Simulation technology for resuscitation training: a systematic review and meta-analysis.

Abstract

To summarize current available data on simulation-based training in resuscitation for health care professionals. MEDLINE, EMBASE, CINAHL, PsycINFO, ERIC, Web of Science, Scopus and reference lists of published reviews. Published studies of any language or date that enrolled health professions' learners to investigate the use of technology-enhanced simulation to teach resuscitation in comparison with no intervention or alternative training. Data were abstracted in duplicate. We identified themes examining different approaches to curriculum design. We pooled results using random effects meta-analysis. 182 studies were identified involving 16,636 participants. Overall, simulation-based training of resuscitation skills, in comparison to no intervention, appears effective regardless of assessed outcome, level of learner, study design, or specific task trained. In comparison to no intervention, simulation training improved outcomes of knowledge (Hedges' g) 1.05 (95% confidence interval, 0.81-1.29), process skill 1.13 (0.99-1.27), product skill 1.92 (1.26-2.60), time skill 1.77 (1.13-2.42) and patient outcomes 0.26 (0.047-0.48). In comparison with non-simulation intervention, learner satisfaction 0.79 (0.27-1.31) and process skill 0.35 (0.12-0.59) outcomes favored simulation. Studies investigating how to optimize simulation training found higher process skill outcomes in courses employing "booster" practice 0.13 (0.03-0.22), team/group dynamics 0.51 (0.06-0.97), distraction 1.76 (1.02-2.50) and integrated feedback 0.49 (0.17-0.80) compared to courses without these features. Most analyses reflected high between-study inconsistency (I(2) values >50%). Simulation-based training for resuscitation is highly effective. Design features of "booster" practice, team/group dynamics, distraction and integrated feedback improve effectiveness.