The ethics of self-experimentation
Case text (Anonymised)
The author was the subject of his study. He depleted himself of a vital nutrient until he had overt clinical and biochemical signs of the deficiency. He monitored various biochemical parameters as he became more deficient and submitted two manuscripts presenting his results: one detailing the biochemical changes and one detailing the differences in results obtained from different commercially available assays for the nutrient.
Reviewers of the first manuscript raised some concerns about the experimental model used and also concerns about the ethics of the study, particularly the lack of any oversight from an institutional research board. The second manuscript was not reviewed and both were rejected on the basis of the ethical concerns raised by the reviewer and concerns about the scientific validity of the results obtained from one individual.
The author feels very strongly that his experiment was not unethical and argues for the autonomy of researchers. He provided the following arguments for his study:
— The author is both experimenter and single subject, so the requirement for informed consent does not apply.
— There is no institutional involvement, so there is no possibility of coercion.
— The subject was assessed by a psychiatrist and found to be competent to evaluate the risks and benefits and to accept full responsibility for the conduct of the experiment.
— The Declaration of Helsinki does not comment on self-experimentation; it is concerned with research in patients and healthy volunteers. The requirement for ethics approval therefore does not apply. The Declaration of Helsinki cannot be cited as a reason for rejecting reports of independent self-experimentation. The Declaration was not intended to prevent autonomous independent humans from performing and reporting self-experimentation
— The subject was monitored by a qualified psychiatrist who continually assessed the condition of the subject. It was agreed in advance that the psychiatrist would intervene if, but only if, there was an immediate life-threatening condition.
— The motivation for performing the experiment was ethical in that the subject wanted to investigate the gross differences between the immunoassays he found in an earlier experiment because he was aware of the potential severe consequences of such errors in measurement of this nutrient. There was no conflict of interest.
— The need for IRB or ethics committee approval would totally exclude from publication any self-experimentation research performed by an independent researcher because they will not have access to any ethics committee or institutional review board. This would have prevented the publication of reports of highly useful reports (http://careers.bmj.com/careers/advice/view-article.html?id=20002566 and http://www.bmj.com/content/341/bmj.c7103.long)
The author appealed the decision to reject his manuscripts. Although the appeal was not upheld, we agreed to bring the author's arguments for self-experimentation to the COPE Forum for wider discussion. It is worth noting that the author’s two papers were eventually published in another peer reviewed journal without any negative response over the ethical issues.
Question(s) for the COPE Forum
• Do other editors at the COPE Forum consider manuscripts on self-experimentation without formal IRB approval?
• What criteria do editors use to determine whether such studies are acceptable?
The Forum suggested there are two issues here: the ethical issue of self-experimentation and whether the study warrants publication with an n of 1.
The Forum agreed that the author has the right to perform the experiments on himself, but he does not the right to have his findings published. There have been some highly publicised cases of self-experimentation, which in one case led to a Nobel prize for Barry Marshall after he infected himself with Helicobacter pylori (the causative organism of peptic ulcers). There was also the case of Spurlock’s documentary, Super Size Me, a social experiment in fast food gastronomy.
Regarding the second issue, the Forum agreed that this has to be a judgement call for any editor. Does the study warrant scientific publication? The author has the option of blogging about his experiments if he wants to disseminate his results.
The editor told the Forum that they were interested in learning if any journals had a specific policy on this issue, but among the Forum audience, no journal had such a policy. The discussion ended in agreement that this is a topic that needs further debate.