Stability of standardized patients' performance in a study of clinical decision making

Standardized patients (SPs) have been used extensively in teaching, but their reliability for use in research has been infrequently addressed. This study analyzes the reliability of performance of 13 SPs during 228 doctor-patient encounters in a year-long study related to the diagnosis of depression.

Patient scenarios were based on real patient cases. Four of the five cases had major depressive disorder. Two to three SPs were coached to enact each of the five case scenarios. Medical encounters were videotaped. Interview content was extracted onto a standardized checklist. Interaction between physician and patient was measured by the Interactional System for Interview Evaluation. Tests of SP performance reliability included the: 1) consistency of symptoms volunteered, 2) stability of affect and behavior, and 3) association of SP performance to detection of depression.

The mean number of SP performances was 20.8 (SD = 5.8), with a range of 6 to 28. Problems with reliability emerged in one of the five patient cases. Results otherwise revealed high intra-performance and inter-performance reliabilities. Detection of depression was consistent across SPs and with the rates reported in the literature.

This study provides evidence that performances, within and among SPs, remained consistent, even when intervals between performances were as long as 3 months.