Designs to Deal with Participant Preference

Abstract

Randomisation equalises characteristics of participants between the two or more randomised groups, but it does not deal with people’s hopes and treatment expectations. In open trials the participant is informed of treatment assignment, which can lead to psychological expectations and introduce bias. A key source of this potential bias within a trial is the role of participants’ preferences. When presented with different interventions many participants will ‘prefer’ one of the alternatives. These preferences can introduce bias if one of the treatment options is only available within the context of randomised trial. If, for example, 50 per cent of those being recruited to a trial of A versus B prefer intervention A, and this is only available within the trial, then they are likely to consent to participate. However, half of those 50 per cent of participants preferring A will be disappointed as they will be allocated to treatment B. If participants are randomised to an intervention that they do not want they may consciously or unconsciously perform less well in the outcome measures, in the knowledge that there was a preferable alternative.