The effect of structured reflection on the diagnostic accuracy of postgraduate trainees during real patient encounters
Master Thesis
2016
Permanent link to this Item
Authors
Supervisors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher
University of Cape Town
Department
Faculty
License
Series
Abstract
Structured reflection has been shown to improve the diagnostic competence of undergraduate and postgraduate trainees in a range of experimental settings using written case scenarios. Evidence supporting the use of this strategy during real patient encounters is lacking. This paper reports on a study conducted to determine the effects of structured reflection on the diagnostic accuracy of postgraduate medical trainees during bedside tutorials using real patient encounters. Method Fifty-five postgraduate trainees in Internal Medicine at the University of Cape Town, South Africa, were prospectively studied during 18 beside tutorials using real patient encounters. Each patient encounter was conducted as a 4-stage diagnostic process and a diagnostic accuracy score (DAS) was calculated for all participants at each stage: • DAS 1: immediately upon arrival at the patient's bedside (visual cues only); • DAS 2: after an oral presentation of the interview and physical examination findings (pre-reflection); • DAS 3: after review of the clinical data using a process of structured reflection (post-reflection); • DAS 4: after discussion of the patient facilitated by the attending physician (facilitated reflection). Memory structure and flexibility in thinking of participants were evaluated using the Diagnostic Thinking Inventory (DTI) and compared to their post-reflection diagnostic accuracy scores. Results A total of 212 diagnostic events were studied. Friedman's test demonstrated a significant difference when comparing the median diagnostic accuracy scores (DAS) of the respective stages of the diagnostic process (χ² (3) = 406.34, p value < 0.001). The Wilcoxon signed-rank test confirmed that there was a significant difference between the immediate DAS (DAS 1) and the pre-reflection DAS (DAS 2) (Z = 8.66, p value < 0.001), the pre-reflection DAS (DAS 2) and the post reflection DAS (DAS 3) (Z = 4.98, p value < 0.001). Linear regression identified a significant relationship between DTI scores and DAS 3 (p value = 0.035), however this explains only a small portion of the variation in the data (r² = 0.093). Conclusion Structured reflection improved the diagnostic accuracy of postgraduate trainees during real patient encounters at the bedside. These data provide support for the suggestion that clinical teachers should consider adding structured reflection to their toolbox of bedside teaching strategies. In addition, DTI scores may help clinical teachers identify trainees struggling with the development of diagnostic expertise.
Description
Keywords
Reference:
Rush, C. 2016. The effect of structured reflection on the diagnostic accuracy of postgraduate trainees during real patient encounters. University of Cape Town.