Predicting clinical aggression in mental health patients (#2716)

Idea Description
Supplementary Information
Innovation 'Elevator Pitch':
A predictive model for clinical aggression providing a daily ranked list of high, medium, and lower-risk patients - giving insight to clinical staff to develop a risk mitigation plan.
Overview of Innovation:
Traditional data in mental health is based on last episode, meds, structured questionnaire etc. However, nurses written observations provide valuable unstructured data (e.g. refused meds today, had no visitors, unresponsive etc.) for predictive analysis.
This solution uses Content Classification on historical data to correlate to aggression. The output file is fed into SPSS Modeller to create a predictive model. The model provides the most important factors (which can be confirmed by clinicians).
The model runs against the client’s data and provides reports identifying patients at risk of becoming aggressive – the nurses and clinicians can then take actions to prevent this. In a pilot client (in Canada) the model predicted the exact number and names of patients who became aggressive.
Stage of Development:
Market ready and adopted - Fully proven, commercially deployable, market ready and already adopted in some areas (in a different region or sector)
Similar Content
No similar content found

Created by

Share and Follow