Abstract
The increase in Gestational Diabetes Mellitus (GDM) in New Zealand especially in the Auckland region has increased the workload on clinicians. However GDM for a large part can be self-managed. This requires healthy lifestyle interventions like diet and exercise. Mobile wellness apps have potential as a dietary assessment tool. At present, most mobile apps have the potential to share their data. While patients are free to use a mobile app of their choice, data from such heterogeneous sources cannot be stored in one single database system available to clinicians. The aim of the proposed body of work is to develop a wellness data integration prototype to store such data from heterogeneous sources (apps) which can be useful to clinicians.
The main research question is to determine whether a diverse “constellation” of apps can be used together to add value and robustness in a real clinical application area. An ecosystem was built to integrate data from various mobile wellness apps to a developed prototype. Data from various mobile wellness apps was self-collected data that is normally not shared with clinicians. If shared with clinicians through email, the data is isolated and not stored long term for further analysis. Most of this data is non-clinical and has no provision to be stored in existing Health Information Systems. The only possibility is to save these as clinician comments. The text-based comments are not easy for writing any query on them and generating reports.
The current body of work involved investigating the nature of clinical and non-clinical data required for GDM. Standard medical terminologies were of interest to code clinical data. Non-clinical wellness data such as food and exercise, were resolved through data interoperability solutions by mapping data from source to a global schema for all types of data. Blood glucose readings and diet with required nutrition and carbohydrate intake were part of the data captured from wellness apps. The target schema was capable of capturing data from various mobile wellness apps. Design science research methodology shaped the research process of building an artefact for setting up an ecosystem with participating mobile wellness apps data. User-centred design principles guided the design of the wellness data integration prototype built in iterations as part of the ecosystem. Interviews and Think Aloud protocol sessions with clinicians in the early and final stages of the prototype provided necessary feedback to design considerations.
The results demonstrated the perceived usefulness and perceived ease of use of the prototype (artefact) in building an ecosystem of consumer mobile wellness apps. Usability issues were resolved through Think Aloud protocol sessions. The artefact demonstrated the proof of concept and its acceptability in a clinical context. The use of open source software and tools demonstrated the ability of the prototype to store wellness data in clinical systems using health terminologies and health information exchange standards. The criteria for selecting mobile wellness apps for the ecosystem guided the building of a framework for the evaluation of these apps.
