The presence and potential impact of psychological safety in the healthcare setting: an evidence synthesis

Introduction: Psychological safety is the shared belief that the team is safe for interpersonal risk taking. Its presence improves innovation and error prevention. This evidence synthesis had 3 objectives: explore the current literature regarding psychological safety, identify methods used in its assessment and investigate for evidence of consequences of a psychologically safe environment. Methods: We searched multiple trial registries through December 2018. All studies addressing psychological safety within healthcare workers were included and reviewed for methodological limitations. A thematic analysis approach explored the presence of psychological safety. Content analysis was utilised to evaluate potential consequences. Results: We included 62 papers from 19 countries. The thematic analysis demonstrated high and low levels of psychological safety both at the individual level in study participants and across the studies themselves. There was heterogeneity in responses across all studies, limiting generalisable conclusions about the overall presence of psychological safety. A wide range of methods were used. Twenty-five used qualitative methodology, predominantly semi-structured interviews. Thirty quantitative or mixed method studies used surveys. Ten studies inferred that low psychological safety negatively impacted patient safety. Nine demonstrated a significant relationship between psychological safety and team outcomes. The thematic analysis allowed the development of concepts beyond the content of the original studies. This analytical process provided a wealth of information regarding facilitators and barriers to psychological safety and the development of a model demonstrating the influence of situational context. Discussion: This evidence synthesis highlights that whilst there is a positive and demonstrable presence of psychological safety within healthcare workers worldwide, there is room for improvement. The variability in methods used demonstrates scope to harmonise this. We draw attention to potential consequences of both high and low psychological safety. We provide novel information about the influence of situational context on an individual’s psychological safety and offer more detail about the facilitators and barriers to psychological safety than seen in previous reviews. There is a risk of participation bias - centres involved in safety research may be more aligned to these ideals. The data in this synthesis are useful for institutions looking to improve psychological safety by providing a framework from which modifiable factors can be identified.