In response to demands to enhance the eciency and accountability of health systems, a range of dierent information technologies are being promoted. These technologies include integrated hospital systems, community health information networks and data repositories. However, the record of such technologies inside and outside the health industry suggest that such technologies cannot necessarily be relied on. The reason identi®ed is that information systems are inherently logical and rational systems, and often come into con¯ict with the less rational social systems of organizations.
Health information is identi®ed in terms of three basic dimensions of information; that associated with managers; with professionals; and, with patients. The information of these dimensions are focused on very dierent objectives, have dierent structures and functions and are controlled by very dierent social processes. The information is also very complex and diverse within the dimensions. In the clinical encounter the clinician draws on specialist expertise, satis®es administrative requirements, and provides a clinical record. Thus, these dimensions converge at that point. However, collecting information is costly, and an ecient service demands economy in data collection.
However, the technologies being promoted demand `complete' data acquisition based on consistent and stable data de®nitions and data structures. The article argues that there is, thus, a con¯ict between the requirements of these technologies, and the realities of providing ecient services within a changing organizational, professional and social environment.