Shock Patients: Interprofessional Communication to Enhance Diagnosis (SPICED)

PROJECT SUMMARY Over 1.3 million Americans develop shock annually and 30-50% do not survive depending on the underlying cause of shock. Most research has focused on recognizing shock but has stopped short of identifying the cause of shock. Differentiating between causes of shock such as infection, heart failure, bleeding, dehydration, or lack of stress hormone, is extremely challenging. The complex presentation of a patient in shock and the many different potential causes of shock requires many expert interprofessional clinicians, such as nurses, pharmacists, emergency medicine, critical care, infectious disease and cardiology doctors, to weigh in. Yet, there is no way for these clinicians to actively share their assessments, build a shared understanding of the cause of shock, and reach a consensus on treatment. Poor communication among clinicians is a leading source of harm in acute care hospitals, yet remains extremely understudied. This application’s long-term objective is to improve the safety of critical care medicine by addressing poor interprofessional communication. The Specific Aims of this application will be to: 1) understand why interprofessional clinicians disagree on the cause of shock; 2) design a systematic process for interprofessional clinicians to share information, their assessments and build a shared understanding of the cause of shock; and 3) put into clinical practice and measure how this systematic process improves agreement among clinicians as well as timeliness and accuracy of the identification of the cause of shock. The central hypothesis is that creating a process, informed by the interprofessional clinicians who would use it, to bring together all experts in order to determine the cause of shock quickly after a patient shows signs of shock, can improve agreement among clinicians, timeliness and accuracy in identifying the cause of shock. We will apply a variety of methods to test this hypothesis including machine learning, critical incident technique, user-centered design, implementation science and interrupted time series analysis. Identifying modifiable healthcare system factors that lead to disagreement among clinicians can inform the design of a systematic process to identify the cause of shock, determine how to treat it and reduce the high risk of death.