Trial By Error: My Most Recent Exchange with Bristol

By David Tuller, DrPH

Last month, I sent Sue Paterson, Bristol University’s director of legal services, a couple of e-mails seeking answers to several questions. Those e-mails can be read here and here. I cc-d a few other people.

On Friday, May 10, I received a response. This morning, I wrote back. I cc-d those I had previously cc-d, since they did not seem to have been cc-d on Bristol’s response to me. I have posted this recent exchange below.

**********

Dear Dr Tuller

In your emails of 4 and 29 April you ask several questions, most of which do not relate to recorded information held by the University of Bristol, but rather seek to elicit opinions or to engage in debate. The University is not obliged to respond to such questions.

The University would respond to your questions about the confidential review of publications commissioned by the HRA if submitted as a Freedom of Information request. The following information is therefore provided, the text of which has been approved by the HRA.

The review panel is made up of two senior professorial members of the University of Bristol’s academic staff, both from different faculties and with no connection with CFS/ME research, and an independent chair nominated by the HRA, who is a distinguished senior academic from another Russell Group university. The review is expected to be concluded in June 2019 and will report to the Pro Vice-Chancellor for Research at the University of Bristol and to the Chief Executive of the HRA, who will determine any further action to be taken, including to whom the outcome should be made known.

As for your original Freedom of Information request about consent forms, you have received the University’s response confirming that the information requested is not held. The University has nothing to add to this response and will not respond to any further enquiries on this matter.

Yours sincerely
Sue Paterson

**********

Dear Ms Paterson–

Thank you for the information about Bristol’s ongoing review of up to at least eleven studies that were exempted from ethical review under the guise of being “service evaluation.” Although I have been covering this issue on Virology Blog since 2017, I read about Bristol’s review itself in a public document. The document noted that this review would be “independent” but did not provide details about the panel members, the process and the timeline. I appreciate the additional data and am glad to hear the review is expected to be concluded in June.

About my other questions…In the course of examining problematic research, it is standard practice–at least in the field of journalism–to ask university officials for comment on possible or documented violations of methodological and ethical norms. Bristol must have experienced this phenomenon last year, when a high-profile cancer investigator resigned amid charges of academic misconduct.

A university can of course refuse to answer difficult questions, but that doesn’t make them go away. Difficult but unanswered questions tend to stick around. In this case, such difficult but unanswered questions include: Why didn’t investigators disclose in their paper on the pediatric Lightning Process trial that they had recruited participants before registration and engaged in mid-study outcome-swapping? And can anyone produce consent forms for the 100+ in-person interviews conducted with minors and their families for research on the potentially fraught topic of school attendance? Under the circumstances, it is not unwarranted to raise broader concerns about Bristol’s research oversight strategies.

Bristol’s pediatric research has exerted significant influence on public health policy and medical practice in this domain, whether the relevant illness is called CFS, ME, CFS/ME, or ME/CFS. Policy and practice should be based on reliable and valid evidence from studies that conform to accepted scientific principles, perhaps even more so when children are involved. In drawing attention to some difficult but unanswered questions, I am seeking to protect a highly vulnerable population from the potentially harmful consequences of recommendations arising from flawed studies. Doing so seems critical at this juncture because the National Institute for Health and Care Excellence is currently developing new clinical guidelines for the illness.

As an example of a recent paper that illustrates the problem, Current Opinion in Pediatrics last month published a major review of the illness in children–causes, treatments, prognosis, and so on. The review’s abstract highlights the Lightning Process as having been shown to be “effective.”

This review does not note that the investigators of the referenced study recruited more than half their participants before trial registration, swapped primary and secondary outcome measures based on the early results, and omitted these details from the published paper. The review also fails to mention that Archives of Disease in Childhood, which published the Lightning Process paper, has posted an editor’s note essentially acknowledging these missteps. (The editor’s note is not visible from the paper, for reasons that the journal has not adequately explained; that means most readers likely have not noticed it.) Despite the Lightning Process trial’s deficiencies, Bristol’s reputation enhances the study’s perceived credibility and likely encourages other researchers to cite the published findings uncritically–with unknown but possibly deleterious impacts on children’s health.

In any event, thank you again for your response. To ensure transparency, I am cc-ing those I cc-d on my previous e-mail.

Best–David