What's Happening?
Researchers from the University of New Mexico School of Medicine have developed a machine learning method to detect hidden self-harm histories in veterans' health records. Analyzing over 1.3 million electronic health records from the Veterans Health Administration,
the study found that traditional diagnosis codes captured only 1.85% of patients with self-harm histories. In contrast, the machine learning approach estimated this figure at 7.9%, revealing a significant underestimation. The study, published in the Journal of Medical Internet Research, highlights the potential of machine learning to improve the accuracy of health data and inform mental health service planning.
Why It's Important?
This study underscores the limitations of relying solely on diagnosis codes for mental health surveillance and planning. By revealing a substantial visibility gap, the research suggests that many veterans with self-harm histories may not receive the necessary mental health services. The use of machine learning to uncover these hidden cases could lead to more accurate assessments of mental health needs and better resource allocation. This advancement is crucial for improving mental health care for veterans, a group that often faces unique challenges and higher risks of mental health issues.
