Effects of Study Design and Allocation on participant behaviour - ESDA: study protocol for a randomized controlled trial

A three-arm parallel group randomised controlled trial design will be used. All recruitment, screening, randomisation, and follow-up will be conducted on-line among university students. Participants who indicate a hazardous level of alcohol consumption will be randomly assigned to one of three groups. Group A will be informed their drinking will be assessed at baseline and again in one month (as in a cohort study design). Group B will be told the study is an intervention trial and they are in the control group. Group C will be told the study is an intervention trial and they are in the intervention group. All will receive exactly the same brief educational material to read. After one month, alcohol intake for the past 4 weeks will be assessed.The experimental manipulations address subtle and previously unexplored ways in which participant behaviour may be unwittingly influenced by standard practice in trials. Given the necessity of relying on self-reported outcome, it will not be possible to distinguish true behaviour change from reporting artefact. This does not matter in the present study, as any effects of awareness of study design or allocation involve bias that is not well understood. There has been little research on awareness effects, and our outcomes will provide an indication of the possible value of further studies of this type and inform hypothesis generation.Australia and New Zealand Clinical Trials Register (ANZCTR): ACTRN12610000846022The "Hawthorne effect", also known as reactivity, refers to the possibility that people may change their behaviour simply because they know they are participating in a study [1]. The name derives from studies done in the workplace at the Western Electric Plant in Hawthorne, Illinois from 1924-32. It is well accepted that there is unintended reactivity by participants in randomised controlled trials [2-4]. Placebo control conditions have been developed in pharmacological trials and elsewhere to control for the effects of d