Effectiveness of a proficiency-based progression e-learning approach to training in communication in the context of clinically deteriorating patients: a multi-arm randomised controlled trial.

OBJECTIVE: To determine the effectiveness of proficiency-based progression (PBP) e-learning in training in communication concerning clinically deteriorating patients. DESIGN: Single-centre multi-arm randomised double-blind controlled trial with three parallel arms. RANDOMISATION, SETTING AND PARTICIPANTS: A computer-generated program randomised and allocated 120 final year medical students in an Irish University into three trial groups. INTERVENTION: Each group completed the standard Identification, Situation, Background, Assessment, Recommendation communication e-learning; group 1 Heath Service Executive course group (HSE) performed this alone; group 2 (PBP) performed additional e-learning using PBP scenarios with expert-determined proficiency benchmarks composed of weighted marking schemes of steps, errors and critical errors cut-offs; group 3 (S) (self-directed, no PBP) performed additional e-learning with identical scenarios to (PBP) without PBP. MAIN OUTCOME MEASURES: Primary analysis was based on 114 students, comparing ability to reach expert-determined predefined proficiency benchmark in standardised low-fidelity simulation assessment, before and after completion of each group's e-learning requirements. Performance was recorded and scored by two independent blinded assessors. RESULTS: Post-intervention, proficiency in each group in the low-fidelity simulation environment improved with statistically significant difference in proficiency between groups (p<0.001). Proficiency was highest in (PBP) (81.1%, 30/37). Post hoc pairwise comparisons revealed statistically significant differences between (PBP) and self-directed (S) (p<0.001) and (HSE) (p<0.001). No statistically significant difference existed between (S) and (HSE) (p=0.479). Changes in proficiency from pre-intervention to post-intervention were significantly different between the three groups (p=0.001). Post-intervention, an extra 67.6% (25/37) in (PBP) achieved proficiency in the low-fidelity simulation. Post hoc pairwise comparisons revealed statistically significant differences between (PBP) and both (S) (p=0.020) and (HSE) (p<0.001). No statistically significant difference was found between (S) and (HSE) (p=0.156). CONCLUSIONS: PBP e-learning is a more effective way to train in communication concerning clinically deteriorating patients than standard e-learning or e-learning without PBP. TRIAL REGISTRATION NUMBER: NCT02937597.