Page - 213 - in Applied Interdisciplinary Theory in Health Informatics - Knowledge Base for Practitioners

6. Does health informatics have any theory of its own? So far, we have focused on the first of our three objectives: to show where and how interdisciplinary theories have been applied in health informatics. Most of the theories have been developed in other fields. We now turn to the question of theory developed specifically within health informatics, to consider where further work is necessary to develop theory-based approaches. We suggest that there areonly three examples of “native” health informatics theory in the textbook: Distributed Cognition (as it is specifically about “information processing in sociotechnical systems”) Deterioration Communication Management Theory (as its aim is “to improve the design and implementation of ICT systems for communication to and from junior hospital doctors”) The NASSS Framework (as it is focused explicitly on “technologies in health and care organizations”). What else is out there? Arguably, the oldest theories in health informatics are the “determinant frameworks” (in Nilsen’s terminology [45]) relating to the structure and content of patient records. This is unsurprising as it is perhaps the most obvious overlap between healthcare and information. In 1605, Francis Bacon harked back to the narrative case histories of the school of Hippocrates as the ideal [20]. Later, Thomas Sydenham, the ‘English Hippocrates’, wrote in 1676 that an “exact history” of every case of disease would improve therapyby making it empirically obvious how to proceed [20]. Francis Clifton proposed to the Royal Society in 1731 that medical observations should be recorded in a particular tabular format to simplify record-keeping and facilitate comparative analysis [50]. In the 1960s, Larry Weed famously proposed problem-oriented medical records “to guide and teach” [63; 64] and this approach has been adopted in some electronic health record systems. Recent health informatics work has included the development of detailed clinical information models of re-usable concepts in representations such as archetypes [43] and Fast Healthcare Interoperability Resources (FHIR) [58]. While that modelling work is for the purpose of technical implementation not informatics theory, there is still an implied hypothesis that such shared concepts are sufficiently stable, definable and comprehensible to be safe and meaningful as a common language of healthcare. The most basic theory in modern health informatics seems to be Friedman’s fundamental theorem [25]. Friedman asserted that “A person working in partnership with an information resource is ‘better’ than that same person unassisted”, with three important corollaries: (1) That informatics is more about people than technology; (2) In order for the theorem to hold, the resource must offer something that the person does not already know; and (3) Whether the theorem holds depends on an interaction between person and resource, the results of which cannot be predicted in advance. The theorem has been questioned [40]and modifications to the wording have been suggested [30;39], but the common sense of Friedman’s theorem seems generally accepted. Another quite basic proposition is the “first law” that van der Lei proposed: “data shall be used only for the purpose for which they were collected”. The continuing validity of this has been questioned [54] and it is expressed as a normative principle rather than P.J.Scott etal. /ReflectingandLooking to theFuture 213