Recommendations on recording harms in randomised controlled trials of behaviour change interventions.

Harms are possible from behaviour change interventions, such as the worsening of a health behaviour intended for change (rebound effect), improving a health behaviour but with subsequent worsening of another behaviour (risk compensation), and participants feeling targeted or stigmatised by an intervention. The processes and definitions originally designed to record harms within drug trials are typically followed to record harms in trials of behaviour change interventions owing to the lack of alternative guidance. Therefore, important harms could be missed in the evaluations of behaviour change interventions or irrelevant harms data may be recorded, leading to inefficiency. This paper presents evidence informed recommendations on how to record harms in randomised controlled trials of behaviour change interventions.