Self-experimentation.

Except in certain cases of unusual risk, self-experimentation should not be encouraged. It is usually scientifically inadequate for lack of proper controls and sufficient subjects to generate meaningful results. It is also inadequate as an ethical test because even if lay persons are also enrolled, self-experimentation is neither necessary nor sufficient to establish that they may participate. It is not necessary to establish that lay persons may participate because institutional ethics review and informed consent are better ways to determine this. It is not sufficient because the investigator may be more risk accepting or not medically typical. Moreover, because scientific research is now done in teams, self-experimentation may involve undue influence when junior investigators participate as research subjects.