Fidelity of a very brief pedometer-based intervention to promote physical activity in preventative health checks

Abstract

Background: We conducted a randomised controlled trial of a very brief (<5 minutes) pedometer-based intervention for physical activity (‘Step It Up’) targeting adults aged 40-74 years attending health checks in primary care, and found no evidence of a positive effect on physical activity. The aim of this study was to assess intervention fidelity to illuminate the trial findings. Methods: Control participants received the usual health check, and intervention participants received the health check plus ‘Step it Up’. A sample of five consultations per practice was randomly selected for audio-recording. A standardised form was used to assess intervention duration, delivery intervention components and any contamination. Findings: Audio-recordings of 37 intervention and 26 control consultations were obtained from 13/23 practices. Preliminary findings show that intervention duration ranged between 16 seconds and 7 minutes. Fidelity of intervention delivery was 60% in the intervention sessions, meaning that on average 9 out of 15 intervention components (e.g. feedback) were delivered. Fidelity varied considerably between practitioners (range 20% to 87%). Contamination in control consultations was minimal. Discussion: Fidelity of delivery was moderate and varied widely among practitioners, whilst contamination was minimal. Mean intervention duration was less than five minutes. Obtaining audio-recordings in a pragmatic trial in primary care was challenging, and we do not know whether the sample of recordings was representative in terms of fidelity levels. The findings suggest that the training of practitioners in very brief physical activity advice can be improved.