Identifying active ingredients in complex behavioural interventions for obese adults with obesity-related co-morbidities or additional risk factors for co-morbidities: a systematic review

Abstract

Reducing obesity is an important preventive strategy for people who are at increased risk of major disabling or life-threatening conditions. Behavioural treatments for obesity are complex and involve several components aiming to facilitate behaviour change. Systematic reviews need to assess the components that moderate intervention effects.

Electronic databases and journals were searched for randomised controlled trials of behavioural interventions targeting dietary and/or physical activity change for obese adults (mean BMI≥30, mean age≥40 years) with risk factors and follow-up data≥12 weeks. A reliable taxonomy of theory-congruent behaviour change techniques (BCTs; Abraham & Michie, 2008) was used to identify programme components.

Meta-regression suggested that increasing numbers of identified BCTs are not necessarily associated with better outcomes. The BCTs provision of instructions (β =− 2.69, p=0.02), self-monitoring (β = − 3.37, p<0.001), relapse prevention (β = − 2.63, p=0.02) and prompting practice (β = − 3.63, p<0.001) could be linked to more successful interventions. Studies including more BCTs aimed at dietary change that are congruent with Control Theory were associated with greater weight loss (β = − 1.13, p=0.04).

Post-hoc ratings of intervention components in published trials can lead to the identification of components and theories for behaviour change practice and research.

Keywords:

• systematic review
• obesity
• behavioural interventions
• behaviour change techniques
• meta-regression

Notes

1. Theory congruence refers to the link between BCTs and a particular theory of behaviour or behaviour change. Note that the concept of ‘theory congruence’ is thus less restrictive than the concept of ‘theory base’ as it does not necessarily assume a theory as the underlying basis for intervention development, but still notes overlap between the design of an intervention and the mechanisms of action as specified by a theory. There are different ways in which one can determine theory congruence between BCTs including previously established evidence, using expert agreement methods (e.g., Michie, Johnston, Francis, Hardeman, & Eccles, 2008) or creative linkage (e.g., Ajzen, 2006). For the purpose of the current paper, theory congruence is defined by the linkage between BCTs and theories as published in the Abraham and Michie (2008) taxonomy.

2. Included studies marked with * in the reference section.