James, Gillian Anne; (2000) Failures to diagnose clinical dysfunctions: The use of a cognitive model to design aids for student physiotherapists. Doctoral thesis (Ph.D), UCL (University College London).

Text Failures_to_diagnose_clinical_.pdf Download (29MB)

Abstract

This thesis explores the contribution Ergonomics can make to the design of clinical worksystems. Such worksystems typically involve interactions between a health care system (eg a doctor and medical technologies) and a patient, in order to fulfill the goals of health care (comprising, for example, improvements in the patients health). A specific health care profession (physiotherapy) is utilised as the area within which the activities described here are based. The deficit, which gave rise to the research is the apparent failure of undergraduate students to prescribe a treatment programme, which will reverse the patients dysfunction. In terms of Ergonomics, such failures to achieve the worksystem goals fall legitimately within its practice. The first issue addressed in the thesis concerns developing a description of the activities involved in treatment planning in Physiotherapy. The treatment planning deficit appeared focused on diagnosis of dysfunction - which was rudimentary or absent in students. A conceptual framework for modelling diagnosis, based upon the blackboard framework for speech recognition, is described. The framework for modelling diagnosis comprises a set of knowledge sources and a central blackboard with symptom, pattern and picture levels. Knowledge sources are classified into anatomical, pathophysiological and clinical categories and communicate with each other via the blackboard. The order in which these knowledge sources are allowed to write to the blackboard is specified by a control mechanism. The second issue examines the differences between experts and students on the premise that experts should present a standard that students are aiming to achieve. Performance differences are described, relating to the quality of the final diagnosis and the activation of knowledge sources, when diagnosing the same case. Differences are explained with respect to the blackboard level from which treatment prescription flows. The final issue applies the models by re-designing the clinical worksystem with the addition of diagnostic training aids to support knowledge acquisition. A blackboard analysis, of students using these aids, suggests performance improves with respect to accuracy of diagnosis and increased knowledge source recruitment. It was also evident that, in terms of the blackboard framework, aided students prescribed treatment at a higher blackboard level. The research offers support for the use of model-based reasoning to enhance work performance. Qualifications exist regarding the control knowledge used to prioritise the blackboard entries and the precise diagnostic knowledge forming object level knowledge sources.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as CSVXMLJSON

Archive Staff Only

View Item