Simulated Patient Role-Plays with Consumers with Lived Experience of Mental Illness Post-Mental Health First Aid Training: Interrater and Test Re-Test Reliability of an Observed Behavioral Assessment Rubric

Abstract

Mental Health First Aid (MHFA) training teaches participants how to assist people experiencing mental health problems and crises. Observed behavioral assessments, post-training, are lacking, and the literature largely focuses on self-reported measurement of behaviors and confidence. This study explores the reliability of an observed behavioral assessment rubric used to assess pharmacy students during simulated patient (SP) role-play assessments with mental health consumers. Post-MHFA training, pharmacy students (n = 528) participated in SP role-play assessments (n = 96) of six mental health cases enacted by consumers with lived experience of mental illness. Each assessment was marked by the tutor, participating student, and consumer (three raters). Non-parametric tests were used to compare raters’ means scores and pass/fail categories. Interrater reliability analyses were conducted for overall scores, as well as pass/fail categories using intra-class correlation coefficient (ICC) and Fleiss’ Kappa, respectively. Test re-test reliability analyses were conducted using Pearson’s correlation. For interrater reliability analyses, the intra-class correlation coefficient varied from poor-to-good to moderate-to-excellent for individual cases but was moderate-to-excellent for combined cases (0.70; CI 0.58–0.80). Fleiss’ Kappa varied across cases but was fair-to-good for combined cases (0.57, p p < 0.001). Recommended modifications to the rubric, including the addition of barrier items, scoring guides, and specific examples, as well as the creation of new case-specific rubric versions, may improve reliability. The rubric can be used to facilitate the measurement of actual, observed behaviors post-MHFA training in pharmacy and other health care curricula.

Keywords

• reliability
• mental health education
• assessment
• simulated patient
• observed behavioral measurement