Aleksandra Sarcevic

Introducing RFID technology in dynamic and time-critical medical settings

We describe the process of introducing RFID technology in the trauma bay of a trauma center to support fast-paced and complex teamwork during resuscitation. We analyzed trauma resuscitation tasks, photographs of medical tools, and videos of simulated resuscitations to gain insight into resuscitation tasks, work practices and procedures. Based on these data, we discuss strategies for placing RFID tags on medical tools and for placing antennas in the environment for optimal tracking and activity recognition. Results from our preliminary RFID deployment in the trauma bay show the feasibility of our approach for tracking tools and for recognizing trauma team activities. We conclude by discussing implications for and challenges to introducing RFID technology in other similar settings characterized by dynamic and collocated collaboration.

Leadership structures in emergency care settings: A study of two trauma centers

BACKGROUND Trauma resuscitation involves multidisciplinary teams under surgical leadership in most US trauma centers. Because many trauma centers have also incorporated emergency department (ED) physicians, shared and cross-disciplinary leadership structures often occur. Our study identifies leadership structures and examines the effects of cross-disciplinary leadership on trauma teamwork. METHODS We conducted an ethnographic study at two US Level-1 trauma centers, one of which is a dedicated pediatric trauma center. We used observation, videotaping and interviews to contextualize and classify leadership structures in trauma resuscitation. Leadership structures were evaluated based on three dimensions of team performance: defined leadership, likelihood of conflict in decision making, and appropriate care. FINDINGS We identified five common leadership structures, grouped under two broad leadership categories: solo decision-making and intervening models within intra-disciplinary leadership; intervening, parallel, and collaborative models within cross-disciplinary leadership. CONCLUSION Most important weaknesses of different leadership structures are manifested in inefficient teamwork or inappropriate patient care. These inefficiencies are particularly problematic when leadership is shared between physicians from different disciplines with different levels of experience, which often leads to conflict, reduces teamwork efficiency and lowers the quality of care. We discuss practical implications for technology design.

Teamwork Errors in Trauma Resuscitation

Human errors in trauma resuscitation can have cascading effects leading to poor patient outcomes. To determine the nature of teamwork errors, we conducted an observational study in a trauma center over a two-year period. While eventually successful in treating the patients, trauma teams had problems tracking and integrating information in a longitudinal trajectory, which resulted in inefficiencies and near-miss errors. As an initial step in system design to support trauma teams, we proposed a model of teamwork and a novel classification of team errors. Four types of team errors emerged from our analysis: communication errors, vigilance errors, interpretation errors, and management errors. Based on these findings, we identified key information structures to support team cognition and decision making. We believe that displaying these information structures will support distributed cognition of trauma teams. Our findings have broader applicability to other collaborative and dynamic work settings that are prone to human error.

Information handover in time-critical work

Information transfer under time pressure and stress often leads to information loss. This paper studies the characteristics and problems of information handover from the emergency medical services (EMS) crew to the trauma team when a critically injured patient arrives to the trauma bay. We consider the characteristics of the handover process and the subsequent use of transferred information. Our goal is to support the design of technology for information transfer by identifying specific challenges faced by EMS crews and trauma teams during handover. Data were drawn from observation and video recording of 18 trauma resuscitations. The study shows how EMS crews report information from the field and the types of information that they include in their reports. Particular problems occur when reports lack structure, continuity, and complete descriptions of treatments given en route. We also found that trauma team members have problems retaining reported information. They pay attention to the items needed for immediately treating the patient and inquire about other items when needed during the resuscitation. The paper identifies a set of design challenges that arise during information transfer under time pressure and stress, and discusses characteristics of potential technological solutions.

Coordinating time-critical work with role-tagging

A Level-1 US trauma center introduced role-tags in their trauma resuscitation rooms to help team members identify respective medical functions, and to limit the number of people in the rooms to required staff only. We use this in situ experiment with a paper prototype to investigate the role-driven nature of coordination and to identify system requirements for computerized support of role-based coordination in time-critical work. While role information is useful in coordinating time-critical work, our findings show that the current low-tech solution did not provide significant improvement in team coordination. The situations that were most in need of role-identification were the least likely to achieve it because role-tags required work by trauma team members. Similarly, because role-tags allowed workarounds and misuse, they proved ineffective in controlling the number of people in the room. We suggest technological ways of identifying roles to help coordination in the trauma bay.

Who's scribing?: documenting patient encounter during trauma resuscitation

With healthcare moving towards electronic health records, it is important to understand existing work practices to design effective systems. We conducted an observational study in a Level I trauma center to examine the documentation process and the role of the nurse recorder in trauma resuscitation. We identified several difficulties with current recording practices, including the late arrival of the nurse recorder, parallel activities of the trauma team, and multitasking by the recorder. Our observations showed that the recorders role extends beyond archival responsibilities. The recorder, with the help of a paper record, manages the resuscitation process, rather than passively documenting it. Our findings highlighted the complexity of the recorders role and the need to consider documentation in the broader context of trauma teamwork. We proposed a set of design challenges that emphasize important aspects of trauma care to be considered when designing technologies to support the documentation process.

Activity recognition for medical teamwork based on passive RFID

We describe a novel and practical activity recognition system for dynamic and complex medical settings using only passive RFID technology. Our activity recognition approach is based on the use of objects that are specific for a given activity. The object-use status is detected from RFID data and the activities are predicted from the statuses of use of different objects. We tagged 10 objects in a trauma room of an emergency department and recorded RFID data for 10 actual trauma resuscitations. More than 20,000 seconds of data were collected and used for analysis. The system achieved a 96% overall accuracy with a 0.74 F-score for detecting use of 10 common resuscitation objects and 95% accuracy with a 0.30 F Score for activity recognition of 10 medical activities.

Deep Learning for RFID-Based Activity Recognition

We present a system for activity recognition from passive RFID data using a deep convolutional neural network. We directly feed the RFID data into a deep convolutional neural network for activity recognition instead of selecting features and using a cascade structure that first detects object use from RFID data followed by predicting the activity. Because our system treats activity recognition as a multi-class classification problem, it is scalable for applications with large number of activity classes. We tested our system using RFID data collected in a trauma room, including 14 hours of RFID data from 16 actual trauma resuscitations. Our system outperformed existing systems developed for activity recognition and achieved similar performance with process-phase detection as systems that require wearable sensors or manually-generated input. We also analyzed the strengths and limitations of our current deep learning architecture for activity recognition from RFID data.

Understanding visual attention of teams in dynamic medical settings through vital signs monitor use

The purpose of this study was to understand how vital signs monitors support teamwork during trauma resuscitation -- the fast-paced and information-rich process of stabilizing critically injured patients. We analyzed 12 videos of simulated resuscitations to characterize trauma team monitor use. To structure our observations, we adopted the feedback loop concept. Our results showed that the monitor was used frequently, especially by team leaders and anesthesiologists. We identified three patterns of monitor use: (i) periods with a low frequency of short looks (glances) to maintain overall process awareness; (ii) periods with a medium frequency of long looks (scrutiny) to monitor trends in patient status; and (iii) peaks with a high frequency of glances to maintain attention on both the patient and monitor during critical tasks. Approximately 75% of looks were 3 seconds or shorter, but many looks (25%) ranged between 3 and 26 seconds. Our results have implications for improving displays by presenting the status of the patients physiological systems and team activities.

Decision making tasks in time-critical medical settings

We examine decision-making tasks and information sources during fast-paced, high-risk medical events, such as trauma resuscitation. Interviews with surgical team leaders and ED physicians reveal several environmental aspects that make decision making difficult, including diagnostic tradeoffs, missing and unreliable information, and managing multiple patients simultaneously. We discuss the implications of these findings for the design of wall displays to support decision making in time-critical medical settings.