Second-Order Adaptive Network Models for Shared Mental Models in Medical Teamwork for Neonatal Resuscitation

Abstract

A medical error indicates that the expected action has not been completed in the medical treatment. Patient safety defined as ‘the prevention of harm to patients’ is a vital concept in medical teams and is essential in improving quality of medical treatment. Safety culture is defined as “the product of individual and group perceptions, attitudes, values, competencies, and patterns of behavior that decide the commitment to, and the style and proficiency of, an organization’s health and safety management”. The terminology for safety has evolved from a focus on error to patient safety to adopting a safety culture with adequate shared mental models. That evolution reflects a process of shifting responsibility from individuals to the whole organization. More recently, scholars is advocating medical organization to get rid of blame culture and embrace just safety culture. Blame culture is a set of organizational norms and attitudes characterized by being reluctant to accept responsibility or to take risks for errors because of being afraid of criticism or management blame. Just safety culture refers to a culture that encourages open dialogue to promote safer practices. On the face of it, healthcare organizations should quickly abandon blame culture and choose a just culture. However, it’s not easy to deploying a just safety culture in reality. Good implement of just safety culture in a medical team requires much teamwork effort. When it comes to teamwork effort, the shared mental model has received a lot of attention in the literature on medical team performance. A mental model means the intelligence that imitates relation structures of external processes. A shared mental model of a team means a large or complete overlap of the team members’ individual mental models. A useful method to analysis shared mental models of a team is adaptive network-oriented modeling, which refers to modeling complex processes by adaptive networks. The AI technology has the potential to help a medical team deploy just safety culture. But research on how AI can involve in the workflow of existing medical team is limited. Thus, in hoping of solving the problem: How can AI participate in the application of just safety culture to neonatal resuscitation? This study chooses these points to explore: how AI improves the efficiency of team communication, how AI leverages organizational learning, and how AI points out defects and prevents errors. From the perspective of result, these converge into how AI can help improve the performance of the medical team. That leads to the main research question of this research: How to use an adaptive network-oriented modeling method to analysis the AI-coach’s contribution in a medical team’s ventilation operation to save baby in danger?...