Measuring clinical skills in agenda-mapping (EAGL-I)
Gobat N., Kinnersley P., Gregory JW., Pickles T., Hood K., Robling M.
© 2015 Elsevier Ireland Ltd. Objective: To develop and validate the Evaluation of AGenda-mapping skilL Instrument (EAGL-I). Methods: EAGL-I was constructed after a literature review and piloting. Simulated consultation recordings were collected in a workshop with third-year medical students at three time points: once pre-teaching, twice post-teaching. Three raters used EAGL-I to assess student agenda-mapping. We examined reliability, ability to detect change and predict full expression of patients' agendas. Results: EAGL-I scores represented reliable assessment of agenda-mapping (Ep<sup>2</sup>=0.832; ϕ=0.675). Generalizability coefficients across items (Ep<sup>2</sup>=0.836) and raters (ϕ=0.797 two raters) were acceptable. A one-way repeated measure ANOVA with post hoc analysis found a statistically significant difference between the pre-teaching occasion of measurement and each post-teaching occasion (p<0.001) and no significant difference between the two post-teaching occasions (p=0.085). Multilevel logistic regression show scores predict expression of scripted hidden agendas irrespective of occasions, or patient scenario (n=60, p=0.005). Conclusion: Evidence of measure validation is shown. Reliability is optimised when two or more raters use EAGL-I and agenda-mapping has been taught. EAGL-I appears sensitive to change. Higher scores predict the likelihood that a patient will disclose their full agenda in a simulated environment. Practice implications: A validated tool for measuring agenda-mapping in teaching and research is now available.