Evaluation of the psychometric properties of the Consultation and Relational Empathy (CARE) measure in Final Year Family Medicine Students

Abstract

Introduction: Good interpersonal skills are essential for an effective family medicine (FM) consultation. The Consultation and Relational Empathy Measure (CARE) has been used by patients to assess doctor’ empathy. This study aimed to establish whether the CARE is valid for assessing medical student empathy. Methods: 158 medical students performed a consultation on a simulated patient (SP) for their final year FM exam. SPs completed the CARE, a global rating score (GRS), the Jefferson Scale of Patient’s Perception of Physician Empathy (JSPPPE), and a history-taking checklist. Observing examiners also completed the CARE. Structural construct was assessed using exploratory (EFA) and confirmatory factor analysis (CFA). Reliability was assessed using intraclass and Spearman’s correlation. Results: EFA identified one factor onto which all 10 CARE items loaded significantly, which was supported by CFA. The CARE strongly correlated with both convergent measures: GRS (ρ = 0.794, p < 0.001) and the JSPPPE (ρ = 0.771, p < 0.001), and weakly correlated with the divergent measure: history taking score (ρ = 0.277, p < 0.001). SP CARE ratings weakly correlated with academic results (ρ=0.274, p=0.001). Female SPs were more likely to give higher CARE ratings (adjusted OR=5.38, p<0.001). The inter-rater reliability between SPs and examiners was low (ICC=0.228, p=0.002). Conclusion & Take-home message: The CARE measure appears to be valid for use in undergraduate Family medicine clinical examinations. The weak inter-rater reliability between SP and examiner CARE ratings suggests that assessments should only be conducted by the surrogate patient and not by the clinician observer.