This is Science Fictions with Stuart Ritchie, a subscriber-only newsletter from i. If you’d like to get this direct to your inbox, every single week, you can sign up here.
Last week, the NHS Covid app was finally, officially switched off. In case you’d forgotten all about it, as I had, it was the app that “pinged” you when you’d been near someone who later reported a positive Covid test. It was very helpful during the scary time when cases were everywhere, and especially before widespread vaccination, but was increasingly a relic of a bygone age, relegated to a folder on the third swipe-screen of your iPhone.
But at the same time as many of us are saying “Pandemic? What pandemic?”, there are reminders that not everyone has been so lucky. Sufferers of long Covid – the extended after-effects of infection, which can include fatigue, breathing difficulties, “brain fog”, and a very wide range of other symptoms – are still affected.
And long Covid is very much back on the agenda, thanks to an article published recently in The Atlantic. In the heavily-shared piece, science writer Ed Yong argued that “Long Covid is Being Erased – Again”.
You might wonder about this headline. As Yong himself states, hundreds of studies have now been done on long Covid. Governments have dedicated many millions of dollars to researching it. It’s mentioned regularly in the media, and even those who aren’t particularly medically minded know that it’s a serious condition. If it’s being “erased”, the erasure isn’t particularly successful.
But Yong argues that in fact, although most doctors, and indeed most people in general, know long Covid exists, they’re now downplaying it: arguing that it’s less severe than we first thought, and refusing to listen to sufferers about how bad their symptoms can get.
One of the major issues with long Covid is its definition. We don’t even truly know whether there is one condition, “long COVID”, or many different conditions with different causes. Because scientists are still so unsure about exactly which biological process causes each of the lingering symptoms, and because those symptoms vary so much from person to person, manifesting in numerous ways across numerous bodily systems and organs, it’s extremely difficult to measure, diagnose, treat, and study.
And depending on the definition you use, you get very different estimates of how many people have long Covid. For example, Yong cites a Scottish study where the headline finding was that 42 per cent of those infected with Covid were still dealing with symptoms 18 months later, while 6 per cent hadn’t recovered at all. Does that mean 42 per cent of people have long Covid? Or 6 per cent? Somewhere in between?
Looking at the study, one sees a big problem with those percentages: they come from a survey that was entirely subjective, and which didn’t ask people how bad their symptoms were. As was pointed out by the sociologist Zeynep Tufecki (who thought other aspects of the study were nonetheless useful), the questionnaire lumps together people who might’ve said “my 5km running time never quite recovered after I had Covid” with those who said “I am bedbound months later with severe breathing problems”. So even though a lot of people have lingering symptoms, we don’t – from this study at least – know if they’re life-changing or just a minor irritation. Those kinds of nuances don’t come across in Yong’s article, which cites this and other subjective surveys without noting the inherent problems.
Table of Contents
A rather jarring argument
Indeed, Yong isn’t just not concerned about subjective reports of symptoms – he thinks they’re the best way to assess long Covid. He makes what I found to be a rather jarring argument in favour of listening to patients over and above what any “objective” data from blood tests or medical records say:
“The patients’ stories are the ground truth against which all other data must be understood. Gaps between the data and the stories don’t immediately invalidate the latter; they just as likely show the holes in the former.”
This view is understandable: it reflects the strong desire to take people seriously and treat them with respect. But in a medical context, it seems at best naïve, and at worst actively anti-science. That’s because we know that people can be mistaken about what might cause or treat their symptoms. Automatically assuming that their evaluation of their own situation is the “ground truth” – rather than considering all the available evidence, including but not limited to their self-description – could lead doctors and scientists astray.
Simply agreeing with people when they tell you what they think caused their symptoms, and smoothing over all the complications and uncertainties about the condition, is not necessarily the best way to help them – and might be actively counterproductive, especially if it means that they pour time, money, and hope into specific therapies that might not work (for more on this, take a listen to the recent Blocked & Reported podcast on long Covid).
Nevertheless, it’s clear that very many long Covid patients—as well as those suffering from other chronic conditions – have had bad experiences with their doctors. It’s also true that science is far behind where it should be with long Covid. Another recent article on the science website STAT News pointed out that although the US government has dedicated a billion dollars to study long Covid, there’s been very little to show for it. The main US study has stalled, failing to recruit very many participants. It hasn’t contributed anything of note to our understanding of long Covid.
Can exercise help or worsen long Covid?
It seems like the study has gotten wrapped up in bureaucracy, and has suffered from being without a single main scientist whose job it is to drive it forwards. This is a significant failure on the part of the US government’s health agencies.
But it must also be said that another reason for the study’s slow progress seems to have been interminable arguments between the researchers and patients and patient advocates. This is particularly true for the idea that exercise might help some people who have long Covid. This is anathema to many advocates, since a lot of sufferers – as in conditions like chronic fatigue syndrome – experience what’s called “post-exertional malaise”, where even light exercise makes them feel even worse. As the STAT News article relates, advocates have pressed the NIH to abandon trials of exercise therapy in long Covid because they think it could do more harm than good.
But recall one of the main features of long Covid: different sufferers have wildly different sets of symptoms from one another. Some people might well benefit from exercise, even if it harms others, and the only way we can find out who benefits is by doing research on well-informed, consenting long Covid patients. And it’s the same for psychological therapies, which are also a source of controversy in long Covid: although some patients might not benefit, others very easily could.
I worry that Yong’s style – which is at this point verging on advocacy rather than journalism – will help close people’s minds to taking part in research about their condition, and thus hold back our understanding of long Covid. After all, if you think we should simply ask sufferers what their symptoms are and privilege their views over any medical tests, why should we bother doing medical tests in the first place?
And it’s actually a bit worse than that. Reading the article, one doesn’t just get the impression that sufferers know best: one gets the impression that Ed Yong knows best. At several points he implies – at least to my mind – that very many people have long Covid but won’t admit it, or are in denial about it. Maybe you have it too.
Long Covid became politicised
I don’t think this is healthy science journalism. Writing accurately about science means expressing a great deal of uncertainty about results, theories, and even medical conditions. Turning yourself into an activist-advocate for long Covid patients, and darkly hinting that the condition is an awful lot worse and more widespread than we might believe, is the opposite of expressing scientific uncertainty. And that’s especially true when the evidence from studies is so murky on this topic.
What seems like a very long time ago, in 2021, I wrote about long Covid. Specifically, I noted that those of us who were trying to push back against “Covid Sceptics” who were minimising the impact of the pandemic, were wont to use long Covid as a debating tactic. “How can you say a Covid infection isn’t a big deal,” we’d argue, “when it could potentially cause long Covid too?”.
There’s nothing necessarily wrong with the argument in and of itself: long Covid is a genuine worry, and it was high on the list of reasons to avoid catching the virus, especially pre-vaccination. But it became just one more thing you’d say in a debate; something you’d express certainty about, when in fact nobody could really be certain about the definition, prevalence, or impact of long Covid.
In other words, long Covid became politicised. And when a scientific topic becomes politicised, the science almost always suffers. And where politicisation distracts from calm, sober science, the people who’re losing out most of all are the patients.
Other things I’ve written recently
This has become grimly inevitable now, but another very famous psychology study—this time the one from 2012 on how higher social-class people are more unethical—has failed to replicate. I wrote about it here.
How worried should we be about the impact of highly partisan TV news, like Fox News and MSNBC? In a new paper, two political scientists argue that the answer is probably “more worried than you previously thought”. I assessed the evidence in an article, and made a terrible confession about my TV-watching habits…
Everyone’s talking about AI – including the legendary computer scientist Geoffrey Hinton, who quit his job at Google this week in order to be able to focus on warning the world about the potential dangers of AI. What are those dangers? Well, as I argued earlier this week, the media provided a rather misleading picture of Hinton’s argument.
Science link of the week
Here’s an article from WIRED on the very high levels of particulate pollution in subway systems – including the London Tube. I thought it was interesting, and particularly well-written since it doesn’t stoke panic or go beyond the data. Instead, it raises the possibility that the pollution might have health effects but urges us not to “jump to conclusions” before we have better evidence.
Thanks for reading what I hope wasn’t too controversial an edition of Science Fictions. See you next week. In the meantime I’m on: [email protected]
This is Science Fictions with Stuart Ritchie, a subscriber-only newsletter from i. If you’d like to get this direct to your inbox, every single week, you can sign up here.