Critical care decisions in fetal and neonatal medicine: ethical issues
Appendix 6: Randomised controlled trials in clinical medicine
Doctors often must decide between a new treatment or procedure which may or may not be superior compared with the standard one. It is rarely obvious that one treatment is very much better than another. It is also known that different individuals respond to the same treatment differently (for example males and females may have different responses). In order to determine whether a treatment is better than the standard treatment (that is the one usually given) a process known as a randomised controlled trial (RCT) is used.
The aim of a randomised trial is to compare outcomes for two groups of patients, one of which receives the new treatment, and the other the standard treatment. Each group is sufficiently large to account for the differences between individuals and to demonstrate the desired effect of the treatment, if it is successful. The process of random allocation to the standard treatment or a new treatment is done to ensure that each group will randomly contain individuals with different responses (for example equal numbers of males and females) so that the treatment can be used in the future across all patients.
The process of deciding which treatment an individual gets is done by a sophisticated technique of randomisation, which in essence is like tossing a coin: that is, which side ends upwards is a matter of chance. A doctor can only enter a patient into a trial if the patient (or parent) gives permission for the doctor to do so, except in very special circumstances where permission would be impossible, for example during resuscitation when the patient is unconscious.
All studies are reviewed by members of a research ethics committee who must satisfy themselves that the study is well designed, that it will contain enough patients to be able to detect a potential difference between treatments and that the study procedure does not place undue pressure on patients to consent to the trial. Patients are always free to withdraw from the trial at any time. Most trials are also externally monitored to ensure they are not producing unexpected results, in which case the trial should be curtailed.
This system of using randomised trials is thought to be fair and allows medicine to move forward in the confidence that its treatments are effective. People almost always make the assumption that any ‘new’ treatment is necessarily better, but in many trials the differences may be less than expected and in some trials unexpected adverse effects occur. A patient cannot request or insist on a new treatment, which is usually only available within the context of the randomised trial, until it is effective and safe. Once this has been established, the medicine is normally licensed by the appropriate authorities.
Cochrane Reviews are analyses of randomised trials and other relevant evidence in certain areas of medicine that aim to determine the effectiveness and appropriateness of a given treatment. These are published in the Cochrane Database of Systematic Reviews (www.cochrane.org); there are currently over 190 reviews in the library relating to trials carried out on babies.