1997

Dr A submitted an article to journal X that was published in 1996. Dr B wrote to the editor in January 1997, pointing out an error by Dr A. Shortly afterwards, Dr B submitted a longer editorial to the journal discussing the issue raised by this error in a much wider context. Dr B then withdrew the article and submitted it to journal Y at the end of March, with a covering letter in which he wrote: “We are unaware of any other papers that have described this problem ….The paper is not under consideration elsewhere.” At the beginning of May, journal Y sent Dr B’s article for peer review. Meanwhile, the letter from Dr B to journal X was accepted at the beginning of June. A revised paper from Dr B was returned to journal Y on 10 June. The editor wanted Dr A to reply and asked Dr B if he had objections to this. Dr B faxed back the following reply on 1 July: “Please feel free to ask Dr A to respond if you wish. We had corresponded with him when we found the error. Note, however, that we wrote a letter to the editor of X about this error and expect that he and his colleagues will respond to that.” The editor of journal Y decided not to pursue a reply from Dr A on the grounds that Dr A would eventually have a reply in journal X. On 15 July, the letter from Dr B was published in journal X, together with a strong rebuttal from Dr A. On 8 November, Dr B’s article was published in journal Y. No mention of either his letter or Dr A’s response in journal X was made, and the editor of journal Y had seen neither piece of correspondence. The editor of journal X complained to his colleagues at journal Y that this episode represented dual publication, for which he plans to publish a notice to this effect. (1) Is he right to do so? (2) What more should he do? (3) What should be the response of the editor of journal Y?

A paper was submitted that described the use of a non-licensed investigational drug. One of the paper’s reviewers drew attention to the fact that none of the investigators in the study had been supplied with the drug since 1992/3. The drug is produced exclusively by one manufacturer who has operated an extremely restrictive policy regarding availability of the compound. I contacted the clinical director who confirmed that the group had not received supplies of the drug since 1992 and that the material supplied at that time had an expiry date of 1993. What should I do?

We had provisionally accepted a randomised controlled trial of an exclusion diet given to young children with a particular condition. The trial design was that one group started the exclusion diet a month before the second group. In other words, both were given the “treatment.” One part of the trial was that children who were thought to have an allergy to a particular food were rechallenged with that food. We discovered rather late in the day that the study had been done without ethics committee approval, despite the fact that the trial had been carried out in a major British hospital. The reason the authors gave for not submitting the study for ethics committee approval was: “because all children received advice on dietary exclusion. The only difference between the control and the trial group was that the dietary advice was delayed for a month. Thus all the children were given a trial of an exclusion diet as part of routine management. No drugs were involved. In view of this, ethics committee approval did not seem necessary. I discussed this informally with a member of the ethics committee who felt that this was a reasonable decision.” We rejected the paper on the grounds that ethics committee approval had not been sought, but should have been. But should we do more? Is this an example of research misconduct?

A researcher has written to us to point out that a paper published in a German journal in 1993 was put together almost verbatim from articles published in the BMJ in 1989 and the New England Journal of Medicine in 1992. About three quarters of the material in the article in the German paper comes from these two journals. It may be that the data are original but it seems unlikely. What should we do?

A reviewer expressed suspicion that data were manufactured. We wrote to the authors saying that our reviewer would like to see the original data. The author replied that this research was carried out in the USA. We then wrote back suggesting that his co-workers in the US would probably be delighted that this work was being submitted for publication and would happily send over the data but that without this we could no longer consider the paper. The line went dead. What should we do next?

A journal published a letter from a student only to discover that it was not written by him. The editor has written to him and his dean apologising, and the journal is publishing a piece saying that the letter was not written by the student. It seems most likely that the piece was written by one of his fellow students. Should we encourage the dean to hold a full investigation?

A group of researchers are conducting a study of whether women aged 65 to 69 years will accept screening for breast cancer. They plan to invite these women for screening in the same way as they invite younger women for screening but will not know that they are part of a research study. The authors want advice on whether journals would be willing to publish their results, despite the fact that the women will not have given informed consent to be part of the study.

We have received a paper in which the authors have exposed a group of babies to physiologically unnatural circumstances. These circumstances do however arise quite regularly in some peoples’ lives. None of the babies had anything wrong with them, but some of them were siblings of babies who had died. Some of the babies showed physiological changes in the unnatural circumstances, which raised the possibility that they might suffer severe consequences if exposed to those circumstances in their ordinary lives. The exact meaning of this research is hard to interpret, but it does suggest that these physiologically unnatural circumstances might have severe consequences for babies. The hospital ethics committee approved the research, and the parents of all the children gave informed consent. Our editorial committee was, however, worried that this was non-therapeutic research and that the parents of the children whose siblings had died might find it very difficult to refuse consent for the research. We therefore commissioned an ethical commentary, in which the author argues that the research is unethical—partly because of the problem of consent we had identified and partly because the researchers had no clear prior hypotheses and had not done a power calculation. They have thus produced research that is very hard to interpret. (1) Is the research sufficiently unethical that we should not publish it? (2) Can we publish the research with a commentary arguing that it is unethical? If we so, should the authors be given the right of reply? (3) Or should we, as the authors argue, write the commentary ourselves rather than commission an outsider? In our commentary we could voice our ethical doubts, but say that we thought the paper was nevertheless publishable.

A reviewer has written to complain that a review he sent to us on which he wrote “In confidence—not for transmission to author” was transmitted in part to the author. He had made some rather derogatory remarks which had been edited out by the editor before he had sent back the comments to the author. The review that remained was critical but unremarkable.

(1) Is it acceptable for reviewers to send comments to editors that the reviewer is not willing to allow the editor to pass on to the author? Isn’t there something unscientific and indeed unjust about the editor having access to comments that are denied to the author?

(2) If editors are sent comments by reviewers that are marked confidential, are they obliged to respect that confidentiality?

(3) Should editors make entirely clear whether they want comments that can be passed directly to authors, comments that are only for them or a combination of the two?

A doctor has submitted an account of how his daughter falsely accused him of having abused her as a child. His daughter is another British doctor. We would like to publish the account as part of a package of articles on false memory syndrome. The questions we are considering are: (1) Can it ever be right to publish something that describes the intimacies of a family conflict, to illustrate a subject? (2) Do we need the daughter’s consent even if we are going to publish the paper anonymously? (3) Should we give the daughter a right of response? (4) Should we go ahead and publish the paper even if the daughter refuses to give her consent and declines to respond?