Impacto de estratégias de treinamento com base em simulação na retenção de competências em ressuscitação cardiopulmonar: estudo randomizado

Abstract

Introduction: The quality of in-hospital cardiopulmonary resuscitation (CPR) depends not only on the initial acquisition of competencies but also on professionals’ ability to sustain complex skills over time, including leadership, decision-making, and execution of time-critical actions in cardiac arrest scenarios. Simulation-based training has become a central strategy in emergency education; however, gaps remain regarding post-training skill retention and the comparative effectiveness of different educational approaches. Objective: To compare the immediate and delayed performance of physicians and nurses exposed to three simulation-based training strategies—simulation followed by feedback, rapid-cycle deliberate practice (RCDP), and action-linked phrases (ALP)regarding global efficiency, time-critical actions in the cardiac arrest algorithm, and chest compression fraction (CCF). Methods: Cluster-randomized controlled educational trial conducted in a high-fidelity simulation environment with 36 professionals from a private hospital in Rio de Janeiro. After a standardized Advanced Life Support course (online theory + in-person practice), participants were randomly allocated into three groups: Control (simulation with feedback), ALP (use of action-linked phrases at key scenario moments), and RCDP (rapid-cycle deliberate practice with immediate feedback and targeted repetition). Team-leader performance was assessed in a standardized adult cardiac arrest scenario at two points: immediately after training and approximately 30 days later. Assessment included validated checklists, blinded video review, and measurement of critical times: T1 (start of compressions), T2 (first defibrillation), T7 (return to compressions after rhythm analysis), and CCF. Statistical analyses included descriptive and inferential tests with a 5% significance level. Results: No statistically significant differences were found among the three strategies at either assessment point, indicating similar global performance across Control, ALP, and RCDP groups. However, there was a marked decline in team-leader global efficiency from immediate (89.0%) to delayed evaluation (73.1%; p<0.001), with large effect sizes, demonstrating substantial loss of retention over an average interval of 33.5 days. This decline was also reflected in delays in recognition of cardiac arrest, defibrillator request, time to first shock, return to compressions after rhythm analysis, and reduced CCF in part of the sample. Additional analyses showed that male participants exhibited higher global efficiency and better delayed CCF, whereas previous CPR training or real-life arrest experience did not predict better retention. Conclusion: All three simulation-based strategies were effective for immediate acquisition of CPR leadership competencies; however, none prevented performance deterioration after one month without reinforcement. Findings highlight that retention of competencies—particularly those tied to time-critical actions and continuous high-quality CPR—is highly vulnerable when training is delivered in a single dose. The study reinforces the need for continuous educational models involving brief, frequent sessions, distributed deliberate practice, structured feedback, and integration of time-and-quality metrics (including CCF) into workplace-based training and Rapid Response Team initiatives. By comparing and problematizing RCDP and ALP in a simulated context, this research contributes to the refinement of healthcare simulation programs and to the advancement of continuing education strategies aimed at patient safety in in-hospital cardiac arrest.