Journal A is dedicated to communication about practical treatments related directly to patient and personal experiences. These ongoing discussions have been part of this specific medical profession for the past 50 years and journal A is a platform for these discussions.
Regarding new treatments and new developments, permission from the local medical ethical commission is mandatory as well as patient written permission for publication. For all other cases concerning standard practise reports about practical treatment issues, only written informed consent from the patient is needed before publication. As expected, only successful practical treatment cases are submitted for publication to journal A.
The society related to the journal organizes a biannual complication meeting which is attended by a small audience of around 120 doctors at which participants present their complication cases. At this meeting, many basic complications are discussed, which can be related to lack of education, lack of knowledge of materials, lack of knowledge about patient disease, or insufficient training, some of which have devastating outcomes. Many of these complications are avoidable.
As an editor and a doctor, I would like to publish some of these cases to improve patient safety and healthcare in general. However, the authors are often afraid of legal repercussions and patients are often not willing to provide consent after bad experiences. My solution would be to publish a series of these case reports as one publication with all authors in alphabetic order. I would leave out specific patient demographics, such as age and gender, and I would do so without written patient informed consent.
Question(s) for the COPE Forum
Would this publication violate COPE guidelines? If yes, how could this be resolved?
Is there any other way to resolve these issues?
The Forum agreed that the idea of presenting cases which do not result in improved patient safety or actual patient harm for learning opportunities is an interesting one and worthy of exploration. If any of the cases in this series have pending legal issues within an institution, this would be a significant consideration to protect the legal process. The editor must consider the legal environment in the country of origin, including regulations on informed consent.
As there are 120 cases, are there also 120 authors? If there are many authors, it may be easy to anonymise the cases, especially if the authors could be added in alphabetical order to help disguise identities. Otherwise, it is difficult to anonymise the data and retain essential elements for learning purposes. A suggestion was to publish one case per issue, deidentifying the data and to make all 120 presenters (or however many there are) the author(s). Another suggestion was to have a recurring column in a journal that has an ethics committee, who would review the cases and make necessary changes for protecting subjects’ identities. The journal could also seek institutional review of the cases to ensure that blinding was sufficient.
The Forum noted that changing the data is not acceptable unless that is clearly indicated in the text; if the case is a composite, this must also be stated. The reason being that researchers might be interested in doing composites of these cases, which would be based on false data unless the original cases are clearly labelled as composite or fictional scenarios.
If the editor wants to develop this project, he needs to have a strategic plan to publish these cases that considers the larger picture, and involves a sound, rational approach, which might be very valuable to readers. The Forum encouraged a careful review of the potential legal and privacy issues that must be addressed.
Authors Virginia Barbour, Muhammad Irfan, Deborah Poff and Michael Wise on behalf of COPE Council Version 1 December 2016 How to cite this COPE Council. Journals’ Best Practices for Ensuring Consent for Publishing Medical Case Reports. Version 1. December 2016 https://doi.org/10.24318/cope.2019.1.6
Our COPE materials are available to use under the Creative Commons Attribution-NonCommercial-NoDerivs license https://creativecommons.org/licenses/by-nc-nd/3.0/ Attribution — You must attribute the work in the manner specified by the author or licensor (but not in any way that suggests that they endorse you or your use of the work). Non-commercial — You may not use this work for commercial purposes. No Derivative Works — You may not alter, transform, or build upon this work. We ask that you give full accreditation to COPE with a link to our website: publicationethics.org
We received a paper which describes genotyping results from a large number of individuals (>50) from five unrelated families, in which family members had various blood and liver conditions. On submission we noted that the paper included specific details regarding the clinical histories of individuals in each family. Some individuals were described in substantial detail, others only briefly. For example, information about probands included age at presentation, sex, ethnicity, clinical history, occupation, clinical complications (some quite specific), clinic attended (for some individuals), history of alcohol consumption, ages of relatives, clinical details for relatives, age at death etc.
The authors responded that when initially obtaining consent to the research project, patients consented that “[t]he results from studies on the research samples may be published, but individual patients will not be identified in the publications”. They claim that details included are not identifying, and that it would be impractical now to trace all living relatives. They asked whether our consent form has been approved by an IRB, and say that before using it they will need to have it reviewed by their IRB.
We are unsure how to proceed but feel that when consenting to the research project the individuals may not have realised they would have been described to this level of detail, and that we should respect their privacy. One option might be to ensure the authors seek consent to publication from all probands, but then remove extraneous detail regarding the relatives. However, it is possible that the IRB may have useful input regarding the publication of individuals’ details from this study.
The editor was asked if the personal information was essential to the study. If it were deleted, would the paper still have value? The editor confirmed that the information was crucial to the study and could not be deleted. All agreed that the editor needs to weigh up the need to publish against any harm that might be caused by publication. Most believed that the harm involved was probably quite severe and so outweighs the need to publish. One test that can be applied is to consider whetheran investigative journalist would be able to identify the patients from the study (in this case, most Forum members felt that they could, therefore the risk of identification is high). The Forum did not believe it was that impractical for the authors to obtain consent and so they thought it was reasonable to ask the authors again to obtain consent. Did the authors make it clear to the patients how much detail and history would be published? Also, some noted that with pedigree studies, it can be very easy to identify patients. As with all such cases, the editor needs to make a judgement concerning the harm that might be caused to the patients in publishing and the benefits of publishing. Most believed that the risk of identification was high and that the editor should not publish without consent. The advice was to go back to the authors and ask them to obtain consent. COPE was asked about whether it could produce a generic consent form for all journals (to avoid problems of having to use multiple forms if a manuscript is reject by one journal and submitted to another). COPE Council agreed to consider this suggestion.
The editor informed the authors that the case had been taken to COPE, and the agreement of those who attended was that patients and family members would need to give consent to publication of the case(s) having read the paper. The authors responded after several weeks to say they had traced the patients and obtained consent to publication. They updated their paper to state this in the revised text, and also updated the text to minimise some details of family members, and to state that details were minimised to maintain anonymity but that further clinical details could be made available to other researchers on request.
A was a researcher in C’s lab for 1 year, during which time they published a joint research paper in a third party journal (journal S). After leaving C’s institute (henceforth called institute X), A published in the journal (journal T as a sole author). The affiliation provided by A on the paper was institute X. All of the data reported in this paper were obtained while A was still employed at institute X, of which C is a senior faculty member and in whose laboratory A was a research associate.
C contacted journal T with several allegations regarding A’s paper in journal T. These were: (1) A submitted the paper after having left institute X (2) A never discussed with C his intention to publish these data from C’s laboratory (3) Material published in journal T is “fully based” on data they had previously published together (4) A did not have approval to use material published in journal T (5) Questions two of A’s methods/techniques and the data that resulted (6) A never acknowledged a funding body
An editor on journal T reviewed these allegations and believes that A’s paper was a follow-up paper, and that it appears to take further the research A and C had previously published in journal S. The paper had undergone peer review on journal T by one specialist referee, who provided a full and penetrating report.
A admits to claims: (1) The work was conducted at institute X (6) A admits to a mistake here
A disputes points: (2) A claims to have raised publication multiple times with C, C’s role in generating the “results were almost non-existent” but indicates that C claimed the data were C’s for C to decide (3) The paper is a follow-up to their previously published work (4) A personally analysed the material with the material produced by a technician (5) A claims that the materials are different images of different samples
Journal T submitted relevant correspondence to the dean at institute X and asked them to investigate. Institute X’s dean responded 4 working days later endorsing C’s position on the basis that A had duplicated material, had misrepresented its novelty and did so both without permission or agreement from C.
The investigation by institute X failed to communicate with A or ask for A’s response to the matter. When the findings were presented to A by journal T, A strongly refuted its key points.
Journal T’s position has been to ask institute X to investigate fully the above allegations (including communicating with A) and to make an official statement linked to the case.
However, we seek COPE’s recommendation whether this should be our position or whether they would advise a different approach.
The Forum acknowledged that the institute failed to handle this case correctly and that their investigation is of little value. However, there is little that the editor can do about the institutional failing other than reporting it to someone more senior than the Dean of the institute. The Forum believed that there was a case for asking institute X to carry out an unbiased independent investigation. In the absence of an independent review, the editor could publish a “notice of dispute”, informing readers of the situation and explaining that it is not for the editor to decide the rights and wrongs of the case. All agreed that the editor had handled the case very well and agreed with his course of action.
The editor requested an independent investigation from the institute. It is the editor’s belief that once he asked for an independent investigation, one that is truly independent and would have taken some effort on their part, that the institute decided it was not worth the trouble. The editor suspects that he will not hear from them again.
We would appreciate COPE Forum’s advice with regard to consent and confidentiality in the publication of case histories.
Our journal is published three times a year and has a relatively small circulation but also goes to some postgraduate medical libraries, so is potentially available to a much wider audience.
Our subscribers are doctors working with patients presenting with sexual difficulties in a variety of settings. Training is experiential, through the presentation of cases in seminar groups with a designated leader. Case presentation at our meetings and in the journal articles also provides learning opportunities and peer review.
As well as addressing (in an ordinary medical way) the physical aspects of the sexual problem presented, our work explicitly uses the study of the doctor–patient relationship to shed light on the presenting difficulties. In other words, the doctor–patient relationship is one of our key treatment ‘tools’ allowing connections to be made between the body and the mind in relation to the sexual problem.
These consultations almost always include intimate and sensitive information. Therefore, it had previously been felt that asking for permission to present cases would disrupt the very doctor–patient relationship on which our work depends. As a result, we have tended instead to seek to ensure anonymity by disguising personal details in presentations at meetings and in the journal rather than asking for consent from the patient.
We recognise that this practice is out of line with that recommended by the BMA and the GMC.
Our main concern relates to the possibility of limiting the availability of suitable case material for teaching purposes in presentations and in the journal if doctors felt unable to comply with a requirement to obtain patient consent. This would seriously interfere with the training.
We would prefer to continue using our current guidance where consent is recommended rather than mandatory. Indeed, we would like to suggest that where the doctor is satisfied that the information can be anonymised, he/she does not need to try to obtain consent.
We have considered adding a statement in the journal for patients who might read it, and recognise or believe that they recognise themselves, or their story. One of our founder members addressed this very idea in the preface of a book, asking patients to accept that no-one but themselves would be able to recognise them and to understand that, although each is unique, there are many like them who share similar problems; they are not alone.
We would appreciate your advice on the following points:
1) Is there any justification for continuing the current practice of ensuring anonymity by altering elements of the case history—for example, name, occupation, family details, appearance, geographical location, etc—and continuing to avoid an absolute requirement to obtain consent from patients? As a training body can we claim ‘special case status’ in view of the need to use case material to further training aims for future generations of doctors?
2) It has been suggested by a lawyer who is however not a medicolegal expert that we might interpret the GMC guidance and the NHS Code of Practice on Confidentiality to mean that consent is not needed where the doctor is satisfied that he or she can anonymise the data to be used (bearing in mind that the symptoms might be an identifier).
3) If you feel that we do need to change our practice and require authors to obtain consent from patients whose case histories they wish to present; it would be very helpful to have your advice on how this might be implemented.
All acknowledged that this is a difficult situation. This issue has come before the Forum many times, particularly in the case of child protection procedures. Such cases have proved almost impossible to publish in the UK and the public interest defence (that public health interests in knowing the information are greater than the requirement for confidentiality) does not seem to hold up.
The Forum warned that if enough information is available in the published report that a good investigative journalist can identify the patient, then the editor is in breach of the Data Protection Act. The Forum also warned that the editor, if on the UK medical register, could be in breach of the General Medical Council (GMC) regulations and was advised to consult the GMC’s documents Good Medical Practice and Patient Confidentiality and Consent.
The Forum also advised contacting the GMC for advice and approaching the Information Commissioner for the DPA. The Forum noted that even producing educational and training leaflets can be contentious and consent should be sought and a disclaimer added to the effect that patients should accept that no-one but themselves would be able to recognise them and that patients need to understand that, although each is unique, there are many like them who share similar problems. PLoS and the BMJ have policies on these issues and it was suggested that the editor might be able to adapt these.
While some advised seeking consent or publishing vignettes of anonymised cases, it is almost impossible to fully anonymise case studies and often such heavily anonymised cases can be educationally useless.
As recommended, we have spoken to the Information Commissioner for the DPA. He advised us that anonymised information could be published, and said that if the patient is not 100% sure that he/she is being described, then this is anonymised information. He said that it would nevertheless be best to ask for consent to publish a case history in its anonymised form.
We filled out the questionnaire from the GMC consultation on their new draft guidance on confidentiality. We have noted that the GMC proposes to change the recommendation that consent is obtained whether or not you think the patient can be identified, to recommending that consent is obtained if the information cannot be anonymised. The glossary now adds the word “reasonably” when discussing identification of information.
Pending the outcome of the GMC Consultation, it is likely that we will change our Guidelines to contributors to recommend that consent is obtained prior to publication of case reports in our journal, even when these have been anonymised. In situations where it is not possible to obtain consent, authors would be asked to confirm that, in their opinion, there is adequate anonymisation of the patient’s details; we have also produced a form for this purpose.
We raised these issues with the members of our organisation at a meeting in November 2008. We explained the background and provided a sample consent form for their use, together with our plans to provide sample patient letters to request consent, clinic information leaflets and suggestions on how to ensure anonymisation. We found that most were in agreement with our suggestions.
We received a paper reporting on the outcomes of treatment of an individual with obsessive–compulsive disorder and body dysmorphic disorder, which seemed not to respond to standard treatment. Following ethical review and approval, and individual consent, the individual was treated with several distinct courses of an experimental therapy. The individual’s clinical and family history, and their outcomes in response to the experimental therapy, were detailed in the paper. The editors felt that sufficient details were included to potentially identify the patient, so as outlined in our editorial policies, we asked the authors to obtain signed consent to publication from the patient.
The authors responded that they had attempted to remove all identifiers from the paper (although the editors felt that very substantial and identifying details remained), and that revealing the patient’s identity to the editorial office would violate their clinical protocol, and the principle of patient anonymity. The editors discussed this and felt that it was still very important that the patient have read the paper and agreed to its publication; secondly however, we were concerned about the journal having access to (and our ability to keep confidential) the patients’ names. We also wondered whether under HIPAA law it would even be legal for the authors to give us the patient’s identity.
Therefore, the editors proposed that we would ask the author to have the patient (or patient’s parents) sign our consent-to-publication form, and then include in the paper a statement that they had done so, but that we stop asking authors to send us copies of these forms and simply rely on the authors’ word instead. Is this acceptable?
The Forum agreed that this was a curious case and raised issues relating to HIPAA law. Some argued that there is no need for a journal to know the identity of the patients reported in case reports. The authors could be asked for a statement that they have followed the correct procedures and obtained consent from the patient. But there is no need for the journal to see the consent form and hence no need to reveal the identity of the patient. If a problem arises in the future, the journal would have a reasonable defence as the onus would be on the authors to have obtained consent. The editor could request a signed assertion from the authors that they have obtained consent. Although the journal’s own consent form can be used, the Forum noted that it would be very useful to have a generic consent form so that patients need only sign a form once in the event that the paper is rejected by one journal and submitted to another journal. The Forum agreed that the editor’s proposed course of action is acceptable.
The authors were asked to confirm in their paper that the patient had seen the manuscript and agreed to its publication in an open access journal, and also to require the patient to sign our consent form, which should then be filed in case notes, but not sent to the journal. The authors did so and the manuscript went out to peer review. It was later rejected on scientific grounds. The editors discussed changing journal policy in light of this case to require that authors document, for such papers, that patients have seen the submitted paper and agreed that it could be published, and to cease requiring that authors send copies of consent forms which had been signed by patients. It was agreed to change policy along these lines and the changes are currently being made to our instructions for authors.
A manuscript was submitted from UK authors. The study was a case series of infants with a particular condition. A table in the manuscript contains descriptive data which are critically important for the readers with respect to understanding the risk of this condition in young infants and the likelihood of abuse.
The question is whether this table violates the law with respect to confidentiality/privacy. No IRB was obtained (case series/QI project). No consent was available from individual patients.
COPE’s advice would be appreciated.
Similar to case 07-10, members of the Forum felt the same issues of confidentiality and public interest applied in this case. Most were in agreement that consent would not be obtained, and as there was a chance that one or more infants could be identified from the data, the paper should not be published. Some of the members of the Forum suggested that perhaps the paper could be presented as a multivariate statistical analysis, removing individual identifying data. If this is not possible, the Forum again advised that the editor should consider not only the ethical and legal consequences involved in deciding to publish, but also the data protection issues.
A case report was submitted to journal X reporting on a child who had been admitted to hospital suffering an injury, which the doctors suspected resulted from a deliberate cigarette burn. This was not proved until the child returned to hospital with other non-accidental injuries, and following a full criminal investigation the child’s parents were convicted of child abuse.
Patient consent was not sought from the child’s legal guardians as the child was placed with the social services and lost to follow-up.
The authors believe the case report should be published to raises awareness among clinicians about potential symptoms of non-accidental injury in children, even though they cannot provide consent from the child or the child’s guardians to publish. The editorial staff of the journal however has some concerns.
It is possible that the child could be identified, because the location of the hospital is given in the authors’ affiliations, the injuries are clearly detailed and the parents have been convicted of child abuse (it is not known if there was any press coverage of this case).
The editors of journal X have informed the authors that they are awaiting advice from COPE before deciding whether or not to proceed with peer review.
Does this case meet the criteria laid out in the COPE code of conduct to publish without patient consent:
“It may be possible to publish without consent if the report is important to public health (or is in some other way important); consent would be unusually burdensome to obtain; and a reasonable patient would be unlikely to object to publication (all three conditions must be met).”
Is the child’s right to privacy in this instance more important than the right to publish?
Should the journal proceed with peer review of this manuscript, or should the report be rejected due to lack of consent?
Some members of the Forum argued that a paper containing identifiable details of a patient cannot be published without consent unless the patient is dead or the patient has given informed consent to publication. In this case, consent cannot be obtained and therefore the presumption should be that the case report should not be published. The Forum advised that the editor should consider not only the ethical and legal issues involved in deciding to publish, but also the data protection issues. In terms of whether the pubic health value of publication should overturn considerations of consent and confidentiality, the general consensus of the Forum was that it is unlikely that a single case report is so medically important that it must be published, and that if he or she wishes to use a public interest defence, the editor must be prepared to defend their decision in the event of a complaint by the patient or a regulatory body.
In summary, the advice from the Forum to the editor was that in deciding to publish, the editor should consider carefully the legal and data protection issues and also the medical value in publishing the case report.
Following discussion of the advice from COPE with the Editor-in-Chief of the journal, this case report was rejected from the journal on the grounds of lack of patient consent. The journal has since made a number of improvements to the instructions for authors to clarify the journal's position on consent for publication of case reports.
Our journal has received a submission regarding clinical trial results. The authors wish to include the “raw data” as an appendix to the manuscript. The study was completed several years ago and was controversial at the time. The authors wish to publish the raw data to allow the public to view the findings and make their own decisions about the trial. We do not know the best way to handle this situation.
Does anyone already publish anonymised raw data in their journal and, if so, how does one handle the ethics of such a publication:
Do you assume that the original ethics approval covers this new publication?
Do you require the authors to go back and obtain ethics approval to publish raw data? If so, does the author have to go back to the original ethics committee?
Also, does one allow the authors to publish only “relevant” data or does one require all of the data generated from the trial to be published?
The discussion centred on the definition of anonymised data. If there are three or more patient identifiers, strictly speaking the data are considered pseudo-anonymised, as theoretically patients could be traced. The committee acknowledged that ethics committee approval should be sought. If this is not possible from the original ethics committee, the author’s current ethics committee should be contacted. Regarding the question of publishing only “relevant” data, the committee agreed that this decision could be made by the editor.
The data were never submitted by the authors, and the manuscript was not published.
A group in a developed country performed an audit of pregnancy and its outcome in a group of 250 women with congenital heart disease. There were four maternal deaths of women in the series. It was decided to write up the audit. The paper reports aggregated statistics classified by diagnosis, but also contains a brief account of the factors associated with the four deaths (an account of one of these deaths has already been published in considerably more detail in the confidential enquiry into maternal deaths). Because of the unusual and complex nature of the lesions, these cases will probably be identifiable to medical staff involved in their care, and to close family members. Permission for publication cannot be obtained from the patients, for obvious reasons.
Do the committee consider that the local ethics committee should review the paper before it is submitted for publication? The policy of the local committee specifically excludes audits from its remit.
This case provoked a lively discussion on the legal and moral rights of the dead. Although legally the right to privacy dies with the patient, the committee felt that morally, consent from the next of kin should be sought. In this case it would seem that there is a strong public interest in publishing the cases in an attempt to prevent future avoidable deaths. Balancing the ethical considerations against the public interest, the majority of the committee thought that the audit should be published if it were impractical to obtain consent from the next of kin.
As recommended by the COPE committee, the paper has been referred to the local ethics committee for their comments. The journal is awaiting their comments before proceeding.
The journal was unable to follow-up the case and it is now considered closed.