Effectiveness of nontechnical skills educational interventions in the context of emergencies: A systematic review and meta-analysis.

INTRODUCTION: In recent years, the importance of training healthcare professionals in nontechnical skills using effective methodologies has been increasingly recognised as a means of preventing clinical errors in the practice of health care. The aim of this study was to evaluate the effectiveness of educational interventions on nontechnical skills in the emergency medical services and/or critical care unit settings. METHODS: A systematic search was carried out in the PubMed, SCOPUS, CINAHL, and Web of Science databases according to predetermined inclusion and exclusion criteria. After the initial search, 7952 records were selected after duplicates removed. Finally, a selection of 38 studies was included for quantitative analysis. Separate meta-analyses of standardised mean changes were carried out for each outcome measure assuming a random-effects model. Cochran's Q-statistic and I2 index were applied to verify study heterogeneity. Weighted analyses of variance and meta-regressions were conducted to test the influence of potential moderators and funnel plots using Duval and Tweedie's trim-and-fill method, and Egger's regression test were used to examine publication bias. RESULTS: All the variables analysed had a significant effect size, with the exception of situational awareness (d+ = -0.448; 95% confidence interval [CI] = -1.034, 0.139). The highest mean effect size was found for knowledge (d+ = -0.925; 95% CI = -1.177, -0.673), followed by the mean effect sizes for global nontechnical skills (d+ = -0.642; 95% CI = -0.849, -0.434), team nontechnical skills (d+ = -0.606; 95% CI = -0.949, -0.262), and leadership nontechnical skills (d+ = -0.571; 95% CI = -0.877, -0.264). Similar mean effect sizes were found for attitude (d+ = -0.406; 95% CI = -0.769, -0.044), self-efficacy (d+ = -0.469; 95% CI = -0.874, -0.064), and communication nontechnical skills (d+ = -0.458; 95% CI = -0.818, -0.099). Large heterogeneity among the standardised mean changes was found in the meta-analyses (I2 > 75% and p < .001), except for self-efficacy where I2 = 58.17%, and there was a nonstatistical result for Cochran's Q. This great variability is also reflected in the forest plots. DISCUSSION: The use of simulation interventions to train emergency and critical care healthcare professionals in nontechnical skills significantly improves levels of knowledge, attitude, self-efficacy, and nontechnical skills performance.