Objective: To evaluate the miniclinical evaluation exercise (mini-CEX), which assesses the clinical skills of residents.
Design: Observational study and psychometric assessment of the mini-CEX.
Setting: 21 internal medicine training programs.
Participants: Data from 1228 mini-CEX encounters involving 421 residents and 316 evaluators.
Intervention: The encounters were assessed for the type of visit, sex and complexity of the patient, when the encounter occurred, length of the encounter, ratings provided, and the satisfaction of the examiners. Using this information, we determined the overall average ratings for residents in all categories, the reliability of the mini-CEX scores, and the effects of the characteristics of the patients and encounters.
Measurements: Interviewing skills, physical examination, professionalism, clinical judgment, counseling, organization and efficiency, and overall competence were evaluated.
Results: Residents were assessed in various clinical settings with a diverse set of patient problems. Residents received the lowest ratings in the physical examination and the highest ratings in professionalism. Comparisons over the first year of training showed statistically significant improvement in all aspects of competence, and the method generated reliable ratings.
Conclusions: The measurement characteristics of the mini-CEX are similar to those of other performance assessments, such as standardized patients. Unlike these assessments, the difficulty of the examination will vary with the patients that a resident encounters. This effect is mitigated to a degree by the examiners, who slightly overcompensate for patient difficulty, and by the fact that each resident interacts with several patients. Furthermore, the mini-CEX has higher fidelity than these formats, permits evaluation based on a much broader set of clinical settings and patient problems, and is administered on site.