Editor's Choice | This Week in BMJ | Press releases
BMJ No 7102 Volume 315 Letters Saturday 26 July 1997 Informed consentPatients' knowledge that they are participating in trial may not bias resultsEditor,In his commentary on his and colleagues' study Martin Dennis puts forward arguments as to why patients in a trial of the effect of a stroke family care worker were not asked for their consent before being entered into the trial.(1) We question whether seeking consent would necessarily have biased the results. We are currently involved in a randomised controlled trial in Oxford of a family support organiser for patients with stroke and their families. As in Edinburgh, patients are randomised in our study before consent has been obtained. At the time of randomisation, however, we write to the closest carers of all patients, inviting them to take part. The letter explains the purpose of the study and that whether or not the carers see the family support organiser will be determined by chance. Altogether, 18 of the 179 families contacted so far have elected not to take part, either in response to the letter or when contacted by a researcher six months after the stroke. The proportion of families not taking part is the same (10%) in both the intervention and control group. At the follow up visit the researcher does not specifically remind patients or their family that the purpose of interviewing them is to evaluate the possible effects of a family support organiser. Our study was approved by the Central Oxford Research Ethics Committee. The theoretical concern is that patients and their families will realise which group they are in and that this might influence the results.(1) Our experience, however, suggests that patients and families do not discriminate between different community services. Prompted by the issues raised by Dennis, we decided to do a limited interim analysis of our (ongoing) study. At the end of the follow up interview the researcher asks what services have been received since the stroke. Only eight of 80 families in the intervention group have mentioned the family support organiser at this time. The researcher then records which of the two groups she thinks the family was in. She has so far guessed correctly for 102 (64%) out of 159 families (95% confidence interval 57% to 72%). While higher than would be expected by chance, this is not significantly more than the 59% recorded by the researcher in the Edinburgh study. This provides circumstantial evidence that many families were effectively blind to their treatment allocation. Therefore, it would seem that consent can be obtained in trials of this sort without compromising the validity of the results. Jonathan Mant
Simon Winner
Judy Carter Derick T Wade Funding: The Oxford family support organiser trial is supported by the Stroke Association. References
1 Dennis M, O'Rourke S, Slattery J, Staniforth T, Warlow C.
Evaluation of a stroke family care worker: results of a r
Home | Current issue | Past issues | Classified ads | Career Focus | Feedback Collections | About this site | About the BMJ | BMA | Medline
|