HealthMore in Health→
Machine Learning Uncovers Hidden Self-Harm History in Veterans' Medical Records
Key Takeaways
- Standard diagnosis codes capture only about one-fourth of clinically documented self-harm history in VHA records.
- Machine learning analysis of 1.3 million veterans' records estimated 7.9% had documented self-harm, vs. 1.85% from codes alone.
- Problem lists also show gaps: only 22.6% of those with a self-harm code had it on their problem list.
- Better measurement can improve mental health service planning and clinical awareness.
DE
DT Editorial Team··via medicalxpress.com