how we’re going to be able to track that, to be honest. But we’re going to have to work on that in the coming months for sure.

This interview has been condensed and edited for length and clarity.

 Ontario Hubs are made possible by the Barry and Laurie Green Family Charitable Trust & Goldie Feldman. 

Related tags:
Author
Thinking of your experience with tvo.org, how likely are you to recommend tvo.org to a friend or colleague?
Not at all Likely
Extremely Likely

Most recent in Hamilton