Alex McCarron: Hello, and welcome to Escape from Lockdown, the show all about how we got into this madness and how we are going to get out of it. Now today I have another of the great pathologists. Very early on I interviewed Dr. John Lee, and his interview set the podcast on fire and set the whole lockdown escape community on fire, really. It had real crossover. And today I think is going to be no different or even better. I’m speaking to a very brilliant person who is doing some incredible work and really putting themselves out there, exposing some really terrifying conclusions that she’s come to, to all of us. It is of course the pathologist Dr. Clare Craig. Clare, how are you doing? Welcome to the show.
Dr Clare Craig: Thank you very much for having me.
Alex McCarron: Can you tell the listeners a little bit about your professional background and how you got to be where you are?
Dr Clare Craig: I’m a consultant pathologist. I’ve worked in the NHS as a consultant pathologist for many years, and I moved to work on the cancer arm of the 100,000 Genomes Project for a couple of years, and then I’ve moved into AI more recently. But, I’ve experience with laboratories, with testing, and understand what false positives mean in medicine.
Alex: So you knew what false positives were before they got big, basically.
Dr Craig: Yes, I would say that my professional career has been around those kinds of problems.
Alex: Can we sort of jump straight into the fact that everybody who’s sort of been looking at the data knows that there’s this thing called the casedemic, but your works shows that actually the problems with the casedemic are actually much more profound than people, even us, quite realize. So can you tell us what’s going on?
Dr Craig: I can try. I mean, a lot of people try to find some data point that they can trust because one by one these data points are being questioned. And so people put a lot of faith in COVID death counts. They think, “Well, they must be true because, you know, how on earth can you misdiagnose someone’s death?” But I’m afraid that even the death count, you have to have a bit of skeptical about because of how we are testing and how we are diagnosing. And there’s a phenomenon that’s worth considering when we’re looking at the situation that we’re living through at the moment, which is called a false positive pseudo epidemic.
There are a few key factors to understand about that, one of which is when you’re living through it, everybody involved believes they’re in an epidemic because the data looks like an epidemic, which is why it’s got that name. But there are a few things that start to show up in the data that you can unpick to figure out that actually this isn’t the case. What starts to happen is that because the data points are related to testing and not to each other, they start to do really funny things.
So one of the things that’s a relatively easy image to understand is looking at ITU admissions compared with deaths, and ICNARC which do ITU audits have just published on this. They show a graph with a familiar spike in the upturn of the ITU patients and then coming back down, followed after a period of time by a spike in deaths coming back down. That was in spring. And you see these two lines followed in parallel all the way through. And then they’ve superimposed what’s happening now on this graph, and you can see a much more shallow line of increased patients in ITU, and below that in parallel the increasing number of deaths.
But in the last couple of weeks that line of deaths has done a sharp upturn, and it looks like it’s going to overtake the line of the number of patients in ITU. And so there are other ways to look at the data that back this up as well, but the point is that we’ve got to a situation where the number of people dying per case diagnosed is on the rise compared with the summer, but the number of people with a severe case (being admitted to hospital, being on ITU) has fallen since summer, which is just slightly baffling, you know.
How can you get to a situation where the severity is reducing but the deaths are increasing? That is quite difficult to get your head around. I don’t think we need to go over it again, but there is this discrepancy that doesn’t make sense, and it especially doesn’t make sense when you realize that 80% of the COVID deaths at the moment are in hospital. So if they’re in hospital, they should be in the hospital admission data, they should be on ITU, and they’re not showing up in that data.
Alex: So basically, if I can put it in a way that is often ridiculed, the former secretary of defense, Donald Rumsfeld famously gave this speech where he talked about there are known knowns, and there are known unknowns, and there are unknown unknowns, the things that we don’t know that we don’t know. That was often widely mocked at the time, but I believe that’s the cleverest thing any politician has publicly said ever because it was a tacit admission of the way that knowledge works and the way that we find things out.
To me, you seem almost half scientist, half detective, almost like sort of forensically going through the data. And it seems to me we have these known knowns, which we established over the summer which was sort of time between infection and death, the kind of general makeup of the disease. But, are you saying that they’re not the same anymore? The data, you know, these things that we depend on are suddenly going crazy, and there’s no relation between what’s coming in and what we thought we knew?
Dr Craig: The way to look at it is to use percentages. You can see the percentage of deaths for people who were admitted 10 days before to hospital. You then look at all the hospital admissions and see 10 days later how many deaths there were, and those two figures should have a set relationship. But what we see is that in the beginning it was quite high, and that was partly because we were not diagnosing everybody, and it comes down, and it carries on coming down right the way until August. In August, if you were admitted to hospital, you had the best chances compared with earlier. And, you know, we’re being told that that’s because of better treatments and what have you.
But then since August, the percentage has started to rise, which is a worry. You worry that things have got worse. Has it come back? Have we been misdiagnosing in the summer but now it’s come back? That pattern is repeated in other data points. The number of deaths per occupied bed on ITU has started to rise, and the number of deaths per case diagnosed has started to rise. All of those data points, they look like things are getting worse. But the other data points of cases to admissions look like things are getting better.
Alex: So what’s going on here? Is it that, are we just in the middle of a kind of orgy of testing and it’s throwing up all this crazy data?
Dr Craig: The thing about testing is that at the beginning we had to get some tests out really quickly, and I really admire the work that was done to get that to happen. Manufacturers turned new tests around as fast as they could, and, you know, it was all about speed at the time. So the fact that they compromised on some of the checks that would normally have been there is entirely justifiable. And then the labs got set up and were scaling right from the beginning. They were scaling what they could do. And we got to a point in May where the UK labs were doing 50,000 tests a day, which is absolutely phenomenal, and at the time it was world-beating, and it was more than enough to get a grip on the situation that we had.
But we carried on with that strategy of volume and speed, volume and speed, and we ended up now we’re doing 200,000 tests a day and with a couple more labs, but it’s essentially the same labs. They’re just being scaled and scaled and scaled. And when you’ve got a laboratory, there are three things a laboratory can do, and they can only do two of them well. They could put through a huge number of tests per day- they can do volume. They can do speed and get the results out as quickly as they can. Or, they can do quality tests. But you have to pick which two you’re going to focus on. We’ve focused on volume and speed. And again, that’s totally justifiable at the start of an epidemic when you’re trying to stop spread, and a small percentage of mistakes along the way are just really irrelevant to the situation that you’re in.
But from a pathology point of view, epidemiology 101 is when you get to peak deaths, you switch your testing strategy. You start with high volume, fast, and as sensitive as possible because you want to find every possible case. At peak deaths you switch strategy to quality testing and being specific because you want your results to be accurate at that point. We haven’t switched strategy. And the only way to switch or to do that, to get quality results is not to put the labs under even more pressure and shout at them and get cross. The only way to get a quality result from a lab is to compromise either on the volume going through every day or in the speed at which they have to turn them around.
Alex: What sort of evidence do we have that the accuracy has been compromised? What blips are we seeing in that data that tell us that these tests aren’t quite what they say they are?
Dr Craig: There’s a beautiful piece of evidence that’s just been produced by a physicist in Scotland called Christine Padgham, who is a force of nature and has gone carefully through all of the Scottish data. Public Health Scotland have been much more open with the data that they’re publishing, and they include in their publications the daily positive numbers, the daily negative numbers, the total number of tests done, and so you can actually get a percentage of positive tests per day that’s accurate. And when you look at the percentage of positive tests per day in Scotland, the percentage of positives is twice as high at the weekends than it is on a Monday. Now that cannot be anything to do with the disease.
That’s to do with the laboratories being under extraordinary pressures. It’s to do with people. The PCR testing, which is the test that we use for COVID, can be an incredibly specific test with a low false positive rate, but it can also be incredibly difficult to actually do because the first step is to translate your RNA to DNA and then you double the amount of DNA in the sample. You double it, and you double it, and you double it until you’re at a billion or a trillion times the amount you began with. What that means is that even the tiniest, tiniest amount of cross-contamination from other things in the lab can mean that you get the wrong result.
Whenever you run a test, you’re going to definitely put a certain positive in that test so you can make sure that the test worked properly. Every time they run a test, a positive control sample is being used. And if a little bit of RNA from that sample gets onto a glove, and gets onto the fridge door or something else around the lab, then every person that touches that fridge door is going to get contaminated, and the samples that they touch will get contaminated. The difference between a weekend and a Monday in a lab is that at the weekend you’re short-staffed, and people are tired, and the labs had all the problems that built up over the week hanging over. On a Monday, people come in fresh-faced. They’ve had a rest over the weekend, and the lab is thoroughly cleaned, and then you get out new chemicals that are all brand new and clean, and you start again.
Alex: So basically people are just kind of turning up hung over on Saturdays and Sundays.
Dr Craig: Oh no, I think that’s really unfair. I think you have to appreciate that if you’ve increased testing to that degree, people have worked their socks off. They are working so, so hard. I don’t think they’ve had time to have a drink. So they’re exhausted.
Alex: I’m imposing my own fecklessness on doctors who I’m sure are doing a very good job. I’m sort of damaging my ability to get new work now. So there’s other data you brought up which was really interesting, which was there’s this correlation which you never see anywhere in biology which I think is…it relates to the number of tests performed and the number of infections that we’re getting. What was it?
Dr Craig: It was a period of time where the hospital tests done related to the number of hospital COVID deaths, and it was a really tight correlation. The hospital tests have ramped up much more gently than the community tests, but we’re still doing a lot. And we got to a point where every admission could be tested, which was great. And then we exceeded that point. So there was the ability to test people more than once. And understandably, if somebody comes in with a broken leg, you’ll test them once as protocol. We don’t normally test them again.
But if somebody is coming in coughing, you might use your spare test to test them again. If somebody is coming in in respiratory failure, they’re going to get more than one test. So there comes a point where the increased number of tests are no longer proportional to the increased number of people tested. You get to a critical mass, and then any further increased tests are used on people who are more sick. Then you start to see this relationship between the number of excess tests done in a hospital and the number of COVID related deaths in the hospital.
Alex: Wow. So basically the implication here is that…is nearly everything that we’re seeing a false positive test, even if it’s in hospital?
Dr Craig: I would hold back from saying that, but I would say that cannot be excluded. The reality is that we have a problem with false positives, and the only way to clear that problem up is to start to carry out confirmatory testing and to sort out the labs. We need to put gateways in and say we’re not going to test everybody, we’re not going to test asymptomatic people so that the volumes decrease so that the laboratories can get on top of it. But only once you’ve got on top of it and you’ve done your confirmatory testing you can actually see what’s real out there. Because at the moment, the numbers that aren’t real are overshadowing the real ones, if they’re there.
Alex: And when the false positive story kind of broke a month or two ago, I think Julia Hartley-Brewer famously questioned it on her radio show. The BBC and, I think, maybe the Huff Post as well. I think Tom Chivers wrote something on this. Basically the determination was rather to examine and delve further into potential, you know, corruption of the data was to poo-poo the notion of false positives being effective data at all, which tells us a lot about the journalistic priorities and the cognitive biases that people fall in.
You know, there’s a famous saying. It’s very difficult to make a man understand something if his job depends on not understanding it. And there’s just a real commitment to rubbish any of the questions, to shut down the questions rather than to investigate what they’re saying, I think at least. So one of the things that people often say is, “Oh, your false positive rates, they don’t really count if the people you’re testing are symptomatic,” you know, because that doesn’t [inaudible 00:19:11] with data as much. I would ask you, does it now?
Dr Craig: The trouble is with COVID that the definition of what it is was back to front. The way that you set up a diagnostic test is you define a disease based on symptoms and signs and what it looks like to a doctor, and then you find a test. You work out if the test is any good by seeing if it can pick up this picture. But in COVID it was back to front. We defined the test, and then the symptoms were worked out after we decided who was positive with the test.
So the list of symptoms is as long as your arm, and you’re allowed to be asymptomatic as well. Anyway, leaving that aside, there are a lot of symptoms that count. With that many symptoms you’ll find a lot of people have those symptoms. I mean, we know from the ONS survey data, when they published who was symptomatic and asymptomatic, that 11% of the people were symptomatic with some symptoms at any one time because, you know, they’re common symptoms. So if your rate in the asymptomatic population is lower than in your symptomatic population, that does still make it look like you found something, right?
But the way that the testing works is that you’re looking for the sequence of letters in the RNA that is unique to COVID, and it’s a great test when it’s done well, as I said before. But when it’s done badly, other sequences of letters can cause a positive. DNA binds certain letters very, very accurately. A binds to T, C binds to G, and they’re really tight binding. But there’s a certain amount of binding that can happen to the wrong letters, so if you’ve got a misspelling that’s a few letters out, it can still bind, and you can still get a positive result. That’s especially true if you’re doing all these extra cycles before deciding whether or not it’s positive.
What that means is that there could be other viruses out there that cross-react with COVID testing and produce a positive test in someone with symptoms when actually it’s a different virus causing the symptoms. And, you know, we know that this is a risk, so when you make a new test you check for that. And what we would normally do is check by getting virus samples and running the tests and seeing if any of them go positive.
But what’s mostly being done for COVID is people have checked DNA databases and have looked to see how many letters match or don’t match, and say, “No, we’re okay. We can run with this,” which as I said before is, you know, justifiable. And then the laboratories, before setting up their testing, they did do wet lab testing, so all of the labs individually will have tested against samples of other viruses. But they’re testing against a range of other viruses, and it’ll be one sample of each type of virus. And that’s fine when you’re testing a high provenance population and you’re testing people who are likely have it.
But when you move to doing mass population screening, which is what we’re doing, you have to have a different threshold for your accuracy. And the only way you can be certain that we’re not getting cross-reactions with other viruses is if you test hundreds of samples of each of those viruses because you’re only going to see that, say, five percent of, you know, a cold virus is going give you a positive if you’ve tested hundreds of those samples. We’ve tested tens.
Dr. Craig: Well, when I says tens, I mean, like, 10 or 20.
Alex: So effectively we’re just… I mean, we all knew false positives were an issue, but I didn’t realize it was this much of an issue. There was an article that came out, I think it was in “Full Fact” recently that was saying, “No, it doesn’t pick up the common cold. It doesn’t pick up coronavirus.” But it seems to me that they weren’t really asking, they weren’t really addressing the right question. They were saying the test isn’t meant to pick up other common colds or other viruses, but what you’re saying is basically, you know, the test just occasionally, unintentionally, and very rarely does.
Dr Craig: I think it can do. So let me tell you a story about a false positive pseudo epidemic. This is a lovely story. It’s my favourite one.
Alex: I was looking forward to this.
Dr Craig: It’s a hospital in New Hampshire, and one of the doctors had a cough. It was a really bad cough. It’s one of those coughs where you cough a lot, and then you have a sharp intake of breath at the end because you’ve, you know, been coughing for so long. They were at lunch with a doctor colleague who thought, “Oh, hang on a minute. That reminds me of whooping cough.” So whooping cough in children, the whoop is after a really, really long period of coughing where they’ve run out of air, and they gasp for air. That’s why it’s called whooping cough. Right? So they said, “This could be whooping cough. We ought to check.”
So they went off to the lab and did a PCR test to see if this doctor had whooping cough, and it came back positive. This set off this kind of panic, and they just decided they’d better start screening the hospital because they had vulnerable babies and vulnerable old people who might catch this horrible bacteria. Not a virus, but anyway. And they started to test members of the staff and patients who had symptoms, and they found some more positives. And then they tested more, and they found more positives. By the end, they had tested 1000 people. They had got 146 positives back, so a 14.6% positive rate.
But one of the doctors was careful and clever enough to say, “Let’s have a backup and try to culture the bacteria from these samples as well.” So as well as testing for PCR, they tried to grow it in the lab and see if any of them would grow. None of them grew. None of them. All of that 14.6% were false positives, and it looked for all the world like an epidemic. After the news had broken that the testing was wrong, it took a long time before people could get their heads around what had happened because there was this collective delusion that they were all in. And, you know, I’m a bit scared when that happens here, actually, what the results will be.
Alex: Well, I think I can tell you. The results will be they bring in heavier and heavier restrictions. They ramp up testing even more, and it will throw up even more false positives. And when people try and question it, they’ll try and shut them up. It’s just a guess.
It’s worth talking about here, actually. So I just did a quick Google of whooping cough for false epidemic, and you got two articles come up straight away. One is in “The New York Times,” which have a wonderful title here called “Faith in Quick Tests Leads to Epidemic That Wasn’t.”
Dr Craig: Yeah, that’s the one. But there are others. There’s another whooping cough one where the false positive rate was 74%. The thing about this is that in retrospect people say, “Well how did it go so wrong?” And that 74% went so wrong because of very high cycle thresholds. But the 14.6%, I’m not sure exactly how it did go so wrong. People speculated that there was a problem with one of the reagents or there was some kind of cross-contamination issue, but they don’t actually for certain know exactly how it went so wrong. But the point is it can, and the only way to be sure that we’re getting the right test results is confirmatory testing.
Alex: Can they just test one PCR test against another done in a different lab?
Dr Craig: No, because if there’d been any problems up until the point where the swab reaches the lab, then that’s going to still be a false positive.
Alex: So how do you do a confirmatory test then?
Dr Craig: You have to have the confidence to say, “We’re not going to diagnose any patients until they’ve had two positive tests, separate days, separate positives.”
Alex: See, the thing is though, what you said was the way that they cracked this terrible problem of the fake whooping cough epidemic (and that is surprisingly difficult to say) is that the doctor decided to grow a lab culture, which to me sounds like very much like a gold standard test, because either it’s going to grow or it’s not. And it’s just 100% extremely accurate. COVID doesn’t have, as far as I know… Actually, I’m going to phrase this question differently. Does COVID have this alternative test we can test it against? Are we stuck with PCR?
Dr Craig: No, there is another test. You can also culture a virus. So what you do is you put the material in with some cells in a lab, and a virus will go into the cells and replicate, and then it will burst the cell open. So you just measure for the cells bursting open. And that has been done. That’s absolutely being done, but it gets done in, like, really kind of high tech, safe laboratories, and it’s hard to do. So you can’t do that at scale, but you can do that on a sample of positive tests and prove the point.
Alex: And do you know if that’s being done at all?
Dr Craig: I don’t know.
Alex: I mean, it probably isn’t.
Dr Craig: Actually, there’s one thing that is being done, which I think is why that’s not being done. The thing that is being done is that we’re doing whole genome sequencing on some of these samples. What that means is that instead of looking for just part of the RNA of COVID, the sample is amplified up in the same way, the doublings, and then you read the letters of every last bit of DNA in that sample so you can see what’s in there. When you do whole genome sequencing you can compare what’s going on, what mutations have happened over time, and you can fit it into the sort of family tree of COVID. If you’re getting samples through that have got positive whole genome sequencing results, it’s really convincing that it’s real. But, of course, if it’s cross-contamination from the false positive control, it’s still going to get a whole genome sequence.
Alex: Because you’d think, the thing that surprised me with this crisis, I don’t like calling it a pandemic because that suggests that we’re still in it, and I’m not sure that we are. But the thing that surprised me is with the £300 billion that we’ve already spent, surely they could set aside, you know, a measly sort of half a billion to sort through these confirmatory tests or to sort of test what they’re doing. It doesn’t seem to be a priority at all.
Dr Craig: No. I mean, if you look at the testing priorities, the priority continues to be to ramp it up and to aim for the moonshot and to have a million tests a day and have us all be tested every morning. It’s completely, like… they clearly have not had advice from somebody who understands this testing. And the people on SAGE that are giving advice are predominantly physicists, chemists, and mathematicians. And for physicists, chemists, and mathematicians, a false positive rate is the lowest positive you’ve ever had in your testing. The fact is the kind of work they do is on really, really accurate testing equipment, and they have really low false positive rates, and it’s a constant. And that’s not the situation in medicine.
Alex: And basically, and this data, these rates are potentially changing all the time. You said yourself they change from a Monday to a Saturday in Scotland.
Dr Craig: Yes.
Alex: How are we going to get out of this? I’m a little bit worried.
Dr Craig: Well I think, to be honest, I’m optimistic because…
Alex: Oh really?
Dr Craig: Yes. The data will start to do crazy things. It’s already started to do crazy things. So as well as the deaths being out of proportion to the severe cases, one of the other things that’s starting to happen is that the number of predicted cases is starting to be lower than the number of cases diagnosed. It’s not quite there yet, but that’s the trend that we’re headed in. When you really do have COVID, PCR testing is reliable for about 20 days. Obviously we’ve heard stories about it going on and on for months, but in most patients you have a 20 day window of it being picked up. And the number of predicted cases in the East Midlands is the sort of number of new cases per day that you would see over the course of a week.
And if you go and look at that week and say, well, how many cases did we diagnose? Assuming that you can have any be picked up in any one of those 20 days during the course of the illness, then we’re pretty much on par. So more crazy things will happen with the data that will be undeniable nonsense. And then, you know, once you get to that stage, people have to start thinking differently because you can’t make sense of these things. There was a lovely article in “The Daily Mail,” and I’m sure it was from the best of places, but it shows how crazy stuff has got where the news broke that the time to death had got worse. Right? It had been an average of two weeks between diagnosis and death in hospital, and it was one week. And they managed to say that this was because treatments had improved. Am I getting this the wrong way around? Let me have a think.
Alex: I think they probably got it the wrong way around.
Dr Craig: No, they said treatments had improved, right? Because patients who would have died after a while are now surviving because of these brilliant treatments…
Alex: Oh, so only the very ill ones that are dying.
Dr Craig: Yes. You’re like, that’s such convoluted thinking. It’s such convoluted thinking, and we’re going to hear more and more convoluted thinking like that because unless you realize the reason that it’s changed is because you’re diagnosing something else completely, then you have to have convoluted thinking to make sense of that kind of data.
Alex: I just find it everywhere. I find it constant, the convoluted thinking. Even the non-pharmaceutical interventions, i.e. the lockdowns, the circuit breakers, all of that stuff, it just results in convoluted thinking. You know, the Welsh thinking, “Yeah, we’re gonna ban books. That’ll do the trick.” And this phenomenon of long COVID as well, it’s as if they’ve kind of lost the battle on the infection fatality rates, and they’ve had to concede that it is lower than they thought it said. But now they’re saying, “Well, you know, this could cause, you know, long term disability.” You just have to say, well, A) no one has had it for more than six months anyway, so how could you possibly know that? And, B) I mean, you’re the pathologist. Don’t all viruses have this?
Dr Craig: Pneumonias are horrid. If you get a pneumonia, you’re going to be sick for six months no matter how old you are. It’s a really, really horrid thing to happen. It takes a long time to get better from. And I think you have to wait six months before assessing whether there’s anything more. And, yes, you know, this was a horrible illness. And actually I disagree with you about the infection fatality rates. I’m kind of an outlier in the community that have written on this. I think the infection fatality rate was higher than we now think it was.
Dr Craig: Because the calculation done more recently have been diluted with false positives.
Alex: Oh, right. Okay.
Dr Craig: When COVID hit in spring, it was a really horrid killer, and we’ve kind of forgotten quite how bad it was. If you go back and try and remember how we were feeling in March and how the news came out and how… So let me take you through the timeline, actually. The 21st of March, news broke that 21 year old Chloe Middleton, who was healthy, had died at home of COVID, which had us all slightly on edge, I think. And then on the 28th of March, Martin Egan, who was a bus driver, died. And the first NHS surgeon who was working had died. The next day was another death of a bus driver, and a 55 year old healthy NHS physician died. And by five days later, we were told five Transport for London bus drivers had died. The next day five NHS staff had died. It was really quick, and it was killing young people who should not have been dying, and it was worth being scared of in March and April.
Dr Craig: I think when we actually managed one day to filter out what was real and what was not real, we’ll see that it did have a significant infection fatality rate. It’s just that since then, what we’ve diagnosed is not it.
Alex: But then fundamentally though, the prevalence can’t have been as big, and it can’t ever have been as big because it’s largely passed through the population now. I mean, the big key metric here to look at is excess deaths, right?
Dr Craig: Right. Let’s come back to excess deaths though because the thing about prevalence is that I totally agree it passed through the whole country. Every part of the country had excess deaths in spring. Liverpool has the same 14% excess deaths this year as London. This kind of story we are told that it infected some areas more than others doesn’t really match with that data of excess deaths. But the way you calculate your infection fatality ratio is based on how many people were symptomatic. That’s what we mean by who had it. And, you know, we’re never going to know for sure because we weren’t testing, and so we don’t know for certain. We don’t have great antibody testing to know for certain.
But what it doesn’t measure when you’re calculating this is people who were immune already. And I think we had significant numbers of people who couldn’t catch it. And when it passed through the country, it wasn’t 100% of us that were susceptible. It just wasn’t. There’s prior immunity from other things that we’ve seen. Our immune systems are amazing, and they work for most of us. You can see that also in the data.
There was a nice match analysis published on household transmission. So people who had a positive COVID test, they went and looked in their households (this is around the world) and found out how many of the people they lived with caught it. And the range was huge. It was from 50% of household contacts catching it to 5%, which seems rather low, almost as if maybe you’re not testing correctly. But going back to the 50%, 50% of household contacts catching it, it means that the rest are immune. They must be immune, especially when we know how quickly this disease spread. It was a very contagious disease, there’s no question about that, and how quickly it went through our country. So it’s a contagious disease that not everybody catches.
Alex: Well famously there was the Diamond Princess, which is your kind of perfect petri dish to see how it affects, because I remember stories about this, infections coming out of cruise ships. You get these norovirus infections and stuff, and they would totally tear through the whole ship because if you want an environment where a disease could spread, a boat is pretty much as good as you’re going to get. But what was it, a huge proportion of people, I can’t remember, they just didn’t get it.
Dr Craig: No. They also at that point had the stories breaking about patients testing positive who had no symptoms, and some of those patients went on to get symptoms, which, you know, means that they probably had it, but others never had symptoms. There’s been so much confusion about this asymptomatic thing that, we’ve just gone into some other world which is different for any other disease. Yes, you can have a positive PCR test and be asymptomatic. Yes, you can even have a positive viral culture and be asymptomatic. So that means that there’s live virus that can get into cells, and people can have that in them and be asymptomatic.
But that does not mean that they’re infected. It doesn’t mean that they’re diseased in the way that we normally talk about disease because they have no symptoms. It means that they’re immune. That is what immunity is. Immunity is when a virus invades, it doesn’t bother you. The stories in the scientific literature about transmission, which is what we should worry about, say yes, these asymptomatic people can have the virus. But can they spread it? And that’s the critical question.
There are two schools of thought on that. So if you take all the scientific literature published about transmission you can put them into two piles: one that shows they do not transmit (you can’t spread it unless you’re coughing, which sort of makes biological sense) and the other that says it’s a serious problem. But if you look again at the pile of papers that say it’s a serious problem, they were all published in China, and I think we just have to have a little bit of skepticism about that when all the other literature contradicts it.
Alex: Well, regular listeners of my podcast will know that nothing coming out of China should be trusted related to this on anything. And as Michael Sanger, one of my former guests, showed, not everything that comes out of China is obviously coming from China. And that is a real danger. I think I said to you off air I don’t get conspiratorial about this. I do think we are in a storm of cognitive biases and motivated reasoning. And even the great reset stuff and all of that, it’s just the people who sort of spout on about this stuff and have been doing so for years, just seeing this as opportunity. It’s no different.
But if there is one bad actor that is certainly the Chinese Communist Party, and they have the motive and the reasoning to do that. Although, having said that, this podcast is more talking about scientific issues rather than politics, so I should try and keep them separate. So we didn’t actually quite go into a little bit, but what’s the thing that tells us that the epidemic has passed through the population? Is it those excess death figures? Which is quite a nasty little blip. It’s a good, you know, what, 20, 25 years since we’ve had something that bad kind of hit the population?
Dr Craig: Is your question really, how do we know it’s over?
Dr Craig: The one thing to look at is when hospital deaths peaked around the country. You can look at by hospital trusts, you know, each of them have their own little Gompertz curve with a maximum. And you can say, “Well this is when peak deaths happened.” And the first peak was in Brighton on the 28th of March, way too soon for lockdown to have had an effect. And then it spread not in a kind of south to north way. It was all over. I think there were lots of different seeding events.
But the last places to spike have a death peak in their hospitals were Hull, Rotherham on the 24th of April, Bradford on the 26th of April, and West Suffolk on the 28th of April. And the thing about those places is that when you do pandemic modelling, they are the places that get the disease last. And they were getting it so long after Brighton that you can see that it was just spreading throughout lockdown. Lockdown didn’t have any effect at all. You can confirm that it’s come, gone, killed people, and then just disappeared because it hasn’t come back. That’s fundamentally the test of immunity. Is it coming back? It should’ve come back at the VE celebrations, or in the marches, or when the beaches were packed. You can’t keep saying, “Well, it’s going to come back tomorrow.” It didn’t come back because it’s gone.
Alex: It’s gone. But we’re still stuck in this situation.
Dr Craig: And it’s not gone forever, you can’t get rid of a virus forever. It’s not gone gone, but the epidemic part of it is gone. So after an epidemic has come and gone, then the population is no longer susceptible because either people have been killed or become immune. It’s just the reality of it, harsh though that is. And therefore, if the virus does, you know, have a winter prevalence, and in the winter there may very well be cases again through the winter, but it’s a different story. That’s just like flu every year. It’s a seasonal infection. It’ll come, but it’s not coming into a susceptible population anymore. It’s coming into a population that has a bit of immunity.
Alex: I suppose that’s the, how can I put it, the slander that the anti herd immunity advocates say is that herd immunity means the eradication of a disease whereas that’s not actually the case, is it? I think [inaudible 00:47:58] calls it the epidemic equilibrium where it just kind of sinks back into the background.
Dr Craig: Yes. The herd immunity deniers keep talking about measles saying, “Well, you know, we only got control of the measles because of vaccination.” And that’s kind of true. The thing with herd immunity is that the number of people who have to be immune depends on the R value, the R0 value. So how contagious is this disease? Measles is really, really contagious. It’s got an R value of eight, so you need 90% plus to have herd immunity. And the problem with measles is that there are babies arriving all the time, and they’re not immune, so in order to have herd immunity you have to keep that vaccination level up really high. But the R0 value for COVID, you know, it’s debatable. In fact, the range is quite massive for what people think it was, but there seems to be a reasonable guess, and three is how you get to the 60% immunity, herd immunity figure, which also seems reasonable. And so, no, you don’t have to have every single person in the community being immune.
Alex: Right. So we’ve spoken for quite a while. If there’s something that could, I’d like to ask you personally is, so I’ve been sort of kicking around in this lockdown skeptic world for I think probably since April. But you’re a real newcomer. It’s amazing. Your Twitter account has only been around since September, and you’ve already got, you know, quite a large following already, which to me sort of encourages me a lot because the podcast where I interview scientists always get really, really high views or listens, rather.
And, you know, your Twitter account has got a lot of information on it, and it shows there’s a real hunger for that. So it’s really one in the eye for these kind of media commentators who think everything has to be dumbed down, which I actually think is quite hopeful for the future. It shows there is appetite to sort of digest this stuff and to disseminate it. So why did you decide to speak out, and why did you speak out when you did?
Dr Craig: That’s a really reasonable question. I realize I’m the latecomer to the party, and a lot of people have been speaking out since…you know, they spotted it way earlier than I spotted it. Essentially, I have four children, and I was really, really busy. I was trying to homeschool four children. And we went through the summer holidays, and then finally September came, and they went back to school.
The kind of little questions I’d had niggling at the back of my mind about what was going on and were we just getting false positives through the summer when the positive rate was flat, you know, I suddenly had time to explore it. I started digging into the data and testing the data and saying, look, if these were false positives, what does that mean? Can we see in the data changes like when COVID deaths happen they were 60% male? And in the summer, the deaths labeled COVID were 50/50? That sort of is suspicious, and so I kept going at that, testing it, and concluded for myself that they were false positives over the summer. And then I wrote to Carl Heneghan, who I was at medical school with, who I haven’t spoken to for 20 years.
Alex: Really? Don’t just pass that. What was he like as a young man?
Dr Craig: In university?
Dr Craig: In his way, he was much cooler than me.
Alex: I bet, was he into, like, The Stone Roses and stuff like that?
Dr Craig: I wouldn’t comment on his musical tastes.
Alex: I bet he went to gigs. He must’ve done.
Dr Craig: Sure. He was a good guy.
Dr Craig: Yes. So I wrote to him and I said, “Look. I think I found this. What should I do?” And he said, “Just get on Twitter, get it out there.” And so that’s when I joined Twitter. It was sort of mid-September, trying to spread the messages. And since then I’ve been digging and digging and digging through the data. I feel like actually I need to change tack. We need to. Now, well, there’s enough evidence now. There’s enough.
What matters is communicating it. I don’t think I’ve been terribly good at communicating it, even though you’ve said flattering things, because I communicate with graphs and with numbers. I communicate as a scientist, which isn’t accessible to everybody. And I think I need to just concentrate on making this…getting the message out in a way that everybody can understand because while we’ve… You know, my followers are physicists and mathematicians, and that’s not the only people. We need to get the message out to the powerful people.