By Jerome Burne
22 Apr 2019
In September 2016 I posted a blog here about a research bombshell that had just exploded at the heart of one of the most bitter disputes in medicine – how best to treat a condition known as ME/CFS that involves relentless fatigue.
A team of senior academics had just reanalysed the trial, known as PACE, that supported the standard NICE-approved, psychology-based treatment and found that the claimed benefit didn’t reflect what results had shown – that the treatment was virtually worthless.
I thought it was going to be a huge story and have all sorts of repercussions but instead it was largely ignored both by the press and by mainstream medics. I’ve come back to it now because there are signs that opinion professional and popular is turning against the treatment. These include:
**At the end of a debate on ME/CFS in the House of Commons last January, the House supported the proposal that use of this standard treatment should be suspended.
** Hilda Bastian, a veteran public health researcher and key opinion leader who has worked for organisations such as the Cochrane Collaboration in the UK, the German Institute for Quality and Efficiency in Health Care and the National Institutes of Health in America, has recently come out in favour of a change.
In a recent post, she comments that: ‘I think the balance should, and ultimately will, tip towards the ME/CFS consumer movement.’ In other words, it doesn’t work.
** A key feature of this dispute is the hostility between activist patients and their doctors. Once this was rare, now it is emerging in other areas of medicine along with calls for a better was of handling such disputes. This was the topic of my recent blog.
Angry and insulted patients
It should have been big story back them because ME/CFS (Myalgic Encephalomyelitis / Chronic Fatigue Syndrome/) is a horrible condition that affects an estimated 250.000 people in the UK, 10% of them children. Patients can become bedridden for years and among the 60 symptoms linked to the disorder are a feeling of constant exhaustion – like a dead battery’ – muscle pain, swollen lymph nodes and being very sensitive to sound, light, heat and cold. It costs the NHS and other services billions.
The cause is unknown – although there is no shortage of theories – and there is a deep disagreement about whether it is a real neurological illness, as patient activists claim, or whether it is the result of the patients’ mistaken belief that they cannot exercise.
This puts them in the hands of psychologists who use CBT (cognitive behaviour therapy) to help them change this ‘unproductive belief’ and to exercise a little more each day with GET (graded exercise therapy).
A proportion of patients have long felt deeply angry and insulted by the psychological diagnosis with its implication that they were inflicting this relentless cascade of pain and distress on themselves. The psychologists claimed that the patients resisted the treatment because they didn’t want to be stigmatised as having a mental illness.
The reanalysis of the PACE trial has changed that. ‘For most patients now, the real issue has become that the treatment just doesn’t work,’ says Jonathan Edwards, Emeritus Professor of Connective Tissue Medicine at University College London and an expert on ME/CFS. ‘That’s what a careful reading of the PACE trial shows.’
Highly critical report ignored for years
Faced with this serious challenge, however, the authors of the original trial, called PACE, published in the Lancet in 2011, effective ignored the re-analysis and simply declared that both CBT and GET, did work for some people. And that was it.
Although the reanalysis was immediately available online and later published in the journal Health Psychology, it was clear that only a very few psychologists and medics were interested to know more. And so, the stalemate has continued until very recently.
What’s moving the story on now is that the researchers who did the re-analysis – including senior independent academics from Stamford, Chicago and University College London – have just published a response to the response by the authors of the PACE trial. The PACE authors repeated the claim that their treatment was effective and there was really no problem with the trial. However, the way the critics have responded makes it hard to see how claims that PACE is OK can continue to go unquestioned.
Patient activists fight to get hidden data from therapists
If you are familiar with the long-running debate over statins, you may be wondering how on earth the reanalysis of data from the big ME/CFS trial was done? For decades the data on the statin trials has remained hidden and requests it be made available ignored. The way the ME/CFS patient activist got hold of hidden data to analyse is the most dramatic moment of this civil war.
The re-analysis had only been possible because patient activists had obtained the raw data on the PACE trial held by Queen Mary’s University by sending in repeated Freedom of Information requests. The University then spent £200,000 challenging an FOIs that was accepted and only complied after a tribunal rejected their reasons for withholding it. The report on the tribunal can be found here.
So back in 2016, it was game-set-and-match to the patient activists and yet this apparently was not a news story. The strong determination by clinical researchers to keep raw data on trials secret is another parallel with the response to criticism by those holding the statin data. An FOI request is probably not relevant for statin data as that is commercial, while PACE received public funding.
How the definition of ‘recovery’ was changed
So, what did the critics discover was wrong with the trial? The published version had found that 22% of patients recovered with either GET or CBT vs 7% in the no-therapy group. But the re-analysis by the independent academics discovered that this had only been possible with some fancy statistical footwork. The true figures were dreadful, barely better than no-treatment; 4%, 7% and 3%.
The authors of the PACE had got their respectable results by changing the way they defined recovery. If you want to show people have benefited from treatment, you need to define what recovery looks like. Is it walking two hundred feet unaided or one hundred? Scoring 60% on a test or 40%.?
The trial had originally chosen a stricter measure for recovery and then changed it part way through to a more lenient one. It was the results from the lenient one that were published. The reanalysis figures came from applying the original standards. The lenient versions really were lenient; a patient could enter the trial, based on the original measures, get worse on two out of four criteria, and then still be rated as ‘recovered’ on the new version.
What other data is being fudged in this way?
This was particularly shocking because the PACE trial had seemed copper-bottomed. It had been approved by the Research Ethics Committee and been peer reviewed by the Medical Research Council and was rapidly accepted by NICE as reliable enough to recommend the CBT and GET combo as the treatment for ME/CFS.
If the reanalysis is only half true, the fact that this was the best a trial can get, should set alarm bells ringing for the reliability of evidence-based medicine as it is currently practised.
The re-analysis also picked up on an inherent problem with trials of this sort of therapy – how to prevent patients knowing if they were getting the treatment or not – in other words in ‘blinding’ it. You are much more likely to say you benefitted if you know you got the treatment.
But the original trial made no attempt to deal with this issue. The psychologists were very enthusiastic about the treatment, telling the patients it was ‘powerful and safe and had been shown to be very effective in ME/CFS.’ Those on the comparison treatment of standard medical care were told nothing positive about their treatment at all.
Improvement in physical ability minimal
Even so, on clear tests of improvement in physical ability, the results for those getting the treatment were dismal. After a year of it, they were only able to walk 30 metres further than the control group. Another unambiguous test of improvement would be needing fewer government benefits; after a year of treatment, the percentage of patients getting benefits actually increased. What’s worse was that these negative results were only published years after the publication of PACE’s positive results.
As damning as the re-analysis was, it doesn’t give much of a sense of what it was like being involved with the trial. This was provided by one of the speakers in a debate on ME/CFS in the House of Commons in January 2019. Carol Monaghan MP, SNP spokesperson for the Armed Forces, reported what she had been told by patient groups about their experience of PACE-type treatment.
‘Many who were able to walk when they embarked on a course of graded exercise dropped out of the treatment in wheelchairs or bedbound. Furthermore, patients were pressurised to describe improvements that they did not feel. As the trial progressed and the results did not meet the authors’ expectations, they simply lowered the threshold to define improvement. In some cases, those whose condition had deteriorated were classed as “recovered”. That is simply not good science.’
Therapists comfortably hold two contradictory beliefs
The two recent papers by the PACE trial authors and the critics went into more and fascinating detail on the issue of changing of the definition of ‘recovery’. The PACE authors claimed that there was no good reason to prefer the original definition to the new ones; the critics responded that in any clinical trial, sticking to the way you measure changes is vital if results are going to mean anything.
Then the PACE authors came up with a remarkable justification for their view. They said they: ‘preferred their modified definition because it generates similar rates of recovery to previous studies and is also more consistent with our clinical experience.’ The critics pointed out that the aim of a trial is not to confirm what you expect to find but to do the opposite – you test to discover if your expectations are correct. They add sternly: ‘It is not appropriate to loosen the definition of recovery simply because things did not go as expected.’ Indeed.
Several of the disputed issues are too technical to go into here, certainly for me, but one that has always been at the heart of the battle is whether ME/CFS is an organic disorder or a psychological one. The PACE authors claimed that they didn’t have an opinion either way; all they cared about was whether the treatment worked. The critics showed why this couldn’t be true because PACE trial report assumed the exact opposite.
It stated, ‘that there was no underlying disease process,’ because what was keeping the patients ill was purely psychological: ‘their thoughts, feelings and behaviour.’ The CBT manual used in the trial took the same line, telling patients: ‘there is nothing to stop your body from gaining strength and fitness.’ A claim that would not be true if the root of the problem was, some sort of neurological malfunction.
The looming problem of compensation
As a result of such arguments, the tide of battle over the causes and treatment of CFS/ME is turning in favour of the patient activists. As well as support for suspending treatment with GET and CBT as means of treatment the Commons debate also called on the government to provide increased funding for biomedical research.
NICE – the body that approves medical treatments – currently recommends CBT/GET. But later this year, it is due to holding hearings about whether to continue recommending it. The committee hearing the is reported to be evenly balanced between supporters of each side.
If they do vote for a change, it will raise the thorny issue of compensation. If the theory that it is the patients’ beliefs that are keeping them ill is rejected, then will patients or the parents of children with the disorder, who have routinely been have been threatened with sanctions if they don’t undergo treatment to change their opinion, be eligible for compensation?