Clinical Trials Past, Present and Future

The modern concept of testing new treatments in randomised control trials began in 1948 when the Medical Research Council in the UK reported the results of a trial of streptomycin for pulmonary tuberculosis which had been organised by Drs D’Arcy and Daniels. However, even before this trial started in 1946, another randomised trial had begun to assess the effect of immunisation for whooping cough. In both studies, subjects were randomly allocated to the two arms of the trial and every attempt was made to ‘blind’ both the subject and the Physician so that assessment could be entirely objective. Even in the early years, there was considerable debate about the use of a placebo. The concept of randomisation was not new and was described in 1662 by van Helmont who advocated allocation to treatment groups by tossing a coin but it appears not to have been adopted into clinical practice. The idea lay largely dormant until R.A. Fisher began a series of experiments on agricultural crops in 1926 using methods of randomisation. The introduction of such methods back into clinical medicine was due to Bradford Hill who initially advocated in 1937 that treatments should be allocated to alternative patients but then in 1946 recommended randomised allocation to allow random error to be calculated and to avoid bias in selection.