You are here: EHealth Web>PreviousWork (21 Aug 2007)

Previous Work and Literature Survey

Healthcare institutions are looking at ways to increase their efficiency by reducing costs while providing care services with a high level of safety and confidentiality. Thus, hospital information systems have to support quality improvement objectives. It has been recognized in the literature [6][11][12] that data integration is important towards meeting these objectives, but it is not, by itself, sufficient. Aspects of activity, such as where, what (else), and why are poorly represented by the data model alone.

An approach to healthcare quality does not only require the measurement of the outcomes of the processes of care (such as clinical outcomes, and patient satisfaction), but also the description and assessment of how these activities are performed throughout the business process. Continuous traceability of all healthcare activities is deemed mandatory by [12] in order to reduce variations in practices, and to detect, measure, and prevent adverse events occurring during health care delivery. Thus the demand has shifted toward process-oriented healthcare information systems [11][12].

Business process modeling (BPM) is a structured method for describing and analyzing opportunities of improving the business objectives of multiple stakeholders. BPM usually involves identifying the roles of users involved in the process, and the definition of activities (often referred to as workflows or services) that contribute to the satisfaction of well-defined business goals [13]. In [12], processes and their related set of activities are considered as scenarios according to which a user interacts with its environment (other users, patients, context of practice), has a set of responsibilities, and reaches goals.

Existing work on business process modeling in healthcare information systems has focused on modeling data and control flows in the chain of care, for example, through hierarchical and structured design techniques [12]. However, more work is required to relate the business process model to the quality objectives supported by the process. Process models help answer the where, what, who, and when questions, but do not sufficiently model why an activity, or particular sequence of activities is performed. A better way of modeling the performance (in terms of quality objectives met) of the business process is required, as well as a way of designing or evolving business processes based on prescribed quality objectives is needed.

The User Requirements Notation (URN) [3] includes two main components that can be used to connect goals, quality, and business processes: a goal-oriented notation (Goal-oriented Requirement Language - GRL) and a scenario-oriented notation (Use Case Maps - UCM). Although URN was mainly applied to telecommunication systems, it is general enough to support BPM in a variety of contexts. For instance, the scenario notation (UCM) was used to elicit requirements by identifying the different responsibilities and the demands on spatial resolution associated to the actions of each administrative unit in a health information system [8].

Aspect-oriented modeling is a new way of looking at and composing desirable qualities and functional views in systems. It promises an improved separation of concerns in models, and aspect-oriented extensions to UML are already being proposed [4]. In a requirements and BPM context like URN, links between aspects, goals and scenarios require further exploration.

Daniel Amyot's research and teaching areas are software engineering and requirements engineering. He has worked on the specification and validation of various types of reactive and distributed systems. He is the Rapporteur responsible for requirements languages at ITU-T, where he leads the standardization of the URN. He is also leading the development of tools for URN, and their integration to requirements management systems (e.g. Telelogic DOORS) [5]. Dr. Amyot recently applied URN to the modeling and evolution of business processes (for supply chain management), in collaboration with M. Weiss [13][14]. He is attempting to extend these concepts to the healthcare domain and to the aspect-oriented modeling area.

Michael Weiss's research interests include web services, software architecture and patterns, and business model design. He has made contributions to non-functional requirements analysis, in particular related to modeling patterns, service composition, and business processes (the latter with D. Amyot [13][14]). He has also worked on the Tropos approach to modeling agent-oriented information systems, in particular, issues of security and privacy in information systems [7]. He has taught courses on e-commerce technologies, software quality assurance, and distributed programming. Dr. Weiss previously worked at Mitel where he managed a group responsible for developing an agent-based environment for communication services.

Doug Angus has research activities at the international, national, provincial, and regional levels, particularly in the areas of health care reform, strategic management, health economics (cost analysis and economic evaluation of health programs), and health policy. Dr. Angus was Director of the Queen's-University of Ottawa Economic Projects on Cost-Effectiveness of the Canadian Health Care System, and is the lead author of its major report, Sustainable Health Care for Canada [1], where he noted that "The major emphasis throughout health care systems worldwide is on better and more efficient use and allocation of resources that has a position between planning an administration, and is the bridged between two sides of the issues: it helps covert goals into action. [...] Quality of resource allocation decisions depends, to a large extent, on the quality, timeliness, and accuracy of available information."

Liam Peyton's research has been focused on software architecture frameworks that integrate business process automation, metrics, and data warehouse technology to manage compliance with enterprise objectives related to performance management and regulatory compliance. He has identified a special category of applications that manage delivery of services by reporting and configuring services based on data collected in the delivery of services [10]. He has analyzed technical support for collaborative on-line processes in support of regulated access to sensitive data in a hospital setting [9]. Dr. Peyton is co-directing the Intelligent Data Warehouse Laboratory at the University of Ottawa.

Alan Forster's research focuses on patient safety, health services, and hospital quality improvement. He conducts studies evaluating the incidence of adverse events following discharge from hospital [2]. He is interested in the use of hospital clinical and administrative data to monitor and prevent adverse events, as well as to measure hospital quality. To this end, he is leading the effort in managing and exploiting the data warehouse at TOH. Dr. Forster has also conducted several studies evaluating the effect of hospital occupancy on several indicators of quality of care using hospital administrative databases.

-- Daniel Amyot - 03 Feb 2006

Topic revision: r5 - 21 Aug 2007 - 10:15:03 - Daniel Amyot
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback