Back

Development and validation of the Behavioral Health Acuity Risk model: a predictive model for suicide prevention through clinical interventions

Digumarthi, V.; Strange, H. E.; Norman, H. B.; Ayers, D.; Patel, R.; Hegarty, K. E.

2022-12-22 health informatics
10.1101/2022.12.21.22283796 medRxiv
Show abstract

Common suicidal ideation screening tools used in healthcare settings rely on the willingness of the patient to express having suicidal thoughts. We present an automatic and data-driven risk model that examines information available in the medical record captured during the normal course of care. This model uses random forests to assess the likelihood of suicidal behavior in patients aged seven or older presenting at any healthcare setting. The Behavioral Health Acuity Risk (BHAR) model achieves an area under the receiver operating curve (AUC) of 0.84 and may be used on its own or as a component of a comprehensive suicidal behavior risk assessment.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.