BMJ Ethics Committee Meeting, 7 April 2003

Anonymised minutes

Present:

Sandy McCall-Smith
Liz Wager
Ann Sommerville
Tom Wilkie
Peter Singer (by phone)
Asad Raja (by phone)
Rachel Fetches

Apologies:

Jeff Tobias
Derek Wade

1.1 Minutes from the last meeting

The Committee accepted the minutes of the last meeting with some minor amendments from LW. AS requested an amendment to item 3.3 to reflect her comments more clearly.

1.2 Matters arising

Sandy McCall-Smith announced that after 3 years he was retiring from the role of Chairman of the BMJ Ethics Committee. SMS said that he had found the work of the Committee very enjoyable, and that there had been some exceptionally interesting cases during his tenure. SMS felt that the Committee was working very well and that the time was right to move on.

Richard Smith echoed the sentiments of the entire Committee in thanking Sandy for his contribution to the Committee. RS said that the Committee was hugely grateful for the work SMS had done on the Committee as Chair.

RS queried what the Committee felt was the correct course of action to find a replacement Chair? It was decided that the post will be advertised and that interviews will be conducted by a panel. It was suggested that RS, a member of the Committee, and an independent person could conduct the interviews.

ALL to suggest independent person for panel

The Committee also welcomed Dr Asad Raja as the newest member of the Committee. Dr Raja is a Professor in the Department of Surgery at the Aga Khan University, Karachi, Pakistan. Dr Raja has an MSc in Bioethics from the University of Toronto, and is the editor of the Pakistan Journal of Medical Ethics.

1.3 Action points

SMS agreed that he would still write an article about the Committee�s work for publication in the BMJ.

2. Update on existing cases

2.1 Case 02/001 A new surgical operation without evidence of either ethics committee approval or consent from the patients to be part of an experiment

Background: Authors from a developing country submitted a study in which they describe a new surgical operation using an established technique in a new context. They say in their paper that in the treatment of the particular disorder described "no operation stands out in comparison to the others". During subsequent correspondence the authors stated they had "taken an informed consent from all the patients undergoing the procedure. They were well informed before they were subjected to the procedure." The authors also stated that their country "did not have any ethical committee whose permission is needed to carry out any new procedure." The authors felt that their paper was about a new technique and not an experiment. The editorial committee felt that even though the technique was established, it had not been used in this context before and therefore the authors could not have known prior to treating the patients whether the technique would have been effective.

The authors were sent the Committee�s discussion of 24/04/02 and the national guidelines on research and consent from the authors� country.

The authors replied and RS asked a senior BMJ editor to comment on the authors� reply and the case. The authors had previously stated that he had not received approval from an ethics committee but that they had received consent from all patients involved. The authors contested that his use of the established technique in the context described was experimental.

AS stated that she felt that the paper showed an innovative use of established technology�was there be an obligation to write up the findings? TW pointed out that the novel technique was carried out in a private clinic. LW stated that the authors were the main health care providers and also researchers which was a difficult position for patients. RS felt that to publish such a piece was to simply publish an account of clinical experience with no controls. AS stated that there was no evidence of the authors being unethical. RS felt that the broader question was whether this established technology had been used in this way before? Viewing this more broadly, if this had been a known drug being used in a novel way then there would have to be an randomised controlled trial (RCT) to determine if this was appropriate and effective. AS felt that there was a different test for a medical device versus a medicinal product. TW stated that it was very difficult to draw the line between a doctor�s discretion and research. TW commented that there is no history of RCTs in the surgical field. RS stated that there was no clear evidence of misbehaviour but that the approach adopted by the authors was probably not the best way to do such a trial. There remains the difficulty of what patients are told at the beginning of the innovative use of established technology. LW agreed this appears to be a case where the novel technique is used a few times and then followed up by using it for more patients. LW queried if there was equipoise? Did the patients in the clinic have a real choice?

RS summed up the Committee�s feeling that there was probably no case for doing more on this case. It had been a useful interchange and it would be useful to debate with the readership where clinical innovation ends and research begins. SMS agreed that any problems with the work was unintentional and below the threshold of work requiring censure. AS agreed, and queried whether there was a differentiation about this topic in surgery? PS pointed out that as far as the law was concerned that a clinical professional was accountable for innovation.

No further action

2.2 Investigation of possible research/publication fraud by a university professor in a developed country

Background: BMJ had strong suspicions that a paper was fabricated. This paper was a follow up on a paper published in another prominent journal in the 1990s. University investigation has found no evidence. The consensus was that there was insufficient reassurance that a proper investigation had been conducted, and there remains a significant possibility that the whole paper might be entirely fraudulent. RS had written a follow up letter to the University authorities expressing the concern of the committee that the raw data had not been examined during the investigation. After a few chasing letters a senior university official had written back and informed RS that the author in question had been on leave without pay for the majority of the first half of 2002 and had not responded to the university�s queries. The author had then tendered his/her resignation. The university have made it clear that despite his/her resignation, they still expected a response to their inquiry. The university stated that at that stage it did not feel that it was in a position to "judge what might have transpired in the conduct of the research in question". It then transpired that the paper under suspicion at the BMJ had been published in a third journal. Following the Committee�s further discussion on 28/10/02 RS contacted the editors of both the journal that published the original paper and also the journal that published the follow up paper.

RS updated the Committee that following the Committee�s previous discussion of the case that he had contacted the editor of the journal that published the original paper to express concerns about the author�s initial study. RS also contacted the editor of the journal who had published the study that the BMJ had brought to the Committee because of the suspicions about the data. RS noted that a Medline search on the author showed that he/she was the sole author on a number of RCTs. RS stated that he had received a letter from a set of authors who had been concerned with the articles in both the journal that published the original paper and also the journal that published the follow up paper, and had written about their scientific concerns to both journals. Their letters to the editor were scheduled for publication. RS expressed that it was extremely unsatisfactory for such severe anxieties based on strong evidence to be kept outside the public domain.

It was noted that a company was marketing products based on the research and using the publications as scientific evidence for their effectiveness. SMS wondered if there was a legal aspect of consumer protection against misrepresentation, however, the products involved are not so highly regulated. TW felt that there was an issue of misleading people using this suspect literature. LW stated that as far as the role of the BMJ, there was a need to disentangle the issue of research integrity from the product being sold.

RS stated that he was keen to tell the story to the readers. PS felt that the integrity of the medical literature was at stake and that the journal editors should seriously consider retracting the papers. It appeared that there was a sufficient level of proof to write to Medline/PubMed. RS agreed that a retraction was appropriate where there were serious doubts that the research was fraudulent. RS noted that there had been examples of papers being retracted where inadvertent misinterpretation had rendered published results as meaningless and there had been no suggestion of fraud. AS queried whether the journals were considering retracting the papers? RS pointed out that there were concerns over libel actions from retractions where there was a suggestion of fraud. Richard Horton, the editor of The Lancet had once suggested the principle whereby editors could resolve this problem by a recall of aegis on the basis that if an editor knew what he or she knew now then he or she would not have published a paper. PS queried whether the issue should be shelved until the journals in question published the letters? RS felt that there still remained serious doubts about the work. The paper submitted to the BMJ was published in another journal. Such serious doubts about the work might suggest that some of the author�s other research requires close scrutiny.

RS also felt that the university�s response was inadequate as they appeared to feel that they were effectively unable to do more now that the author had resigned. PS suggested that RS write again to the university stating that he was still concerned. The university seem to imply that RS�s letter after their initial investigation raised new concerns about the author�s work when such concerns had been conveyed originally. The Committee had been particularly concerned that the university�s original investigation felt that there was no need to request the raw data; this was inadequate in light of the seriousness of the allegations raised. RS agreed to communicate further with the two journals with copies to the university. LW queried whether RS would consider adding something to the debate. RS was considering this as he was concerned that a letter to the editor would not be noticed by readers or that with such a device it might not come through that there were serious concerns. RS stated that it was not the BMJ�s policy to conduct a trial by media unless all other avenues had been exhausted but that this was an option if necessary.

RS to write to the university again�what course of action?
RS to write to both journal editors

2.3 Case 02/003 A case reporting multiple errors in treatment where the patient died

Background: The authors describe the errors in treatment of the particular patient but stated that they did not have consent of the relatives to publish the case. It appears that the relatives and the "authorities" do not know about the errors that occurred with this patient. The authors write that there is no non-punitive voluntary reporting system in place in their country (a developing country).

In an earlier discussion (28/10/02) the Committee had decided that the authors needed to get the family�s consent in order to publish the piece in the BMJ. However, in light of the volatile nature of the conditions in the country that it would probably be inadvisable for the authors to pursue this course of action. The authors did however contact the family and detailed the process of getting consent: "We downloaded the consent form from BMJ's web site and went to the house of the patient twice with a gap of two weeks between the two visits. In the first visit, the authors identified themselves and recalled the entire unfortunate episode of death of the patient. The authors expressed their desire to get the case reported in the BMJ and showed the family a copy of the journal. We informed them that the focus in the article is the mistakes that occurred during the management of the patient. The authors showed the family a copy of the article and requested the written permission of the father of the patient. The authors explained to the family the consent form and its contents and also explained that they were at liberty to refuse consent. The authors told the family that they could have a copy of the published report if they so desired. With this the first visit was terminated and consent form was left with the family to think about whether to agree consent or not. After 15 days, the authors revisited the family and repeated what had been said during the first visit. An opportunity was given to the family members to ask any questions if they felt necessary. None were posed. The consent form was then signed by the father of the patient."

The Committee considered the authors� detailing of the attempt to gain the relatives� consent and decided that it was genuine. As the consent was lost in the postal system, the Committee decided that they would forgo the bureaucracy of requiring the physical proof and to trust the authors� version.

Case would be acceptable for publication without the physical proof of consent

3. New cases

3.1 An attempt to induce an editor to publish

RS gave the background on this case: the BMJ had introduced their policy that they would not publish an economic evaluation without also being sent the clinical paper for publication. This group of authors had published the clinical paper elsewhere and submitted the economic evaluation to the BMJ just before the policy change was announced. Somebody�probably a representative of a drug company�rang an editor on behalf of the authors of a study to say that the company would guarantee to buy 1000 reprints if we would continue to consider for publication a study that conflicted with the policy on economic evaluations that we had just introduced. "And", he/she said, "I will buy you a dinner at any restaurant you choose."

SMS felt that RS should contact the drug company about this matter as people sometimes are over enthusiastic and don�t realise that what they are offering is an inducement. AS felt that the drug company should make a clear policy and transmit this to their employees that such action is unacceptable. TW & AS stated that offering inducements to print unacceptably twists the peer review process. It was noted that such an inducement went against the Association of the British Pharmaceutical Industry�s (ABPI) code of practice. TW agreed with the COPE Committee�s advice that this should be taken up with the drug company concerned. TW also felt that this could also be written up in the journal.

RS to contact the drug company
RS to consider writing a piece on this

4. Submissions from developing countries

PS introduced the two papers as stating that there is a role for the BMJ and other journals to show international leadership in opening access to Southern readers. The has been started by allowing Southern readers free access to Northern medicine but there also need to be progress on publishing Southern medicine. There needs to be a better balance between the publishing of Northern and Southern authors� work. This was really a policy issue. Some empirical research in 2003 showed that a follow up of the 2002 work the BMJ, The Lancet, The New England Journal of Medicine, and JAMA had remained static or showed some increase in Southern content.

PS suggested that it would be useful to have a session with the editorial board members to reach a consensus on the problem and determine priorities for moving a policy forward. RS suggested editing PS/AR�s pieces into an education & debate piece. RS was concerned about setting a "quota" for Southern content and felt that a target would be better. RS thought it would be useful to take specific suggestions to the Education Management & Strategy meeting to discuss. RS suggested that at the BMJ Board meeting it would be useful to have a workshop on the direction the BMJ should take. The BMJPG had taken on board Richard Horton�s suggestions for broadening editorial boards so that they are more international.

TW stated that the BMJ & The Lancet were ahead of the US journals in this respect. It would be quite useful to have a breakdown of the readership of the BMJ to place the discussion in a broader context. RS felt that there is a split between the print version of the BMJ which goes to a primarily UK audience, and bmj.com which has a different, international audience.

RS, PS, & AR to consider workshop with Board
PS/AR to write up pieces as E & D for possible publication

5. Updating the BMJ Instructions to Authors

5.1 Research ethics

After the discussion at the previous meeting about whether verbal consent was acceptable for papers submitted to the BMJ, TW had written to suggest that in order for verbal consent to be considered acceptable there needed to be an audit trail that verified that consent had been properly sought and given. As a more general point, TW pointed felt that the bigger issue irrespective of whether consent was verbal or oral was whether the consent was genuine. RS had brought to the attention of the Committee clause 22 of the Declaration of Helsinki which states "� If the consent cannot be obtained in writing, the non-written consent must be formally documented and witnessed." This formal requirement should be fed back to the editor who submitted the paper.

Need to add link to Declaration of Helsinki

5.2 Criticising doctors

AS updated the Committee that she had put the BMJ Ethics Committee�s discussion to the BMA Ethics Committee and they were pretty much in agreement but were worried about any discrepancies with the GMC�s position. The GMC state that it would be a breach of confidentiality to disclose any patient information without consent. SMS pointed out that the issue here was whether by publishing an account of his or her treatment and patient has given "implicit" consent for the doctor to refute allegations? AS thought that there were problems with this as this is apparently not the GMC�s view of the issue. LW stated that there are two issues that the GMC highlights: (1) breach of confidentiality and consent, and (2) concern about debate in the press. It was unclear what the GMC thought about the notion of implicit consent. TW felt that there were broader issues, such as why did patients bypass the NHS complaints procedures? Explicit consent solves the problem of the doctor responding but implicit consent is problematic. SMS flagged the last sentence of the MEC meeting: "Although there was concern about the use of public media as a forum for resolving disputes, the committee concluded that it could be seen as unreasonable for patients to tell their story to the media but to deny permission for the doctors concerned to respond." AS pointed out that in the example given of Mrs Rose Addis this was more complicated because it was not the patient herself who went to the media but her family.

AS felt that doctors were really tied by what the GMC�s line was. LW suggested raising the issue with the GMC as a point of concern from the BMJ Ethics Committee. SMS stated that he felt that the GMC were unlikely to move on this point.

SMS pointed out that with the move to more patient control and ownership of data relating to their health care it was more likely that patients would attempt to tell their stories. RS indicated that the BMJ would consider publishing patients� stories but as they did not wish to engage in trial by media then it would be difficult to do so without the other side of the story from the medical team. TW pointed out that the BMJ publishes accounts by doctors of bad practice so to not allow patients� to tell their stories creates an imbalance. SMS felt that perhaps the BMJ should only publish where both views could be obtained. LW pointed out that this created the danger of a doctor "veto" over publication. SMS agreed and stated that patients should be able to publish their stories so long as the opportunity for the other side to respond had been consented to. If the "other side" had decided not to publish this would not affect publication of the patient�s view.

5.3 Other suggestions for inclusion

The Committee agreed with the suggested insertions and for additional areas not yet covered to be adopted.

REF to arrange posting of guidelines

6. Publications from the Ethics Committee

SMS agreed to write a piece on the Ethics Committee�s work to date.

PS & SMS to write up the process of revising the consent guidelines and submit for publication.

SMS/PS to send articles to Ethics Committee

7. Any other business

7.1 Ethics committee approval for audit.

DW to submit discussion paper for OCT meeting

7.2 Consent for genetic pedigree studies

TW to contact Angus Clarke/Peter Harper about this issue

8. Dates of next meeting

Monday 14 July 2003 (CANCELLED)