This article describes three kinds of information constraints in medical encounters that have not been discussed at length in the medical ethics literature: constraints from the concept of a disease, from the diffusion of medical innovation, and from withholding information. It describes how these limit the reliance rational people can justifiably put in their doctors, and even the reliance doctors can have on their own advice. It notes the implications of these constraints for the value of informed consent, identifies several procedural steps that could increase the value of the latter and improve diffusion of innovation, and argues that recognition of these constraints should lead us to devise protections which intrude on but can improve these encounters.
