University of New Mexico Study Reveals Underreported Self-Harm Histories in Veterans Using Machine Learning
The University of New Mexico School of Medicine conducted a study analyzing electronic health records of over 1.3 million patients from the Veterans Health Administration. The study, published in the Journal of Medical Internet Research, found that traditional diagnosis codes identified self-harm history in only 1.85% of patients. However, a novel machine learning method developed by the researchers estimated that 7.9% of patients had documented self-harm histories, indicating a significant underreporting. The study combined machine learning with expert chart review and statistical calibration to achieve these results. Additionally, it was found that among veterans with a diagnosis code for self-harm, only 22.6% had self-harm listed on the VHA problem list. This discrepancy highlights potential gaps in mental health service needs and planning. 
