Referrals from the emergency department (ED) to inpatient teams are routine but high-stakes interactions, yet little is known about how they are accomplished in practice. Prior work often treats referrals as information transfer and emphasises structural tensions between departments, paying limited attention to the interactional detail of referral calls. To address this gap, we draw on seventeen recorded referral calls collected during a year of ethnographic fieldwork in a metropolitan ED. We show how clinicians manage jurisdictional seams through fine-grained conversational moves, and identify navigation strategies: tentative problem framing, preference-sensitive questioning, implicit boundary negotiation, offering of assistance and follow-through, and balancing displays of competence and humility. We demonstrate how the patient's case is not treated as a fixed record but is reshaped in talk to align with different specialties, operating as a dynamic object. These insights extend accounts of boundary work by offering a micro-interactional view of how clinicians negotiate responsibility at jurisdictional seams, and contribute training implications that treat referral talk as interactional expertise, and design directions for improved referral tool technologies that surface seams, scaffold rather than script calls, and keep automation in service of clinical judgement.

When given the opportunity, people tend to try to reach coarse breakpoints for work interruptions. Coarse breakpoints are frequently associated with less effort when resuming the task. We investigated how supporting task resumption with augmented reality (AR)-cues affects this behavior. In a mixed factorial experiment, 50 participants performed a physical sorting task that included deferrable interruptions with varying distances to a coarse breakpoint, either with or without an AR-cue indicating the next correct step after interruption. Participants with AR-cue accepted interruptions at fine breakpoints more frequently than those without a cue, except when the coarse breakpoint was one step away, and reported less stress. Our findings indicate that AR-cues attenuate but do not eliminate the need for specific task resumption strategies, such as reaching a coarse breakpoint, and reduce the stress. Considering AR-cues for task resumption may be particularly beneficial for time-critical interruptions and fast-paced work environments.

Interpreting electrocardiograms (ECGs) is an important but complex and error-prone task. While diagnostic support algorithms exist, how support is displayed and how clinicians interact with ECG diagnostic and clinical decision support systems in general remain underexplored. In this preregistered experiment, we studied how providing clinicians with different versions of diagnostic support affects ECG interpretation. All four support types improved diagnosis accuracy compared to a no-support control condition, but the most effective was support offering visual ECG trace markings. User experience, in the form of psychological need satisfaction of competence and security, was highest when clinicians first viewed the ECG independently and then received support in a second stage. The latter two-stage support also resulted in the shortest diagnosis times. We conclude with design and research implications for creating clinician-algorithmic support interactions that improved user experience, efficacy, and effectiveness in the present study, and may ultimately contribute to patient safety.

Effective information support tools are challenging to design for fast-paced, information rich, and difficult to predict circumstances, particularly when information is fragmented and sources are dispersed. To explore, we conducted a field study on handover and the associated information work, which included 40 visits and 75 hours of observation and interviews with doctors in a metropolitan emergency department (ED). Beyond information exchange, we found that handovers highlight doctors' proactive approach by anticipating information needs, managing uncertainties arising from dynamic information, and developing patient care plans through multiple contingencies. Expanding on the idea of handover as a multifaceted process rather than a single event, we reinforce existing calls for greater flexibility emphasising that the ascertainment of pertinent information is an ongoing, adaptive process. This work demonstrates that deciding what constitutes relevant information is a priori indeterminate when designing information systems and support tools in environments such as EDs. We propose the preservation of specific ‘relativities’ of information—such as uncertainty, particularity, incompleteness, and temporality—in designing information support tools for dynamic, critical and multi-disciplinary work environments.

Recent advancements in artificial intelligence have sparked discussions on how clinical decision-making can be supported. New clinical decision support systems (CDSSs) have been developed and evaluated through workshops and interviews. However, limited research exists on how CDSSs affect decision-making as it unfolds, particularly in settings such as acute care, where decisions are made collaboratively under time pressure and uncertainty. Using a mixed-method study, we explored the impact of a CDSS on decision-making in anesthetic teams during simulated operating room crises. Fourteen anesthetic teams participated in high-fidelity simulations, half using a CDSS prototype for comparative analysis. Qualitative findings from conversation analysis and quantitative results on decision-making efficiency and workload revealed that the CDSS changed team structure, communication, and diagnostic processes. It homogenized decision-making, empowered nursing staff, and introduced friction between analytical and intuitive thinking. We discuss whether these changes are beneficial or detrimental and offer insights to guide future CDSS design.

Many work domains include numerous interruptions, which can contribute to errors. We investigated the potential of augmented reality (AR) cues to facilitate primary task resumption after interruptions of varying lengths. Experiment 1 (N = 83) involved a computer-based primary task with a red AR arrow at the to-be-resumed task step which was placed via a gesture by the participants or automatically. Compared to no cue, both cues significantly reduced the resumption lag (i.e., the time between the end of the interruption and the resumption of the primary task) following long but not short interruptions. Experiment 2 (N = 38) involved a tangible sorting task, utilizing only the automatic cue. The AR cue facilitated task resumption compared to not cue after both short and long interruptions. We demonstrated the potential of AR cues in mitigating the negative effects of interruptions and make suggestions for integrating AR technologies for task resumption.

Research has explored head-worn displays (HWD) in various professional contexts. However, evaluations have been limited by short-term use, a focus on the person using the HWD, and on performance variables. In a field study, we evaluated a monocular, opaque HWD for multi-patient monitoring, which supervising anesthesiologists wore for 8-10 days each. We investigated the effect of prolonged HWD use on the experience of the supervising anesthesiologists and their co-workers using interviews and repeated observations. A reflexive thematic analysis showed (1) interaction and mindset changes over time, (2) information on the HWD is more than numbers, (3) the HWD affects co-workers' collaboration with supervisors, and (4) distraction depends on the point of view. Using activity theory, we discuss the fact that HWD use develops and changes over time and that even a single-user HWD influences the collaboration with co-workers. We conclude with implications for HWD design, implementation, and evaluation.

Technology use in acute care has so far been studied with a predominant focus on safety and efficiency rather than the experience of acute care staff. In the present paper, we report a qualitative study including 130 hours of participant observation on acute care teams and retrospective interviews with nine anesthesiologists on their experiences with technology. Our approach is theoretically guided by four modern HCI perspectives, namely psychological need satisfaction, activity theory, embodied interaction, and media equation theory. We analyzed the interview and observation data using a reflexive thematic analysis and identified four themes. Technology as necessary (evil) describes the pervasiveness of technology and how acute care teams depend on technology. While technology as second patient covers the additional load for anesthesiologists on a functional and emotional level, technology as problem solver and safety net highlights relieving aspects of technology use. Technology as artificial limb focuses on so-called transparent tools that extend the body and capabilities of users. We then discuss the identified themes concerning the previously presented theories. Based on our findings and theoretical perspectives, we report general insights for research and design in acute care, such as the benefits of mixed methods or the importance of the gut feelings of experts. Finally, we propose possible directions for future work, such as developing a UX questionnaire adapted to the context of safety-critical domains, such as acute care.

Abstract Air traffic control (ATC) is a safety-critical, cooperative work domain, which faces usability challenges due to technology driven development in the past. In this work, we followed a user-centered design process to explore how novel interaction concepts increase user experience in ATC. Based on controllers’ needs we envisioned one unified interface together with three possible interaction concepts (the mouse interface, flight-dial and tangible interface) addressing different aspects of ATC. We prototypically implemented the interaction concepts and iterated each prototype based on feedback from 24 controllers. Qualitative data from these iterative formative evaluations indicated that controllers prefer interfaces that are efficient to use, minimalistic, customizable and context sensitive. A summative evaluation (N = 12) showed that the hedonic quality of all three concepts were higher compared to the system currently in use. Our results and insights can provide guidance and inspiration for the future design of ATC interfaces.

Supervising anesthesiologists oversee junior anesthesiologists in several operating rooms to instruct and provide advice, and hence ensure high patient safety. Supervisors benefit from good situation awareness but accessing information about multiple patients can be challenging. In a user-centered design process, we have developed an interface concept for a head-worn display (HWD) to support multi-patient monitoring. The design includes data about the patients’ physiology, vital status, vital trends, and medications. We conducted an exploratory evaluation of a prototype in a high-fidelity medical simulation with six supervising anesthesiologists. Our results show that the supervisors’ situation awareness of all simulated operations improved because participants noticed and responded to multiple events over the course of the scenario. However, the supervisors reported increased mental workload due to the constant availability of patient data. We discuss insights about the HWD display, the design process, and the experience and views of staff on using the HWD.

Despite many advances, clinical decision support tools (DSTs) often suffer from implementation and acceptance problems in the actual clinical context. We suggest that considering psychological needs-based and embodied user experience theories in the design of DSTs could help to overcome these problems. To examine this idea, we iteratively developed a DST called Cassandra supporting anesthetic teams in crisis management, specifically focusing on psychological needs and fluent interaction with the social and physical environment. We preliminarily evaluated Cassandra in a medical simulation, requiring anesthetic teams to handle a crisis. Although not all features of Cassandra had the intended effect, the results indicated that interacting with Cassandra supported the fulfillment of the identified needs for autonomy, competence, and relatedness and was seamlessly integrated into existing diagnostic processes. Considering user experience theories for the design of DSTs seems a promising way to overcome implementation and acceptance problems and eventually improve patient safety.

To increase capacity and safety in air traffic control, digital strip systems have superseded paper strips in lower airspace control centers in Europe. Previous ethnographic studies on paper strip systems anticipated a radical change in work practices with digital strip systems, but we are not aware of any studies that evaluated these predictions. We carried out contextual inquiries with controllers and focused on face-to-face and radio communication, interactions with the digital strip system and the workspace in general. In turn, we contribute (1) detailed descriptions of controllers' work practices, such as using tacit information from radio communication and 'standard advocates vs. tinkerers' operation modes, (2) respective implications for design and (3) discuss how the observed work practices are similar or different from the reported practices in the literature of the two preceding decades. Our key insights are, that documentation speed is faster with digital strips, although a high load in the case of radio frequency persists. Controllers retrieve tacit information from the radio communication and combine it with scattered cues from several displays to form empathic decisions that sometimes exceed the standard protocol. We conclude that the role of tacit information holds opportunities for future flight systems and should be considered in a holistic approach to individualized workspaces for controllers.

In visually rich working environments, it is important for operators to distribute their visual attention in an optimal fashion in order to operate safely. Computational models can provide a systematic method of investigating the attention distribution of humans. In this article, we reanalyze eye tracking data from anesthesiologists when inducing general anesthesia to test whether the so-called expectancy value version of the salience, effort, expectancy, value (SEEV) model can accommodate the visual attention distribution of anesthesiologists, and to investigate the effect of case (real versus simulated cases) and experience (junior versus senior) on the expectancy value model fit. The overall model fit is good (predicted-observed percentage dwell time correlation of 0.810, R2= 0.656). We observe that the model fit is better in simulated cases than real ones. In addition, the model fit is good for junior anesthesiologists independent of the case, but that there is an even better model fit in simulated cases than in real ones for senior anesthesiologists (case × experience interaction). Overall, the expectancy value model can be validated. However, at least within the context of anesthesiology, the full SEEV model may be needed to capture the large and distractive visual work environment of anesthesiologists. From a practical point of view, previous research suggested that anesthesiologists pay more attention to monitoring in simulated cases. However, the SEEV analysis suggests that anesthesiologists do not pay extra attention to monitoring equipment in simulated cases, but may not be able to pay enough attention to monitoring equipment in real cases.

Cognitive aids - artefacts that support a user in the completion of a task at the time - have raised great interest to support healthcare staff during medical emergencies. However, the mechanisms of how cognitive aids support or affect staff remain understudied. We describe the iterative development of a tablet-based cognitive aid application to support in-hospital resuscitation team leaders. We report a summative evaluation of two different versions of the application. Finally, we outline the limitations of current explanations of how cognitive aids work and suggest an approach based on embodied cognition. We discuss how cognitive aids alter the task of the team leader (distributed cognition), the importance of the present team situation (socially situated), and the result of the interaction between mind and environment (sensorimotor coupling). Understanding and considering the implications of introducing cognitive aids may help to increase acceptance and effectiveness of cognitive aids and eventually improve patient safety.