Covid skeptics: Podcast Extract

N/B: This is a fragment of a script for a podcast episode first released back on June 8. I am publishing it now to express my concern that Prof Gupta, who appears to be on the fringe of COVID-19 scholarship and persistently comes up with theories to downplay its significance and importance, even when they contradict the evidence, has been invited to recent discussions at Downing Street. It seems they are looking for contrarian views: no bad thing in itself, I suppose, but when they’re this muddled, I do wonder.

You can see something similar happening with COVID. One example that I will name is Professor Gupta from Oxford, who had another piece out lately that circulated in my orbit. This is the same person behind the group that released the paper on March 23rd which suggested that half of the UK had already been infected with COVID-19. That paper — which really expressed the most wildly optimistic case for a virus that spread very very quickly but was very mild in most cases — I gave some benefit of the doubt, despite disagreeing with its conclusions, because it was couched in terms of “we should do lots of antibody testing to reduce uncertainty and figure this out right away”, and not “50% definitely have COVID.”

Since that paper was published as the “Oxford Model”, nearly two months ago as I write this, it’s become increasingly clear that any idea that the 50% of people were already immune in March was wrong. The serology testing that was asked for has now happened, and the most recent results suggest that around 5% of the UK had had COVID by May (rising to 17% in London.)

But Professor Gupta has now changed her argument — I’m taking this all from Unherd where she was interviewed — and is suggesting that the serology testing that was considered to be so important is actually not relevant, and doesn’t contradict her thesis, and that instead people are just immune due to cross-immunity from other coronaviruses, such as those that cause the common cold.

So the argument appears to have changed from “the disease has already spread through the entire population and therefore probably has a very low fatality rate, let’s just check with some antibody testing” to “actually most people are already immune to the disease because of other coronaviruses, antibody testing is useless and doesn’t contradict my theory” Now, ignoring for a second that the Professor, after being proved wrong in one argument, has changed to another that contradicts it — because if everyone was immune already, it’s hard to see how the virus could spread this quickly and to so many places — you would think that this would require a change on at least one point.
If, in fact, most people have not been infected with COVID — because they are already immune — then presumably the fatality rate should be higher than she first suggested. It just happens that infections are less likely because of existing immunity in the population. But Professor Gupta doesn’t concede this point, instead arguing that “the infection fatality rate is probably between 0.1% and 0.01%, probably closer to 1 in 10,000.”

Given that at least 36,000 have died of COVID in the UK already, if the IFR was really 1 in 10,000, it would imply that 360 million people in the UK have had COVID-19 — an impressive feat for a country with a population of just 70 million. Let’s be charitable and say that “closer to 1 in 10,000 than 1 in 1,000” lies at the midpoint, 0.05%. Then we’d still need everyone in the country to have been infected — and this by a virus that apparently most people are magically immune to anyway.

In response to places like New York, where the fatality rate is already running at 0.15% of the entire population — assuming everyone was infected — she simply says that “where vulnerable people are gathered together, the disease can “rip through” them more easily.” So apparently the entire state of New York is a population that’s more than 10x more vulnerable than she expects for the average person, and for some reason, the people there don’t get colds or don’t have this inexplicable, “natural immunity”, while the people of Denmark, which began social distancing earlier, or Beijing or Seoul, where cases have been much more strictly controlled from the start, somehow do.

She further argues that evidence for this is that “the course of the virus has been the same in countries around the world, whether there has been a lockdown or not, which is most easily explained by a build-up of immunity in the population.” So not pre-existing immunity, but immunity that arises once you’re infected; what else could a “build-up of immunity” mean? And how can one claim that the course of the virus has been the same regardless of whether and when countries locked down, when countries that implemented early social distancing measures, testing and tracing, like South Korea, has 5 of deaths per million population, while the UK has more than five hundred per million population?

I think I’ve spent enough time on this. Professor Gupta made her mind up two months ago — and, in the rest of the interview, where she talks about the social harms and economic cost of lockdown, you can see her motivations for making this case more clearly — and is now just casting about for different arguments, even when they contradict each other, to support the general theme: “this virus is less severe than everyone else says”, a scientific point that masks the underlying political/implication point “the lockdown is more costly than it is beneficial.”

There’s a place for debate about whether the lockdown is more costly than beneficial, just as there’s a place for talking about heterogeneity in networks, and a place for talking about the possibility of cross-immunity from other coronaviruses — these are all really interesting, potentially crucial topics in understanding the disease. What I don’t like to see is people choosing their camps and refusing to be swayed by any new evidence or even address their critics directly.

And so you can have the COVID gish gallop. This disease isn’t going to be a problem, it’s being contained, and anyway it’s not that bad, much less fatal than you think it is, or everyone is already immune to it anyway, and if it isn’t then there’s a cure, and if the cure doesn’t work then there’s nothing we can do because lockdowns don’t work either. On the extreme end, you have the deaths are fake (smacks of the temperature records being fake!), or that everyone would have died anyway (which is much like the argument that anthropogenic climate change actually comes from natural variability). So you can see that we are used to these parallels and the willingness of people to rapidly switch between arguments. The difference is that a rapidly-unfolding pandemic tends to prove people wrong faster than climate, which takes years and decades to get reliable data on. Yet it doesn’t seem to change people’s minds once they’ve decided what camp they’re in. Funny, that.
In the wingnut fringe, we see the same things. Global warming is a conspiracy by Al Gore to sell books, or COVID is a conspiracy by Bill Gates to sell vaccines. Ridiculous but this is where people’s minds go. And meanwhile the poor old epidemiologists are stuck saying: “We are forced to tell people things, even though we wish they weren’t true, because of public safety.” Yeah.