Physical activity measurements affected participants' behavior in a randomized controlled trial

Background and ObjectiveAssessing levels and determinants of physical activity as outcome measurements might have an independent effect on participant's physical activity behavior. The objective is to study this effect in a randomized controlled trial (RCT) promoting regular physical activity in Dutch general practice.MethodsUsing a Solomon four-group design, participants were randomized twice. After randomization to a control or intervention-condition at general practice level (N = 29), participants were randomized to a group participating in measurements at baseline, 2 and 6 months (3M-group, N = 361), or a group only participating in measurements at 6 months (1M-group, N = 356). Outcome measures assessed at 6 months included: level of physical activity (self-reported and objectively measured with accelerometry), meeting ACSM/CDC guideline for regular physical activity, stage of change, and determinants of physical activity.ResultsFollow-up data on 635 participants (89%) was collected. Statistically significant measurement effects were found for meeting the ACSM/CDC guideline (self-reported), self-efficacy for resisting relapse, knowledge, and on awareness. Other outcome measures showed positive trends, except stages of change.ConclusionMeasurements of physical activity affect participant's physical activity behavior, possibly triggered by a raised awareness about their own physical activity level. Implications for future research are discussed, as well as methodologic limitations of the study design.