Interruptions in the healthcare workplace: a sociotechnical systems approach

Abstract

Interruptions to workflow have been a source of concern in the health services community. Interruptions have been associated with lost time, staff and patient frustration, and they may contribute to preventable medical error. Despite extensive efforts by researchers to mitigate the effects of interruptions, interventions are sometimes ineffective and do not fit the intended work system. In this thesis I investigate why interruptions occur in the first place, whether they are needed, and what, if anything, we should do about them. In a large-scale critical review of the healthcare interruptions literature I examined the research traditions that have motivated and guided interruptions research in order to impose some structure on the complex and conflicting body of literature. Among other insights, the review revealed that investigations into interruptions are often focused on a single clinical discipline, and that they fail to consider inter-disciplinary dependencies and other contributing sociotechnical system factors. Next I present two methodological innovations consistent with a systems approach. The first methodological innovation is the Dual Perspectives Method, an observational technique developed in order to collect multiple perspectives about interruptions that can be used to draw conclusions about their purpose in healthcare or the burden they impose. The Dual Perspectives Method was used to conduct observations of 46 participants at a 30-bed ICU at a large tertiary care hospital in Brisbane, Australia. In contrast to many other interruptions studies, participants were sampled from all operational levels and all work roles in the ICU work system, including medical staff, nurses in clinical and non-clinical roles, support staff, and allied health staff, in order to explore a full range of coordination needs and motivations for interrupting. The second methodological innovation is the application of social network analysis and modeling techniques to the observational data collected using the Dual Perspectives Method, making it possible to analyze patterns of interaction and interdependencies across all roles in a work system. Together the Dual Perspectives Method and social network analysis allowed me to identify targeted systems-based interventions that may reduce unnecessary interruptions while at the same time avoid unintended consequences that impose additional burden on ICU staff. Finally, I present a guide for identifying appropriate interventions, designed to help nurse unit managers improve the management of workflow interruptions in their respective units by guiding them towards system-based vs. person-based interventions, where appropriate. The concept behind the guide was based on the Institute for Safe Medication Practice’s Hierarchy of Intervention Effectiveness. Semi-structured interviews were used to evaluate the guide and indicate improvements to future iterations. In contrast to a single-discipline approach, this program of research employed a systems approach to reveal vulnerabilities in the clinical work system and the sociotechnical factors that must be considered when developing interventions that better fit clinical workflow. The results indicate that simple counts of interruptions do not reveal enough information to lead to effective interventions. Instead, it is imperative to include a broader sampling of work roles in investigations. The findings further highlight that systems-based interventions are possible that could improve work coordination at ‘blunt end’, while at the same time avoiding unintended consequences to the rest of the work system. These findings may discourage clinical leaders and policy makers from simply recommending that interruptions be minimized without considering workflow issues and the role of staff in preserving the quality of their work.