Use this URL to cite or link to this record in EThOS:
Title: Optimising the development of effective mobile health behaviour change interventions : text messages to support smoking cessation in Thailand
Author: Kingkaew, Pritaporn
ISNI:       0000 0004 7660 6738
Awarding Body: University of Leeds
Current Institution: University of Leeds
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Background: Tobacco smoking is recognised as a leading threat to global population health. Rigorous evaluations of mobile health (mHealth) behaviour change interventions for smoking cessation were reported to be mixed due to the diverse and complex nature of these interventions. There is a lack of evidence in guiding intervention designs to maximise effects of mHealth interventions. The Multiphase Optimisation Strategy (MOST) is an approach which aims to optimise and evaluate multicomponent interventions consisting of screening, refining, and confirming phases. It can be applied to develop and test complex interventions. Objectives: 1) to systematically identify effective components of mHealth behaviour change interventions (e.g. behaviour change techniques (BCTs), modes of delivery, functionality) associated with improvements in smoking cessation; 2) to design mHealth behaviour change interventions that contain effective components to support smoking cessation among Thai smokers; and 3) to simultaneously test whether effective components in mHealth behaviour change interventions improve smoking cessation rates among Thai smokers. Methods: For the first objective, a systematic review, a meta-analysis, and a meta-regression of randomised controlled trials (RCT) of mHealth interventions for tobacco cessation were conducted to identify the effect sizes of mHealth interventions and to quantify the association of the characteristics of mHealth interventions with effect size. For the second objective, mobile text messages were designed to provide support for smokers aimed at three theory-based behaviour change components, namely: 'Capability', 'Opportunity', and 'Motivation'. The development involved three steps: 1) selecting BCTs and constructing text messages; 2) testing for the inter-coder reliability of the BCT-enhanced text messages; and 3) validating the acceptability of BCT-enhanced text messages among stakeholders in Thailand using a structured face-to-face focus group discussion. For the third objective, an RCT employing a 2×2×2 full factorial design was conducted to simultaneously assess the effectiveness of the BCT-enhanced text messages for smoking cessation individually, and in combination, among Thai smokers. Effect sizes are presented using odds ratios (OR) and 95% confidence intervals (CI). Kappa's statistic (k) was used to quantify the level of agreement between the two BCT coders. Results: For the first objective, there were 24 mHealth studies identified from the systematic review with the majority being SMS-based interventions. The effect size (OR) of mHealth intervention for smoking cessation was 1.41 (95% CI: 1.19 to 1.67) at 6-months follow-up. From the meta-regression analysis of 23 studies, interventions reported BCTs in the following BCTTv1 domains: 'Feedback and monitoring' (OR 1.39, 95% CI: 1.08 to 1.78), 'Comparison of behaviour' (OR 1.36, 95% CI: 1.12 to 1.65), 'Comparison of outcomes' (OR 1.37, 95% CI: 1.13 to 1.66), and 'Antecedents' (OR 1.29, 95% CI: 1.09 to 1.54), Covert learning' (OR 1.83, 95% CI: 1.21 to 2.75) were associated with an increased odds of smoking cessation. Interventions reported BCTs mapped onto all three theory-based behaviour change components (OR 1.30, 95% CI: 1.05 to 1.59), use theory to inform an intervention (OR 1.51, 95% CI: 1.14 to 1.99), use theory to develop an intervention (OR 1.42, 95% CI: 1.15 to 1.74), and tailoring interventions to participant's needs (OR 1.56, 95% CI: 1.26 to 1.94) were also associated with an increased odds of smoking cessation. These results suggest further research to make efficient, causal conclusions about components as well as about packages of components. For the second objective, text messages were designed based on 39 evidence-based BCTs mapped onto three behaviour change components. Inter-coder reliability for BCT coding suggested that there was a substantial level of agreement (k = 0.78) between the two BCT coders and none of the discrepancies fell into different behaviour change components. However, only 32 BCTs were found to be acceptable among the Thai expert panel involved with tobacco control and was included in the final set of text messages. For the third objective, a total of 1,571 smokers were randomised to receive one of the eight combinations of BCT-enhanced text messages (Placebo, C, O, M, CO, CM, OM, and COM) twice a day for 30 days. 1,260 participants (80%) received all 60 text messages as intended and 94% of the participants reported that they had opened and read the text messages. The overall 7-day self-reported smoking abstinence rate was 40% (n = 521) at 1-month follow-up. Providing BCT-enhanced text messages aimed at supporting smokers' capability to quit (OR 1.20, 95% CI: 0.77 to 1.86), smokers' opportunity to quit (OR 1.05, 95% CI: 0.67 to 1.64), or smokers' motivation to quit (OR 1.13, 95% CI: 0.73 to 1.77) did not significantly improve the 7-day smoking abstinence rate at 1-month follow-up. The additional components of BCT-enhanced text messages (two or more) showed a trend of decreasing the odds of quitting, which suggested an antagonistic interaction effect. Conclusion: This study optimised and evaluated multicomponent mHealth behaviour change interventions in a resource-limited country with emerging mHealth technology. Though a meta-regression suggested a promising result of combinations of BCT-contained mHealth interventions, the interventions failed to provide a significant improvement in cessation rates in a trial setting. Moreover, the addition of two or more behaviour change components decreased the effect size suggesting the importance of experimental studies for decision making. Understanding the effects of these fine-grained behaviour change components rather than a whole set of interventions as a 'black box' will advance knowledge in this field of research using a factorial design.
Supervisor: Glidewell, Liz ; Walwyn, Rebecca ; Wyatt, Jeremy C. Sponsor: Health Promotion Economic Evaluation Collaborative Center ; Thailand ; Health Intervention and Technology Assessment Program (HITAP) ; Thailand
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available