The BMJ?s ethics committee: What does it do, and how does it work?
 
BMJ Ethics Committee Members 2006 

Attending:

Sandy McCall Smith

Liz Wager

Tom Wilkie

Richard Smith

Anne Sommerville

Derick Wade

Jeffrey Tobias

Alison Tonks

Peter Singer

Item 1. Introductions

Item 2. Review of committee?s brief and aspects of its working methods

• Competing interests. We agreed that ethics committee members should fill out the standard BMJ form declaring their competing interests. RS to update members on our competing interest policy for editors at the next meeting. Committee members to declare competing interests in relation to specific items in advance of each meeting.

• Power of the committee. Committee will be mainly advisory with the proviso outlined in ATs original paper dated Aug 29 2000: If editors decide to over ride the committee?s advice they will justify their actions in writing to the chairman. The chairman can appeal to the journal committee in the event of a dispute.

• The committee?s workings must be as transparent as possible. Decisions will be posted on the bmj.com in an anonymised, abridged form with some indication of the size of the majority.

• Access to confidential papers: Papers need not be anonymised if BMJ authors and potential authors have fair warning that an ethics committee might scrutinise their paper. A paragraph in "guidance to contributors" on bmj.com is a satisfactory solution .

• The committee will meet 4 times a year. Advice will be given at other times facilitated by an email server. The Chairman will take decisions alone only in exceptional circumstances.

Case 1

Two ophthalmologists from a tertiary referral centre report a case referred to them by another hospital where, they imply, her care was suboptimal. She never recovered her vision and is now nearly blind. Two external reviewers saw the report and both agreed that her eyesight could and should have been saved. One responded that the patient had grounds to sue the initial hospital for negligence.

RS wrote to the authors from the tertiary referral centre, sending them the reviewer?s reports and asking whether they thought further action was warranted. They did not reply.

Discussion

We agreed that the editors had a duty towards the wronged patient, although we were unable to define the limits of that duty. The editors have clear advice that the patient was treated poorly and should sue for negligence. The editors have no legal obligation towards the patient, however.

Committee?s comments, paraphrased:

AS: The journal has hearsay evidence about this patient, so their position is weak. The authors are much closer to the case and the responsibility to the patient lies mainly with them. We should encourage them to take the matter up with the patient or the referring hospital.

PS:This patient is the victim of a medical mistake. Naming, shaming, and blaming (ie telling the patient she has grounds to sue) is the wrong approach if we want to move on from the blame culture in medicine and acknowledge that mistakes happen because of faulty systems, not people. Taking steps to change the system is a more progressive approach

TW: True, but the only way this patient can get the compensation she needs to alleviate her disability is by suing the hospital for negligence.

Decision

RS to pursue the authors for a response to his letter, and a robust reassurance that they will report to the initial hospital their concerns over this woman?s care. He might also suggest that we will if they don?t.

Case 2

After an editorial on cheating at medical school, two junior doctors wrote to the editor reporting that some 60% of the students at their medical school had known in advance about topics that would be examined in the objective structured clinical examination part of their finals. The medical school did not know. More students than normal were awarded distinctions that year.

RS responded that he was now implicated by his knowledge of the affair, but reassured them that they would remain anonymous whatever action the journal decided to take

Discussion

The Committee on Publication Ethics (COPE) has already advised that RS tell the medical school what has happened. This committee agrees unanimously. PS commented that the medical students were not yet whistleblowers but they would be if we informed the medical school of the contents of their letter. We therefore must ask their permission first.

Decision

RS to inform the medical school of the incident after asking for permission from the two junior doctors who wrote in. Reassure them that they will remain anonymous, and congratulate them for doing the right thing by writing to the BMJ in the first place.

Case 3

A doctor from outside the UK submitted a paper describing a new theory of disease. It made little or no scientific sense. The covering letter said he had treated himself and many patients with a dietary treatment (unspecified) arising from his theories, and that the treatment had been universally successful for a range of diverse diseases.

Discussion

This paper prompted a discussion on the editors? "duty to warn". It was agreed that editors have such a duty, but that it has limits. Those limits are still undefined but the committee?s deliberations will be an evolving "case law". Authors should be aware that the editor?s duty to warn may over ride their rights to confidentiality (in other words, we may alert the authorities to their research), and AT will add a paragraph to "advice to contributors" making this clear. Circumstances that might drive us to breech confidentiality will be clarified as the committee gains experience. There will eventually be a formal policy. Authors can follow developments on bmj.com.

RS noted that our policy of confidence to authors was based on a doctor?s duty of confidence to a patient. He felt this was na?ve, and probably inappropriate.

SMcS commented that the ethics committee has a broader public responsibility than others who might also see these papers, including reviewers.

Decision (unanimous)

Refer paper to the author?s national regulatory authorities with a non-judgmental statement to the effect that we are not sure what to make of the paper or its author, but feel someone should take a look at what he is doing. Write to the author informing him of our action. Some committee members thought he might be a danger to the public, others that he was probably a harmless crank.

Case 4

A plastic surgeon submitted a Minerva picture reporting a procedure which was later described by a reviewer as dangerous. RS put this to the author who responded with several articles showing that the procedure was used in America, and might not be dangerous. The quality of the evidence was poor. RS offered to publish a debate in the journal on the best way to evaluate experimental surgical procedures, using this case as an example. The plastic surgeon declined.

The committee was divided over whether or not this author?s technique was established enough to make it acceptable practice. AT, PS, and JT felt it was no less acceptable than many other procedures that are not supported by good evidence. Other members worried about the nature of the consent given by patients, since it was unlikely that the author told them the technique was controversial and poorly evaluated. RS was also worried that the author might be selling the technique (inappropriately) as the latest big thing from the USA.

AS suggested that we would be satisfying our concerns about the technique if we asked the ethics committee of the Royal College of Surgeons to take a look at the evidence and issue guidelines about it.

Decision (unanimous): Take no action. RS can take an editorial decision whether or not to approach the ethics committee of the Royal College of Surgeons. Consent from the author is not required, but would be polite.

Case 5

A private practitioner submitted a paper reporting a case series of over 600 patients. He given them a treatment which many would regard as scientifically dubious. In addition, some had been treated with increasing doses of a new drug that has been evaluated in randomised trials and found to work. The study is effectively a dosage study of the new treatment. The author did not seek ethics committee approval for his study, and the results, in our judgement, are effectively meaningless. It would be impossible to conclude anything with confidence..

The men in the study no doubt gave their consent to treatment, although there must be doubts about how informed this consent was. It?s not clear whether or not they knew they were part of a study.

RS referred the case to the General Medical Council

Discussion

The GMC agreed that the research was scientifically meaningless, but couldn?t do anything about it because there?s no system of ethical review of research in the private sector. Research ethics committees only consider research that was done on NHS patients, or premises, or notes. The committee felt that the journal should some how bring this issue to the attention of readers.

Decision. The GMC has already ruled on this case. No further action necessary. RS to decide how to debate the issue of private sector research in the journal.

Case 6

A nurse submitted a personal view in which she describes being assaulted by a violent patient. We would like to publish the case as a debate including several commentaries from different experts, but would not normally publish patient information without consent. It would not be possible to approach this patient for consent, and if it were, he would be unlikely to give it. Can we publish without?

Discussion and decision: The committee agreed that the editor should go ahead and publish anonymously if: The author can reassure editors that the subject?s violence was not a symptom of his illness and that he had insight into what he was doing, and that he has been charged and found guilty of assault (or similar). If the first is true, this is a case of common assault in the workplace and we owe him no duty of confidentiality. If the second is true, the case is already in the public domain. Removing the nurse?s name and details from the case reduces the risk of identification substantially anyway.

Case 7

A leading Chinese researcher submitted a report of a small phase I study in which patients were deliberately given a communicable disease as a treatment for another communicable disease. The study protocol had been approved by several American review boards then by similar boards in China. This paper reported only toxicity data. Data on efficacy were sent to another journal The BMJ rejected it, largely on ethical grounds and a spirited correspondence with the authors ensued in which they argued, convincingly, that the study and the treatment were ethical, necessary, and could be beneficial, particularly to the developing world.

Discussion and decision

After consideration of the authors? appeal and all the accompanying reference material the committee agreed that there were no unresolved ethical issues over this research. It is an editorial decision whether or not to publish in the BMJ preliminary toxicity data on a treatment for a high profile infectious disease. The publication would be widely read and cited in the lay press. TW felt publication might be irresponsible but not unethical.

Case 8

A doctor from outside the UK wrote up a report of a clinical error that may have contributed to a patient?s death, but probably didn?t. The BMJ agreed to publish the case report anonymously at the centre of a debate about error and how to deal with it. The author, who committed the clinical error agreed to the anonymous publication, but wanted to write a signed commentary to go with the others. If he did this we would have to conceal from readers that he is commenting on his own error, and publish a commentary written as though he were an impartial observer. Is this an acceptable deceit?

Decision (unanimous): The author of the case report must comment anonymously, or not at all. Deceiving our readers that he is an impartial observer would be unacceptable. We have already decided that he must remain anonymous to protect the identity of the patient and her relatives.

Action

The author accepts our decision and will incorporate his comments into the anonymous case report.

Case 9

A case report describing in detail an encounter with a patient in a remote part of developing world country. We asked the author to gain consent for publication from the patient, but he replied that this would be impossible as the patient is thousands of miles away and untraceable. Can we publish the report without consent?

Discussion and decision:

The committee advises that we don?t publish this case report. Consent is impossible and we cannot relax the rules for people who happen to live in the developing world. There was strong dissent from PS who argued that there was a "public interest" case for publishing something that illustrates global inequalities in health so starkly. The public interest arguments, he says, over ride the subjects? right to confidentiality. Others disagreed equally strongly.

Case 10

A letter of complaint about the journal?s biased treatment of a commercial company with global interests.

Discussion and decision: The chairman will write to the correspondent asking for evidence of a systematic bias against the company by the BMJ. He will also ask for evidence that RS refuses to discuss it. The committee will then consider the evidence and decide whether or not the journal needs to act (apologise for example).

Item 4. Next year?s meetings. Deferred

Item 5. Any other business. None

Minutes prepared by Alison Tonks, BMJ
 

The BMJ?s ethics committee: What does it do, and how does it work?

The BMJ has an ethics committee of nine members that meets quarterly. It develops editorial policy, as well as advising on specific issues as they arise. It is an independent committee recruited through open advertisement in the lay press and the BMJ. Editors selected the chairperson after interviewing four candidates. The chairperson subsequently helped editors select the other members from over 150 applicants. Collectively, members have a broad expertise including clinical medicine, research, journalism, bioethics, law, and medical editing. They are listed below. Members are appointed for a probationary period of one year.

Alexander McCall Smith, professor of medical law, University of Edinburgh (Chair)

Derick Wade, professor of neurological disability, University of Oxford

Liz Wager, head of international medical publications (UK), GlaxoWellcome

Peter Singer, professor of medicine and director of joint centre for bioethics, University of Toronto

Anne Sommerville, head of medical ethics, British Medical Association

Tom Wilkie, adviser in bioethics at the Wellcome Trust, and editor of Scientific Computing World

Jeffrey Tobias, consultant in radiotherapy and oncology, University College Hospital, London

Richard Smith, Editor BMJ

Alison Tonks, assistant editor, BMJ
 

Limitations of membership:

The committee has a large majority of British members. This was not deliberate and we have plans to expand the international presence soon. There is an urgent need for input from the developing world.

What does the committee do?

1 Clarify, review, and develop existing editorial policies such as:

Policy on consent to publication of material arising from the doctor patient relationship

Policy on competing interest for authors, editors, reviewers, and ethics committee members

Policy on prior disclosure of results to research participants

Editors duty of confidentiality to authors

2 Formulate new policies?on, for example, the journal?s response and responsibility to whistleblowers, for example, or research participants from resource poor countries.

3 Advise editors on ethical questions that come up during their routine work, including scrutinising occasional papers referred by editors worried about some aspect of their conception, design, conduct, presentation, peer review, or authorship. Committee members will have access to unabridged papers that have not been anonymised. The BMJ?s guidance to contributors on BMJ.com is being changed to make it clear to authors that their work may be seen by an ethics committee.

4 Advise editors on their moral duties and responsibilities to patients, research subjects, authors, reviewers, publishers, and other editors, especially where they are in conflict. We hope that as the committee gains experience tackling the kind of problems that come up in the day to day running of a general medical journal, their deliberations and decisions will evolve into "case law" that will inform what editors do in response to common problems. A good example is the notion of an editors "duty to warn": BMJ editors read over 6000 papers each year. They are exposed to information about a range of borderline activities including poor patient care, experimental treatments, and bad judgement in clinical practice or research. The decision not to publish is easy, but do they have a duty to report their concerns to the research subjects, the authors, or the author?s institution? If they do have such a duty, what are its limits?

5 Act as a resource to help editors enhance the coverage of bioethics in the BMJ

6 Keep editors informed of developments in research and publication ethics.

How does the committee work and what powers does it have?

The committee meets in person four times a year. Between meetings, editors can consult members by email. Responses are coordinated by the chairman and a decision is made by majority. The chairman will make decisions without consulting other members only in exceptional circumstances

The committee is advisory. The editor can ignore its advice, but must justify his action to the chairman in writing. The advice and the editor?s response would both be published. In the event of a dispute, the committee?s chairman can appeal to the journal committee of the BMA.

How will decisions be reported?

Minutes of all meetings, including decisions made on specific papers will be posted on BMJ.com. The minutes will be edited first to remove all identifying features from papers.

Comments that are minuted will be attributed to particular members, and the nature of the final decision will be recorded.

The committee will publish an annual report, again in an anonymised form, summarising its work over the previous year. Important decisions may also be reported in the paper edition BMJ. Alison Tonks, BMJ
 

BMJ Ethics Committee Members 2006

Iona Heath, general practitioner, Caversham Group Practice (chair)