Effective behaviour change techniques to promote physical activity in inactive adults: systematic review and meta-analysis

Abstract

Background: Low physical activity levels are predictive of poor health outcomes. Building an evidence base of the best approaches to change the behaviour of inactive individuals is crucial in reducing mortality and morbidity. Methods: Seventeen databases were searched from 1990-2014. Studies were restricted to randomised controlled trials aiming to increase physical activity and/or reduce sedentary behaviour in inactive adults without a chronic condition, with a minimum of six months post-intervention follow-up. Risk of bias, ‘Template for Intervention Description and Replication’ (TiDieR) guidelines, and behaviour change techniques (BCT) were coded independently by two reviewers. A meta-analysis pooled continuous data for key outcomes. Meta-regression and subgroup analyses analysed how BCT presence impacted effectiveness. Findings: Twenty five studies were included, with 16 pooled for meta-analysis. Physical activity interventions were effective in the short term and long term. Two studies measuring changes in sedentary behaviour were not effective in the short or long term. In the short term, the BCTs ‘Biofeedback’, ‘Demonstration of the behaviour’, ‘Behaviour practice/rehearsal’, and ‘Graded tasks’ were associated with increased effectiveness. In the long term, interventions were more effective if they included ‘Action planning’, ‘Instruction on how to perform the behaviour’, ‘Prompts/cues’, ‘Behaviour practice/rehearsal’, ‘Graded tasks’, and ‘Self-reward’. Discussions: Good quality evidence was found for the lasting effects of physical activity interventions, but not for interventions targeting sitting time. Closer adherence to the TIDieR guidelines would be beneficial for the replication of successful interventions. The effectiveness of a number of BCTs provides intervention designers with techniques to target inactive adults.