Mastery Learning in Critical Care

To the Editor:

We read the Willi and colleagues commentary, “Mastery in Simulation in Critical Care before Transitioning to Practice: Are There Drawbacks?” with interest (1). The commentary addresses two essential issues in critical care medicine (CCM) education: 1) feasibility and 2) drawbacks of the mastery approach. Via this letter, we aim to enrich discussion about CCM education and to amplify statements about how simulation-based mastery learning (SBML) and traditional supervised practice are complementary.

Willi and colleagues assume that traditional supervised practice is the clinical education “gold standard.” We disagree. Decades of medical education research presents a different conclusion, which is that traditional clinical education produces clinicians with variable skills that may lead to unsafe patient care (2).

One critique of SBML mentioned by Willi and colleagues is the “time . . . require[d] in simulation for skills training before transitioning to supervised bedside practice.” They calculate that at least 23 hours of CCM simulation-based procedural training is the minimum requirement. In the U.S. model, a 3-year Pulmonary/CCM fellowship has approximately 3,000 training hours. Thus, 23 hours accounts for 0.0076 of total time, a brief 3-day spell.

To make the case for CCM SBML, we address four critiques distilled from the narrative of Willi and colleagues, each with an annotated comment.

• 1.SBML addresses isolated clinical skills, standardized learning measured using checklists, or global rating scales with a set minimum passing standard. We assert that acquisition of key clinical skills that save lives (intubation, central venous catheter insertion, and ventilator management) requires targeted learning and high standards. Clinical education and learning should be stepwise, in which basic and then complex skills are mastered in a simulation laboratory before moving to the clinical environment with real patients (2). For example, the American Heart Association now endorses mastery learning for skill acquisition in areas such as advanced cardiac life support (3).
• 2.SBML does not capture complex real-life clinical situations that involve adaptation and judgment under pressure. We agree that simulation-based education can never capture the richness of clinical work. As practicing clinicians, we teach a small sample of clinical problems and try hard to measure complex learning outcomes objectively. This resembles what Ericsson and Pool specify as measuring outcomes in a “highly developed field” (4). A thematic, sustained, and cumulative research program demonstrates that mastery of central line insertion by residents in the simulation laboratory transfers to improved intensive care unit patient care, including adaptation and judgment under pressure, and better patient outcomes (2). More research is needed to conclusively link SBML to improved outcomes in complex clinical events.
• 3.SBML ignores transfer of learning from the simulation laboratory to real clinical environments. Multiple studies in obstetric, hospital floor, intensive care unit, and operating room settings show that SBML learning outcomes transfer to improved clinical outcomes and substantial return on investment (2). These and other emergent translational medical education outcomes represent the same idea as transfer of training in the behavioral sciences, expressed in clinical language (5).
• 4.Achieving competence in simulation will lead to missed opportunities for learning under supervision, which is needed to develop expertise. Traditional “see one, do one, teach one” clinical education falls short of rigorous mastery learning with deliberate practice (2). Nobel laureate Daniel Kahneman stated, “The acquisition of skills requires a regular [controlled] environment, an adequate opportunity to practice, and rapid and unequivocal feedback about the correctness of thoughts and actions. When these conditions are fulfilled, skill eventually develops, and the intuitive judgments and choices that quickly come to mind will mostly be accurate” (6). SBML meets or exceeds these clinical training conditions and has produced skill acquisition and clinical outcome data about a variety of invasive and noninvasive procedures (2).

In SBML, the focus of attention is the medical learner. By contrast, in clinical education settings, the patient is the center of attention. Clinical skill mastery in CCM simulation is a necessary prerequisite for patient care and is a powerful foundation for supervised clinical training. Although we may disagree about the limits of SBML, we concur that SBML plus supervised practice, “parallel integration of simulations and clinical experiences” according to Willi and colleagues (1), is the best approach to CCM education.