Video ethnography of leadership during emergency department resuscitation
Background: Leadership during emergency department (ED) resuscitation has traditionally been conceptualised as a singular and hierarchical phenomenon, often with one doctor being designated as the leader of all aspects of the resuscitation episode. This is set against a positivist backdrop which seeks to quantify a leader’s behaviours or traits. Emerging research, however, suggests that leadership can be dynamically shared amongst team members who each contribute to leadership depending on the needs of the group. This research study positions leadership as a teamlevel concept. Introduction: The thesis will observe how leadership is expressed in ED resuscitation teams – who initiates leadership, how it is enacted, how it evolves over time – through the lens of shared leadership theory. Aim(s) To study leadership during emergency department resuscitation using a teamlevel approach. Objective(s) (1) To develop a team leadership behavioural marker tool that can be used as a framework to identify who displays leadership and how they do this. (2) Using a team behavioural marker tool as a framework, conduct a videobased ethnography of leadership based on observations of real patient care episodes. Methods: This study is a mixed-method ethnography located at a large ED in Scotland, triangulating data from (1) questionnaires, (2) interviews and (3) video recordings of ED resuscitation. Using these methods, a team leadership taxonomy was developed to provide a framework for conducting ethnographic observations of leadership during resuscitation. Data collection was from July 2015 to July 2018. Specifically, the methods included: (1) one hundred questionnaires were distributed to ED resuscitation staff. (2) twenty interviews were conducted with Consultants (n=5), Charge Nurses (n=5), Registrars (n=3), Staff Nurses (n=4) and Foundation Year Doctors (n=3). (3) twenty resuscitation episodes were reviewed using video with full ethnographic field notes. Over 300 resuscitation cases were observed in total to immerse the researcher in the resuscitation environment. Analysis: Based on critical realism, an interpretive analytic framework was used in keeping with ethnography. Thematic analysis was used to develop themes from interviews and observations. Where relevant, descriptive statistics were used when providing quantitative data. Findings: A team leadership taxonomy was developed using the three discrete data sets with substantial inter-rater reliability (κ= 0.72, 95% CI: 0.62–0.82) (Objective 1). Building on this behavioural taxonomy, four main themes emerged in the ethnography (Objective 2): (1) Leadership is conceived and practiced as a plural concept – during twenty resuscitation episodes a total of 323 leadership behaviours were observed (avg=16 per case). In 75% of cases, leadership was shared between 3 or more team members; this finding was corroborated by interview statements, such as “everyone in that room can have a leadership role”. (2) Leadership is contingent and adaptive – leadership was observed to evolve in response to situational factors, such as patient acuity, the experience of the team and who was present in the room. (3) The meaning of leadership is heterogeneous – staff conceptualised leadership differently, with ambiguous semantic boundaries and the identity of the ‘leader’ having multiple meanings. (4) There is a distinction between nursing and physician leadership – nurses and doctors exhibited leadership differently. Whilst Doctors overtly display leadership, transcripts from interviews and observations highlighted that Nurses’ leadership is often ‘silent’, ‘subtle’ and ‘in the background’. Conclusion: To view leadership of complex resuscitations as the product one of individual leader risks negating the subtleties of adaptive team performance. Leadership, as reported by clinicians and observed on video during this mixedmethod ethnography, is best conceived as a collective group process where individuals dynamically contribute leadership actions depending on situational factors. This conclusion shifts the focus of leadership and its theory away from the individual leader and towards leadership as a shared team construct.