A Mixed-Method Exploration of Behaviour Change Techniques for Children Aged 5 to 15 Years Living with Overweight or Obesity

Background: High prevalence of overweight and obesity in children and adolescents and the associated health risks are a global concern. Behaviour change interventions have been investigated in many studies to address excess weight and to improve health. These use a range of structured methods to change behaviour, which are referred to as behaviour change techniques (BCTs) and include components, e.g., goal setting. As behaviour change interventions are often complex and interrelated, a common language is needed to optimise communication and to effectively evaluate and replicate interventions. This has been done through the development of a series of taxonomies, and the most comprehensive and widely used is known as BCT Taxonomy version 1 (BCTTv1), which defines 93 individual BCTs. This taxonomy facilitates the systematic review and meta-analysis of behaviour change interventions in adults and has the potential to be used in evaluating studies in children living with overweight or obesity. This thesis uses BCTTv1 for the first time to examine BCTs used in weight management programmes in children both in ‘research’ and ‘real-life’ settings. Aim: To identify the use of BCTs and effectiveness of published interventions and in one family weight management programme in a real-life setting for children aged 5-15 years living with overweight or obesity in England. Methods: A four-phase mixed method approach and the BCTTv1 were used across ‘research’ and ‘real-life’ settings: (1) a systematic review and meta-analysis of randomised controlled trials (RCT) of behaviour change interventions for managing overweight and obesity in children aged 5-15 years following PRISMA guidelines; (2) a qualitative exploration of BCTs used in an intervention described by ten staff who delivered a family weight management programme in a public health setting in England using audio-recorded semi-structured interviews and analysed using thematic analysis using a codebook; (3) a quantitative comparison of BMI z-score before and after an intervention using a database provided from the same family weight management programme using SPSS statistical software; (4) a qualitative exploration of the perspectives of two families who participated in the family weight management programme using audio-recorded semi-structured interviews and analysed using thematic analysis. Results: (1) Four studies met the systematic review inclusion criteria and, in these research settings, twenty-eight different BCTs were used across the studies (range 12-19 BCTs per study). Meta-analysis showed that the interventions were effective in improving BMI z-score at six months after intervention finished compared to control group, with an overall standardized mean difference -0.27 (95% CI = -0.52 to -0.02, I2=0%, p=0.03). The most frequently used BCTs were goal setting (behaviour), problem solving, self-monitoring of behaviour, social support (unspecified) and demonstration of behaviour. However, the four RCTs were identified as either being at high risk of bias or of some concern of bias and the evidence was rated as having low or very low certainty. (2) In a real-life setting, twenty-four BCTs were described as used in the weight management intervention and two groups of BCTs were often used together, i.e., goal setting (behaviour) group and instruction on how to perform behaviour group. The programme was delivered face-to-face to families with children aged 5-15 years by UK-registered nutritionists and sport coaches over twelve weeks with one two-hour session each week and online support between sessions. (3) The programme was associated with improved BMI z-score in children living with overweight and obesity, with a median (range) reduction in BMI z-score -0.08 (-1.13 to 0.71, p<0.001. (4) The two families who participated in the programme and in semistructured interviews described their satisfaction in the programme and valued the practical ways in which they were supported to change their families’ behaviour in a non-judgmental manner. Conclusion: There were similarities and differences in BCTs used in the ‘research’ and ‘real-life’ settings. The ‘real-life’ setting focused more on BCTs relating to the behaviour itself and less on those relating to the measurable outcomes of the behaviour which were used more in ‘research’ settings. Using BCTs in combination were effective in helping to improve overall BMI z-scores both in the ‘research’ and ‘real-life’ settings. However, more research is needed to confirm which combination of BCTs is most effective in helping children living with overweight or obesity to change their behaviour toward a healthy lifestyle.