Shedding light on research participation effects in behaviour change trials: a qualitative study examining research...

The sequence of events in a behaviour change trial involves interactions between research participants and the trial process. Taking part in such a study has the potential to influence the behaviour of the participant, and if it does, this can engender bias in trial outcomes. Since participants' experience has received scant attention, the aim of this study is thus to generate hypotheses about which aspects of the conduct of behaviour change trials might matter most to participants, and thus have potential to alter subsequent behaviours and bias trial outcomes

Twenty participants were opportunistically screened for a health compromising behaviour (unhealthy diet, lack of exercise, smoking or alcohol consumption) and recruited if eligible. Semi structured face to face interviews were conducted, after going through the usual processes involved in trial recruitment, baseline assessment and randomisation. Participants were given information on the contents of an intervention or control condition in a behaviour change trial, which was not actually implemented. Three months later they returned to reflect on these experiences and whether they had any effect on their behaviour during the intervening period. Data from the latter interview were analysed thematically using a modified grounded theory approach.

The early processes of trial participation raised awareness of unhealthy behaviours, although most reported having had only fleeting intentions to change their behaviour as a result of taking part in this study, in the absence of interventions. However, careful examination of the accounts revealed evidence of subtle research participation effects, which varied according to the health behaviour, and its perceived social acceptability. Participants' relationships with the research study were viewed as somewhat important in stimulating thinking about whether and how to make lifestyle changes.

These participants described no dramatic impacts attributable to taking part in this study. This study demonstrates the likely value of well conducted qualitative studies of subtle research participation effects, which may be particularly important to explore for alcohol. Separating unintended influences in trial participation from the effects of behaviour change interventions being evaluated therein is necessary for valid estimates of intervention effects.