A mixed methods study to explore the effects of program design elements and participant characteristics on parents' engagement with an mHealth program to promote healthy infant feeding: the growing healthy program

Copyright © 2019 Taki, Russell, Lymer, Laws, Campbell, Appleton, Ong and Denney-Wilson. Purpose: Mobile health (mHealth) interventions have great potential to promote health. To increase consumer engagement in mHealth interventions it is necessary to address factors that influence the target demographic. The Growing healthy (GH) program is the first obesity prevention program delivered via a smartphone app and website offering evidence-based information on infant feeding from birth until 9 months of age. This sub-study aimed to explore how the design features, quality of the app and participant characteristics influenced parents' engagement with the GH app. Methods: A sequential mixed methods design was used. The GH app participants (225/301) were considered for this sub-study. Participant app engagement was measured through a purpose-built Engagement Index (EI) using app metrics. Participants were categorized as low, moderately or highly engaged based on their EI score upon completing the 9 months program and were then invited to participate in semi-structured telephone interviews. Participants who used the app program, given an EI score and expressed interest to participate in these interviews were eligible. The interviews explored factors that influenced app engagement including delivery features and quality. Thematic analysis networks was used for analysis. Results: 108/225 expressed interest and 18 interviews were conducted from low (n = 3), moderately (n = 7), or highly (n = 8) engaged participants based on purposeful sampling. Participants defined as highly engaged were likely to be a first-time parent, felt the app content to be trustworthy and the app design facilitated easy navigation and regularly opened the push notifications. Participants defined as having low or moderate engagement were likely to have experience from previous children, felt they had sufficient knowledge on infant feeding and the app did not provide further information, or experienced technological issues including app dysfunction due to system upgrades. Conclusions/Implications: This study demonstrated a novel approach to comprehensively analyse engagement in an mHealth intervention through quantitative (Engagement Index) and qualitative (interviews) methods. It provides an insight on maximizing data collected from these programs for measuring effectiveness and to understand users of various engagement levels interaction with program features. Measuring this can determine efficacy and refine programs to meet user requirements.