Mental Health Adverse Events Predictive Tool
TEAM
Canada: Andrea Waddel (PI), Christo El Morr (PI), Elham Dolatabadi (Co-PI)
Research Assistants: Valentina Tamayo Velasquez, Abdul Hamid Dabboussi, David Wen.
Research Assistants: Valentina Tamayo Velasquez, Abdul Hamid Dabboussi, David Wen.
Project
Patient and staff safety are critical priorities in healthcare, particularly in mental health settings where the risk of adverse events is disproportionately high. Despite significant advancements in patient safety tools across other medical fields, mental health and forensic psychiatry lag behind in leveraging predictive technologies to mitigate risks such as self-harm, aggression, violence, and other safety incidents. Current methods, including voluntary reporting and risk assessment tools, often fall short in accurately predicting and preventing adverse events.
This project seeks to address this gap by developing a machine learning (ML)-based predictive model tailored for mental health environments. By utilizing electronic medical record (EMR) data from a large mental health hospital with forensic and non-forensic units, the model will predict patient deterioration and adverse events, enabling real-time alerts and fostering timely interventions. Inspired by successful applications like CHARTwatch in general internal medicine, this project aims to adapt ML-driven early warning systems (EWS) to meet the unique needs of mental health care.
This project seeks to address this gap by developing a machine learning (ML)-based predictive model tailored for mental health environments. By utilizing electronic medical record (EMR) data from a large mental health hospital with forensic and non-forensic units, the model will predict patient deterioration and adverse events, enabling real-time alerts and fostering timely interventions. Inspired by successful applications like CHARTwatch in general internal medicine, this project aims to adapt ML-driven early warning systems (EWS) to meet the unique needs of mental health care.