RESEARCH REPORT PLATFORM PRESENTATION

DIAGNOSING UNILATERAL NEGLECT: HYPOTHESIS GENERATION AND PATTERN RECOGNITION IN PHYSIOTHERAPISTS’ CLINICAL REASONING. Plummer-Damato P¹, Morris M², Denisenko S³; ¹University of California Los Angeles, Los Angeles, USA. ²University of Melbourne, Melbourne, Australia. ³La Trobe University, Melbourne, Australia

PURPOSE: To identify the clinical reasoning processes used by physiotherapists in the assessment of unilateral neglect (ULN) and to ascertain whether therapists generate hypotheses about the different types of neglect. RELEVANCE: Understanding clinical decision making processes is important in identifying how particular behavioural symptoms are recognised, interpreted, and classified in clinical practice. PARTICIPANTS: The participants were 14 physiotherapists recruited from the Victorian branch of the Neurology Special Interest Group of the Australian Physiotherapy Association. METHODS: Participants observed a 27-minute videotaped assessment of an experienced physiotherapist assessing a patient with left ULN following stroke. The participants were not told that the patient had ULN. At predetermined intervals, the videotape was paused and the participants documented their thoughts pertaining to the case at that point in time. There were 5 segments of the assessment: initial interview, sensory, motor function, balance and gait, and standardised tests for neglect. ANALYSIS: A transcript of responses was prepared for each participant. Each transcript was read by two researchers independently several times and patterns in the participants’ thinking were noted. Data were then displayed in a matrix to facilitate comparison across participants for each segment of the assessment. The data for each section of the video were then examined for evidence of hypothesis generation. Only hypotheses pertaining to ULN were analysed. RESULTS: Physiotherapists used both hypothetico-deductive reasoning and pattern recognition in their thinking related to ULN. A hypothetico-deductive clinical reasoning approach, which is characterised by specific testing to confirm a hypothesis, was evident amongst therapists who, after forming a hypothesis about neglect, thought about strategies to test their hypothesis. Pattern recognition, which does not involve explicit testing of a hypothesis, was evident in a small group of clinicians who evaluated their early hypotheses of ULN through observation of the patient’s non-verbal behaviour during the interview; they did not wait for the hypothesis to be formally tested. This clinical reasoning behaviour is consistent with models of pattern recognition in which clinicians use knowledge from prior experience (i.e. recognise a pattern of symptoms) to make a diagnosis. Hypotheses concerning specific types of neglect were infrequently generated. Physiotherapists concentrated on verifying the presence of neglect and quantifying the severity of the syndrome, rather than differentially diagnosing the specific types. CONCLUSIONS: Physiotherapists collect a large amount of data through observation and they use this information to formulate and evaluate clinical hypotheses about neglect. Existing knowledge and previous experience may influence whether a therapist uses a pattern recognition or hypothesis testing model of clinical reasoning. The lack of hypothesis generation for different types of ULN suggests that clinicians do not typically distinguish between the different types of neglect in routine clinical assessment. IMPLICATIONS: Failure to explicitly identify the nature of ULN implies that the primary factors contributing to a person’s functional limitations may not always be addressed specifically in treatment. Education in the different types of ULN may help clinicians to refine their neglect-related hypotheses. KEYWORDS: clinical reasoning, diagnosis, unilateral neglect. FUNDING ACKNOWLEDGEMENTS: This research was not funded.

ETHICS COMMITTEE: La Trobe University Faculty Human Research Ethics Committee.