It is often said of military planners that they spend their time preparing to fight the last battle, not the next one. The same could be said of regulators. Take research ethics. Recently I was with the WHO in Geneva looking at the regulation of research during health emergencies. The question we were invited to consider was whether ordinary ethical standards could be modified in times of crisis. Pandemic flu was clearly in mind, although several spoke of viral haemorrhagic fevers in all their fear and drama.

A number of participants outlined the history and genesis of current ethical standards in research involving human subjects: the Declaration of Helsinki; guidance from the Council of Europe and from CIOMS. It was striking how much of it was owed to a single generative trauma, an event that has never stopped haunting the conscience of the medical profession: the research undertaken by Nazi doctors – if doctors we have to call them – during the Second World War. It is to this moral monstrosity that the guidance is addressed, to the murdered and mutilated victims. Never again.

As a result, two themes, two indelible principles run like a watermark through all subsequent guidance: informed consent and the protection of vulnerable individuals. Participants must be able to choose and those who cannot choose must be protected. Explicit consent must be given both to participation in research and to the uses the research is to be put. It extends beyond the subject to the subject’s identifiable data.

Before Geneva I had not given the regulations much thought. Aside from some unease among researchers about the hurdles that confidentiality could impose, they seemed to have served us well. But at Geneva I started to think a little more. At first, and somewhat against my instincts, I wondered whether a case could be made for the sheer horrific exceptionalism of Nazi Germany. Should such extremity govern our practice in less extreme times? But then, as I discussed in an earlier post, there was nothing extreme about the conditions prevailing in Tuskegee during the infamous syphilis trials. Scientific curiosity allied to a certain human indifference meant that penicillin was secretly withheld from black sharecroppers so that researchers could study the natural history of the disease. Research abuse can thrive as well under a democracy as a tyranny.

The meeting in Geneva nonetheless put before us a genuine ethical dilemma. A researcher into haemorrhagic fevers gave it at its most forceful. The diseases he studies are vicious. They can kill more than eight out of ten of those infected. Little is known about them. So far they have favoured remote locations and have been successfully contained. But the past does not guarantee the future. The people he treats – and the people he would like to involve in research – are in extremis. They are unconscious and haemorrhaging blood. They are in isolation. They cannot consent. But the use of their existing samples for research could deliver enormous potential benefits. Was the requirement for explicit consent excessive?

One would wish no insult to the victims of past abuses, but I wondered if we might find help in this dilemma by digging into the research regulation and going back to those deeper principles. Of course principles can be monolithic things, and bashing them out in practice can  require sophisticated interpretation, balancing and application. This case is no exception. Briefly put the purpose of medical research involving human subjects is the provision of generalisable new knowledge. Ordinarily the benefits accrue not to the research subject but, if useful knowledge is acquired, to unknown others. The research subject shoulders the burdens, strangers reap the benefits. Regulation addresses itself to this asymmetry, and in doing so it tries to minimise the harms. The primary principle is familiar: above all do no harm. And maybe this is the key. In the case of haemorrhagic fevers, what harms accrue to the patient from the use of existing samples? Of course there is more to harm than immediate physical pain. There are injuries to dignity. If later use of samples provides rich profits then there could be economic injury. But in an emergency, surely the calculation of such distant and uncertain harms could reasonably be deferred.

Could ethical standards be adapted in an emergency? In my view, yes. In the absence of immediate harm to the patient, consent to the use of existing samples could be set, temporarily at least, aside. The past is not always the best guide to the future.

Julian Sheather is ethics manager at the British Medical Association. The views expressed are his own.