Medical knowledge advances rapidly and the and support the patient in tracking their own journey and can help to improve performance in many and perhaps all of these areas growth causes major problems of accessibility. Hundreds of thousands of papers, tens of thousands of books and reports are published every year; a systematic search for up to date information about a clinical condition and the best treatment for a specific patient given current knowledge is so difficult it is frequently not even attempted. Clinicians have traditionally depended on a number of tools to focus their search for information, but in the modern world these tools are becoming impractical for clinical use.
General reference texts tend to become out of date quickly and are time-consuming to consult with little guarantee that they contain the information required in a particular clinical setting. They are expensive, and often remain on the shelf with little use for years after they were published.
Medical journals and research articles are the primary source of new scientific findings and reports of clinical trials, but few clinicians have time to follow developments even in restricted specialties.
Clinical guidelines are written with the aim of summarising current evidence-based practice for diagnosis and treatment. However, they tend to be very general and are arguably more useful in setting up or reviewing clinical services than as tools for supporting individual decision-making at the point of care.
Modern electronic media are powerful and quick to use, but still have major shortcomings.
Online journals (e.g. Pubmed) and search engines(e.g. NHS Evidence) can be searched exhaustively and instantaneously but finding the critical bit of information still takes time, and reading and absorbing the results of a search takes just as long as on paper so on-line searching is arguably not a practical tool for use at the point of care.
Specialised repositories (e.g. Map of Medicine) can include a large, well structured and easily navigable body of practical and medidcal information. However, navigation over web pages and flow diagrams seems can still be time consuming and still does not deliver information that is specific to the patient.
Although electronic media have valuable uses their limitations mean that the medical informatics community has been seeking new ways of supporting quality and safety of decision-making at the point of care.
Machine-executable guidelines can support real-time decision-making and workflow in a patient-specific manner. However, applications are normally stand-alone, and cover only one clinical service (or a few at best). Many different applications would be needed for a single clinical setting, compounding the content management problem.
The demands of staying up to date with current knowledge are to some extent reduced through specialisation but these methods do not scale: the development and maintenance of these services by a single organisation, even one of the size of the NHS, is unsustainable.
The Safe and Sound team aim to demonstrate a distinctive approach to dealing with these challenges through an incremental development programme.
First, we well demonstrate web-based applications that are able to deploy machine executable clinical guidelines and other clinical services written in Oxford's PROforma language. We will show how this can be done using a variant of a traditional client-server approach. Second, we will show how suites of coordinated PROforma services can be provided for clinical users with different roles in different places at different times. In this service orchestration approach service development and management is centralised and closed to outside service providers. Third, we will develop a distributed framework, in which machine executable clinical decision support and workflow management services, includingg external 3rd party services, can be dynamically connected to each other as needed. In this service choreography approach service development and management is open to outside service providers. Providers retain responsibility for content development and maintenance and the model is far more scalable and flexible than the orchestration framework.
A service orchestration demonstration can be reviewed here .
A service choreograpy demonstration will be installed here by the end of the project.
