How I became a human plague – and stumbled into one of the most astonishing scientific stories I’ve ever encountered.
By George Monbiot, published in the Guardian 14th April 2021
Rejoice! A mystery has been solved. We now have an explanation for long Covid, a condition afflicting many thousands of people. A super-spreader has been identified. Important as this finding is, I’m reluctant to call for the vector to be eradicated. Why? Because it’s me.
In a presentation to the reinsurance giant Swiss Re, Michael Sharpe, a professor of psychological medicine at the University of Oxford and founder of a Long Covid clinic, proposed that one of the causes of the syndrome was “social factors”. The social factor at the top of his list was an article I wrote for the Guardian, describing the suffering of patients with the condition.
I listed the symptoms of Long Covid and compared some of them to myalgic encephalomyelitis or chronic fatigue syndrome (ME/CFS), the debilitating condition that afflicts around a quarter of a million people in the UK. Press coverage like this, Sharpe claimed, as well as the work of support groups and sympathetic doctors, could induce people to believe they had the illness, thereby spreading it. Long Covid, he appeared to suggest, is partly a psychological condition, so “the best treatment is psychologically informed rehabilitation”. This, we can only hope, will cure people of the fearful pox of Guardian journalism.
I was bemused by the fact that none of the references he gave at the end of the presentation supported these claims. So I wrote to Sharpe asking for his sources. He told me the evidence consisted of “patient reports”, and that “we are seeing many improve with reassurance about the absence of damage and with supported rehabilitation”. But assertions like this do not meet the standard of scientific evidence. Unfortunately, he told me, “I am unable to engage in further correspondence”.
A scientific assessment of the use of cognitive behavioural therapy (CBT) – a form of “psychological rehabilitation” that Sharpe has repeatedly championed – suggests that it’s of no use in treating other post-viral syndromes, and is unlikely to “reduce disability or lead to objective improvement in long Covid”.
But what if, despite the lack of evidence, he happens to be right? What if, by discussing the problem, I’ve caused it? As I look back on my work, my heart sinks. I’ve covered many terrible issues, and the more I’ve written about them, the worse they’ve got. I might be responsible for more human suffering than the entire cast of Alvin and the Chipmunks.
Until now, I’d never heard of Michael Sharpe. But as I began to investigate, I stumbled into one of the most astonishing scientific stories I’ve ever encountered. Sharpe was one of the authors of the famous Pace trial, part-funded by the UK’s Department for Work and Pensions, and published, in 2011, in the Lancet. It claimed to show that CBT and graded exercise therapy (GET) were effective treatments for ME/CFC.
Patients questioned this. When they and other researchers began exploring the methods behind the trial, they discovered some remarkable anomalies. According to an article in the Journal of Health Psychology, after the Pace trial began, the investigators altered assessment thresholds at which they claimed patients had improved or recovered. When the original markers were used, the effectiveness of CBT and GET fell from the reported 59% and 61 %, respectively, to just 20% and 21%. Results were also contaminated when investigators promoted their treatments as successful to the people they were studying.
A group of patients launched a five-year campaign to obtain the trial data, which should have been in the public domain from the outset. Their requests were repeatedly refused until a tribunal ordered Queen Mary University of London to release it. In 2016, the patients, with the help of academic researchers, reanalysed the data and found it did not support the conclusions of the Pace trial: there was no statistically significant difference between the outcomes for people who received CBT and GET and those who did not. Their findings were published in two peer-reviewed journals. It was an astounding victory for citizen science. The suppression of the data and subsequent collapse of the study’s claims has been dubbed PaceGate in the scientific press. In November last year, the National Institute for Health And Care Excellence published its analysis of the Pace trial findings and other claims that CBT and GET can treat ME/CFS. It found without exception that the quality of the science was either “low” or “very low”. As a result, it has now changed its draft clinical guidelines, advising doctors not to promote CBT or GET as treatments or cures for ME/CFS.
There seems to be a strong case for the retraction of the Pace trial papers. In 2019, Sharpe did reflect on Pace’s scientific processes, but doubled down on its conclusions. When Carol Monaghan, a Scottish National party MP, questioned Sharpe’s work, he said her behaviour was “unbecoming of an MP”. And when the Journal of Health Psychology sought to engage the trial’s authors in a reanalysis of their findings, its editor reported “a consistent pattern of resistance to the debate”.
In the presentation in which he named me as a cause of long Covid, Sharpe promoted the original Pace trial paper, without explaining that its findings had been undermined. In a paper in 2019, he claimed that chronic fatigue syndrome was an “illness without disease”, prompting other researchers to accuse him of “medical indifference” and of failing to acknowledge that CBT and GET had been rejected as treatments. Now he seems to have transferred his claims about treatments for ME/CFS to long Covid.
It feels to me like the entirely apocryphal story about Bono. This claims that he once announced from the stage: “Every time I clap my hands, a child in Africa dies.” Whereupon someone in the audience shouted, “Well, fucking stop doing it then.”