Simulation for learning and teaching procedural skills: the state of the science.

INTRODUCTION: Simulation is increasingly used to support learning of procedural skills. Our panel was tasked with summarizing the "best evidence." We addressed the following question: To what extent does simulation support learning and teaching in procedural skills?
METHODS: We conducted a literature search from 2000 to 2010 using Medline, CINAHL, ERIC, and PSYCHINFO databases. Inclusion criteria were established and then data extracted from abstracts according to several categories. Although secondary sources of literature were sourced from key informants and participants at the "Research Consensus Summit: State of the Science," they were not included in the data extraction process but were used to inform discussion.
RESULTS: Eighty-one of 1,575 abstracts met inclusion criteria. The uses of simulation for learning and teaching procedural skills were diverse. The most commonly reported simulator type was manikins (n = 17), followed by simulated patients (n = 14), anatomic simulators (eg, part-task) (n = 12), and others. For research design, most abstracts (n = 52) were at Level IV of the National Health and Medical Research Council classification (ie, case series, posttest, or pretest/posttest, with no control group, narrative reviews, and editorials). The most frequent Best Evidence Medical Education ranking was for conclusions probable (n = 37). Using the modified Kirkpatrick scale for impact of educational intervention, the most frequent classification was for modification of knowledge and/or skills (Level 2b) (n = 52). Abstracts assessed skills (n = 47), knowledge (n = 32), and attitude (n = 15) with the majority demonstrating improvements after simulation-based interventions. Studies focused on immediate gains and skills assessments were usually conducted in simulation. DISCUSSION: The current state of the science finds that simulation usually leads to improved knowledge and skills. Learners and instructors express high levels of satisfaction with the method. While most studies focus on short-term gains attained in the simulation setting, a small number support the transfer of simulation learning to clinical practice. Further study is needed to optimize the alignment of learner, instructor, simulator, setting, and simulation for learning and teaching procedural skills. Instructional design and educational theory, contextualization, transferability, accessibility, and scalability must all be considered in simulation-based education programs. More consistently, robust research designs are required to strengthen the evidence.