Ethics in psychiatric practice hinges on the physician's obligation to the patient.
In psychiatry this brings up a real puzzle -- we imagine the psychiatrist as treating 'the patient", however, in psychiatry, the "patient" as an agency or person, is often in a transitional frame of mind.
Moreover, such a question raises the difficult issue of the capacity of the patient to plan clearly for the furture.
Are we asking our patients too much by asking them to plan for the furture ? Should we expect them to have that capacity.
Instead, I would argue that ethical reasoning should be based upon the tendency of the capacity that we imagine being fully restored to the patient, and their subsequent wishes.
The most serious difficulty in establishing an approach to ethics in practice in psychiatry (EPP) is the problem is the defective nomenclature used.
In order to make sense of problems in EPP, our first task is to examine commonly used anguagem and determine if there is a common understanding of meaning.
In fact, that is one of the deficiencies of any approach to diagnostics that relies on an inadequantley defined nomenclatus, such as the DSM approach.
The problem with terminology begins at the very outset. The term "behavorial health" to categorize patients in a psychiatric setting, is typical of such terminology. Similarly the term "mental health" presents problems. "Behavioral" implies that "behavior" is a separate entity from the enitre domain of physiologic phenmena in some undefined but distinct way - it uses a term that adds no information and instead, provides the peril that mutiple disucssants will try to reach a conclusion without agreeing on the terms of their discussion.
In trying to make sense of EPP, we have to consider a few questions that will occur again and again. What is the obiligation towards the patient ? Is there anything that overrides that obligation ? How do we go about quantifying risks ? The same question, put differently - how do we establish the value of different clinical descisions ? How do we correct the calculus of risks with the liklihood of available treatments ? How we work with or utilize estimates of uncertainty about risk ?
That is perhaps one core argument I will make, which is that ethics in psychiatry must be quantitatively based, and that such a quantitiatve basis cannot be established without reasonable numerical models.