Self experimentation and the Nuremberg Code

The Nuremberg Code is the cornerstone of human experimentation law and ethics.1 Nonetheless, the suggested exception of self experimentation in its article 5 has never been persuasive: no experiment should be conducted if there is an a priori reason to believe that death or disabling injury will occur, except, perhaps, in experiments where the experimental physicians also serve as subjects⇓.

• Download figure
• Open in new tab
• Download powerpoint

Duncan Smith

The judges should have stopped at the word “occur.” Why is the exception there, and is it justifiable to put the lives of others at risk because the investigator is willing to risk his or her own life? The answer is that the prosecution at Nuremberg (and apparently the judges as well) thought that this exception was necessary to prevent the Nazi doctors from arguing that previous US government military experiments—most notably the Walter Reed yellow fever experiment—had also knowingly risked the lives of subjects. This explanation is supported by the originally suggested wording of article 5 by each of the two principal doctors who worked for the prosecution at Nuremberg, Leo Alexander and Andrew Ivy. Alexander suggested adding yet another clause: “such as was done in the case of Walter Reed’s yellow fever experiments.”2 Ivy would have replaced the existing clause with: “except in such experiments as those on yellow fever where the experimenters serve as subjects along with non-scientific personnel.”3

As Alexander and Ivy had anticipated, the 1900-1 yellow fever experiments did come up in the cross examination of Andrew Ivy. Ultimately, however, the studies did not play a crucial role because Ivy testified that he could not recall their details. Instead, he made a different point, that unlike the Nazi experiments, these were not performed under government or military orders. Because of the common equation of Walter Reed with self experimentation, it is of interest that Walter Reed himself, unlike his research team, was not a subject in the experiments. Similarly, although the yellow fever experiments made an appearance at Nuremberg, nothing in the testimony suggested that the willingness of an investigator to be a subject could serve as an ethical justification to put other humans at risk of death.

Neither of these historical footnotes has much relevance today, not least because few, if any, contemporary researchers are willing to risk their own lives to prove the value or safety of their research.4 In the rare contemporary cases of self experimentation, today’s problem is whether self experimentation (an experiment done by the investigator on himself or herself only) must be reviewed and approved by an ethics committee before it is conducted.4 Three linked articles help clarify the ethics review question.

Groves (doi:10.1136/bmj.c6801) studied his bicycle riding and only his bicycle riding, and he has no plans to conduct bicycle riding research. He justifies not having his “n=1” bicycle riding study reviewed because he is the “sole investigator and subject” of the study and that it was “conducted . . . using his normal mode of transport.”5 This misses the central point. Groves was not doing research at all—he was not seeking generalisable knowledge by testing a hypothesis. He was simply trying to decide which of two bicycles to ride to work. The fact that he wrote up his “findings” in the form of a journal article does not make his decision making process a research project. It is a single anecdote that makes interesting reading, but it is much more suited to a cycling magazine or the newspapers than to a medical journal.6 Of course ethics review committees do not and should not review consumer product choice procedures (although someone might caution Groves about the increased morbidity and mortality rates of cycling in the rain and snow).

The Danish research team (doi:10.1136/bmj.c6812) don’t say why they didn’t seek ethics committee review for their study on whether alcohol can be absorbed through the feet, although it was probably because all three investigators were physicians in good health, they understood the study, and they reasonably believed that it carried no risk, except perhaps of embarrassment.7 This is responsible. But it is also responsible for an institution (and a medical journal) to require that the “no risk” determination be made by an ethics committee. As the Danish team acknowledges, a slight modification to their protocol, such as including “eyeball drinking,” could radically alter the risks of their research.7

The PARCHED investigators (doi:10.1136/bmj.c6761) did seek and obtain ethics committee approval for their investigator subject study. This is appropriate because their group contained more than one investigator subject (no plural exists for self experimentation), and because they recognised the risk of death posed by compromises to renal function.8 The ethics committee was overly cautious in its requirement of a data monitoring committee, but it curiously did not require documentation of fluid intake by the subjects, suggesting a cursory review at best. The investigators appropriately note this failure as a limitation of their study, but this simply makes their recommendation to drink more water while on shift all the more puzzling.

Where does this leave us? The reasonable conclusion is that, contrary to article 5 of the Nuremberg Code, for life threatening research the participation of the researcher as a subject adds nothing to the ethical analysis of whether the research can be justified at all. The Walter Reed studies should have been characterised as unethical at Nuremberg (because they predictably would cause the deaths of non-investigator subjects), rather than weakly defended, and they certainly provide no ethical basis for their repetition today. Informed consent is a necessary, but not sufficient, condition of ethical experimentation. Self experimentation is neither necessary nor sufficient. Even where risks are minimal, prior ethics committee review of research in which investigators are subjects should be sought, if only to confirm the reasonableness of the risk assessment. If, on the other hand, an investigator proposes to experiment only on him or herself, that activity is not properly categorised as research at all, but as self indulgence (or, some may say, self abuse). Nuremburg continues to teach us serious ethical lessons. Trivial interventions masquerading as research studies are primarily a source of amusement.