Presented By: Jenny Fortun, Florida International University Herbert Wertheim College of Medicine
Co-Authors: Christopher Day, Florida International University Herbert Wertheim College of Medicine
Aaron Gomez, Florida International University Herbert Wertheim College of Medicine
Juan Manuel Lozano, Florida International University Herbert Wertheim College of Medicine
Ligia Perez, Florida International University Herbert Wertheim College of Medicine
Justin Shaw, Florida International University Herbert Wertheim College of Medicine

Purpose 
Diagnostic Reasoning (DxR) exams are progressive disclosure case-based assessments that aim to mimic clinical encounters. While studies have explored the relationship between examination formats and student performance, associations between preparatory strategies and performance remains unclear. Understanding study approaches employed by students for DxR and NBME examinations has the potential to inform future course development and curriculum design.  

Methods 
A retrospective cohort study was performed via secondary analysis of data obtained from three cohorts of second-year medical students through end-of-course surveys for three preclinical courses. The surveys explored students' preferences of using a top-down or bottom-up approach to prepare for DxR and NBME examinations. A top-down approach was defined as starting with patient presentations and then learning about diseases. A bottom-up approach was defined as starting with diseases and then reviewing correlating patient presentations. Frequencies and 95% confidence intervals were calculated for students' preferred study approach by examination type. McNemar tests and odds ratios were used to assess differences overall and stratified by course and cohort. Finally, we compared grades achieved by students of all cohorts and courses on DxR and NBME exams according to the preferred study approach. 

Results 
855 unique survey responses from 332 students were included. Students preferred the top-down over the bottom-up approach to prepare for DxR compared to NBME examinations (OR: 2.14, 95% CI: 1.53 to 3.00, p<0.001). These findings were consistent in stratified analyses by course and cohort. Students scored similarly on DxR and NBME exams regardless of preferred study approach. We observed no difference in study approach across quartiles for grades obtained in DxR and NBME examinations. 

Conclusion 
Students preferred a top-down over bottom-up study approach when studying for DxR examinations in comparison to NBME exams. However, no association was found between students' study approaches and exam performance in either testing modality.