Validity Evidence for the Core Physical Examination in Medical Students

Purpose: The Core Physical Exam (CPE) has been proposed as a basis for the Core + Cluster curriculum for teaching and assessing physical examination (PE) skills in medical students. This study provides initial validity evidence for a modified, institution-specific CPE as an assessment of PE skills in medical students. Methods. The University of Colorado School of Medicine (UCSOM) CPE was developed as a 25- item version of the published CPE. Validity evidence for the UCSOM CPE was gathered using data from two classes of approximately 180 medical students from to September 2015 to December 2018. Validity evidence specific to content, response process, internal structure, relationship to other variables and consequences was gathered. Results: Content and response process evidence was based on the use of content experts to develop the UCSOM CPE and the extensive rater training in the institution’s clinical performance center. Correlations of performance on the UCSOM CPE to other assessments of the PE were generally low in the range of 0.14 to 0.23, consistent with correlations between stations of objective structured clinical examinations. The 90% pass-fail cut point determined by a modified Angoff approach resulted in a fail rate of 10% to 13% for the UCSOM CPE in first year and 36% to 38% in second year. Clinical skills course directors selected a consensus passfail cut score of 80% as a defensible and practical threshold for the UCSOM CPE in its current educational context for entry into supervised clinical practice in the context of a preceptorship experience. Conclusion: Initial validity evidence supports the use of UCSOM CPE as a useful educational strategy for teaching physical examination and as a formative assessment of PE competence in medical students.