United StatesHealthcare and life sciences2017 – ongoingVeterans Health Administration
REACH VET suicide-risk prediction and clinical outreach
VHA outreach to patients in the highest 0.1% suicide-risk tier was associated with fewer suicide attempts
Problem
VHA needed to identify veterans at high suicide risk and connect them with additional care.
Approach
REACH VET used a suicide-risk prediction algorithm to identify patients in the highest local risk tier. Designated clinicians reassessed treatment and contacted patients to improve care. The national program paired prediction with safety planning and closer monitoring.
Results
Takeaway
The evaluation covered a clinical intervention triggered by risk scores. It found associations with better engagement and fewer attempts, without detecting a mortality difference.
Categories
- Size
- Large enterprise
- Industry
- Healthcare and life sciences
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Built in-house
- Goal
- Quality and risk
- Evidence
- Independent study
Sources
Source summaries
VA reproduced a peer-reviewed cohort evaluation covering REACH VET entries beginning in March 2017. The study associated the program with improved engagement and fewer suicide attempts, but found no detectable difference in suicide or all-cause mortality.
Full details are in the original sources.