Factors Influencing the Use of a Web-Based Application for Supporting the Self-Care of Patients with Type 2 Diabetes: A Longitudinal Study

BACKGROUND: The take-up of eHealth applications in general is still rather low and user attrition is often high. Only limited information is available about the use of eHealth technologies among specific patient groups. OBJECTIVE: The aim of this study was to explore the factors that influence the initial and long-term use of a Web-based application (DiabetesCoach) for supporting the self-care of patients with type 2 diabetes. METHODS: A mixed-methods research design was used for a process analysis of the actual usage of the Web application over a 2-year period and to identify user profiles. Research instruments included log files, interviews, usability tests, and a survey. RESULTS: The DiabetesCoach was predominantly used for interactive features like online monitoring, personal data, and patient–nurse email contact. It was the continuous, personal feedback that particularly appealed to the patients; they felt more closely monitored by their nurse and encouraged to play a more active role in self-managing their disease. Despite the positive outcomes, usage of the Web application was hindered by low enrollment and nonusage attrition. The main barrier to enrollment had to do with a lack of access to the Internet (146/226, 65%). Although 68% (34/50) of the enrollees were continuous users, of whom 32% (16/50) could be defined as hardcore users (highly active), the remaining 32% (16/50) did not continue using the Web application for the full duration of the study period. Barriers to long-term use were primarily due to poor user-friendliness of the Web application (the absence of “push” factors or reminders) and selection of the “wrong” users; the well-regulated patients were not the ones who could benefit the most from system use because of a ceiling effect. Patients with a greater need for care seemed to be more engaged in long-term use; highly active users were significantly more often medication users than low/inactive users (P = .005) and had a longer diabetes duration (P = .03). CONCLUSION: Innovations in health care will diffuse more rapidly when technology is employed that is simple to use and has applicable components for interactivity. This would foresee the patients’ need for continuous and personalized feedback, in particular for patients with a greater need for care. From this study several factors appear to influence increased use of eHealth technologies: (1) avoiding selective enrollment, (2) making use of participatory design methods, and (3) developing push factors for persistence. Further research should focus on the causal relationship between using the system’s features and actual usage, as such a view would provide important evidence on how specific technology features can engage and captivate users.