This was a fascinating, illuminating, and enraging conversation with Dr Richard Denniss, Executive Director of The Australia Institute, and Professor Margaret Hellard, Deputy Director of the Burnet Institute, about why we’re not collecting covid data, and why it matters. This one is really important, so please listen, and share.
MH: “People and their communities are at the centre of this, and one of the things that I think we did really poorly all the way through, and continue to do poorly, is speak to people, about what their wants and needs, and their communities’ wants and needs are. It’s just really important to keep people and community at the centre of it, so that when we’re collecting data, we are actually collecting data that addresses the critical wants and needs of those communities. So that we go back and ask communities: “How has this impacted on you?” So that we keep that at the centre of our thinking.”
RD: “Don’t get me wrong. This is really important, and we really should be collecting a lot more data, and encouraging people to do the simple things that the data we do have says would work, but this is not the only big public policy problem that our policy makers are wilfully blind to. It would be wrong to think this is a covid conspiracy. This is a well worn path. You know, we have terrible data on greenhouse gas emissions by industry in Australia, because hey! Maybe we don’t want to embarrass anybody. We have terrible data on lots of big problems and we have really really good data on some things that powerful groups are really interested in. So unfortunately the map of what we have data on is as interesting as the maps within data. A lot of people think that these things are just accidents. I have a very simple predictor for what we have data on, it’s groups that were powerful ten years ago, because there’s a long lag on data.”
RD:”Why aren’t we collecting the data? Because they don’t want to admit failure. They don’t want to make it easy for me to tell you what the cost to GDP of this heroic approach to covid has been.”
MH: “The story of vaccination, and getting those levels up, is really important, because that’s what’s going to protect us into the future, from really yucky outcomes of covid into the future, as we have waves continue to come through. So this is why it’s really important that we understand what happened. So that people don’t think it was all a waste of time. It wasn’t a waste of time, we saved a lot of lives, but good public health measures remain important into the future.”
Transcript
Linda: Welcome to another episode of Make Me Data Literate. This one’s a little bit different in that we don’t have a single guest answering the same old questions. This time I want to look at the issue of the COVID data that we’re not collecting and what impact that has on how we understand what’s currently going on. And I’ve been concerned about this for some time as a data nerd, but I hadn’t heard anyone publicly comment on it until Richard Dennis was speaking on a panel in January and was like, well, we don’t know what’s happening with COVID because we’re not collecting the numbers. I was like, finally, somebody’s actually put a voice to it. And from that, the idea for this podcast was born. So I’m very grateful to have Dr Richard Dennis, the executive director of the Australia Institute and Professor Margaret Heller, deputy director of the Burnett Institute and epidemiologist. Thank you both for coming. This is awesome.
Richard: Thank you.
Margaret: Pleasure to be here.
Linda: So I want to start with you, Richard. The World Health Organization has declared the COVID-19 emergency over, but the rest of the message that the pandemic is very much ongoing seems to have been lost. Why do you think governments are so keen to pretend that COVID is over? Because it’s easy to and because in Australia, I can only speak to the Australian politics. I think, you know, for the current federal government, everyone remembers that the previous Liberal government made a mess of it towards the end. They coped with the beginning of it quite well from an economic point of view. There was a lot of stimulus. You know, they closed the borders. There were lockdowns. You know, they took the scientific advice.
And then by the end, you know, the disaster with, well, it took too long to roll out the vaccines. And then there was the debacle with the rapid antigen tests. So, you know, and to the extent that the public think about it, they remember that the Liberals made a mess of it.
So, for Labor now, if Labor were to set any targets for anything, place pressure on anyone to do anything, then if they missed the targets, they’d fail too. Or if they suggest people do something unpopular, they’d get attacked too.
So, unfortunately, you know, everyone likes bipartisan consensus. Everyone likes it when the political parties agree. The real danger when the major political parties agree in Australia is that important issues become invisible. And, you know, that’s the case, you know, with a wide range of things. You know, why don’t we ban junk food advertising? Neither party want to talk about it.
You know, so unfortunately, we’ve got bipartisan consensus that maybe we don’t really want to emphasize this problem anymore. So, then you get this disconnect between what’s politically convenient and what the science says we need to do. And, you know, measuring obviously is part of that. How many people have got COVID today? Dunno. How’s the air quality at my kids’ school? Dunno. What’s happening? We don’t know. And because there’s no numbers to respond to, there’s no pressure. So, the lack of data and the political consensus that we can ignore it are not accidental. They are fully integrated.
Linda: I would really like to head the desk at this point. It’s very frustrating to me. You know, I like data and I like to know what’s going on and I have a science background and I want all the information so that I can make informed decisions. It feels like we’re missing it. Margaret, do you have anything you wanted to add to that?
Margaret: Yeah. I mean, I think you sort of alluded to the word that they’re sort of wanting to pretend it’s not… I just think it’s… I agree. It’s not like we’re saying… And I don’t even think governments are saying it’s over or not over. They’re just not talking about it. And if you don’t talk about it, you don’t give oxygen to it. So therefore… So I agree with Richard’s point there. Absolutely.
Also I think even whilst we had community concern about COVID, we’ve not actually had a COVID plan in our country. And it’s something that I’ve often asked for when we were doing the modelling, for a long time. So, early on there was a brief period where we had a plan. But if somebody said… And I would ask this question because I’m a very simple girl so I have to ask simple questions. If you want me to model something, what am I modelling it for? Like what’s the outcome measure that we’re trying to… What are the guardrails that we’re trying to stay between? Are we wanting to ensure deaths stay below a certain number? Are we wanting to ensure hospitalisation so we don’t overwhelm our hospital system? Even if you don’t want to count total numbers. Are we wanting to talk about economic impact of COVID and its sequelae, both sequelae like long COVID, but also unintended consequences of public health actions that seriously impact on people? What are the things you’re wanting to measure or to know about as this is a critical point where we will act or not? And to be perfectly honest that’s not been there almost forever. Which means that you don’t even know what you need to measure most accurately. Because to me, you measure stuff, when you’re trying to get data and say what am I going to measure, you’re measuring it for a reason. Otherwise it’s just measuring. And as I say to people, most of us learnt to count to 10 and learnt to count quite early on. So really when you’re going to measure something, you actually say what do you want me to measure it for? Because that makes a difference to what I measure and how I measure and when I measure and what I measure. Because then I will be able to help you with your decision making.
But really, because we’ve got no target at all as to what is a good or a bad outcome over the next couple of years due to COVID, and I mean by that both the disease but also the impact of any measure that you want to put on. It’s a really interesting thing to observe that it doesn’t really matter. And as Richard said, we’re currently kind of just wandering along.
Linda: At the height of the lockdowns, before we had vaccines and before we really understood the ongoing implications of COVID to the extent that we understand them now, reporting a positive test was mandatory and we had a lot of PCR testing and everything. So there were fairly high testing rates. But the numbers we had were still not the numbers of COVID in the community. They were the numbers of positive cases reported, which is.. we hope was close but not necessarily the same. How helpful do you think that data was for managing the outbreaks and how wide was the gap between the data we had and the reality of the situation?
Margaret: So early on and so if you know during 2020, during 2020, once we got tests and 20 because you might remember there was a period when we actually didn’t have any tests, so very many. And so during 2020, second half of 2020, 2021, probably pretty good in terms of: were the proportion of tests being done, the proportion of tests that were positive and the proportion of tests that were being reported as positive, reflecting what was going on. Probably pretty good.
Once we began to do vaccines, it meant that we probably had more asymptomatic infection, which would make it then that somebody quite reasonably might not test because they just didn’t recognise that. So you’ve always had that component of it. But even during, so the Delta outbreak, we were pretty confident we were reasonably close and we had a clear idea. By the time we were wandering into the December, January of 2021 with the Omicron outbreak and where we had rapid tests coming in and the antigens, that’s where things got really much more difficult to work out what was going on.
So we had a combination of factors, a lot more infection, some of that effect. So we had vaccination background, so some people might not be getting symptoms. We had a lot more infection. We had a scarcity of tests for a period, which impacted on even if and we did work on this. I was involved in modelling of COVID, but also we’re following a cohort of people, the optimized study where I was asking people what was going on. So every month they would tell us what was going on and we know we did what we call a snapshot survey to say we want to know what… we know what you did last summer, where we could see people were trying to get tests and struggling to get tests because we did both quantitative and qualitative work there.
So it’s probably in the beginning of that 2022 coming off the Omicron where we began to really have much less of an idea of how much the case numbers were being, how much the reporting of cases were being, were the truth. And if you said to me now, do we know what the truth is, we certainly do not know what we would call the truth. We have guesstimates and estimates of how under we are in terms of what gets reported to what’s the truth. Ways we try and extrapolate that from hospital data and from death data, but even the hospital data can’t with confidence say it’s necessarily reflecting the full truth.
And people, this is a podcast, so I’m going truth and then doing that inverted commas signal around the truth because it’s very rarely do we understand the truth. So now it’s a real struggle. But that’s sort of the narrative of how close we were and then when we began to not be as close and the reasons for that, which are multiple factors.
Linda: Yeah. Richard, do you have anything you want to add there?
Richard: Oh, look, I mean, economists use rubbish data all the time because it’s better than nothing. So, when you hear about the Reserve Bank managing the economy, it’s economists say it’s like trying to drive a car looking in the rear window. The data is usually three to six months old. And by the way, we’re always updating the data. So the data that we were using three months ago, we’ve updated it since then because more came in. So knowing things is a choice, not knowing things is a choice. Some data is better than none. And your initial question was about, well, we had data on the number of people who were testing positive. We’d never had data on the number of people who had it. But look, if you know there’s a relationship between those two things, if you think it’s really unlikely that the number of people testing for COVID would double when, and there’s no theoretical or empirical reason to think that the number of people with it hasn’t doubled. Like, as long as you’ve got a good map in your head of how the data, imperfect as it is, how it maps to reality as you understand it, then imperfect data is still a very useful tool, especially in a rapidly changing environment. And to kind of wait for perfect data rather than act on imperfect data, would an economist would say that would be a heroic approach.
So, you know, like, we have, let’s be, let’s be clear, like, when we choose to collect enormous amounts of data on GDP, it’s very expensive to collect, on the consumer price index, very expensive to collect, that data is imperfect. Our data on the consumer price index is only collected in capital cities. So you know what we know about the price of groceries in Newcastle? Nothing. You know what we know about the price of groceries in remote Western Australia? Nothing. But do we still use the consumer price index? Of course we do.
So we could spend more money to collect better data on prices across the country. We’ve chosen not to do that. Most people live in the capital cities. We think there’s a proxy relationship between the two. You know, we as, you know, we as scientists, we as citizens of a rich country know how to collect more data on things we care about, how to draw a line about where to stop collecting the data, and how to use imperfect data without confusing ourselves that the map is the reality.
When it comes to COVID, let’s be clear, we don’t want to know. Right. We don’t want to know. And we could be running, you know, as they were during the UK, maybe still are, I don’t know, during their crisis. You know, you can run big samples out there. Right now we could be out there paying 10 bucks to sample people on the street. Right. We could. We could do anything we want. What are we choosing to do? Nothing.
Margaret: There are, rather than the UK, which to be perfect, we should say a big thanks, shout out to the UK for collecting amazing data at the beginning of their pandemic, which was used all around the world. And that’s the other thing is that that data was used. We were all using that UK data, but it was wickedly expensive as to how much it was costing them every day to collect that data. When I asked for that amount of money, because I’m not sure that, so I’m not going to say the exact amount, but I’m just going to give it a comparison. So the daily cost of that UK data collection, I would have been, you know, I asked one of our governments if I could have something like that over a sort of a six month period and I couldn’t get it. So that’s Richard’s right. It’s a matter of what you’re prepared to spend.
But also I do think we, as I said, there’s no point in, I mean, we collect, we look at data and there is data that we can extrapolate from and we continue to do modelling. But if somebody said to me, what does the government want to do with that data? Governments stopped probably wanting to do a lot with that data for a while now. And that’s the issue. It’s people, you know, in the websites, like around the world, that were collecting data, have all, you know, where you’d go and have a look at what was going on. A lot of them are now, they don’t get the funding that they got from the universities or the governments to do it, because there’s been a decision that that’s just of not, of is greater importance anymore. And part of that is if you are not going to act on data, as governments are choosing not to, then don’t collect it, because it is wasting money to pay me to collect data if you don’t want to act on it.
Now, we could have then that discussion of, well, don’t you want to act on it? And why? But as I said, that discussion hasn’t happened in my view, since… I remember doing a model, I’ll be really clear about it. I remember doing a model in and got into, and all sorts of people said rude things about the quality of my models, my team, in May of 2021, where we were showing what would happen when we got vaccinated and the level of vaccines and some assumptions around that. And then we said that if you just let it run, this would be what happened. And if you public health measures and people looked at it and said, that’s just, I can remember it was like, no, no, no, have a look at this model. This is really important.
What it means is that if we want to keep some level of control of this pandemic, even following vaccines, you will need to have, and I’m not talking lockdowns, but some moderate public health measures. And we got knocked around for that piece of work. I’m giving a talk tonight to some public health registrars about work we did. And I was looking at that model and the predictions of that model, we were spot on. Like, I feel like going, you know, nobody cares now. And it’s like, oh, we were right, you were wrong. But what I’m saying is we were spot on. And I look at it. And I think, but people weren’t interested in knowing that because in May of 2021, we were interested in vaccines being rolled out to stop lockdowns and to stop having to worry and talk about this anymore.
And the government narrative was that, even though the Doherty put out a model where it talked about what needs to happen, they also said to keep some level of control and to stop certain things happening, you would need to maintain some public health measures, some surveillance measures and these things. That never got imparted in a clear way by the government speaking of that or many people speaking about that model. And yet it said exactly the same thing. All of the models said exactly the same thing.
But most people got to a stage and this is where government is pushed by community and community is pushed by government and a lack of information, as Richard said, is people just don’t want to talk about that at the moment because the thought of some of those public health measures and my view, not even all the time, but as we’re going into a wave so that we can stop quite so many cases so we can stop quite so many deaths is something that most people don’t want to have a discussion about.
Linda: Yeah, I get very frustrated as someone who’s at high risk and has already had long COVID. The lack of discussion about masking and air filtration, which seemed like very simple and low invasive, you know, we’re not, as you say, we’re not talking lockdowns, we’re talking about air quality and simple measures that can be taken to keep people safer and we’re just not talking about them at all.
Margaret: It’s a really interesting kind of question and it’s to me, I look at certain other things that we do in response to issues. So nobody says that a seatbelt stops a car crash, but it reduces the impact of a car crash. Nobody says airbags stop a car crash. Nobody says improving the camber of your roads stop a car crash. Nobody says this. All of these components, each brought in thoughtfully, carefully, some external, some behavioural, reduce the likelihood of a bad outcome following a car crash or reduce the likelihood of the car crashing.
But not one of those things alone stops any of that. But it’s as if people say, oh yeah, but you talk about air quality, but you know, it doesn’t work. No, it doesn’t work 100%, but it works this amount and it can be, people talk about masks and again, I totally agree masks don’t work perfectly and I know that some people really struggled to wear masks. But what we clearly saw is under certain circumstances when explained properly, the vast majority of people were prepared to wear a mask under certain circumstances, not all of the time and I totally get that. And in certain places maybe, for certain periods during a wave, maybe they would if it’s explained in a way where “this is not perfect, but the likelihood of you not getting sick or somebody that you know not getting sick or a total stranger…”.
But we want it to be all or nothing, we’re making it all very black and white. When we know in many other things, nothing is perfect, but we work towards improvement.
Linda: I think at the beginning, when we were doing the contact tracing, it was quite easy for people to understand why it was important to know who had COVID because they could track who they’d been in contact with and who else might have COVID and try to nip the spread and the but once the outbreaks grew and contact tracing became impossible, why was it important to keep collecting the data? Like how does that feed in? You talk about the models that you make, how important is that data to the modelling that you do? Why does it matter?
Margaret: Well, Richard alluded to this before. The better data you have, the more likelihood that your models will be accurate and the, kind of the error bars around what you’re doing go down. So then the nature of the advice you can give with the level of what I’ll say, certainty, as in this is closer to the truth than not, that really helps.
It doesn’t mean we can’t model at the moment. We do stuff where we make assumptions, we make calculations, we do this and that, but if somebody said to me, how certain are you? There’s the strength of my recommendation of the interpretation in my view goes down. And so then that’s just the reality.
But to me, our models and other people’s models were being used to try and get an understanding of, really, and this is where making sure you spend money in the right time, in the right place, in the right moment, pushing out vaccines as fast as possible, making sure that key areas are getting those high levels of vaccination up before you remove restrictions, particularly with the Delta virus and those. because otherwise, morbidity, deaths, they would be higher.
So there was the ways that you could model that to say, if we get the timing right, it’s the Paul Keating, soft landing after a recession or whatever we’re meant to have after economic, whatever, you’re aiming for a soft landing, not a perfect landing, but a softer landing often. And the modelling, that’s essentially what we were trying to work out, is the better we could have data, the more we could say, well, we can begin to reduce these restrictions.
What people don’t realise, and in fact, we’re sort of in the process of getting that data out in terms of being into the public domain, published, because people say, oh, those lockdowns, it did nothing, it was all terrible. Thousands and thousands and thousands and thousands and thousands of lives were saved in Victoria alone because of the lockdown that we had. And people say, well, those people might have died anyway. No, we’ve looked at that. So we’ve been kicking the can down the road.
So the lockdowns saved lives, significant numbers of lives before we had vaccines come out. And I think it’s really important that people are aware of that so that they don’t think it was… because sometimes they get the feeling people think, oh, that was all just a waste of time. No, it wasn’t. We stopped people from dying who are alive today and well and doing things because of our actions. And I think our governments should be complimented on that because it was tough and brave decisions.
But I don’t think that that narrative is out there in the public domain as it should be, as to the number of lives that got saved by those actions. And people should be complimented and thanked for that properly. We’re in a different circumstances now in a predominantly vaccinated population. But critical, I think, is that the story of vaccination and getting those vaccination levels up and getting out is really important because that’s what’s going to protect us into the future of really yucky outcomes of COVID into the future as we have waves continue to come through.
So to me, this is why it’s really also important to make sure that we understand what happened and that the story is clear so that people don’t think, oh, that was all a waste of time. It wasn’t a waste of time. We saved a lot of lives. But vaccine, good public health measures remain important into the future.
Linda: Richard, the Australia Institute’s goal is to change minds. What are we missing in the, in terms of changing minds around COVID, in terms of people actually protecting themselves? And we’ve seen vaccination rates falling, even though the bivalent vaccine is available now. People just aren’t taking it up in the same numbers. What are we missing in that changing minds?
Richard: Oh, we’re missing leadership. We’re missing political leadership. We’re missing a Prime Minister or a Health Minister talking to people about simple steps that Australians could take to save themselves and their loved ones and their workmates and their community, an enormous amount of pain and suffering.
I mean, you referred to a piece I wrote, I think late last year. I mean, in December, the Health Minister put out one press release about COVID. The federal minister, Mark Butler, put out one press release about COVID, one press release about Japanese encephalitis, and I think one press release about Monkey Pox. One each.
So guess what? People don’t think that this is a big deal. People don’t think that there’s much they could be doing. Why do they not think that? Because their elected representatives, responsible for their health, don’t want to talk about it. And as I said at the beginning, why don’t they want to talk about it? Because there’s no political upside in talking about it.
Now anyone that says we should do something runs the risk that the thing they propose upsets someone, which it probably will, or doesn’t achieve its goal, which it can’t be guaranteed. So why take the risk? As Margaret said before, seatbelts don’t stop car crashes. But don’t forget, Margaret, how hard the car industry fought against making seatbelts compulsory. Don’t forget how hard the car industry fought against. Not just once they were installed, they didn’t want them to be worn compulsorily.
Because they don’t want people thinking about safety when they’re thinking about cars. They want people thinking about prestige and being attractive to the opposite sex and buying freedom. And no, all that’s – they’re in the job of making money from selling fancy cars. And if people were thinking about safety when they’re buying a car, then they wouldn’t be thinking about flashy extras. And the car company didn’t want to say, do you want an optional seatbelt with that? They wanted to say, do you want a spoiler on the back or something?
So let’s be clear. Capitalism is really good at getting us to focus on some problems. And people that sell stuff to make profit are really good at saying, don’t worry about the downsides here. So the retail industry, the entertainment industry, there’s a whole bunch of people who don’t want people wearing masks in the theater because it makes some other people wonder, should I be at the theater?
So our debate has radically changed. When we were in the middle of COVID, I’ll sound like an economist here, rational people were hoping that other people would wear masks because other people might make them sick. And even if I wasn’t sure that the masks would protect me, I knew that I didn’t want the hospital system to be overwhelmed in case I needed it.
So back when I was worried about my health before I was vaccinated and back before I was worried, when I was worried that the health system might be overwhelmed, I wanted other people to behave in a way that made sense. Now that I think I won’t get sick because I’m vaccinated or it’s not that bad, we’ve all had it, now that I’m not so worried about me and I’m not worried that my health system won’t be there when I need it, what do I care how everyone else behaves?
Linda: What, we’re seeing though, there’s a lot of businesses having to curtail their activities because of staff shortages and the education system, schools just can’t get casual relief teachers and they’re constantly having teachers off sick and students off sick and the impact, there must be an economic impact to being to being blasé about COVID that surely overwhelms the impact of actually tackling it.
Richard: Yeah, you’re right. We do need to increase migration rates to address the skills shortage. Yeah. We don’t even link the two. We don’t even link the two. We talk about skills shortages and we don’t talk about the number of people off work with COVID or long COVID. See, there’s no data, so why connect them? Imagine if I had weekly data on the number of people who were off work with COVID this week, you know what I could quickly do and say, well, hang on, that looks like about half our current skilled migrant intake this year. So if the reason that we need to bring in so many people to Australia is to fill a skills shortage, look, it seems that half that skills shortage at least seems to be caused by COVID, but boy, that might require data.
Margaret: And it also requires, as Richard just said, a little bit of winners and losers and it’s not political, necessarily, to like to have the idea of winners and losers. So like you said, if I wear to the movies, a mask, because we’re being asked to – which would make sense coming into a wave – if that meant that the picture theaters weren’t full, they would say, well, they’re the losers and that somebody else is the winners here.
So it’s that sort of nature of winners and losers and compensation for losers at certain stages where early on in the pandemic, we did sort of acknowledge that. But very quickly, we, you know, because of economics and I presume various things and I’m not an economist, so I don’t fully understand these things. Somebody I have always assumed went and did a model that I never saw, because I always felt that a narrative changed around the benefit of interventions to support people, whilst we managed things at sort of key times. I assume that the model showed that there was no benefit because going, we went from having a situation where we were going, how do we balance this a bit to going, pfft, not so fussed anymore. And I always assumed that there was some work done, which might not be in the public domain, that simply said, there is no economic benefit at all to doing anything. But Richard’s – people can’t see this. I can actually see Richard. He’s going “No Margaret, you’re being way too logical here.”
Richard: No, Margaret, as someone who’s, you know, I’m a long-term escapade from academia, but my actual academic expertise is in macroeconomic modelling. And I assure you that anyone who pretended to undertake the modelling exercise you just described would be completely – choosing my words here – making shit up. And so I know that no such exercise could be done, even more confident that no exercise was done. I think it’s complete political calculus based on the temper of the times. And, you know, politics is a two-player game, hopefully three, four-player game. I think, you know, diversity in parliament is a good, no, you know, we talk about two parties, but, you know, we have the Greens, we have the Teals, we have the National Party, diversity is a good thing.
But the dominant players spend as much time thinking about what the other one is going to do as what they should do. And, you know, there’s no doubt that the political calculus was done by the former government and the current government that actually it was just time, just to let the air out of the balloon and just kind of slowly walk away from the party. And no, I don’t think for a minute any modelling was undertaken. I don’t, I don’t think that, I know that no announcement was ever made that went, look, we’re a bit bored. We’re sick of this stuff.
So, you know, we noticed when state premiers stopped doing their daily updates, right? There were pieces that were really obvious. But, yeah, so some steps away from the party were more visible than others. But neither the government nor the opposition, state or federal now, wants to be the fun busters saying, you know, we assumed all along that people would continue to do those, those simple control measures that Margaret was talking about before, you know, there is a role for masks, there is a role for social distancing, you know, going to theatres in the middle of, in the middle of a spike is a really risky thing, not just for you, but your whole community.
Neither the Labor nor Liberal Party, state or federal, want to be the one to point that out. And that’s, and having data would just embarrass them.
Linda: As it should. We keep mentioning Long COVID and other sequelae. We, I try to do a lot of reading on the kind of consequences of COVID because, as I say, I have Long COVID and I have the ability to read that literature, but I’m not up with exactly what’s going on, Margaret. I’d like to think there’s more studies, systematic studies going on, on sort of the numbers of people who get, or the percentages of people who get Long COVID and the, and the likelihood of other negative outcomes from COVID. How, how much of that research is going on? Do you think we’ve got enough?
Margaret: Oh, I’m a medical researcher, of course, I don’t think we’ve got enough. Look, the Long COVID inquiry was really interesting. And the first thing that was interesting about it was that the terms of reference were not to discuss how to prevent Long COVID, because there’s a pretty simple way how to prevent Long COVID.
Linda: Don’t get COVID.
Margaret: The most simple one in the world. Don’t get COVID.
So when you have an inquiry into a disease where you actually don’t talk about prevention – as a person who spends their life in all of my other diseases, discussing prevention, and then how do you manage well, if somebody actually has the illness at hand, because I do work at many things other than COVID. So to me, it’s always, how do you stop something from happening? Or if something has happened, how do you make sure that the, that it can be diagnosed, it can be managed, it can be, the individual can be supported, that this is a patient-centered approach, but best never to have the patient is what I say to people.
So there’s some really increasing work and increasing acknowledgement of work that needs to be done around the management of people with COVID, going into Long COVID, to stop the likelihood of that Long COVID being as consequential, as severe and all of those things. What can be done? One of the critical things that I’m, it’s, I was on a call sort of around some work that we’d been doing around Long COVID just prior to this podcast, was around the critical importance of actually number one acknowledging it exists.
So at least now there is a general acknowledgement that Long COVID is real, that it’s, you know, as I say to people, I used to say to people with certain diseases, you’re not mad and you’re not malingering, which is a really important for people to understand. But it’s also really important to make sure that that message gets out into the community as well, that somebody with Long COVID is neither mad nor malingering, it is a real entity. So that’s the first thing.
And the next thing you’ve got to sort of do is really make sure that the health workforce also knows that and then is providing adequate acknowledgement, diagnosis and support. And in a really timely way rather than saying, oh, go away. I think, you know, it’s nothing. It’s actually doing something relatively early on may have – and this is where work needs to be done. And what does that look like? – may have benefit in making it that the length of Long COVID or the impact of Long COVID could be reduced by by telling somebody not to push through for certain periods as it is the nature of most people are, I’m feeling not so flash, I’m going to push through though, because I’ve got to do all those things in my life.
It’s actually no maybe having a rest early and not rushing back to work. It’s something – for those on the podcast, I’m an infectious diseases physician by my other work or one of my works – And one of the things I recognize really early on with certain illnesses is the tendency for people to want to get back to work. It’s this notion that we’re all a bunch of slackers. I can remember early on thinking, no, most people are really trying to get back to work. And I can remember trying to explain to a number of people with significant illness, that if you rush back too fast, and too full on and too full time, you will actually have a longer consequence of that – with many, many illnesses, not just infectious diseases – than actually taking time off, allowing your body to recuperate, and then stepping back into things.
This is really, well, we sort of know this, but we often ignore it. And I think it’s really important with Long COVID that we improve our evidence base so that we can really be more convincing of telling people that – the individual, their family, their community, their workplace, government. So we take account of that. So that’s where work needs to be done, as well as, as I would say, that simple prevention stuff, which is trying to say during waves of infection, perhaps.
As I say to people, you don’t even need to do a model on this, it’s maths. We don’t know exactly what proportion of people with COVID go on to get Long COVID, but conservatively, it’s between three to five percent. And some people say it’s significantly more. And then if we define Long COVID as having symptoms beyond four weeks, or look at a WHO definition, or all sorts of definitions, but if we just go, then it’s a significant number of people that it impacts on. And if the more often you get an infection, the more likely you put it within that infection, then you’ll go on to develop Long COVID. Then maybe having four or five infections is something to avoid. And they’re the kind of pieces of work that we need, to really understand what’s going on. Because this notion of, oh, I’ll just get infected every year, maybe it’s not such a good idea. Or maybe it doesn’t matter. That question needs to be properly answered.
Linda: On my list of questions, I have why aren’t we collecting that data? But I think we’ve pretty much answered that. And I still find the answer a bit bewildering, to be honest. In an ideal world, though, what data would we have? And what would that enable us to do?
Margaret: In an ideal world, we would have a few things. In an ideal world, we would have a way of more accurate, more accurately, because you don’t want to go, I totally agree with Richard, you could spend a large amount of money aiming for perfection, you don’t need perfection. But we would have more accurate information about the number of cases in a timely way. So that’s really important. You’d have accurate information about changing genotypes. And in some ways, we get that from sewage waste and those kind of things, but you’d be having that kind of accurate information.
The reason it’s useful to have it in people as well, is to sort of have some level of monitoring around the likelihood of impact of that, depending on vaccination and severity and those kind of things. You would have accurate information around hospitalisations. And interestingly, we don’t have as accurate as you might think, we would have accurate information of
Linda: Why is that?
Margaret: it’s complicated. Do you want me to just go, “Oh, it’s complicated.” Because again, it’s how things get prioritised, how things get measured, if we all choose to measure slightly differently, if we all choose to report slightly differently. And if there’s no consequence of not reporting in the same way, it’s like all of these things. And then deaths. And even that’s not necessarily accurate. And then excess mortality. So there’s those things that you want to measure.
And then if somebody gets COVID, you want to be actually accurately measuring the impact of that COVID on the individual. And then also you want to be looking at things like vaccination. There’s a series of things. And each of those doesn’t have to be 100% perfect and rigorous. But a consistency of measure is really important, because then you can look at trend over time.
Margaret: I should say the fancy researchy stuff as well, the extra bits. But that would be what I would say for government.
Linda: Richard, is there economic data that we’re not collecting that we should be? Well, I think the health data is economic data. I’d like to know how many people are sick and getting and staying sick and how long they’re sick for. Then I could tell you how much of our skills shortage was caused. I’d like to know the demographics of who’s getting sick so that I could anticipate how different interventions that we could still be making may help particular communities. So, yeah, I think economists would find that sort of data very, very valuable.
And from that raw data, we can interpolate all sorts of things. Like, what’s foregone income here? Or do we need to look at, is 10 days sick leave enough in a world where people are getting COVID once or twice a year? Because if we don’t want people with COVID to go to work and we don’t want people to lose their house when they get COVID, at some point we’re going to have to have a conversation about this.
So, yeah, I know that the raw health data would be really good, but it’d also be really good to measure air quality. And if we were doing that systemically in large public buildings or randomly in buildings, we might be able to know what kind of measures building owners are taking so that we could put pressure on them more, perhaps. We might think, well, we’re going to need to regulate because the data we’ve got suggests that proxy indicators of air quality suggests that buildings are getting worse, not better.
Or, like, yeah, I just think that all of the data that epidemiologists would like to have would be of incredible use to economists. And in particular, if they were interested, policymakers in trying to either design and target interventions or just understand the cost of our failure to intervene. And, you know, back to the question you didn’t ask, why aren’t we collecting the data? Because we don’t want to admit the failure. They don’t want to admit, they don’t want to make it easy for me to tell you what, you know, what the cost to GDP of this heroic approach, you know, to COVID has been.
Margaret: And the other thing that I just want to touch on that we haven’t touched on in this conversation yet is what I would call the inequity of it. COVID, like so many other areas of health, is impacting with far greater sort of having far greater impact on all sorts of levels on people from social and structural disadvantage. And that’s also, to me, often in the areas where I do work prior to COVID, I’ve always assumed that one of the things is that we can pop that group of people over in the corner because we really, it’s complicated and it’s a whole of systems approach. And there’s many, many things that we need to do as a community and conversations that we need to have as communities around our support, our efforts to reduce inequity. And COVID is totally again impacting on us, having greater impact on a group of people already at considerable social and structural disadvantage. And to me, my work over the years has suggested again, it’s one of those things that governments, people, us all, because we’re part of communities, struggle to really deal with well.
Linda: Is there anything else either of you wants to add that we haven’t touched on or that you think is important here?
Richard: Just the context, like, don’t get me wrong, this is really important. And we really should be collecting a lot more data on COVID. And we really should be encouraging people to do the simple things that the data we do have says would work. But this is not the only big public policy problem that our policymakers are willfully blind to. It would be wrong to think this is a COVID conspiracy, this is a well worn path.
And we have terrible data on greenhouse gas emissions by industry in Australia. And it’s not timely, and it’s not easy to understand because, hey, maybe we don’t want to embarrass anybody. We have terrible data on lots of big problems. And we have really, really good data on some things that powerful groups are really interested in. Unfortunately, the map of what we have data on is as interesting as the maps within data.
And a lot of people think that these things are all accidents. I suggest that I have a very simple predictive model for what we’ve got good data on. It’s groups that were powerful 10 years ago, because there’s a long lag on data. So we have incredible data on sheep in Australia, because since Federation sheep were really important, and it’s hard to stop collecting the sheep data, and we don’t have good data on COVID because we don’t want to.
Margaret: The thing I just want to bring up, because we’ve talked a lot about data and these kind of things, but at the end of the day, we are talking about people as well. And it’s just super important to go that people are at the centre of this, and people in their communities at the centre of this. And one of the things that I think we did really poorly all the way through, and continue to do really poorly, is speak to people about really what their wants, needs, and their communities wants and needs are. And I think it gets back to some of the stuff that Richard’s been talking about, of whether we really want to, whether we, as in the broader community or governments, want to know all of that. But it’s just really important to keep people and community at the centre of it, so that when we’re collecting data, we are actually collecting data that addresses the critical issues and needs of those communities, as opposed to what, say, somebody like me as the researcher might think is critical. It’s actually making sure we go back and ask communities, how is this truly impacting on you as an individual or community, and keep, sort of, as we go forward, that at the centre of our thinking.
Linda: I think that’s a perfect place to wrap it up. Thank you both so much. It’s been a while in the making, this episode, and it’s just been such a privilege to hear you both speak and to raise these issues. I’m really grateful.
Richard: Great, thank you.
Margaret: Thank you. Thanks for listening to Make Me Data Literate. You can support the work of the Australian Data Science Education Institute at givenow.com.au/adsei. Tune in next time for more conversations with amazing data experts.
