BMJ Ethics Committee Meeting, 21 January 2003

Anonymised minutes

 

Present:

Richard Smith
Tom Wilkie
Liz Wager (Acting Chair)
Ann Sommerville
Jeff Tobias

Derick Wade

Rachel Fetches
Jocalyn Clark (BMJ Editorial Registrar)

 

Apologies

Sandy McCall Smith
Peter Singer

1. Minutes from the last meeting

The Committee accepted the minutes with minor edits from LW.

The Committee went through the outstanding action points from the minutes. In regards to the Guidelines on articles criticising doctors the Committee was unsure whether SMS had contacted the GMC to discuss the issue informally. AS updated the Committee on the BMA Ethics Committee’s discussion of the issue. The BMA Ethics Committee was in broad agreement with the discussion at the BMJ Ethics Committee but there was some disquiet that patients might feel pressured to agree to the doctor being able to put his/her side of the story. RS felt that the BMJ guidelines should include information on this issue but there should be room for discretion in some cases—for example, of a patient reporting abuse. RS thought that the issue should be highlighted in the BMJ so that readers/authors could debate the issue. AS pointed out that moving to the position discussed by the BMA/BMJ Ethics Committees would differ from the GMC guidelines and so place doctors in a difficult position. DW raised the point that we would like a debate from the GMC on the issue not simply a statement on the standard line in regards to confidentiality. AS noted that there is guidance for institutions commenting on celebrities receiving treatment in the institution but not for other patients.

SMS to contact the GMC informally
RS to highlight the issue in the BMJ

In regards to the guidelines on consent, RS has not yet mentioned the guidelines in his editor’s choice column and the editorial from SMS/PS had not been sent through yet. The Committee considered Roger Robinson’s comments on the implementation of the consent guidelines by the BMJ editorial team. RR had suggested keeping a prospective report of the fillers that the team would want journalistically but where consent was unobtainable. The Committee discussed what it felt it would need to monitor the papers. RS queried whether the Committee would want just numbers or a record of what was accepted/rejected with reasons. LW suggested that just the numbers of papers would be sufficient with difficult cases being referred to the Committee for advice. The Committee decided that REF did not need to collate a report on all the decided cases for review by the Committee.

SMS/PS to write article on development/use of revised guidelines
REF to advise RR that Committee would like prospective record

The Committee discussed how to focus a discussion on obtaining consent for genetic pedigree studies and the first series of cases in a new condition. TW highlighted that the issues in genetic pedigrees revolved around whether consent should be obtained from individuals or families. Many support groups for genetic diseases support the sharing of information within families, even viewing this as a responsibility. At the other end of the spectrum, many health care professionals focus on the individual they are treating. RS felt that it would be difficult to reach a definitive policy and it would be useful to debate the issue in the journal. TW suggested that the journal might have to adopt a case by case approach depending on the likely degree of harm and social resonance of each case. RS suggested contacting the editor of the Journal of Medical Genetics to see what that journal’s policy was on the issue. It was also suggested that Angus Clarke, and SMS (as Deputy Chair of the Human Genetics Commission) might be able to provide information in this area. AS highlighted that the BMA had produced a book Human Genetics: choice and responsibility. DW pointed out that there might not be that much information with a focus on the publication of such information.

RS to contact Eamonn Maher (JMG)
REF to ask SMS if there are relevant documents to bring to Committee

LW asked all members who had not completed their competing interest forms to forward to REF as soon as possible. It was noted that most members had written a declaration rather than completing the form provided.

AS/JT/DW/SMS to forward CI to REF

TW queried whether the Committee would be planning a weekend workshop. RS thought that somewhere such as Ashridge would be a good place to look into holding the workshop.

REF to follow up with RS

 

2. Review of BMJ instructions for authors/info to contributors

LW noted that PS/LW had not yet scanned other journal’s to compare the BMJ policies versus other journals. As an introduction to the topic RS explained that at the Barcelona Peer Review Conference the BMJ guidelines were reviewed and compared well to other journals. However, as noted in the Committee’s earlier meetings there were still some areas that could be improved on. DW suggested that it would be helpful to divide the guidelines to authors into two areas: practical instructions and policy items outlining values. RS explained that at present the categories are arranged under headings but relevant areas are cross-referenced.

PS/LW to scan other journal’s instructions to compare the BMJ policies
Guidance on bmj.com to be split between practical and policy items

The first item proposed by PS/LW was an additional section on the editor’s duty of confidence with the suggested title "What we do if we suspect misconduct". The Committee agreed in principle to adopt this suggestion though the precise wording of the "warning" would need to be finalised.

New section "what we do if we suspect misconduct" PS/LW to draft

RS highlighted that there is a facility on bmj.com to complain about the editors and/or editorial processes. The journal has also posted information about the papers triage process and the rejection of many papers without review. JT also stated that it was easier now to track the progress of one’s paper with the online manuscript tracking system, Bench>Press. LW stated that it was important to raise correct expectations on the part of authors’ as many would prefer a rejection after 48 hours rather than 6 months later in the peer review process. RS stated that most authors preferred the open review process that explicitly stated reasons for rejection after review.

LW stated that she had not been able to find the link to the guidelines on consent. REF stated that these had been posted with the minutes of the 17/07/02 meeting on the Ethics Committee part of bmj.com but she would check that all links on bmj.com were correctly linked through. The Committee discussed that the guidelines should be posted as the finalised version in the advice to contributors area. LW pointed out that PS/SMS were writing a piece on the development and use of the new guidelines. This article should be cross-linked to the guidelines.

REF to check all Ethics Committee links on bmj.com
Consent guidelines to be posted in advice to contributors area
PS/SMS to write article on development/use of guidelines

The Committee agreed with item 2 that the section on confidentiality of patient information should be updated to refer to the ethics committee work in this area.

Confidentiality of patient information to be updated

LW admitted a competing interest with item 3 as she edits quite a few papers pharmaceutical companies wishing to publish articles by non-native speakers.

In regards to item 3, that there should be an explicit statement about submissions from developing countries RS stated that the BMJ was open to papers from anywhere in the world. RS thought this point was more a policy point than guidance to authors. RS assured the Committee that the BMJ was very willing to rewrite items that were good science but where the authors’ first language was not English. The only problem was where the quality of the English was so low that the scientific point of the paper was difficult to elicit. TW commented that the issue raised by the accompanying articles was much deeper and related to access of "Southern" countries to health care research and information. The drive to globalise had to be balanced with relevance to the readership, which was an editorial judgement. RS pointed out that the journal’s readership was increasingly global. The Committee felt that the issue should be discussed in its own right at a meeting with PS present. DW noted that it would still be worth putting a statement into the guidance for authors stating that all papers are judged on scientific merit and that copy-editing would be provided if necessary. RS also indicated that occasionally where there was a real language barrier the technical editors would write the article with the author.

Schedule for next meeting with PS present

Item 4 suggested that the information about research ethics should be expanded. PS/LW suggested that the journal should state that it would not publish results from studies unless they have had appropriate research ethics review. The difficulty was that any guidance should also state what types of research do not require research ethics approval and what authors should do if there is no relevant ethical review authority. LW stated that the Committee needed to consider in detail what type of research definitely do and do not require research ethics approval. DW queried to what extent the authors should state that they received ethics approval? Should it simply be a statement that they had, or should it be a more explicit statement of the name of the Committee? LW thought that it would be an interesting research project to see what the most common practice is. RS stated that occasionally a journal would go back to the research ethics committee but this was usually through the author. TW felt that the research ethics review statement should address the substance of the research—that is, the research was conducted ethically. Quite often the review by a research ethics committee was an administrative device with no guarantee about how the research was conducted. Such a statement would make the individual authors responsible for their conduct of the research, and would also catch items such as audit, which are not currently covered by research ethics review. DW supported that suggestion but still felt that there should still be the expectation that researchers should have gone through the research ethics review process. RS queried whether the system of having a requirement for a third party review of research by a research ethics review board was not an important part of the process. TW felt that regardless of this, the individual responsibility should still lie with the researchers not the research ethics review as this was sometimes viewed as a device to protect the researcher and not the patient. There should be an obligation on the individual to undertake ethical research. TW stated that such a requirement would allow the journal to question what steps had been taken to carry out such things as audit ethically. TW indicated that large granting bodies such as The Wellcome Trust, DOH, and Royal College of Obs & Gyn were expanding the scrutiny of ethics review beyond mere process. RS queried would this mean authors having to show that they had addressed potential ethical issues, including how they have thought about solving the problems and what actions they had taken. TW agreed that this should involve the author ensuring that the study was ethical and then referring to a research ethics review. If there were none then the authors could look to relevant codes of practice and research guidance from the Colleges. LW noted that there was currently no obvious reference to the Declaration of Helsinki on bmj.com. It might be useful to have a link to documents such as the Declaration, the DTI guidance, DoH guidance, and the Royal College of Physician’s Guidelines on the practice of ethics committees in medical research involving human subjects (1996). RS summarised: the journal would state that authors are ethically responsible for research, obtaining REB approval would not end that responsibility. Authors must act ethically throughout the research referring to relevant ethical guidance and if questioned be able to defend their actions. JT felt that the journal should help authors by pointing them in the right direction. LW felt that the web site could be a useful place to post resources. AS pointed out that there are almost 50 sets of guidelines in the UK, however, the Committee agreed that providing a starting point would be useful. The Committee discussed a few examples of cases where authors had claimed that there had been no relevant ethical review process. DW pointed out that a statement as summarised by RS above would cover such things as audit where there is no formal ethics review process. RS agreed to draft a piece for the web site. JT indicated that there was a 5-point research governance document that could be included.

RS to draft appropriate statement
ALL forward potential author resources to REF

Item 5 suggested that a statement on editorial ethics could be added and what recourse authors have if they disagree with the opinions/decisions of the reviewers or editors. PS/LW suggested that a more explicit explanation of the appeals process should be posted. TW felt that this item may confuse issues of editorial judgement which were not ethics related. RS pointed out that there was a facility to complain about the editorial processes. There was a first appeals process but not a second. Additionally, a complaint could be made to the Press Complaints Commission or the Advertising Standards Commission. RS agreed that the complaints procedure and the procedure for corrections and clarifications should be listed more prominently.

RS to draft statement on complaining/appeals process

Item 6 suggested that the section on authorship/contributorship needed updating to reflect the revised International Committee of Medical Journal Editors criteria. The new criteria include guidance on analysis and data collection. RS pointed out that the ICMJE’s guidance on authorship is moderately restrictive and that the BMJ doesn’t use this. The BMJ uses the very broad contributorship guidelines posted on bmj.com. LW pointed out that the revised criteria would still be a useful reference document for authors. JT queried whether the notion of contributorship was an adequate way of dealing with large studies where one may have 8 or 9 main people who depend on a very large group of contributors? RS pointed out that the BMJ doesn’t distinguish the boundary between authors and those "involved". Quite often such distinctions were governed by power relationships. The policy is that everyone states what he or she did in the research. JT queried whether the BMJ would be in danger of losing important papers that were the result of a large study as many researchers’ jobs are linked to their publication record? RS pointed out that the journal publishes a list of authors' names at the beginning of the paper and, secondly, lists contributors (some of whom may not be included as authors) at the end of the paper, giving details of who did what in planning, conducting, and reporting the work.

Updated ICMJE criteria to be added as author resource

Item 7 suggested a section on post-publication "review" and comment on how this is handled. It is not clear if the BMJ encourages authors to respond to all comments or if they are alerted when comments are posted to their articles. RS explained that the journal posts all rapid responses that contribute substantially to the debate and are not obscene or libellous. Authors of the original papers are notified that a comment has been posted in regards to their paper. If a letter is received on paper this is forwarded to the authors of the original paper. The aim of the rapid response system is to encourage a timely debate. LW suggested that a section explicitly stating this policy and a suggested time frame for response would be helpful. RS agreed to draft this for posting on the web.

RS to draft section on rapid response policy

Item 8 listed minor alterations to the text.

REF to forward to appropriate person for alteration

 

3. New items

3.1 BMJ research projects (12 protocols)

The BMJ Research and Experimentation group submitted 12 research proposal protocols to the Committee for their opinion. The 12 study protocols covered the following areas:

1. Surveyed authors whose work had been rejected for their preference as to the communication of the rejection.

2. Evaluated the readability of a section of BMJ articles.

3. Surveyed the level of agreement of editors reviewing manuscripts in differing ways.

4. Surveyed the perceptions of authors whose work had been rejected.

5. Surveyed the influence of statements of competing interests on readers.

6. Evaluation of non-standard training grade medical recruitment advertisements.

7. Evaluation of the peer review of references.

8. Evaluation of the authors’ perceptions of electronic long, paper short (ELPS).

9. Evaluation of the value of providing materials in PDA compatible format.

10. Study about the communication of risk to doctors and patients.

11. Evaluation of readers’ & authors’ preferences on presentation of scientific information.

12. Evaluation of author proposed versus editor selected referees.

LW queried what the expectation of the Committee was? RS stated that the BMJ policy was that all research with patients requires ethics committee approval. Currently the only protocol concerning patients is no 10. It was also discussed that any protocols put to the Committee should be in their final form as changes could alter the issues to be considered. TW felt the Committee’s role should be to make observations on the process of the studies and raise any ethical questions that require addressing. In many of the studies there was no danger of breach of confidence or harm to persons. There may be issues of breach of privacy and some studies may impact on the editor’s duty of care to readers, reviewers, and authors. It was noted that PS had forwarded the link to the Canadian Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans, which includes social science research. SMS had indicated at the previous meeting that there was EU legislation in this area.

SMS to forward information on EU Directive to REF

LW asked if there were any concerns on protocols 1–4, which have already been completed? The Committee only had specific comments on protocols 5 and 10.

The Committee first considered protocol 5. This research, which had already been carried out, involved some deception. RS queried what duty the researchers had to the readers involved in the study? TW felt that the researchers should have debriefed the readers about the nature of the deception and the true purpose of the study. This should be done as each stage of the study is completed. RS agreed that it was good practice to go back to the responders and queried whether the researchers should go back to all "readers" who had been sent the article or only those who had responded? TW felt that the duty of the researchers was to meaningful participants—that is, those who had responded.

Responders should be debriefed

The Committee considered protocol 10. RS informed the Committee that the research group had not agreed to fund this study yet and would be reconsidering this protocol in April. LW stated that her major concern was the question of where the patients would come from. The researchers would have to access background information on patients in order to contact them. DW stated that as this would involve patients at multiple sites then the protocol could go before an NHS or Multicentre Research Ethics Committee. RS pointed out that the study would be web based. DW suggested that one of London’s Strategic Health Authority Ethics Committees might be a good starting point to determine who to submit the protocol to for review. The Committee agreed that the BMJ Ethics Committee was not an adequate forum for complete ethics review of this protocol. JC stated that the process had been to first take the protocol for scientific review at committee then progress the scientifically valid protocols to the ethics committee. RS reiterated that this was the first protocol to involve patients so the processes were being developed. The formal submission to a research ethics committee and submission to the Committee could run in parallel. RS felt that it would make sense for the Committee to also look at the final protocol. LW agreed as changes to the protocol could affect the ethical issues raised. RS asked the Committee if any protocols involving patients should go through a formal research ethics committee rather than the BMJ Ethics Committee. The Committee agreed that while they would comment on these protocols a formal process would be required for protocols involving patients. LW suggested that any specific comments should be forwarded to Dr Sara Schroter, Senior researcher, BMJ.

Protocol needs to be formally reviewed by research ethics committee

LW queried how the system of cases coming to the Committee would work? JT also enquired about the likely volume coming to the Committee. RS stated that the scientific committee review would probably allow through about 4–5 protocols at any one time. LW suggested that the Ethics Committee could review once the scientific research group had approved the protocols, however, she queried whether this would delay the research? RS thought that the Committee could act in an advisory role. Any derogation from advice would have to be justified in writing to the Committee. LW agreed that the Committee could not act as an approval process but will comment. RS felt that it would be useful to have the Committee review the protocols as it could flag up difficulties and issues of publication. JT agreed that the Committee could provide practical advice that could help facilitate research.

3.2 Obtaining consent from patients who are not mentally competent

An author has sent a letter asking if the BMJ would consider publishing a picture of a patient who is not mentally competent as part of a description of that patient’s case. The author has consent from the patient’s family but cannot obtain it from the patient.

AS raised several issues to be addressed about consent for publication of information concerning patients who are not competent. Was the image proposed for publication demeaning to the patient or would be contrary to his/her interests? The publication may not be in the patient’s interest but would it be in the interest of people in the same category as the patient? How essential would the picture be? Must the picture be identifiable? JT queried what the role of the relative who had given assent must be—the main carer? DW pointed out that a person might assent to publication of their own information even if it is contrary to others’ interests. RS highlighted that publication of an image would rarely if ever be in a particular individual’s interest. AS pointed out that in the case of a person who was mentally incompetent then one could consider if an action was contrary to his/her interest. There is an argument, however, that publication must be essential to justify publication. JT pointed out that it was very difficult to address these questions without seeing the picture being discussed. RS pointed out that a "story" is unlikely to ever reach the threshold of being in the best interest of a mentally incapacitated person. RS queried whether the Committee felt that there was a difference between a description of and image of a mentally incapacitated individual?

The Committee considered this query using the revised guidelines on consent. The guidelines do mention cases of images being used without consent and stipulate that they must be carefully anonymised. In regards to 3a of the guidelines the Committee thought that the guidelines should be changed to "without an unduly burdensome effort and or it is also impossible…." This would cover the case where a patient was traceable but—as with the case of a mentally incompetent adult—neither the patient nor the next of kin could consent on his/her behalf. Without seeing the photo and accompanying text it was impossible to see if the piece met the conditions of being worthwhile, unobjectionable, and capable of being sufficiently anonymised.

PS/SMS to note suggested changes to 3(a) of consent guidance
RS to write to ask author to submit piece

 

3.3 Acceptability of verbal consent

A randomised control trial has been submitted to the BMJ to determine if an intervention with a nutritional supplement could reduce mortality in a certain age group. The methods state that: "verbal consent was obtained from study participants". Is verbal consent an acceptable form of consent?

LW started the discussion by putting the query in the context of the low proportion of the population who are literate (~35%) in the country the research was carried out. LW asked would it be practical to require written consent? AS noted that in some cases even written consent was not valid. TW agreed, stating that there was a difference between taking consent as an administrative convenience versus obtaining real consent from the participant. RS queried whether the Committee felt that "written" consent involved providing a document for the participant to read or actual documentation recording consent to the research? AS felt that there needed to be evidence of a procedure of some kind. JC indicated that there had been no evidence of documentation. DW felt that documentation would not be conclusive as often it came down to whether the journal believed that consent had actually been obtained. AS stated that in order to give informed consent the participant must have understood the information being given—was this likely? AS thought it likely that field workers would be part of a team so it could be possible to validate that consent had been taken. JC indicated that in this research, there was a strong relationship between the participants and the field workers.

LW summarised that the Committee’s discussion indicated that there was no problem in principle with verbal informed consent but that such consent must be well documented. AS pointed out that written consent was not a legal requirement in the country where the research was carried out. RS pointed out that some physicians have been struck off in the UK for failure to demonstrate that they had written consent. JT felt that if the consent by the participant was verbally given to the field worker it should also be witnessed and documented. AS pointed out that in the environment in question it would be quite difficult to test whether the participant had completely understood the information in order to give a fully informed consent. TW felt that even if one where to obtain a thumbprint on a piece of paper stating that informed consent had been given that it would be unclear whether there was fully informed consent. TW felt that this was a matter of proportionality and referred the Committee to the papers about encouraging research in developing countries submitted as item 3 of the review of the BMJ guidelines. JC pointed out that the main researchers in the study are from a large Western medical research institute so this is not strictly research originating from a developing country. The Committee discussed that even though a large proportion of the population are not literate they would still expect some kind of documentation. This is to demonstrate that the patient existed and that consent was obtained. LW summarised that there should be assurance from the researcher and field workers that consent was obtained (detailing the process) and some kind of mark by the participant that they had given consent. JT felt that this would be too open to fraud and that there should also be verification by a witness as to the process and obtaining of consent. After some discussion it was decided that obtaining a mark from the participant was somewhat superfluous as an unscrupulous researcher could merely ask an illiterate participant to put their mark on the consent form. TW still felt that irrespective of whether the consent was written or verbal the issue still came down to whether the consent was real. For example, in some cultures the researchers explained the trial to the village elders in order to obtain consent from a group. LW asked the Committee what advice they should provide in this particular case. DW/JT/AS felt verbal consent was adequate. TW/RS felt that further information was required about the documentation of the verbal consent. JT pointed out that it would be useful to debate this issue with the readership when discussing the guidelines on consent.

PS/SMS to include this issue on article re consent guidelines
Editor should ask for more information about the verbal consent

 

3.4 Paper from another journal that the BMJ was considering republishing

The piece describes the actions of a doctor with a terminally ill patient who did not wish to have further medical treatment but was overridden by the wishes of the family. The doctor acquiesced to the family’s demands for fear of being accused of not doing enough for the patient. There does not appear to be consent from the relatives to publish the piece and the author is not responding to queries about this.

JT felt that the piece was worthwhile and queried where the journal was in regards to consent? TW queried whether the editorial office had checked the medical register to get a recent address for the author. The BMJ would be interested in republishing the piece but the author is not responding to communications to proof the article or obtain consent. It is likely that the patient’s relatives would not be happy with the article.

The Committee considered the article in the light of the guidelines on publishing without consent. Consent could not be obtained from the patient and it was unclear whether the author could trace the family. The Committee agreed that the piece was worthwhile. However, it was felt that it was likely that the publication of the piece might cause the family distress. The Committee discussed that in a case where a patient disagreed with his/her family then it would be possible for the patient to give prospective permission for publication. The Committee agreed that though the other journal may decide to publish, in this case the author would need to obtain consent from the patient’s family in order to publish in the BMJ. It was queried whether the author’s non-response to communications was deliberate.

RS to relay discussion to journal editor

 

4. Ethics committee approval for audit

DW to provide a focus paper that will be circulated to the Committee members. TW/LW indicated that they would both be happy to collaborate and comment on the focus paper. This will aid the development of the guidance to authors about audit and research.

DW to circulate focus paper for APRIL; Committee to discuss in JULY

 

5. Publications about the Committee’s work

Have not received draft article from SMS, defer to next meeting.

SMS to write piece on Committee’s work to date for APRIL

 

6. Update on current cases

See item 1.

7. Any other business

REF asked the members if they were happy with receiving all items electronically. The Committee agreed that this would be acceptable for future meetings.

8. Date for the next meeting

The next meeting is Monday 7 April 2003


Other confirmed 2003 dates:

Monday 14 July 2003