Sustaining the Usefulness of eHealth Research Software : Lessons Learned in Action Design Research

Research software is vital to advancement in the sciences, engineering, humanities, and all other fields. Scientific research is dependent on the quality of and accessibility to research software. Research software is often developed hastily to solve one-off problems, leading to flimsy code that is not sustainable or usable beyond the lifetime of a given research project and is difficult for researchers, outside of the original context, to use, reuse or extend. It is critical to address the many challenges related to the development, deployment, and maintenance of research software. Therefore, there is a growing concern in the scientific community regarding designing sustainable research software. The academic research context refers to the environment or community concerned with scientific research, sponsored by research grants and public funding. Despite the increasing dependence on research software, software development practices in academia lag far behind those in the commercial sector. Health care relies on a very complex information technology architecture with many different IT components and also has a highly complex governance structure alongside the very rapid technology development. Additionally, there are ever-increasing demands and needs from health care users for more flexibility, more functionality and making the care transparent and patient-centred. Taken together, this poses significant challenges for eHealth and Information Systems researchers, as each artefact, depending on the context, has different quality characteristics to operationalise the requirements under consideration. The research objective is to explore what Information Systems researchers and practitioners need to be aware of for sustaining the usefulness of eHealth research software, in the academic research context. This longitudinal action design research (ADR) project, with its three cases, was conducted in an eHealth research project over a period of six years. Contributions from this research include the identification of quality characteristics and their enactment in the actual organisational settings, as well as empirically grounded design principles and a typology for sustaining the usefulness of eHealth research software, based on a formalisation of learning in the three ADR cases. This dissertation also contributes to the method space with the introduction of the augmented action design research (AADR) method, an extension of ADR, on how to conduct multiple ADR projects that build towards an overarching knowledge aim. Practice contributions are the design and development of internet-based eHealth research software to offer patients psychological treatment and support for issues resulting from physical illnesses, while also providing a chance for researchers to study the effectiveness of the aid provided. The dissertation also contributed in a broader sense to the research software development practice, as the findings extend to research areas in which research software is needed to read and interpret research data, and where software must continue to function so that it allows continued access and use of research data