RANDOMISED CLINICAL TRIALS – MAIN TECHNIQUE
The main technique used in this endeavour is the randomised clinical trial. Here doctors select a fairly uniform group of people—all within a certain age range, with similar types and extent of cancer and about the same degree of physical fitness. To further reduce any possible bias, the people in this already fairly uniform group are allotted randomly (by chance) to one or other of the treatments to be compared. This could be done, say, with the flip of a coin: heads = treatment X, tails = treatment Y. Scientifically-trained doctors believe that such random allocation of people to treatment is the best way of making sure that any differences in results are due to the different treatments and not to any other factor. As I have explained, such techniques are only necessary if we are looking for small differences. For example, trials often include hundreds of people in an attempt to make sure that they do not miss differences of five per cent (one in twenty) in remission rates or of a few weeks or months in average length of life—’statistically significant’ differences. Attention is completely concentrated on ways of expressing results that can be accurately measured and subjected to statistical analysis— things like remission rate and length of life.
*141/40/1*
