Evidence of meeting #22 for Health in the 43rd Parliament, 1st Session. (The original version is on Parliament’s site, as are the minutes.) The winning word was data.
A recording is available from Parliament.
6 p.m.
Professor of Medicine and Public Health, University of Toronto, Chief Executive Officer and Founder, BlueDot
I'm so sorry about that.
We are also working with Global Affairs Canada to support capacity building—as I mentioned in my opening remarks—in 10 countries in Southeast Asia. We're working with the State of California.
In many regards I think we would be very eager to support public health responses across the country and work closely with the provinces and territories. We've had, in some ways, an issue with respect to capacity to do this in the midst of the COVID-19 pandemic. However, we have had engagement at the federal level, and are producing analytics across the entire country on a week-over-week basis, and also with the Province of Ontario.
I'm not sure if that answers your question, but a lot of the analytics are focused on understanding issues related to social distancing and how that is related to epidemic activity. Also, keep in mind that while today we're in a bit of a lockdown, as the economy reopens and we have a highly susceptible population, we're going to have to start.... We may find ourselves in the same place as New Zealand in the future, where we have to start looking outward again and start to think about introductions that could trigger the next wave.
We've been involved in supporting both an internal look and tackling this in our own backyards, as well as monitoring the global situation and potential introductions.
6 p.m.
Liberal
The Chair Liberal Ron McKinnon
Thank you, Dr. Jaczek.
Dr. Kitchen, it's over to you for five minutes, please.
6 p.m.
Conservative
Robert Gordon Kitchen Conservative Souris—Moose Mountain, SK
Thank you, Mr. Chair,
Thank you, everybody, for your presentations today. They've been greatly appreciated.
Dr. Attaran, yesterday I spoke to Dr. Tam and asked her a question about data sharing between the provinces and organizations with the federal government. You've answered a lot of those questions I had for you, but further to that, I was asking her about demographic data, in particular how New York City has come up with a lot more demographic information, etc. She indicated to me that it's on the Public Health Agency of Canada's website, so I took the opportunity this morning to go onto that site. With some help from my staff, I finally managed to find some information on that.
They talk about updating the data as of today and about 4,201 cases of clinical presentations, and of those, 561 cases or 13% were clinically or radiologically diagnosed with pneumonia. My point about that is it provides a lot of information and then, all of a sudden, I find a little bit further down a little statement: “The epidemiology update is based upon information received for 38,746 cases. Not all data fields are complete, only cases with data available are included.” The bottom line is they're providing inappropriate information on the data that we have.
How is it that we ask you or other epidemiologists to come up with data and provide modelling when we put this out with inappropriate information?
6 p.m.
Prof. Amir Attaran
Dr. Kitchen, thank you for a very intelligent question. You're exactly right. You mentioned there were roughly 38,000 cases in the data that you looked at. I'm going by memory here, but I think we've had about 80,000 cases reported in Canada so far, so that's under 50%. What that means is that at the high-water mark, anyone like Dr. Fisman or Dr. Khan doing modelling, or me when I do it in my amateurish way, are working with less than half a deck.
6 p.m.
Prof. Amir Attaran
There are obvious problems with that.
6 p.m.
Conservative
Robert Gordon Kitchen Conservative Souris—Moose Mountain, SK
When the Public Health Agency is making these decisions based on World Health Organization data, which is maybe coming in from China or wherever, which is inappropriate, again how do you come up with that proper information?
Dr. Fisman, do you have any comments?
6:05 p.m.
Professor of Epidemiology, University of Toronto, As an Individual
I'll tell you, my group at University of Toronto call ourselves “data raccoons”, because we've sort of managed to thrive for about 15 years on data that most people regard as garbage, so it's sort of a bit of the normal state of affairs for us with public health data analysis. The stuff we have is pretty good by our standards.
Working with folks here in Ontario, there's been a modelling table convened over the last few weeks. We've been given access to case files. There's a lot you can learn, but there are also a lot of fields that are missing. We could potentially do better, but I think it's also important to remember that those fields are being filled in by very harried front-line public health epidemiologists.
I suspect that what you're seeing from the Public Health Agency of Canada is that they're putting out the data where they have complete fields, and that it's their way of dealing with missing data. Missing data is just part of epidemiologic data analysis. It happens no matter how good the data are that you have. I'd sort of want to know more about how they've made those decisions, but sometimes it's good enough.
6:05 p.m.
Conservative
Robert Gordon Kitchen Conservative Souris—Moose Mountain, SK
That's a challenge, though, when you don't have proper data and you don't understand that.
I'm going to go on a little bit further.
Mr. Ciciretto, you talked about high-quality antibody testing. We've heard a lot from you today and all of the witnesses about testing. Last week Health Canada approved the first serological test for detecting antibodies in those who contracted or may have contracted COVID-19. The approved serological test comes from an Italian biotechnology company.
Do we have the capacity to produce these tests domestically? Do you know that?
6:05 p.m.
President and Chief Executive Officer, Dynacare
The answer to the question is, we do have capacity. The particular test that was approved was from a company called DiaSorin. We don't have that testing platform, in particular, so that's critical. Could we acquire it? Yes, we could acquire it.
There are other companies that we're working with right now, large diagnostics organizations that are looking to get Health Canada approval as well for a serological test. Once that happens, I have 200 collection centres and 850 phlebotomists who can collect those samples and bring them into a laboratory and onto existing test platforms that we have today that could do that testing quickly, efficiently and accurately.
6:05 p.m.
Liberal
The Chair Liberal Ron McKinnon
Thank you, Dr. Kitchen.
We'll go now to Mr. Kelloway.
Go ahead. You have five minutes, please.
May 20th, 2020 / 6:05 p.m.
Liberal
Mike Kelloway Liberal Cape Breton—Canso, NS
Thanks, Mr. Chair.
Hello, colleagues.
I want to say a really special thank you to the witnesses today. I'll echo Dr. Jaczek's that it's interesting to see such a rich series of viewpoints, insights, opinions and also backgrounds in a variety of areas. I really appreciate it. It's very illuminating.
Dr. Khan, I find your work very fascinating in terms of the technology you use. I don't necessarily want to look to the past but to potentially a second or third wave.
Can you talk about how your technology may be able to be used to track and identify, in many ways, a second or third wave? Could you illuminate a little bit what the biggest risk factor is that could trigger the next wave?
6:05 p.m.
Professor of Medicine and Public Health, University of Toronto, Chief Executive Officer and Founder, BlueDot
Maybe the way I could sort of frame what we have been building with the metaphor of a smoke detector and fire extinguisher. For six and a half years, we've been building systems to be able to detect threats early, because we know, as I mentioned, that time is our most valuable resource.
To be able to quickly go from detection to what kind of risks we are facing—not just from the dispersion of the disease, but what kind of disruption might occur—is very important because diseases spread around the world all the time. They don't all cause outbreaks or pandemics. That is a complex requirement because every disease behaves differently. Zika virus is different from Ebola, which is different from COVID-19 or measles for that matter.
We've been spending a number of years building up that capacity to have a bird's-eye view of what's happening around the planet, to be able to relate it to geographies across the planet, and to do this in, really, a matter of seconds.
With respect to once an outbreak starts to spread and is now occurring locally, this is where we have been using—again, I want to underscore, anonymously—just the pings, the digital locations from hundreds of millions of mobile apps and mobile devices.
That kind of information can help us understand. Ultimately, this is a virus that spreads, as Dr. Fisman mentioned, through the movements and interactions of people. These are really rich datasets—over three billion data points a day—that can really allow us to understand how those movements are occurring so that we can then start to anticipate how the epidemic might evolve. It also allows us to generate insights about some of the non-pharmaceutical interventions like physical distancing or recommendations for quarantine. Are those being adhered to at a population level?
I do want to highlight that we're not tracking anyone who is infected or their contacts. We're looking at population movement.
With respect to going forward to the next wave, I think the simple reality is that no one really quite knows what this is going to look like or exactly how it's going to unfold. We are dealing with a completely novel disease. Certainly, we have concerns that as we get into the latter months in the fall.... We know that coronaviruses tend to be in cooler, drier climates where they may be more efficiently transmitted. As that occurs, currently we are relating a lot of this mobility data to understanding how the epidemic curve is evolving. Perhaps there are lessons that we can learn about which geographies and which locations seem to be opening up society in such a way that they can, you know, generate some sense of normalcy and some kind of economic activity without having an exponential increase in the epidemic. I think that's really the $6,400 question. How do we do this gracefully? How do we thread the needle?
These are things that I, candidly speaking, don't know anyone has the answers for just yet. I think it goes back to the point that surveillance, testing and monitoring are critically important, because as we start to reopen society, it is going to be incredibly important for us to be watching very closely what the response is in terms of epidemic activity and transmission.
I hope that I perhaps have given you a little bit of a sense of what we're thinking going forward.
6:10 p.m.
Liberal
Mike Kelloway Liberal Cape Breton—Canso, NS
You have.
I think you mentioned measuring disruption in society. Maybe I'm miscategorizing that.
Can you unpack that a bit?
6:10 p.m.
Professor of Medicine and Public Health, University of Toronto, Chief Executive Officer and Founder, BlueDot
Yes, thank you. It's a really big and very important issue.
The four Ds that we work on are detection, dispersion, disruption and dissemination. Detection speaks for itself, early detection. Dispersion mean, how do these things leap across continents in hours? How do we anticipate the next move? Without getting into a lot of detailed epidemiology, what sometimes is called the infectious disease triangle is a disruption or an outbreak that really lies at the crossroads of the characteristics of the microbe or the germ itself, the characteristics of the population, and the environmental conditions.
The Zika virus is not going to spread here locally in Toronto, because there's no mosquito and it's too cold; it might spread in Miami in July, but maybe not in January. That is a very complex set of data and we're bringing in hundreds of data sources, from real-time satellite data to insect observations, demographics, etc. We can do this for over 100 different diseases so we can try to get a sense of whether the necessary ingredients are there for this to actually cause a disruption, an outbreak.
As you can imagine, this is not a data problem. It requires deep subject matter expertise integrated with deep data analytical expertise and data science. This is the area we're actively involved in. We're well on our path and well on our way, but this is a formidable challenge that really is going to take years.
6:10 p.m.
Liberal
The Chair Liberal Ron McKinnon
We go now to Mr. Thériault.
Mr. Thériault, please go ahead for two and a half minutes.
6:10 p.m.
Bloc
Luc Thériault Bloc Montcalm, QC
Thank you, Mr. Chair.
My question is for Professor Fisman. Perhaps Mr. Schabas can express his opinion as well.
We do not yet have a vaccine or antivirals. Serological tests are just beginning. Faced with the desire for reopening, we have suddenly and a little hastily seen the notion of herd immunity appear. But there is no real certainty about the exact data, about the connection between COVID-19 and herd immunity.
Can you tell us where we are at in terms of knowledge or studies on herd immunity with COVID-19? Can you describe the situation?
If reopening were at an ideal rate, would we achieve herd immunity? At what rate would we need to achieve it to make everything safe?
6:15 p.m.
Professor of Epidemiology, University of Toronto, As an Individual
Thank you very much.
There are a lot of moving parts here. Herd immunity can be approximated as a function of the reproduction number of the disease. The higher the reproduction number, the more people need to be immune if the disease is not to take off. That's why we see measles outbreaks when vaccine levels fall off just a little bit, because the reproduction number for measles in a susceptible population is about 20. You can get about 20 new cases from an old case.
This is a much less infectious disease. The reproduction number is somewhere between two and three. That means you need somewhere between half and two-thirds of the population to be immune to have herd immunity, so that if you bring an infectious case into the population you won't don't have an epidemic.
Where are we right now? We don't know. I've been doing a running meta-analysis on seroprevalence studies as they've come out. I'm up to about 50 of them. You can compare antibody prevalence in populations to what those communities think they have going on in the number of cases they have. It's called a cumulative meta-analysis, just adding study to study to study. The long and short of it is that I think we probably detect about 7% of cases. We have an inflation factor of somewhere between tenfold and twentyfold.
If we look at Canada with 80,000 recognized cases, that would be somewhere between 800,000 and 1.6 million cases in reality. That puts us—I'm going to get hung up in trying to do the math on the fly—at 4%.
If we're there now, New York is well ahead of us. New York has good seroprevalence data. They're at about 15%, but they had to go through hell to get there. They did experience a wholesale collapse of hospital systems in much of the city, including the Bronx and Queens, to get to 15%. That means they might be able to get to 50%, 60%, or 70% herd immunity by going through that a few more times. I don't think they will allow that to happen. They've lost approximately 20,000 New Yorkers of all ages, I would add, to get to that point.
What we have to do right now—a lot of countries around the world, indeed a lot of provinces in Canada, show us that we can knock this disease down to low levels and then we can use good public health practice. I agree with Dr. Schabas that you can't do contact tracing if you're having 200 cases a day, as we are in Toronto. It's just too much. If you're having five cases a day, you sure can. If you're testing a lot, you sure can. You need to use the distancing to knock the reproduction number down. We're still at around one in Quebec and Ontario. I would add that the Canadian epidemic, at this point, is a Quebec and Ontario epidemic. The other provinces have got the job done at this point. If you can do that, then we can start to use other public health measures, like contact tracing, to keep a lid on this and get through the summer and allow the economy to reopen.
We haven't touched on masks at all. There's pretty good ecological evidence at this point that the countries that are doing much better than us are mostly mask-adopting countries. You can argue the science, and we can have a symposium in five years about who was right, or we can use the precautionary principle and move towards masks now, which I think Dr. Tam has started to do.
We can do a lot to keep that reproduction number low and reopen our economy to a degree, and muddle through.
Exciting stuff is happening with vaccines. There are RNA vaccines that weren't on the table 10 years ago. There's a really exciting live virus vaccine from the U.K., where AstraZeneca, the pharmaceutical company, is manufacturing the vaccine at scale while the trials go on. If the trials are a success, they're going to have millions of doses ready to put into people's arms.
We need to avoid mass death situations until we can get through to a point where we can effectively deal with this pandemic. We will, but it's a matter of tenacity, patience and competence, and that's very patchy across the country. Some places have shown it; other places haven't. I'm sad to tell you that I feel that my province, at a provincial level, is one of the places that hasn't shown that, although individual local public health units have really shone and distinguished themselves.
6:20 p.m.
Liberal
The Chair Liberal Ron McKinnon
Thank you, Mr. Thériault.
We go now to Mr. Davies for two and a half minutes, please.
6:20 p.m.
NDP
Don Davies NDP Vancouver Kingsway, BC
Thank you.
Dr. Fisman, last week, you stated, “I continue to be concerned that there hasn't been enough attention given to epidemiology in kids. I know folks are starting to study that in Germany and Switzerland, but we haven't really studied it in North America.” Then you said, “For those of us who have been really concerned about the possibility that children may be important vectors of this disease..”.
Given that we haven't done much research in North America and your concern that children may be important vectors, how do we square that with sending our kids back to school?
6:20 p.m.
Professor of Epidemiology, University of Toronto, As an Individual
Honestly, it's a dilemma.
I think I mentioned earlier that the signature of this disease is that it takes off with big gatherings, so there's a lot you can probably reopen economically, safely, if you stay away from large gatherings of people. The one big gathering that's really, really tough to cancel—and which has huge economic implications—is at schools. That's the hardest thing.
The reason to be concerned about aggregating kids is that we see evidence from other respiratory infectious diseases that kids don't die of them, but they are tremendously good at transmitting these diseases.
6:20 p.m.
NDP
Don Davies NDP Vancouver Kingsway, BC
That being the case, why would we be sending kids, who are vectors of this disease, to gather in large gatherings and to come back to homes where they may be in contact with seniors?
I don't see what the dilemma is there. What is the dilemma?