The relationship between checklist scores on a communication OSCE and analogue patients' perceptions of communication.

Many efforts to teach and evaluate physician-patient communication are based on two assumptions: first, that communication can be conceptualized as consisting of specific observable behaviors, and second, that physicians who exhibit certain behaviors are more effective in communicating with patients. These assumptions are usually implicit, and are seldom tested. The purpose of this study was to investigate whether specific communication behaviors are positively related to patients' perceptions of effective communication. Trained raters used a checklist to record the presence or absence of specific communication behaviors in 100 encounters in a communication Objective Structured Clinical Examination (OSCE). Lay volunteers served as analogue patients and rated communication during each encounter. Correlations between checklist scores and analogue patients' ratings were not significantly different from zero for four of five OSCE cases studied. Within each case, certain communication behaviors did appear to be related to patients' ratings, but the critical behaviors were not consistent across cases. We conclude that scores from OSCE communication checklists may not predict patients' perceptions of communication. Determinants of patient perceptions of physician communication may be more subtle, more complex, and more case-specific than we were able to capture with the current checklist.

Tracking the longitudinal stability of medical students' perceptions using the AAMC graduation questionnaire and serial evaluation surveys.

BACKGROUND: This study examined the longitudinal stability of students' perceptions by comparing ratings on similar survey items in three sequential evaluations: end-of-clerkship (EOC), AAMC graduation questionnaire (GQ), and a postgraduate survey (PGY1). METHOD: For the classes of 2000 and 2001, ratings were compiled from EOC evaluations and comparable items from the GQ. For both cohorts, selected GQ items were included in the PGY1 survey and these ratings were compiled. Matched responses from EOC versus GQ and PGY1 versus GQ were compared. RESULTS: Proportions of "excellent" ratings were consistent across EOC and GQ surveys for all clerkships. Comparison of GQ and PGY1 ratings revealed significant differences in only seven of 31 items. CONCLUSION: Student perceptions as measured by GQ ratings are notably consistent across the clinical years and internship. This longitudinal stability supports the usefulness of the GQ in programmatic assessment and reinforces its value as a measure of student satisfaction.