Effect of simulation on stress, anxiety, and self-confidence in nursing students: Systematic review with meta-analysis and meta-regression

Introduction: Simulation is a promising strategy in health education, with evidence of importance for learning, but the available systematic reviews are still inconclusive about the effect of the strategy on stress, anxiety, and selfconfidence of nursing students, which impact the adherence to and sustainment of this strategy. Thus, better evidence is needed of the impact of simulation on these competences, essential for health professional education. Objective: To evaluate the effect of simulation-based experiences on stress, anxiety, self-confidence and learning of undergraduate nursing students compared to conventional teaching strategies or no intervention. Design: Systematic review with meta-analysis and meta-regression. Data sources: The databases used included: CENTRAL, CINAHL, Embase (R), ERIC, LILACS, MEDLINE, PsycINFO (R), SCOPUS and Web of Science. Additional searches occurred in PQDT Open (ProQuest), BDTD, Google Scholar and journals with a specific scope in clinical simulation. Review methods: This study was conducted by the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions. Experimental and quasi-experimental studies that compared the effects of simulation on stress, anxiety, and self-confidence of nursing students were included. Study selection and data extraction steps were performed independently by two reviewers. Critical appraisal of the studies was managed by means of the risk of bias tools RoB 2 and ROBINS-I, and quality of evidence by means of the GRADE tool. Data summarization was performed by qualitative synthesis with descriptive analysis and quantitative synthesis by meta-analytic methods and meta-regression. Results: Sixty-two studies were included covering an overall sample of 4570 undergraduate nursing students. When comparing simulation with other teaching strategies, simulation showed small effect size for anxiety (d = -0.33; 95% CI: -0.66-0.002; p = 0.051) and medium effect size for self-confidence (d = 0.71; 95% CI: 0.46-0.95; p < 0.001); there was no difference in the effect-size for stress (d = 0.04; 95% CI:-0.91-0.99; p = 0.90). Subgroup analyses showed an effect of methodological characteristics of the studies on students' anxiety and self-confidence. A positive relationship between self-confidence and learning was identified by meta-regression (p = 0.018; R2 = 20.96%). Conclusion: Simulation is an effective strategy for reducing anxiety and increasing self-confidence compared to conventional teaching strategies. Results are still inconclusive for stress. The use of simulation-based experiences in nursing education obtains positive results on anxiety and self-confidence in students, providing support for its implementation in undergraduate curricula to improve the education of qualified nurses. Registration number: CRD42020206077. Tweetable abstract: Simulation is an effective teaching strategy for reducing anxiety and increasing self-confidence in learning.