Medical decision making takes place in a complex and dynamic environment, involving various individual and organisational stakeholders. In such contexts supportive information systems are required to be designed for practical efficacy, emphasising transparency, usability, and usefulness to their primary users. At the same time, they need to demonstrate improved efficiency and diagnostic accuracy. In addition, they also need to provide effective communication of relevant information. This paper describes the use of design science to develop and evaluate a DSS incorporated into online tools to support sleep physician's diagnosis of sleep disorders. The design science approach offers guidance for developing this solution, that ensure the artefacts fit into their usage context, and that development and evaluation principles that generalise to similar problem domains are identified. How this project fits in the design science guidelines is specified, thus demonstrating how this philosophy and methodology advances theory particularly relevant to other similar medical diagnostic domains. The usability and technical evaluation of the set of tools indicated provide clear practical benefits at patient, physician and organisational levels.