Jonathan Edwards testimony must have made an impression. Several stakeholders refer to it in their submission, often attacking him and his reputation in an aggressive way (especially the comments by the RCP are inappropriate).
Here's a selection of what I'd stumbled upon but there might be more (too much to read all, could only skim):
Newcastle upon-Tyne Hospitals NHS Foundation Trust:
We are particularly concerned by the testimony of Prof Edwards […] We are not convinced that Prof Edwards has the necessary expertise and knowledge in clinical trials of non-pharmacological interventions to make this assessment. For example, it is flawed logic to suppose that a trial of CBT or other psychological therapy should somehow attempt to separate the effect of the therapy per se from the effect of the therapeutic relationship. In real clinical practice, it is impossible to have one without the other, and a trial must reflect this reality
With reference to Professor Johnathan Edwards – he demonstrates, in his comments and arguments, a lack of understanding of the factors associated with a positive outcome of therapies, which include developing a personal relationship with caregivers. A positive, optimistic, and empathic relationship with therapists are shown to be important for a positive patient experience and outcome.
More specifically, he states: “The central difficulty for trial design in ME/CFS is the high risk of systematic expectation bias in a fluctuating condition with subjective core features”. This is essentially true, but the same applies to mental health problems (fluctuating condition with subjective core features) where psychological treatments are considered valid options, yet the trials may share some similar design issues.
If the committee are to take into account Professor’s Edwards views on the role of subjective core features in a therapy based research trial, then they also need to take into account the subjective role of the qualitative studies they have claimed as ‘evidence’.
North Bristol NHS Trust - CFS/ME Service:
It is not clear why the expert testimony were included from Professor Jonathan Edwards. It is clear that Prof Edwards’ expertise is in pharmacological research, and not in research into non-pharmacological interventions. It would have been more useful to have an expert witness who understood the issues related to non-pharmacological interventions rather than one who focussed primarily on the problem on non-blinding. At one point (page 6) he criticises studies which do not show objective improvements and then immediately criticises the studies which did show objective improvements: “Some exercise studies show objective changes in indices of fitness but that does not necessarily imply a reduction in illness or disability”, domains which are typically assessed using subjective measures. The impression is of someone who wants to have his cake and eat it.
Prof Edwards states that “unblinded trials with subjective outcomes are specifically considered unreliable.” If NICE as an organisation were to generalise this opinion across all guidelines for example those for back pain, osteoarthritis and multiple sclerosis, what would the impact of this opinion be on these other guidelines? Prof Edwards puts forward the idea that placebo is due to “positive thinking” but does not seem to acknowledge the biological element of placebo, and the element associated with training. It would be advisable to find an expert who could offer a more informed view of non-pharmacological issues, rather than one whose expertise is in a different field.
Register of Lightning Process Practitioners:
Professor Jonathan Edwards claims to have no conflict of interest or bias. However, he does admit to being a board director for a ME/CFS patient forum (Expert testimony p 13) and this may explain his adoption of a, possibly unconscious, bias as evidenced by this statement: “I think it is important to establish certain simple facts. The Lightning Process is a commercial product whose nature is secret and whose value is unproven – thus qualifying as an ‘alternative therapy’. Yours faithfully, Jonathan Edwards, Professor Emeritus, Division of Medicine, University College London”
https://www.virology.ws/2020/05/30/trial-by-error-two-letters-to[1]dagbladet-about-its-me-coverage/. Letters to Dagbladet About Its ME Coverage, 30 MAY 2020: This can also be observed in his desire to devalue the statistically significant result of the SMILE RCT by claiming them to be ‘apparent positive results for the Lightning Process’. This inaccurate commentary on the LP, whilst having no direct experience of it or awareness of the published papers or books on it, brings into question his lack of impartiality and his suitability for the role of expert.
The Royal College of Psychiatrists:
Professor Jonathan Edwards was asked to give evidence on the methodological issues around non pharmacological trials and outcomes. His main contribution concerned non- blinded interventions and what he called “subjective” outcomes (which we prefer to call “patient reported”), contrasting them unfavourably with those that he considered to be “objective” His argument is that trials that are unblinded and have self[1]reported outcomes cannot be trusted or used. Because some unblinded trials are not replicated by subsequent blinded ones, which is true, then all unblinded trials are not replicated, which is not true. His point is that patients who main symptoms are self[1]reported do not give an accurate account of those symptoms after any intervention which is non-blinded. For that reason objective measures, such as biomarkers, exercise testing and so on, which are superior to self-report measures, could be used.
We do not accept this.
We do not accept this. Indeed, the general direction of travel is more in the opposite direction – for example in responding to the views of patients about what matters to them, NICE and others have steadily increased, not decreased, the importance of self-reported outcomes, also known as Patient Reported Outcome Measures (PROMS). The issues around complex interventions that cannot be blinded has been addressed in the MRC Guide to Assessing Complex Interventions, used by most triallists in this field. He also did not mention that these issues are already included in the Grade System as outlined in the Methods Section of this review, including both unblinding and performance bias. Instead he implied that his views represented something new and additional for the committee to take into account.
Do NICE agree?
ME/CFS to the head of medicine at UCH his only comment was that for ME/CFS we would need to be 100% sure of blinding.” His “claims to the contrary” is a systematic review of the placebo effect in ME/CFS
https://pubmed.ncbi.nlm.nih.gov/15784798/. “In contrast with the conventional wisdom, the placebo response in CFS is low. Psychological-psychiatric interventions were shown to have a lower placebo response, perhaps linked to patient expectations”. This finding casts doubt on his main thesis, which is that “subjective outcomes are to be particularly distrusted in trials of CBT and GET in ME/CFS” - the evidence points the other way.
Could NICE please comment?
Royal College of Physicians (RCP):
“There are three expert testimonies. We note that one of the testimonies is from a recognised advocate against trials and evidence in interventions who has served on the Board of Phoenix Rising. Their career and academic standing is based on laboratory research and pharmaceutical interventions, as opposed to guidelines development or pragmatic trial designs for health and social interventions. Our experts believe that the testimony shows a lack of understanding of this area of research. The fluctuating nature of the illness is a problem in RCTs, but if they are sufficiently powered it should not be a problem. Again, these trials cannot be blinded. Respondent bias is recognized in trials, and this is not a reason to dismiss them; and it is possible then to match up those where respondent bias may be a problem, with other measures of functioning or exercise that can be independently assessed. Our experts believe that a lack of understanding on how the theoretical model is simply used to think about the processes that might be in play to explain why the intervention may be effective; in itself, the theory should not be used to dismiss the results of a trial after the trial is completed simply because the reader doesn’t agree with it. Our experts believe that the comment about Cochrane and peer review is misplaced; this review was extensively peer reviewed. Our experts believe that the methods are being criticised here as belief disconfirmation bias is being experienced. Our experts believe that to dismiss all the evidence from trials in ME/CFS is idiosyncratic and that the detail of this presentation is largely anecdotal. Our experts believe that the chair of the guidelines panel or the Director of NICE needs to instruct the panel to dismiss this expert witness as it stands against the whole NICE evidence-based process in general for complex interventions.