Case Specificity and Intervening Variables Affecting Diagnostic Accuracy within Two Similar Clinical Presentations
Date
2013-10-18
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
This study investigated the generalizability of diagnostic performance and problem solving strategy within and across cases from two similar clinical presentations (CP). Many variables have been described in medical education literature that influences the generalizability of diagnostic performance. This doctoral research further examined the effect of intervening variables in diagnostic performance. The variables studied were: disease difficulty, case typicality, CP, and problem solving strategy.
Participants of the study were sixth year medical students enrolled in four universities in Saudi Arabia. Two types of data were analyzed, diagnostic performance scores obtained from written clinical vignettes and problem solving strategies abstracted from written think aloud exercises.
Classical Test Theory (CTT) and Item Response Theory (IRT) were two measurement frameworks used to analyze the data and triangulate study findings. Variable reduction strategies were used to examine the construct of performance scores in comparison to the CP schemes. Cross-tabulation and Chi-square were used to investigate the consistency in using problem solving strategies within and across cases from two similar CPs. Also, Multi-Facet Rasch Modelling (MFRM) was employed to determine degree of impact on diagnostic performance by the independent variables.
This study answered three research questions related to diagnostic generalizability and intervening variables that contribute to diagnostic performance. Most of the diseases included in this study showed high correlation coefficients and high discriminating measures. The estimated measures from two different measurement theories supported predictability of performance from one case to another.
This research provides evidence that the diagnostic strategy utilized can alter diagnostic performance. Further support was found for schema-based instruction in medical schools to enhance student diagnostic performance.
The research model illuminates some factors that influence diagnostic performance. Disease difficulty and type of CP were two variables found to alter diagnostic performance. Further research is required to identify other variables.
Finally, these study findings set the base for future research. Directions for further investigation include: examining performance and problem solving strategy generalizability within and across dissimilar CPs; studying the effect of case typicality on diagnostic performance using MFRM; and including a broader sample that would include participants from medical schools that have different curricular models.
Description
Keywords
Medicine and Surgery
Citation
Munshi, F. (2013). Case Specificity and Intervening Variables Affecting Diagnostic Accuracy within Two Similar Clinical Presentations (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/25221