the point is made especially strongly in respect of the fourth principle, that of consent. Thus: i) Covert coercion: Various proposals have been made for protecting the patient from covert coercion: that consent be requested by a third party (for example, a nurse or doctor not involved in the research); that a relative, or even a 'befriender' or 'advocate', be present; that consent be requested in the patient's home or some other informal environment. Each of these proposals may be helpful in certain circumstances. However, the patient and researcher must eventually come face to face, if not at the outset, then during the course of the research. This is because consent should not be just a one-off event but an on-going state, with the patient retaining the ability to withdraw at any stage. Hence, responsibility for avoiding undue pressures inevitably comes back in the end to the researcher. ii) Disclosure of information: As described above, even though the standards for disclosure of information are higher for research than for ordinary clinical practice, it remains the case that full disclosure is rarely practical. Hence a view must always be taken about how much information is necessary for consent to be valid in a given case. There are various ways in which this might be achieved. An ethics of research committee will always be involved, however, and the committee will be dependent on the information supplied to it by the researcher, as already described. iii) Difficulties arising from specific psychopathologies: The researcher, through his or her specialised knowledge, is likely to be the best person to deal with any difficulties of valid consent arising from the patient's psychopathology. Many of the difficulties of decision and action, especially those related to severely impaired insight, are poorly understood, and there is often little that can be done about them (but see below). Communication difficulties, however, are rather well understood. It is reasonable therefore to expect that the experimental design should include proposals covering both general communication skills (use of appropriate language, repetition, sufficient time for questions and answers, written material for the patient to keep, etc), and also the means for dealing with difficulties arising from particular psychopathologies. The importance of this is not only ethical but also heuristic, since the way in which a patient understands the nature of a research procedure may materially affect the results obtained. Other considerations Important as it is, there is more to the development of good practice in research than a sound 'model' of ethical reasoning. First, there is a need for education. Many medical schools are now providing training in medical ethics and law and in thecopyright. on November 22, 2021 by guest. Protected by http://jme.bmj.com/ J Med Ethics: first published as 10.1136/jme.19.2.85 on 1 June 1993. Downloaded from 90 Ethics of research with psychiatric patients: principles, problems and the primary responsibilities of researchers communication skills that are essential to good practice. These subjects are as important in research as in ordinary clinical care. Then again, there is a need for vehicles for the accumulation of shared experience. The royal colleges could provide this; or perhaps a national research ethics committee, or even an association established on voluntary lines. Above all, though, there is a need for research practice itself to become a subject of research. It will be clear from what has been said in this paper, that if standards of practice in research are to improve, research on research is needed in a number of areas: i) empirical: Covering communication skills, what is helpful and what is not, in general and with specific psychopathologies; ii) ethical: For example, examining what patients, relatives, nurses and other clinical caregivers, actually feel about research. It is generally assumed that research is an imposition. But it can have beneficial effects on outcome (29,30). Moreover, patients may actually welcome the opportunity to take part in research, and thus to help others with problems similar to their own. iii) conceptual: Directed especially towards clarifying the concepts of health and disease. One