Exploring Persuasive System Design Features and Users’ Perceptions of an mHealth Breastfeeding App: A Mixed-Method Study

Abstract

The integration of persuasive technology in healthcare has gained considerable momentum through the proliferation of mobile health applications (mHealth apps). These apps become instrumental tools for providing tailored health support systems, facilitating informed decision-making, and promoting positive behaviours among users in increasingly persuasive ways. Among the various applications of mHealth apps, breastfeeding apps have emerged as invaluable resources for breastfeeding mothers. These apps can offer interactive information, tailored guidance, and support networks. Thus, incorporating persuasive features within these apps would further enhance their efficiency in promoting and supporting breastfeeding practices. The persuasive systems design (PSD) model represents a conceptual framework that informs persuasive technology design. The PSD model highlights four principles for persuasive system content and functionality and provides valuable insights into the key features that can be employed to design persuasive systems that are efficient and impactful. However, despite the prevalence of persuasive features in various mHealth apps, there is a limited understanding of how the intended users perceive these features. Research has primarily focused on the technical aspects of designing and implementing persuasive software features with limited empirical evidence on users' subjective experiences and perspectives regarding these features. This thesis comprises five studies, including quantitative and qualitative, that are thoroughly related and successively build upon each other to jointly achieve the central research aim: an in-depth exploration of user perceptions and experiences of persuasive design principles embedded in a mHealth app designed to support breastfeeding. Firstly, a systematic review was conducted to explore the characteristics of technology-based breastfeeding interventions and their effects on breastfeeding outcomes. Given the rising integration of technology with breastfeeding practices, it's crucial to examine these interventions' design features and impacts on breastfeeding outcomes. The study identified that the most effective features were educational content and personalised support via online forums. These aspects align with the PSD model's primary task support, dialogue support, social support, and credibility support principles. These results support the view that the Milky Way mHealth breastfeeding app would be a feasible tool to support and promote breastfeeding. Therefore, this research utilised the Milky Way app as a novel persuasive approach informed by the PSD model. The app incorporates various design features such as comprehensive information collection, interactive feedback, and social connectivity, all aimed at educating and supporting breastfeeding women. The PSD model comprises software features that have the potential to enhance the persuasive capabilities of mHealth apps; however, there is a limited understanding of how users perceive and experience these features. Recognising the importance of measuring users' perceptions of persuasion, the second study in this thesis aimed to develop and validate an instrument for measuring user perceptions of persuasive design principles in the Milky Way breastfeeding mHealth app. The instrument was designed following best practices and comprised 24 items. A Modified Delphi Study involved experts from academic and clinical backgrounds to assess the items' relevance and clarity. The instrument demonstrated high content validity (CVI), with most items scoring excellent (I-CVI ≥ 0.78), except one. The overall content validity index was moderate to high (S-CVI/Ave was 0.89 and 0.92 for relevance and clarity, respectively), confirming the instrument's validity, relevance, and clarity. In the third study of this thesis, a cross-sectional analysis was conducted on 168 women using convenience sampling to assess the instrument's validity and reliability. Statistical methods, such as Exploratory Factor Analysis (EFA), were employed to investigate the relationship between the PSD categories and app features. Construct validity was assessed with EFA, and reliability was measured using Cronbach's alpha. Results identified a four-factor solution with 23 items explaining 62.71% of the total variance. The instrument showed strong discriminant validity and high internal consistency (Cronbach's alpha α > 0.8). No statistically significant differences in participant responses based on demographic characteristics were found, which suggests that the instrument is applicable across diverse user groups. The study provided valuable insights into the multidimensional nature of the persuasive design principles within the instrument. The fourth study further explored the persuasive aspects of the Milky Way mHealth app's design, functionality, and usability based on 168 women's feedback via an online survey. Quantitative data showed significant positive perceptions of the app's persuasive design features (One-Sample t-Test, p < .001). A qualitative analysis of 288 responses indicated that the app is informative, credible, and supportive of breastfeeding. Additionally, practical design recommendations were suggested. The fifth and final study of this thesis was a qualitative follow-up study consisting of individual semi-structured interviews. These interviews were conducted to gain a comprehensive understanding by empirically exploring participants' experiences with the Milky Way mHealth app's persuasive features. Furthermore, the findings were analysed within the value-sensitive design (VSD) framework to align them with users' values and ethical considerations. Results highlighted the app's reliable educational content and supportive features, with tailored support and evidence- based advice being perceived as significant. A deductive analysis showed a strong alignment between user values and the app's design within the VSD framework. The research findings emphasise the significance of including evidence-based content, tailored support, and social connectivity as persuasive features in mHealth apps. These well-integrated PSD features greatly enhance user experience. The developed research instrument contributes to the advancement of research in the field of persuasive technology and its application in healthcare settings. It addresses a gap in understanding user perceptions of persuasive features. It provides a valuable instrument in establishing the foundation for future studies aiming to explore the persuasive capabilities of mHealth apps or similar technologies and their potential impact on promoting sustainable behaviour change. This research underscores the role of persuasive apps in health behaviour promotion. It also extends the empirical exploration of critical issues related to persuasive systems, specifically examining the consideration of ethical values in designing persuasive systems. The insights contribute to advancing persuasive technology and provide guidance for designing impactful, persuasive health technology aimed at promoting and supporting healthy behaviours. These findings have broader implications for the field of persuasive technology in providing valuable insights for designing useful and user-focused persuasive health technology. They inform researchers and practitioners in developing technologies that efficiently promote and support positive health behaviour change. In the broader context, the study contributes to advancing the understanding and application of persuasive design principles by highlighting the important role of user perceptions of software features in facilitating meaningful behaviour promotion and enhancing the overall user experience in healthcare contexts.