Understanding digital intervention engagement: making sense of large-scale data

Abstract

Digital behaviour change interventions provide convenient and personalised health support for users, and the opportunity to record substantial amounts of data about users’ interactions with the intervention. If analysed systematically, these data are able to explain how the intervention was effective, for whom and in what context, leading to recommendations on how the intervention and future dissemination may be improved. However, the volume of data, their complexity and diversity can become a barrier. The aim of this thesis was to devise and apply a method to support analyses of large-scale usage data. The framework for Analysing and Measuring Usage and Engagement Data (AMUsED) was developed to support researchers in establishing clear rationales for collecting usage metrics and undertaking inferential analyses. The framework was applied to usage analyses of two interventions addressing antimicrobial resistance by lowering unnecessary medication prescriptions: Internet Dr encourages self-care for respiratory tract infections thereby reducing avoidable GP visits; PRIMIT/Germ Defence intervention lowers transmission of viruses by increasing handwashing in the home. The process evaluations identified: what type of engagement was successful; specific improvements for the interventions; and the importance of context for using the intervention. Internet Dr findings revealed that the intervention is effective at raising enablement to self-care for users who are not experiencing symptoms, suggesting the structure and theory-based content are relevant for increasing self-care for other minor ailments. The PRIMIT/Germ Defence findings provide insight for public health campaigns by evidencing the value of targeting multiple handwashing situations where risk of transmission is high. A scoping review was carried out to capture how ‘usage’ and ‘engagement’ with digital interventions are conceptualised. The review confirms the need for a practical and generalisable method for usage analyses, and the AMUsED framework is the only method proposed to do this. The process evaluations structured by the framework demonstrate how it supports researchers in conducting analyses of large-scale data, leading to a better understanding of digital intervention engagement, specific recommendations for improvement, and informs our understanding of applying behaviour change theories to a target health issue. These types of findings will lead to more effective digital behaviour change interventions that provide better support for the people who use them.

More information

Identifiers

ORCID for Mark Weal: orcid.org/0000-0001-6251-8786

Catalogue record

Export record