Trial By Error: A Reuters Update

By David Tuller, DrPH

Last Sunday, I sent a letter expressing my concerns about Kate Kelland’s recent article to the appropriate person–the Reuters global editor for ethics and standards. After considering the issues, Reuters has decided to add my current academic title to the post and leave everything else as is. I appreciate the first decision and obviously disagree strongly with the second.

I note that the story still refers to me, for some reason, as a “former journalist.” This is false. I am both a currently working journalist and a public health investigator. For some reason, Reuters has determined it is impossible to be both.

I might have more to say about this going forward. For now, I’m posting the letter I sent. (In the actual e-mail, I linked to everything; I didn’t bother to do that here.)

**********

Before posting my letter, however, here’s a brief, irritating “pledge break”–a term familiar to anyone in the US who listens to public radio. In my case, it’s a “crowdfunding break.” As of today, I am now more than two-thirds of the way toward my goal, at 67%, with 12 days left to go.

Once again, here’s the link: https://crowdfund.berkeley.edu/project/14941

**********

Dear Ms Freedman:

I am writing to you in your capacity as the Reuters global editor for ethics and standards. I am a senior fellow in public health and journalism at the Center for Global Public Health at the University of California, Berkeley. The Center for Global Public Health is part of Berkeley’s School of Public Health.

I mention this specifically at the outset because readers of a Reuters story published last month about the illness best-known in the US as “chronic fatigue syndrome” would not have learned that relevant detail about my academic and professional status from how the reporter, Kate Kelland, describes my work. (Here is my page on Berkeley’s site.)

Beyond that, Kelland’s piece presents a distorted picture of the heated scientific debate over the issue and is harmful to the interests and well-being of patients. And there are a lot of them–a million Americans and 250,000 are estimated to suffer from this devastating illness.

Kelland is a UK-based science correspondent for Reuters. Her March 13 story, called Special Report: Online activists are silencing us, scientists say, discusses the experiences of some researchers investigating chronic fatigue syndrome, which is also known as myalgic encephalomyelitis, CFS/ME, and ME/CFS–the latter being the name used most often these days by US government agencies. In particular, Kelland highlights a seminal study called the PACE trial. This study was conducted by British psychiatrists and other mental health experts and reported that cognitive behavior therapy (CBT) and graded exercise therapy (GET) were effective treatments. The first results were published in The Lancet in 2011, with further findings in subsequent papers.

I apologize in advance for the length of this letter, and for the fact that it has taken me a few weeks to pull it together. Kelland’s widely disseminated piece represents a significant violation of core journalistic principles to which Reuters and many other serious news organizations subscribe. I say that as someone who spent ten years as a reporter and editor at The San Francisco Chronicle and served as health editor at Salon.com before receiving masters and doctoral degrees in public health from Berkeley in 2006 and 2013, respectively.

I have also contributed regularly to The New York Times going back to the 1980s. In fact, your former Wall Street Journal colleague Tara Parker Pope has commissioned stories from me for the Well section of The Times, including this piece about my octogenarian mom’s recipe for healthy aging and this piece about being a non-traditional dad. Most recently, last December, the Well section ran this piece about the high cost of dental care, which I wrote for Kaiser Health News.

Given my background, I am familiar with your work and reputation. That’s why I have some confidence that you will read what I have to say in full despite its length, assess my concerns with an objective eye and render a thoughtful judgement as to whether Reuters has covered this situation in a balanced manner.

First, I will provide a brief background on the debate and my own involvement in reporting about it. Then, I will document how Kelland’s story offers Reuters readers a biased version of events by misrepresenting both my professional status and the scientific debate.

*****

From 2010-2012, I wrote a series of stories for The New York Times about the since-disproved possibility that ME/CFS was associated with a mouse retrovirus called XMRV. The first PACE trial results were published during that time, and I began to read analyses of the study being written by patients. It seemed to me that patients were raising legitimate and important scientific questions about the research. I vetted my concerns with colleagues in epidemiology, biostatistics and other disciplines, at Berkeley and elsewhere.

These experts were uniformly shocked at the PACE study’s methodological missteps and aberrations as well as the fact that the results appeared in The Lancet and other prestigious journals. In fact, my epidemiology colleagues at Berkeley now use PACE in seminars as a case study of poor-quality research, a fact you wouldn’t learn from Kelland’s piece.

I ultimately published a 15,000-word investigation of the PACE study in October, 2015, on Virology Blog, a well-regarded website hosted by my colleague and friend Vincent Racaniello, a Columbia University microbiology professor. I have since expanded that investigation to include the body of CBT/GET research, publishing more than 150 posts on Virology Blog as well as writing and co-writing articles and opinion pieces about the issue for The New York Times, the journal Health Affairs, the MIT-based science magazine Undark, STAT, and elsewhere. My work on ME/CFS has been mentioned or cited in Science, Nature, The Wall Street Journal, The Guardian, and Slate, among other publications.

(I have cc’d Racaniello on this e-mail because any criticism of my ME/CFS investigation is also by definition criticism of Virology Blog for hosting it. Kelland conducted an e-mail interview with Racaniello but did not include his strong defense of my work in her story.)

My Virology Blog investigation has helped generate international alarm about PACE and the accompanying recommendations for CBT and GET, the treatments promoted for decades by Kelland’s protagonists. Last summer, more than 100 scientists, physicians and other experts from Harvard, Stanford, Columbia, Berkeley, University College London, Queen Mary University of London, Georgetown, Johns Hopkins, London School of Hygiene & Tropical Medicine, etc., signed an open letter to The Lancet denouncing the PACE study’s unacceptable methodological lapses.

This open letter demanded that The Lancet seek a fully independent analysis of the trial data. Given the significance of this development, The Times (UK) and BMJ (the former British Medical Journal) ran news stories about it. Reuters did not. Nor did Kelland think the open letter deserved mention in last month’s special report. I organized this open letter, gathered the signatures (with some help), and posted it on Virology Blog, so it certainly demonstrates the widespread scientific and academic support I have garnered for my negative views of PACE.

For the Reuters piece to have mentioned the open letter with its 100+ expert signatories would have undermined Kelland’s preferred narrative: That PACE and related CBT/GET research is robust science; that those who challenge this research constitute a mob of belligerent patients engaged in an anti-scientific campaign comparable to climate-change denialism; and that this mob of belligerent patients is being egged on by questionable non-patient “campaigners” like me.

Like Sharpe and his colleagues, Kelland does not appear capable of accepting the possibility that those who challenge PACE and the CBT/GET treatment paradigm have valid points to make. As the open letter to The Lancet documents, PACE is bad science–at least in according to many leading experts. That’s why patients reject it. And that’s why they’re right to reject it. To suggest they are like climate-change deniers is not only untrue but very unfair, a low blow delivered against a very ill group of people. (With regards to the article’s charge that researchers have been poorly treated, it goes without saying that nothing justifies abuse, harassment or threats, whether on social media or through other means.)

It can perhaps be hard to understand why anyone would raise objections to what might seem to be benign recommendations for exercise and psychotherapy. Besides the fact that the science is inadequate and does not support the recommendations, a major reason is that patients suffer serious relapses after even minimal amounts of exertion. In a comprehensive 2015 report that recognized ME/CFS as an organic, multi-system disease marked by neurological, immunological, autonomic and energy metabolism dysfunctions, the US Institute of Medicine (now the National Academy of Medicine) called this characteristic symptom “exertional intolerance.” In the medical literature, it is usually called “post-exertional malaise.”

In other words, a program of increasing exercise is contra-indicted for an illness in which “exertion intolerance” or “post-exertional malaise” is the characteristic symptom. And in surveys, patients are far more likely to report having experienced prolonged harms than benefits from a course of GET. In fact, this approach is based on the theoretical but misguided notion that patients’ symptoms are perpetuated by deconditioning, not by an underlying physiological disorder.

As regards CBT, the version tested in PACE and offered to ME/CFS patients is not like that provided to those with cancer, multiple sclerosis and Parkinson’s, which is supposed to help them cope with the realities of having a severe illness. In contrast, the CBT advocated by Kelland’s protagonists is specifically designed to alleviate patients of the hypothesized “unhelpful” beliefs they purportedly harbor about having a debilitating organic illness in the first place.

Patients always knew this was nonsense. Now the scientific and public health communities in the US and elsewhere have come to the same recognition. However, that recognition has been much slower to dawn on the powerful PACE/CBT/GET defenders in the UK medical and academic establishments–and on the journalists who have covered their work approvingly for years. Kelland’s protagonists are now confronting a fast-moving paradigm shift in how ME/CFS is viewed, studied and treated internationally. It is not surprising they don’t like it. I do find it surprising that Reuters has allowed Kelland to present their version of reality while ignoring the main scientific issues involved.

Kelland interviewed me at length in New York last August, shortly after I posted the open letter to The Lancet on Virology Blog. It was clear from our conversation that she had already established her framing and was impervious to facts and data that contradicted it. In this letter, I will next address two specific problems: The story’s misrepresentation of my professional status and its misrepresentation of the scientific debate.

*****

1.The story misrepresents my professional status

Kelland starts her description of me by referring to me as a former journalist. This phrase puzzled me. I was a reporter and editor at The San Francisco Chronicle from 1988 to 1998. These days, I regularly write articles that have nothing to do with ME/CFS for Kaiser Health News, Health Affairs, and elsewhere. I also consider my ongoing examination of ME/CFS research to be an investigative journalism project as well as a public health project. Had Kelland described me as a former newspaper reporter or former full-time journalist at a news organization, that would have been fine. But it is false to call me a former journalist. And in the context of this story, the reference comes across as dismissive.

Second, as I mentioned in introducing myself, I am an academic appointee at the Berkeley campus of the University of California. I have been employed in various academic positions at Berkeley since 2009, at either the Graduate School of Journalism, the School of Public Health, or both. My University of California positions have ranged from half-time to full-time and have always included salary and full benefits. In my current half-time position at the Center for Global Public Health, my main responsibility is to conduct the ME/CFS-related investigation that Kelland has written about. (I have also cc’d Hildy Fong, the executive director of Berkeley’ Center for Global Public Health, who can confirm my current employment status.)

Yet Kelland never mentions that I am a Berkeley employee and that my ME/CFS investigation is an official Berkeley project. In fact, Kelland never mentions that I am employed anywhere at all. Reuters readers would be forgiven for assuming that I am a rogue actor with no apparent current academic or professional affiliation. (Remember, I’m also presented as a former journalist.) Kelland does mention that I received a doctorate in public health from Berkeley at some point; the inclusion of that detail does not justify the erasure of my present academic and employment status.

Kelland highlights my crowdfunding and accurately notes that last year’s total was more than $87,000. Yet she omits the salient fact that I crowdfund on Berkeley’s own crowdfunding platform to support my academic position at the School of Public Health. Berkeley’s crowdfunding platform is reserved for approved Berkeley projects. The money raised goes directly to Berkeley, not to me. The Center for Global Public Health uses those funds to pay my regulated University of California salary and provide me with health insurance and the other benefits I receive as a state employee. In exchange, I investigate and report on research and other issues involving ME/CFS.

In fact, April is a crowdfunding month at Berkeley, so I am crowdfunding right now for the fiscal year running from July, 2019, through June, 2020. I am seeking $95,000 to support my Berkeley position at 60% time. As you can see from this link, this effort is prominently sign-posted as being under the Berkeley umbrella.

Within the University of California system, many non-tenure-track academic appointees, like me, survive partly or fully on outside support for their projects. This is not remotely unusual. The only difference here is that my support comes in the form of hundreds of individual donations rather than larger gifts or grants from traditional institutional sources. In the university’s view, I am doing what academics are expected to do, bringing in funding. Kelland is free to disapprove of my innovative and resourceful fundraising strategy, but that does not give her and Reuters the right to withhold from readers that I have a bonafide academic position at one of the world’s leading institutions of higher education.

When it comes to the PACE/CBT/GET defenders, Reuters does not mention where they received their diplomas but provides their current professional titles and places of employment–information readers have a right to expect. To exclude parallel information about me breaches all sorts of tenets of even-handedness and fairness, and Kelland’s decision to do should raise a few questions about her intentions. The effect of stripping me of my current status in this manner is self-evident–to reduce my perceived credibility as a sharp critic of these eminent scholars, who are presumably my moral and intellectual betters.

After all, as someone with no legitimate academic institution or media organization apparently willing to provide me with a professional home, who am I to challenge Professor Michael Sharpe of Oxford University and call his beloved PACE trial a piece of crap? I acknowledge that I can speak in terms that are at times provocative, inelegant and, to some, off-putting. But Kelland has focused on my occasional use of flamboyant language while refusing to examine the substance of my critiques and failing to identify me properly.

(Shortly after the article appeared, I complained to Kelland and the two editors listed about the omission of my current academic and professional status. Remarkably, they did not respond to this concern, nor did they write me back. I will forward you the e-mail I sent them.)

*****

2. The story misrepresents the current state of the debate over PACE and related research

Kelland generally limits her discussion to what she frames as bad behavior and ignores the substance of the scientific issues involved. Here’s her account of the enormous range of concerns that I (and many others) have raised: His [Tuller’s] argument is that the therapies evaluated in the PACE trial are based on a misguided hypothesis that CFS/ME patients suffer from unhelpful beliefs that they have a biological disease, and that their symptoms of fatigue are made worse by deconditioning due to inactivity. He also says he thinks the trial’s methodology was flawed. The scientists involved reject those arguments.

That’s it–the full extent of what Kelland writes about the flaws of the PACE trial. First of all, to clarify: The PACE trial itself states clearly that the therapies are based on the hypothesis mentioned, so it is not just my personal “argument” that the therapies are based on that hypothesis. It is my “argument” that this hypothesis is “misguided.”

This letter is not the place to recap the voluminous debate over PACE except to make one point that captures some of the trial’s absurdities. In the study, 13% of the participants were defined as disabled for physical function, one of the primary outcomes, based on their answers to a questionnaire, even as they were simultaneously defined as recovered for physical function based on those same questionnaire answers. How could that be? Because the score for being considered recovered on physical function was lower and represented worse health than the score for being considered disabled on the same measure.

You read that previous sentence correctly, but the point is bizarre and hard to grasp so I will repeat it, the score for being considered recovered on physical function in the PACE trial was lower and represented worse health than the score for being considered disabled on the same measure. Specifically, to be considered “disabled” enough on physical function for trial entry, you needed to score 65 or below on the questionnaire’s 100-point scale; to be considered “recovered” for physical function at the end you needed to score a 60 or above on the same 100-point questionnaire scale. So there was a five-point overlap in these critical thresholds.

In other words, participants could be recovered for physical function at the beginning of the study, before any treatment at all–even as the exact same score defined them as disabled enough to be in the trial in the first place. Since these participants were already “recovered” on this key variable, it is hard to understand why they were in the trial in the first place. In fact, patients could drop from a score of 65 for physical function at trial entry to a score of 60 at the end yet still be considered recovered. This anomaly was just one of PACE trial’s many troubling features. Kelland does not mention it.

Moreover, Kelland frames the argument as my personal opinion against those of her protagonists. As she notes, David Tuller “thinks” the study has problems; these researchers understandably “reject” my thoughts on that. Yet in the story, as I have noted, I am portrayed as a rogue nobody while those I am criticizing are highly positioned somebodies. My only apparent allies in this struggle are aggressive, hostile patients akin to climate-change-deniers and one other non-patient campaigner. In addition to omitting any mention of last summer’s open letter to The Lancet with 100+ expert signatories, Kelland overlooks the abundant evidence of extensive scientific and medical opposition to PACE and the entire CBT/GET treatment paradigm.

Here is some additional information that Reuters readers might have found useful in considering the issue:

*Kelland mentions that I was co-author of a critique of the PACE trial. However, she doesn’t mention that this paper was more than a critique, it was a robust reanalysis and full refutation of all the main PACE findings. The reanalysis was based on the PACE authors’ own trial protocol, in which they spelled out before patient recruitment how they would conduct their analyses. After collecting their data, they abandoned their main pre-specified outcome measures in favor of weaker ones that produced better-looking results. Does it matter to Reuters that the scientific literature now includes a convincing, peer-reviewed debunking of PACE in a quality journal, but that Kelland did not provide readers with this information? I would hope so.

*Many dozens of readily accessible experts recognize that PACE is a piece of crap, as I have called it, and have said so publicly (although they have avoided use of the word crap). Kelland could have sought comment from any of the 100+ experts who signed last August’s open letter to The Lancet. Bruce Levin, a biostatistician at Columbia, has called PACE the height of clinical trial amateurism. Racaniello, the Columbia microbiologist who hosts Virology Blog, has called PACE a sham. Jonathan Edwards, a rheumatologist at University College London, has called it a mass of uninterpretability. Ronald Davis, a geneticist at Stanford University, said: I’m shocked that The Lancet published it…I don’t understand how it got through any kind of peer review.

*In 2017, the Journal of Health Psychology published an entire issue with multiple critical commentaries about the controversy it called “PACE-gate.” (Disclosure: I wrote one of these critical commentaries.) Here’s what the journal’s editor, psychologist David Marks, wrote in a subsequent essay called “The PACE Trial: A Catalogue of Errors”: “The PACE Trial team were operating within a closed system or groupthink in which they ‘know’ their theory is correct. With every twist and turn, no matter what the actual data show, the investigators are able to confirm their theory….Furthermore, critical analysis suggests that the PACE investigators involved themselves in manipulating protocols midway through the trial, selecting confirming data and omitting disconfirming data, and publishing biased reports of findings which created a catalogue of errors.” The Reuters story demonstrates Kelland’s transparent allegiance to the same “closed system or groupthink” as the PACE investigators.

*Two days before Reuters published Kelland’s story, I posted an interview on Virology Blog with Steve Olson, medical director at Kaiser Permanente for northern California. He explained that the organization is overhauling its entire approach to ME/CFS because it now recognizes that patients and advocates have been right about the science all along–that PACE and other CBT/GET studies are examples of poor research and should never have been used as the basis for clinical guidelines. Here’s what Olson, a family physician, told me about his change of views: “I began to meet people involved in patient advocacy and found that they were all very grounded, looking at the current information and research about ME/CFS in a pretty objective manner…They rejected the GET and CBT framework because they found the science to be inadequate. I concluded they were right.” Kelland and the CBT/GET defenders have concluded the opposite. They are wrong.

*Kelland conveys, without skepticism, her protagonists’ suggestion that the US Centers for Disease Control and Prevention removed its CBT/GET recommendations and its references to PACE because of patient pressure rather than scientific concerns. Similarly, Kelland allows authors of a disputed review of exercise therapies to claim, without evidence, that the publisher planned to withdraw it because of patient pressure rather than genuine scientific concerns. (Cochrane, the organization involved, is an international collaborative that publishes systematic reviews of medical treatments.) Yet for some reason, Reuters has not provided patients who pushed for these actions at the CDC and Cochrane with any opportunity to explain why they believed such changes were necessary for scientific and medical reasons.

*Regarding the Cochrane exercise review specifically, Kelland does not address any of the methodological concerns that patients, I, and many others have raised. She mentions that I pressed Cochrane on the issue, which is true. She does not mention that I organized and posted on Virology Blog an open statement of support for Cochrane’s decision, which noted in detail the exercise review’s disqualifying flaws; this statement of support for Cochrane was signed by 40+ experts from Harvard, Columbia, University College London, Stanford, Dartmouth, etc. And Kelland does not mention a key document–Cochrane’s own internal report about the exercise review, which makes clear that scientific deficiencies prompted the decision to seek its withdrawal.

*****

I could go on. The point is that Kelland appears to accept at face value that criticism of PACE and the CBT/GET paradigm is inherently irrational and anti-scientific. That is her right, but she falls woefully short of accepted journalistic practice in not presenting relevant material that would have offered a different perspective than the one she seems to share with her protagonists.

Such material was certainly available to Kelland, but she chose not to include it. Although she was assessing and critiquing my job performance as a Berkeley public health academic, she did not touch base with my boss–Arthur Reingold, the head of epidemiology at Berkeley and the co-leader of the Center for Global Public Health, or, as far as I know, anyone else at the university.

She did e-mail Racaniello, the Columbia microbiology professor who hosts Virology Blog. Yet she did not include any part of their exchange in the story. Presumably it would not have served her thesis to quote Racaniello on why he supports my work and has provided me with a high-profile scientific platform like Virology Blog for the last 3+ years.

Here’s part of what Racaniello wrote in his response:

David Tuller is doing important investigative journalism–he is exposing the flaws in the PACE trial for ME/CFS and along the way is encountering incorrect and unethical practices in other studies on the disease…I fully support his work as do the many other scientists, epidemiologists, and physicians who have signed his open letters. PACE was a poorly designed and executed clinical trial.

Racaniello added this: The PACE trial illustrates exactly how NOT to conduct a clinical trial. My Columbia colleague, biostatistician Bruce Levin, recently gave a talk on that issue with PACE as the case study. Yet the authors and the journals refuse to admit the study’s flaws.

In his response, Racaniello also noted that Kelland had asked him nothing about the actual scientific problems I have been exposing. He provided her with contact information for five other well-regarded researchers, all of whom strongly support my PACE investigation and oppose the CBT/GET recommendations as not based on reliable evidence, biostatistician Bruce Levin from Columbia, surgeon Ronald Tompkins from Harvard, rheumatologist Jonathan Edwards from University College London, pediatrician Peter Rowe from Johns Hopkins, and psychologist Brian Hughes from National University of Ireland Galway. Kelland’s article does not include any of these experts.

(While apparently unconcerned about the scientific issues at the heart of the dispute, Kelland did ask Racaniello whether he has donated to my crowdfunding and whether he has family members with the illness. I will forward you their e-mail exchange so you can assess the tenor of Kelland’s questions and the seemingly limited scope of her interests.)

In summary, Kelland’s special report is a one-sided presentation of a highly fraught issue that she did not bother to fully investigate. The story should never have appeared in this form. I deserve better. Patients deserve better. And not incidentally, so do Reuters readers.

At this point, appropriate editorial oversight would involve a major review of Kelland’s piece by top editors. This review should determine whether Reuters believes the article presents an accurate accounting of my professional status, the level of academic support for my position on PACE/CBT/GET, and the current state of the overall scientific debate. If such a review were undertaken, I suspect you would identify substantive deficiencies.

Going forward, I would be happy to provide more details about how and why this situation has occurred and/or to discuss appropriate ways to correct the public record. Thank you for your attention to this matter.

Best–David

David Tuller, DrPH
Senior Fellow in Public Health and Journalism
Center for Global Public Health
School of Public Health
University of California, Berkeley

Comments are closed.

Scroll to Top