BACKGROUND: With the recent emphasis on core competencies, medical schools and residency programs have attempted to monitor and regulate trainees' patient encounters. The educational validity of this practice is unknown. Our objective was to determine whether patient encounter logs correlate with educational outcomes. METHODS: We reviewed patient logs of all 212 neurology clerkship students from the 2005-2006 academic year and determined the number of patients each student saw in five diagnostic categories (seizure, headache, stroke, acute mental status change, and dementia). We compared these numbers with the students' written examination scores (total and category-specific) and clinical evaluation scores using Pearson product-moment correlations. RESULTS: The more patients in a given diagnostic category that students saw, the lower the students' examination subscores in that disease category (r = -0.066, p = 0.03). The total number of patients each student saw did not correlate with the student's total examination score (r = -0.021, p = 0.77) or the student's overall clinical performance rating (r = 0.089, p = 0.23). CONCLUSIONS: Higher numbers of logged patients did not correlate with better clerkship performance, whether the outcome measures were written tests or faculty ratings, and whether the analysis involved total or disease-specific patient counts. Thus, patient census may not be a meaningful index of educational experience or outcome. Considerable time, money, and effort are required to maintain accurate logs of trainees' encounters with patients; based on the current study, this may be an inefficient use of resources.
BACKGROUND: With the recent emphasis on core competencies, medical schools and residency programs have attempted to monitor and regulate trainees' patient encounters. The educational validity of this practice is unknown. Our objective was to determine whether patient encounter logs correlate with educational outcomes. METHODS: We reviewed patient logs of all 212 neurology clerkship students from the 2005-2006 academic year and determined the number of patients each student saw in five diagnostic categories (seizure, headache, stroke, acute mental status change, and dementia). We compared these numbers with the students' written examination scores (total and category-specific) and clinical evaluation scores using Pearson product-moment correlations. RESULTS: The more patients in a given diagnostic category that students saw, the lower the students' examination subscores in that disease category (r = -0.066, p = 0.03). The total number of patients each student saw did not correlate with the student's total examination score (r = -0.021, p = 0.77) or the student's overall clinical performance rating (r = 0.089, p = 0.23). CONCLUSIONS: Higher numbers of logged patients did not correlate with better clerkship performance, whether the outcome measures were written tests or faculty ratings, and whether the analysis involved total or disease-specific patient counts. Thus, patient census may not be a meaningful index of educational experience or outcome. Considerable time, money, and effort are required to maintain accurate logs of trainees' encounters with patients; based on the current study, this may be an inefficient use of resources.