Editor's Choice | This Week in BMJ | Press releases

BMJ No 7102 Volume 315 Letters Saturday 26 July 1997 Informed consent Trials that use Zelen's procedure should be acceptable Editor, In his editorial on the ethics of obtaining consent in trials, Richard Smith describes the Edinburgh evaluation of family stroke care workers as one `in which informed consent was not sought,' a description taken up in the lay press.(1, 2) In fact, Martin Dennis and colleagues make it clear that they did seek consent, using a variant of Zelen's procedure, the single randomised consent design (figure).(3) In her commentary criticising the Edinburgh trial Sheila McLean says: `Anyone who is to be studied must be given the fullest possible information.' (2) Len Doyal seems to agree, using the terms informed consent and fully informed consent interchangeably.(4) What is fully informed consent? Does it include details of all the evidence justifying the mounting of a trial and details of the financing of the study, how the sample size was derived, and the methods that will be used to analyse the results? If taken literally, the idea is absurd; consent can never be fully informed. In any case, an attempt at implementation-that is, at ensuring that everybody knows everything-would defeat its purpose. To paraphrase Zelen, what we want is not fully informed subjects but fully understanding ones. We must choose what information to impart, or we only confuse. It is adequately informed consent that is the hallmark of ethical research. In decisions of what constitutes adequately informed consent, the conflict is not simply between researchers' convenience and the moral rights of subjects. Insisting on consent to randomisation in pursuit of one ethical aim may lead to the conduct of an unethical trial for another reason. As willingness to consent to randomisation is a psychological characteristic it may be associated with other characteristics that themselves determine the outcome of treatment. This applies particularly to trials of psychosocial interventions. If refusals are substantial but the trial is completed, sampling bias will be large but uninterpretable. Research that is useless or yields misleading results because of design faults is unethical, just as much as inadequately informed consent is. It does not help the ethical argument to talk about informed consent, fully informed consent, and consent to randomisation as if they were the same thing. Nor does it help to argue that seeking consent to randomisation is always ethical while not doing so is simply self serving. A good case has not been made for obligatory adherence to consent to randomisation, and until it has, the BMJ should continue to publish trials that use Zelen's procedure. Allan House* Consultant Department of Liaison Psychiatry, Leeds General Infirmary, Leeds LS1 3EX Peter Knapp* Research psychologist Research School of Medicine, University of Leeds, Leeds LS2 9JT *The authors are conducting a trial of psychological intervention after stroke, funded by the NHS research and development programme, that uses a randomised consent design. References 1 Smith R. Informed consent: the intricacies. BMJ l997;314:1059-60. (12 April.) 2 Dennis M, O'Rourke S, Slattery J, Staniforth T, Warlow C. Evaluation of a stroke family care worker: results of a randomised controlled trial. [With commentaries by S McLean and M Dennis.] BMJ 1997;314:1071-7. (12 April.) 3 Zelen M. Randomised consent designs for clinical trials: an update. Stat Med 1990;9:645-56. 4 Doyal L. Journals should not publish research to which patients have not given fully informed consent-with three exceptions. BMJ 1997;314:1107-11. (12 April.) Home | Current issue | Past issues | Classified ads | Career Focus | Feedback Collections | About this site | About the BMJ | BMA | Medline