Retraction Watch

Tracking retractions as a window into the scientific process

UK tribunal orders release of data from controversial chronic fatigue syndrome study

with 23 comments

court caseA tribunal in the UK has rejected an appeal by Queen Mary University of London, who sought to reverse a previous order that they release data from a controversial 2011 paper in The Lancet about chronic fatigue syndrome (CFS).

The decision is one in a long series of judgments about the so-called PACE trial, which reported that two treatments — known as cognitive behavioral therapy and graded exercise therapy — helped alleviate the symptoms of the condition. But ever since The Lancet article and follow-up papers have been published, patients and critics have questioned the conclusions and clamored to see the raw data.

The main criticisms: The findings may prompt some to believe chronic fatigue is a mental, not a physical, disorder, and the PACE program could actually be harmful to patients by encouraging too much exercise. These criticisms were recently bolstered by a re-analysis of the evidence by the Agency for Healthcare Research and Quality, which downgraded its original conclusions about the effectiveness of cognitive behavioral therapy and graded exercise therapy.

In March 2014, Alem Matthees petitioned Queen Mary University of London (QMUL) — home of some of the authors of the PACE trial — to release the anonymized patient data behind The Lancet paper, under the UK’s Freedom of Information Act. QMUL refused, citing concerns doing so would “prejudice its research.”

Matthees appealed to the Information Commissioner, which ruled in October, 2015, that QMUL had to hand over the data. The university appealed that decision to the First-Tier Tribunal, which ruled earlier this month that it would uphold the Commissioner’s decision.

QMUL’s arguments for why it shouldn’t release the data included that they were “sensitive patient data” that could, with some effort, unmask patients’ identities. Furthermore, as the Commissioner’s report notes, QMUL argued:

QMUL has no permission from participants to publish data, and to require disclosure would damage trust and jeopardise future studies…

The institution also argued:

Publicity surrounding disclosure would cause anxiety to participants, particularly when one considers the potential hostility that could be shown to them by certain groups.

Two participants had withdrawn their consent following the disclosure of some 600 pages of material including details of around 50 serious adverse events and reactions recorded during the trial. This caused a delay of four months in the project and increased costs.

The Commissioner, however, ruled that the data had been stripped enough of identifiers, and it was unlikely that disclosing the data would cause a significant drop in patient participation, nor hurt the university’s future reputation or funding.

QMUL appealed this ruling, citing many of the same arguments (you can read them in detail in the report.) Matthees, in turn, cited a petition with 12,000 signatures to release the data, noting “the public interest significantly outweighs the overstated risks.”

Ultimately, the First-Tier Tribunal upheld the Commissioner’s rulings. On the “key issue” — whether patients could be identified by releasing the information — the Tribunal was split, with the majority accepting the Commissioner’s opinion that identification was “an extremely remote possibility.”

Valerie Eliot Smith, a UK lawyer who has been living with CFS (also known as  myalgic encephalomyelitis), has been closely following the case. She explained that this is one of many requests for the PACE data under FOIA since The Lancet paper was published in 2011. What made this request “groundbreaking,” she added, is that it was the first to make its way up the appeal process, including a three-day hearing with witnesses.

She told us:

Many people within the international ME community have been attempting to make sense of the seemingly bizarre claims made by the authors of the PACE Trial since its 2011 publication in The Lancet. This judgment represents a defining moment for the community and for the Trial itself, whatever the final outcome of any further appeal.

This isn’t the only PACE paper for which the raw data have been requested — recently, James Coyne at University Medical Center Groningen in the Netherlands asked for the data from a follow-up to The Lancet paper, published in 2012 in PLOS ONE. Kings College London (which co-owns the rights to the data), called the request “vexatious,” and refused.

Our co-founders Ivan Oransky and Adam Marcus weighed in on the debate last year, in a STAT column titled: “To keep science honest, study data must be shared.”

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.

Written by Alison McCook

August 17th, 2016 at 11:30 am

  • Leo August 17, 2016 at 12:45 pm

    One important aspect not mentioned here is that shortly after the publication of the PACE trial, the media started reporting that the authors were the target of a campaign of harassment, and that they were facing death threats.

    The authors turned down a number of freedom of information requests on the grounds that they were “vexatious”, or in other words, meant to harass them. During the tribunal hearing, they argued, among other things, that disclosure of anonymized data would put PACE trial participants at risk of being identified and harassed by activists.

    The tribunal exonerated patients.

    It was clear that (Professor Anderson’s) assessment of activist behaviour was, in our view, grossly exaggerated and the only actual evidence was that an individual at a seminar had heckled Professor Chalder.
    Professor Chalder’s evidence when she accepts that unpleasant things have been said to and about PACE researchers only, but that no threats have been made to researchers or participants. The highest she could put it was that some participants stated that they had been made to feel “uncomfortable” as a result of their contact with and treatment from her, not because of their participation in the trial per se.

    So right now it looks like these stories about threats and harassment were mostly or entirely a dirty trick used to smear critics and prevent independent scrutiny of the data.

  • Not-yet-knighted August 17, 2016 at 1:13 pm

    “The findings may prompt some to believe chronic fatigue is a mental, not a physical, disorder”

    I don’t think that this is accurate. The primary criticisms that I am aware of:

    – The trial was non-blinded and included aspects likely to encourage particular problems with response bias for the CBT and GET arms.
    – The trial’s more objective outcomes did not support the researchers positive claims, founded upon subjective self-report questionnaire scores, yet these more objective outcomes have been downplayed or ignored by the trial’s researchers.
    – The researchers have failed to release result for the outcomes pre-specified in the trial protocol. The change to their criteria for recovery has probably attracted the most criticism, as patients could be classed as ‘recovered’ even if they had reported a worsening of scores from baseline for the trial’s two primary outcomes. They have also failed to release results for their pre-specified primary outcomes.
    – Participants were not properly informed of COIs involving the insurance industry and DWP.

    The petition cited above complains about factual inaccuracies and protocol deviations. I could see no mention of PACE leading to CFS being classed as a mental health disorder. Nor have I been able to find any reference to this concern in Mr Matthees writing.

    I think that it’s unfair to present this controversy in terms of the classification of CFS as a mental health disorder while ignoring the more substantial and important criticisms that have been made of the trial. Whether or not CFS is classed as a mental health disorder, patients should still have access to reliable information about the efficacy of the treatments available to them. It is wrong for QMUL to continue to fight against the release of information that would allow for the calculation of their own pre-specified outcomes.

    • Samuel Eglington August 17, 2016 at 3:40 pm

      The principal authors are strong advocates of the biopsychosocial model of ME/CFS. The treatments of Graded Exercise Therapy largely developed by the authors and and the form of CBT used in the trial only make sense within the context of this model. Therefore finding that GET and CBT are the most effective treatments supports the biopsychosocial model at the expense of the evidence that this is a physical illness. Also at the expense of patients preferred method of managing this illness which is pacing. If the hypothesis that ME/CFS is primarily caused by deconditioning and false illness beliefs developed during a period of illness due viral infection, exposure to chemicals or prolonged stress is false and ME/CFS is a result of continuous organic illness processes, such as the many energy metabolism problems, release of inflammatory cytokines lack of blood supply to the brain and so on and so forth that have been found over the years. Then the treatments of GET and GBT make no sense and are potentially harmful.

      The PACE trial is the principal evidence supporting the biopsycosocial model. Therefore I would argue that it isn’t possible to separate the classification of the illness from “more substantial and important criticisms” about “factual inaccuracies and protocol deviations”. If “factual inaccuracies” invalidate the paper then the biopsycosocial model is not supported by the evidence and the implications for the treatment both medical and social of those of us that suffer the illness are huge. It would validate our account of what we have suffered and remove competition for research funding from biopsycosocial advocates, hopefully speeding up the development of effective treatments.

      • Not-yet-knighted August 17, 2016 at 4:55 pm

        Regardless of the models which led to the development of CBT and GET, they could both be effective treatments for CFS without CFS being what would conventionally be understood as a mental disorder. Equally, CFS could be a mental health disorder even if CBT and GET were not effective. We do not have a clear way of meaningfully distinguishing ‘mental’ from ‘physical’ disorders anyway.

        If the the history of their development and the models underpinning CBT and GET’s delivery were different, but everything else stayed the same, I do not think that this would make the PACE trial a better piece of research. The biopsychosocial model has also been used to justify poor quality research and seemingly manipulative practices for ‘physical’ disorders like MS and cancer.

        When there are so many clear problems with the PACE trial as a piece of research, the matter of how CFS should be classified seems a needless distraction, and one that has been repeatedly used by the PACE trial researchers to avoid discussing more important details.

        • Samuel Eglington August 18, 2016 at 4:46 pm

          So you can’t meaningfully distinguish between mental and physical disorders in the case of ME/CFS but you can in the case of MS?

          While I welcome your interest in and criticism of this paper it is important to recognise that the law distinguishes between mental and physical and nobody would expect to be sectioned for a physical illness. Nor would they expect a charity to have to fight over 148 court cases of wrongful accusations of child abuse against parents by medical professionals if it was considered a ‘physical’ illness, the designation is important for sufferers. It is also important to recognise that GET flies in the face of all other research by other ‘physical’ branches of medical science. There is no way that anybody looking at this body of literature could come up with GET as an effective treatment for the illness whose principal symptom is a worsening of all symptoms in response to exertion. Failing to include this in the diagnosis of trial participants is arguably one of the biggest problems with the trial, because it doesn’t eliminate sufferers of other ‘mental’ conditions. The involvement of psychologists in this illness has lead to 2 control groups being used trials 1 healthy and 1 depressed. Results consistently show abnormalities in ME/CFS sufferers but not in depressed and healthy controls who are very ‘physically’ similar. Therefore patients with ME/CFS are likely to respond differently to treatment from patients with ‘mental’ conditions and the failure to remove them from the trial biases the results.

          I suggest reading Mary Schweitzer’s comment and trust you will receive your knighthood in due course.

  • Mary Schweitzer August 17, 2016 at 4:15 pm

    Thank you for posting this information. I thought I might add a little information for those unfamiliar with the PACE study.

    It has long been the theory of a group of psychiatrists who call themselves the “biopsychosocial” school of medicine that CBT, combined with graded exercise (together called CBT/GET) does not simply help patients with this disease – it CURES them. The theory is based on the assumption that the symptoms experienced by the patient cannot be explained by physiological evidence. Rather, the patient has “false illness beliefs” that are reinforced by deconditioning caused by “playing the sick role.” For two decades they have claimed that ten weeks of CBT, to teach the patient she really isn’t sick (and it is always a “she”), combined with ten weeks of graded exercise to get her back into condition, will return the patient to normal functioning. [In recent years they have dialed back both the conviction that a cure will be achieved in ten weeks, and that all patients can be cured by it.] At the same time, CBT/GET has become the ONLY treatment offered by NHS, with unfortunate consequences for many patients (in polls, over 70% of CFS and ME patients who have undergone CBT/GET have said that it made them worse, often permanently worse).

    The PACE study was commissioned by the British government (at a cost of £6 million pounds) to prove, once and for all, whether CBT was an effective treatment for “CFS/ME.”

    The study used the “Oxford definition” for this disease, which requires six months of debilitating fatigue WITHOUT physiological explanation. In contrast, definitions formally adopted in the US and Canada require at least some physiological abnormalities.

    Last year, the Agency for Healthcare Research and Quality (AHRQ) of the US Dept. of Health and Human Services concluded in a report on “ME/CFS” that the Oxford definition should be retired – and recently reversed their earlier approval of CBT/GET as an option for patient treatment because the only evidence for the efficacy of CBT/GET is based on the Oxford definition.

    The PACE trial also deviated in critical ways from the original study outline published before the trials began – most importantly, all objective criteria were jettisoned and the study ended up relying solely on subjective questionnaires.

    Since its publication in 2011, both patients and the international medical community as a whole have questioned the methods and conclusions of the trial. All efforts to acquire an anonymized copy of the data set have been denied, with the authors and institutions involved often insisting that these requests were “vexatious” in nature. That is the background for the current tribunal finding that the data (which already is anonymized) should be released. It is a victory for the standard of replicability in medical science and, ultimately, for the continuing search by medical researchers for the etiology of the disease and, ultimately, effective treatments.

    • Ken August 17, 2016 at 9:06 pm

      If you believe that the methodology of the trial is flawed then why do you want the data. If you believe that there are going to be biases then any analysis will be biased. I have nothing against access to data, but it needs to be done with a certain amount of responsibility. In this case, given that previous criticisms seem to be based on a shotgun approach to finding fault, where even minor defects are declared as fatal. All trials have faults.

      I expect once the data is available that a lot of pointless analyses will be performed in the hope that something will be found that might be embarrassing. I suggest looking at the balance of the trial groups after randomisation for every minor demographic variable. Multiple subgroup analyses until one is found where the significant results disappear. Or change how missing data and dropouts are handled.

      • Oppie August 17, 2016 at 10:38 pm

        Methodological problems don’t necessarily mean the data is useless. The way data is handled after collection is just as much a methodological issue as the way it is collected in the first place.

        The reason for the data request is to allow independent analysis as per PACE’s own original published protocol, something that PACE themselves failed to do and continue to refuse to do. I also note that they have persistently ignored or trivialised the objective results from their own study, which collectively fail to offer any support to their therapeutic claims.

        Why are PACE, and people like you – who presumably claim to be ‘defending good science’ – so hostile to that original protocol analysis being done?

        What are you frightened of finding out?

      • Mary Schweitzer August 17, 2016 at 11:15 pm

        Ken commented above: “If you believe that the methodology is flawed then why do you want the data.” That is a fair question.

        I am posting my response here because it disappears if you post it as a reply.

        Oppie answered it well, I believe:

        “The reason for the data request is to allow independent analysis as per PACE’s own original published protocol, something that PACE themselves failed to do and continue to refuse to do. I also note that they have persistently ignored or trivialised the objective results from their own study, which collectively fail to offer any support to their therapeutic claim.”

        There is criticism of the way the data set was collected. There is criticism of the way the study was then conducted and presented to the public.

        I actually have a different problem with the PACE study in its published form in Lancet (2011).

        The researchers show a complete lack of acquaintance with the literature in the field that would contradict their own primary hypothesis. At least, if they are acquainted with it, they do not share that with the public. If all you read is the psychiatric literature on Cognitive Behaviour Therapy and Graded Exercise, you will learn little if anything about the parallel universe of literature that has been conducted by other medical sciences.

        I wrote earlier that the biopsychosocial theory behind CBT/GET as a cure for ME or CFS requires – requires – that there be no medical explanations possible for the somatic symptoms experienced by the patient. But there has been a body of literature by biochemists, immunologists, virologists, and other researchers that has been growing in the 25 years since the CBT/GET prescription first appeared on the public stage, and the rate of growth has speeded up in the past 5 years.

        I do not understand why the editors of Lancet and other scientific journals do not require that the authors offer at least a footnote on the scientific literature that might disprove their thesis. As medical science becomes more and more specialized, it is impossible for everybody to keep up with all fields. There must be a guidepost to literature in other fields on the same subject. To omit that – most importantly, to omit that when the omission is essential to your thesis – does not fit any definition of scholarship with which I am acquainted. To me, this is the most egregious error of all.

        If you have an interest in the scholarly literature on ME, let me suggest starting with this recent bibliography by Leonard Jason et al of DePaul University, focusing on biomarkers:

        How would you apply that scholarly research to the hypothesis that was tested in the PACE trial?

    • Bill Anderson August 18, 2016 at 12:58 pm

      This piece continues the confusion. To refer to the ailment as a disease is to prejudge the issue. The sufferers clearly have a serious illness – they have symptoms which are seriously distressing and which limit their options, causing disability. We do not know how this comes about nor what factors may influence the illness for good or ill. A cohort of sufferers are convinced that the illness stems from a bodily disorder, a disease. That contention has lacked evidential support though that may yet be forthcomng. It may equally turn out that the illness has no source in bodily pathology. That is the contention between a sub group of sufferers and a wide range of clinical investigators and it is wickedly inopportune to assume the conclusion before adequate evidence is to hand.

      • Mary Schweitzer August 18, 2016 at 2:46 pm

        I am guessing you did not read the peer-reviewed published article by Leonard Jason that I suggested above. You might want to do that, so you can have an idea of the depth and breath of research from medical specialties other than psychiatry.

        For example, there are the two-day CPET studies conducted by Chris Snell, Staci Stevens, and others. (The test measures the ability to utilize oxygen and expel carbon dioxide at maximal cardiac effort – it is commonly used in cardiology practice but also by athletes in training.) They found that while deconditioned controls and high-functioning ME/CFS (Canadian 2003) patients scored the same on their first CPET test, the next day the scores of the controls were the same or even a bit better, while the patients’ scores plummeted by as much as 50%. This study has since been replicated by universities on three continents. There are now studies examining the physiological changes in patients at different stages of CPET testing.

        Another area where there has been extensive research is the abnormal cytokine patterns found in patients diagnosed with ME/CFS (mostly Canadian 2003, some Fukuda 1994). The most recent study, by Ian Lipkin a noted virologist at Columbia University, indicates that the cytokine patterns differ WITHIN patient groups when separated by length of illness.

        A third example where research is converging is natural killer cell function, which can not only distinguish between patients and controls, but also can serve as a marker for the severity of the illness.

        There is also a clear subset of patients – perhaps a majority – who struggle with chronic low-grade viral infections. Several have died of myocarditis, with the viruses lodged in the heart muscle. Studies have found active viruses such as HHV-6 and cytomegalovirus active in the spinal fluid of a subset of patients with ME/CFS.

        These are just three examples. At the moment, researchers in several nations are searching for an easy-to-test biomarker that can be generally agreed upon. That will greatly increase the accuracy of further studies.

        Ron Davis at Stanford is in the midst of a big data study, which includes patients with severe ME, along the lines of the human genome project of which he was a critical investigator. Japanese researchers have recently published a study suggesting that the beta human herpesviruses (HHV-5, or cytomegalovirus; HHV-6A&B; HHV-7) can be used as biomarkers for ME/CFS. Don Staines’ group at Griffith University in Australia has been studying intracellular abnormalities, particularly SNPs (single point genetic mutations) that affect transient receptor potential (TRP) function, and have found a consistent abnormality among patients in the ion channel receptor TRPM3.

        All of these researchers are in the hunt for an easy-to-test biomarker that can be generally agreed upon. That will greatly increase the accuracy of further studies. And again, I have only listed a very few – there is much more out there.

        Finally, the highly-regarded Institute of Medicine of the American Academy of Sciences, in its report published last year, Beyond Myalgic Encephalomyelitis/Chronic Fatigue Syndrome, suggested a new name which would explicitly use the term “disease.”

        If you disregard research conducted using the Oxford definition, there is little doubt there is a disease process in ME/CFS. Indeed, to use the Oxford definition to claim there IS no disease process, and/or no physiological explanation for the patients’ symptoms, is to create a tautology: if there IS a physiological explanation, the patient won’t meet the Oxford definition.

        • Bill Anderson August 18, 2016 at 6:19 pm

          The problem is much deeper than I had imagined. The Jasona paper in its title accepts the ME terminology. Myalgia is certainly a frequent symptom but encephalomyelitis? Regularly used but underpinned by significant evidence? A thorough assessment might have eschewed this and stuck with CFS which is concise and precise if not all encompassing.

          • Matt August 18, 2016 at 8:42 pm

            Bill Anderson
            The problem is much deeper than I had imagined. The Jasona paper in its title accepts the ME terminology. Myalgia is certainly a frequent symptom but encephalomyelitis? Regularly used but underpinned by significant evidence? A thorough assessment might have eschewed this and stuck with CFS which is concise and precise if not all encompassing.

            It is about as concise as someone giving a 9 hour filibuster after drinking 5 litres of vodka. I would agree in the sense that definitive proof of encephalomyelitis in living patients is lacking for the time being. ME in itself has historic value much like diabetes does. Diabetes really does nothing to describe the technicalities of the condition. ME goes back to the time of Melvyn Ramsey who approached the condition in recognition of its physical nature (it is now intellectually impossible to argue that large numbers of patients within the often wide and various ME/CFS diagnostic criteria have observable and measurable physical symptoms which cannot be explained by a psychiatric aetiology). CFS is insensitive to patients and can in fact deeply offensive. George Carlin once spoke about how a change of words can deny the severity of illnesses that sufferers are forced to endure. In this instance, Chronic Fatigue leaves out things like, muscle pain, autonomic dysfunction (low blood pressure, pots, dizziness, blackouts, cold sweats), cognitive dysfunction (inability to form sentences, or concentrate on tasks).

            To be frank, as a severe ME patient CFS is an insult, forgiveable of course, people just do not know.

            I must say, I’m a little curious at how easily you could reach the conclusion that CFS is concise and precise. It is neither, it is objectively non-committal and overreaches into a potentially (in my opinion, near-certain) heterogeneous group of patients. Especially since symptoms beyond fatigue have been measurable for over half a century.

          • Mary Schweitzer August 18, 2016 at 11:37 pm

            I am an ME patient (I was a college professor before my collapse in 1994). I had a progressive form of the disease. I most like was in a cluster outbreak at my university in 1990, but I kept trying to push through my increasingly limited capabilities until one day I had a blackout in my office and when I came to, I was the creature from outer space. And yet, it still got worse from 1994 to 1999. By the end of 1998 I couldn’t even brush my own teeth, although I could get to the bathroom and back. I could have gotten worse. There are patients who live in the dark, in pain, on feeding tubes. They – and I – have what we call severe ME.

            I live in the United States, and, fortunately, have never been subjected to CBT/GET. I have had good specialists through most of the past 22 years.

            I have been in a lot of studies: natural killer cell function 2-3%, 37kDa Rnase-L defect, abnormal cytokine profile, abnormal SPECT scans, abnormal Holtar monitor test, inability to pass simple Romberg test, foot drop, VERY abnormal CPET scores (so low as to qualify me for permanent disability as a cardiac patient had we wanted to go that route).

            I suffered from intense pain behind my eyes and in the back of my neck 24/7, and frequently also had migraine-level headaches. I could not turn my neck – I had to move my whole body. My muscles ached all of the time, especially the large muscles. I had ataxia, dyslexia, expressive aphasia, disorientation, very poor short-term memory, blackouts, and profound confusion to the level that I once poured an entire pot of coffee into a silverware drawer. My daughter would fasten my seat belt because I couldn’t remember what you did with those two things.

            I was in the first study of HHV-6A in CFS by Dharam Ablashi, who was the co-discoverer of HHV-6 and its two variants while at NCI. My viral load was 6 times the level necessary for a diagnosis. My illness began with a bad outbreak of EBV at the university where I taught in 1990 – several of the professors, including me, got sick, tho I had already had mono. Through the illness EBV would come and go – but HHV-6A remained. I also would test positive for CMV, HHV-7, and three strains of Coxsackie B. One researcher joked that my blood was a toxic stew.

            In 2009, when in relapse because FDA had taken the drug away when the head of my practice died, we found HHV-6 and Cytomegalovirus in my spinal fluid. That means it was bathing my brain. Which means encephalomyelitis, while a mouthful, might actually be accurate. I don’t really care. As Matt noted, there are a number of diseases that were named before the etiology was known. My late husband used to say, “I don’t care what you call it: fix her.”

            But I have been lucky. I was able to get into an open-label (that means I know I’m on it), compassionate care (I was far too sick to wait for approval), cost-recovery ($16,000/year not counting the testing) study for the Phase III immune modulator Ampligen, and I do well on it. My immune markers return to normal ranges, and my viruses go dormant. My immune system by itself just is not strong enough to be off Ampligen – twice I went off; twice I relapsed severely within a year – the second time, when I knew what was coming, was not unlike being Robert DeNiro in Awakenings, or Cliff Robertson in Charlie (Flowers for Algernon). I have to be close enough to an infusion site to get infusions twice a week and – again – I am very lucky we have been able to continue this.

            On Ampligen, my encephalitic symptoms disappear completely. I do struggle with stamina – and both relapses seemed to have made it harder to overcome that.

            It is intoxicating to be able to WALK – to stride. To walk on a beach. To be able to drive a car – and to walk on a trail away from the sound of cars. To read again. To have a life again.

            So there you go. There is a clearly identified subset of patients who test pretty much like I do – the same symptoms, the same immune biomarkers, and many of the same viruses. There are some alternatives to Ampligen for this group – gamma globulin, Valcyte, Vistide – but not enough. We really need to develop more antivirals and not lean so much on vaccines (IMHO). There has also been some success with a cancer drug, Rituxan – whether those patients have the same biomarkers I do, I don’t really know.

            And again, I have to say how lucky I have been. I have many friends who live on much less than what Ampligen costs. The disease is a all too often a one-way ticket to impoverishment. And I have lost too many friends. (Heart disease, lymphomas, suicide) I am also intensely aware of the 85% of US patients who have no diagnosis. That would be 850,000 by the most conservative estimates of the prevalence of ME and CFS in the US. Where are they? It is troubling.

            From my perspective, the PACE study marks the dead end of a theory that turned out to be wrong. That paradigm has had more than its 25 years. It is time for it to go.

        • Bill Anderson August 19, 2016 at 5:43 pm

          The IOM use of “disease” was ill judged. If PACE is flawed then the rational response is to design a fresh study with a better design. The answer to this issue (if there is a single answer) is most likely to come from the method of science fallible and flawed though it be. That process is ill served by a chorus of condemnation espousing a range of preconditions on what conclusions might be acceptable. The successes of science in addressing puzzling illnesses such as AIDS dramatically highlight the value of this approach. Sufferers from this illness are being denied access to possible advances by the vociferous proponents of one view of the situation which must surely dissuade many investigators and funding bodies from becoming involved.

          • Samuel Eglington August 20, 2016 at 11:16 am

            Possible advances? Do tell! What possible advances could be prevented by shredding PACE? GET and CBT are already available to patients as the NICE recommended treatments. Pacing is what most people find themselves doing after GET and CBT doesn’t work or if they are not offered it and general medical care is what doctors offer to all patients. This paper is a dead end. The only hope of advance comes from continued refining of diagnoses criteria and the identification of sub groups through biomarkers, drug trials and such like. GET is considered harmful by 74% of patients receiving it. CBT is not generally harmful but it is also ineffective from the evidence of multiple patient surveys. Pacing can result in significant improvements over time (it has for me) but is not a cure in most cases (or mine). General medical care can be useful but has a very limited range of options for treatment. A fresh study of these approaches would waste years and money that could be better spent on biological studies. These treatments are a dead end as the follow up study revealed none of them worked effectively, even by PACE standards of effective. It is very disappointing that you should contend that an absence of evidence is evidence of absence when so little has been invested in looking. It requires you to ignore patient testimony as to what our illness involves. After all that I have achieved through pacing and the determination involved I simply cannot accept that my continued inability to do any conventional sort of interval training or regularly participate in sport (as opposed to doing it once every now and then) is the product of subconscious self sabotage the improvement in my health means that I am certainly no longer deconditioned. A fresh study of these treatments offers me nothing. Ending the dominance of the biophsycosocial school however is at least to have hope.

  • Oppie August 17, 2016 at 9:28 pm

    Thank you for reporting on this.

    “The decision is one in a long series of judgments about the so-called PACE trial, which reported that two treatments — known as cognitive behavioral therapy and graded exercise therapy — helped alleviate the symptoms of the condition.”

    PACE went way past claiming to be able to merely alleviate. They are claiming to have delivered “recovery”, back to within the normal range for the healthy population.

    But the objective evidence from PACE itself clearly refutes this claim. That is beyond dispute. Not even PACE are citing any of it in support of their claims, and are instead relying entirely on highly manipulated and marginal results from problematic subjective measures.

    With respect, Retraction Watch is being too polite and soft on these guys, and you need to pick your game up. The whole PACE saga, and the general handling of the condition it purports to be studying, has been an extraordinary failure in the integrity and competence of medical science. The implications of which go well past just this area of medicine.

  • R August 18, 2016 at 1:07 am

    Thank you for continuing to follow the ongoing effort of patient advocates and scientists to force the PACE investigators to share their data.

    As a patient, I want to offer a personal reason why this effort is so important. Graded exercise therapy has been suggested as a treatment for chronic fatigue syndrome for many years, in fact my physician suggested it in 1996 to me. Essentially, it is conditioning. As a student athlete and avid hiker, I was no stranger to getting in shape. My efforts on the treadmill, which would have been easy for an elderly adult, floored me. I spent weeks afflicted with symptoms after my best efforts at exercise. It never got easier, it only sickened me and robbed me of whatever modesty energy I had.

    PACE was published in a high quality journal with a great deal of fanfare. I think they had a press release and access to journalists. Any patient with my experiences would balk at the conclusions of the trial where they boasted a 30% recovery rate and improvements in another 30%. I could only conclude that they netted a different type of patient, or that some statistical gerrymandering took place. Unfortunately for patients, the positive message of the trial would filter down into the lowest levels of clinical care. Even as an American I have heard “the best treatment for you is exercise” many times. PACE reinforces the exercise recommendation as a panacea for “fatigue” and trumps anything a mere patient has to say in the exam room in her defense. Nothing frustrates physicians more than a non-compliant patient. When we fail to exercise, doctors think we are lazy and resistant and don’t’ want to help us. They think we really can exercise if we want to.

    By giving independent researchers access to data, patient advocates and scientists can hope to dissect the innards of this trial and find out what really happened. If statistical manipulation has taken place, then perhaps the idea of exercise for chronic fatigue syndrome can start to be dismantled, and the conveyor belt of CBT/’GET studies coming out of this group can be halted for good.

  • Sasha August 18, 2016 at 2:20 am

    Thank you for covering this important story. As you mentioned, over 12,000 patients petitioned Dr. Richard Horton, editor of The Lancet, to retract “misleading” claims made in the 2011 publication of the main PACE results. In the post-hoc analyses in that paper that were the basis for much-heralded claims of “recovery” in an accompanying editorial pre-approved by the study authors, patients’s physical function (measured on the 100-point SF-36 scale) could worsen during the trial to the average level of a Class II congestive heart failure patient and yet they would be considered to have been effectively treated.

    Patients can’t understand how this paper got published in The Lancet or why Dr. Horton will not retract those analyses. They’re clearly insane, and such appallingly bad science is dangerous for patients.

    I hope that you, as prominent journalists, will try to get a statement from Dr. Horton as to why he ignores this well-founded criticism and fails to act. He has ignored the petition and has failed to act on a letter to him (posted openly on Virology Blog and sent last November) from 42 scientists and clinicians also pointing out this flaw to him.

    It’s his job to ensure that corrections or retractions are made when a paper in his journal includes a mistake. I and thousands of patients, and many scientists and doctors, can’t understand why he doesn’t do his job when it comes to ME/CFS.

    Please ask him.

  • Ollie C August 18, 2016 at 5:34 am

    Thanks to Retraction Watch for having their eyes open and hands on the wheel in the midst of the deafening silence that so often characterises the debate around this terribly damaging trial. It is damaging (literally) to many patients through the treatments themselves, it is damaging to patients because it diverts attention & funding from research looking for the mechanism of the disease, and damaging to the reputation of science. How did the system of research science allow such a flawed study to acquire such influence? Why did peer review fail to spectacularly? How is it acceptable the Lancet editor trumpets the importance of good science, whilst putting his hands over his ears about PACE? Why did ALL the journalism around the publication of PACE in 2011 miss all the many flaws, conflicts of interest & results exaggeration? How is a study known to be so flawed still at the core of the justification for the use of these harmful treatments? These issues stretch far beyond ME/CFS. This study was not designed to find the truth about treatment, but to produce evidence to support the favoured treatments – that is not science we should be proud of. PACE = Poor Attempt at Creating Evidence.

    Inexplicably the investigators decided the best way to promote & defend their work and silence their critics was to attack sick patients, to malign them and label them as aggressive activists and criminal harassers. It is clear looking at the press coverage in 2011 that they ran an organised media campaign with this specific intent. This recent tribunal hearing makes clear its conclusion that these claims were wildly exaggerated.

    For anyone wanting to read further, there is a significant source of information about PACE, its history, related papers, prominent critics, media coverage, and the many calls for data release & independent review (by scientists from Berkeley, Stanford, Columbia and UCL) here

  • Lilpink August 18, 2016 at 5:41 am

    I’m surprised not more is being made of the following (copying from a comment I made on Tuesday):

    From the Tribunal Decision document , an observation by Matthees which surely has to be interpreted that Chalder acknowledges the Cochrane Review wasn’t independent and so (in my words) shouldn’t it also be retracted?
    ” He notes that in Professor Chalder’s witness statement at para.67 PACE researchers have
    supplied “requested, anonymised data to independent scientists, as part of normal research
    collaboration”. This Tribunal notes this suggests an acceptance on Professor Chalder’s part that
    anonymization can be sufficiently carried out for disclosure purposes. Professor Chalder states
    that disclosure to the Cochrane review does not count as disclosure to independent scientists
    as all three of the PACE principal investigators sat on the review panel. “

    • Rocina Nante August 24, 2016 at 10:32 am

      Can you elaborate re disclosure to Cochrane Review, dated a month ago, which reviewed GET with the conclusions of no harm and some benefit to patients. The review was dated just a month ago.

      Are you saying that Cochrane reviewed the patient data before publishing their review of GET and PACE?

  • Post a comment

    Threaded commenting powered by interconnect/it code.