An investigation of minimisation criteria

We developed an automated package for patient allocation which incorporated a simulation arm. We here demonstrate how simulation of data can help to determine the input parameters to be used in a subsequent application of minimisation.Several scenarios were simulated. Within the selected scenarios, increasing the number of factors did not substantially adversely affect the extent to which the treatment groups were balanced with respect to the prognostic factors. Weighting of the factors tended to improve the balance when factors had many categories with only a slight negative effect on the factors with fewer categories. When interactions between factors were included as minimisation factors, there was no major reduction in the balance overall.With the advent of widely available computing facilities, researchers can be better equipped to implement minimisation as a means of patient allocation. Simulations prior to study commencement can assist in the choice of minimisation parameters and can be used to justify those selections.Most medical researchers are aware that it is necessary to perform a randomised controlled trial to effectively establish the usefulness of a new treatment. The aim is that treatments are compared on similar groups of patients. Completely random allocation of patients to treatments does not, however, ensure that the patient groups are similar with respect to prognostic factors. For example, purely by chance one of the treatment groups may have been allocated older or more severely ill patients. If such an imbalance in prognostic factors has occurred then it may be difficult to attribute any differences to treatment, the analyses will require adjustment and the study will have less power.Minimisation [1-3] is a dynamic allocation procedure that ensures treatment groups are similar with respect to a series of pre-specified prognostic factors. As patients are recruited to the trial they are allocated to the treatment group that will 'minimise' the