Sir: Polly Toynbee ("No one really wins in this life-and-death lottery", 29 May) makes the point that randomised clinical trials are necessary if medicine is to continue to introduce new and effective treatments.

However, to suggest that clinicians involved in the Medical Research Council or other trials withhold advice to patients or coerce them into trials simply for "scientific ends" misrepresents the reason for most clinical trials. This fundamental purpose is to establish what works for patients. Your readers should not be left with the impression that clinical trials pose inherent conflict between the interests of science and those of patient treatment.

Trials are only undertaken where there is genuine uncertainty about the best treatment, and where there is a clear prospect that the trial will help resolve that uncertainty. We will not approve any trial where we consider that the potential risks of the treatment outweigh the potential benefits. For this reason we did not sponsor the trial of the preventative use of taxomixfen in healthy women at risk of breast cancer to which Polly Toynbee refers.

The MRC puts considerable effort into making sure that patients are able to make an informed decision before entering one of our trials. At the end of the day the decision always rests with the patient, as it should. But if refusal to participate on any scale were to develop - and happily this is not the case at present - we would be very concerned that our main tool for improving treatments and practice would be jeopardised.

LIAM O'TOOLE

Clinical Trials Manager

Medical Research Council

London W1