ONTARIO SHORES DRAMATICALLY MITIGATES PATIENT AGGRESSION
WITH DATASKILL PREDICTIVE TECHNOLOGY.
The mental health center leverages DataSkill ACUMI to forecast patient aggression with accuracy and reliability.
Violent behavior. Staff and patient injuries. Suicide attempts. Litigation.
Medical institutions, especially mental health facilities, must deal with any number of these issues on a regular basis. According to Medscape.com, the cost to businesses from workplace violence has been estimated at $120 billion a year. The average jury award, in subsequent liability cases where the employer failed to take proactive, preventive measures under OSHA guidelines, is $3.1 million per person, per incident.
Ontario Shores Centre for Mental Health Sciences, a public teaching hospital providing specialized assessment and treatment services to those living with serious mental illness, has a rich history of providing exemplary patient care. The staff delivers this through safe and evidence-based approaches where successful outcomes are achieved using best clinical practices and the latest advances in research. Simply put, Ontario Shores is not willing to compromise when it comes to providing patients with a recovery-oriented environment of care built on compassion, inspiration and hope
Ontario Shores, like other organizations in the healthcare industry, produced an array of structured and unstruc- tured patient data which had analytical potential far beyond its current uses. The existing process for mitigating patient aggression was time consuming and inaccurate. This resulted in extra time and resources dedicated to managing patient health records and, more importantly, mitigating patient aggression. Furthermore, restraints and seclusion are not ideal approaches to deal with aggression because they are reactive and reduce the quality of life for patients. Ontario Shores wanted to provide the optimal, safest environment possible for its patients and clinical staff—on a unified platform—so a search for a proven solution began.
The existing method for tracking and addressing patient violence was misunderstood and not well implement- ed—we needed a more quantitative approach.
-Sanaz Riahi, Senior Director of Professional Practice and Clinical Information, Ontario Shores
Ontario Shores engaged DataSkill to explore a predictive model for identifying which patients would most likely have incidents of aggression toward staff or co-patients. After much vetting, demonstration and testing, Ontario Shores implemented the analytics-driven DataSkill ACUMI solution to meet this critical objective.
ACUMI software predicts and facilitates mitigation measures for clinical aggression by using industry-leading machine learning algorithms and natural language processing (NLP) in a scalable, predictive model. Not only does this revolutionary approach mitigate legal risk, but it improves the quality of life for patients and provides safety for the clinical staff. One of the greatest innovations to this approach is leveraging clinical notes, which indirectly adds the intuition of clinical staff to a host of other predictive factors. The entire process is executed on a simplified, single platform.
The ACUMI predictive model for clinical aggression now provides Ontario Shores with a daily ranked list of high, medium, and lower-risk patients. This report is based on both historical and recent data on the patient, including potential triggers for aggression, giving insight to clinical staff to develop a risk mitigation plan. Today, the center successfully leverages ACUMI to forecast patient aggression by capturing the knowledge of their best clinicians and using that for the best possible decision making across the organization.
DataSkill technology enables Ontario Shores to pursue its ongoing mission to adorn its patients with respect and dignity, not restraints.
- Behavioral Sciences
ACUMI Product + IBM Solution
- ACUMI DataFind
- ACUMI DataLearn
- IBM Watson
- Watson Explorer
- SPSS Modeler
- Ontology Generation
- Concept Extraction
- Unstructured Data Analytics
- Predictive Modeling
- Neural Networks
- Deep Machine Learning
- On Premises installation of hardware and run over network
- Virtual machine must have 300GB of available hard drive space and at least 8MB of ram.
- Virus scanning must be turned off for the folder containing WEX data
- Client is responsible for the process that runs daily, which adds to the SQL database any new data from the past day. This must be a fast process, not removing/replacing all data, but incrementally loading the new data.