Data included in EHRs may provide accurate predictions of a patient's future risk of suicidal behavior, according to new research published in The American Journal of Psychiatry.
Using EHR data from a 15-year period (1998 to 2012), researchers with Boston Children's Hospital Informatics Program developed a model to see if this longitudinal historical data could predict future suicidal behavior. They defined suicidal behavior using expert clinical consensus review, supplemented with state death certificates.
The researchers' model achieved 33 percent to 45 percent sensitivity, or true positive rate, and 90 percent to 95 percent specificity, or true negative rate. Additionally, the model predicted patients' future suicidal behavior an average of three to four years in advance.
Some of the strongest predictors of future suicidal behavior risk include substance abuse, psychiatric disorders, certain injuries and chronic conditions.
"These findings suggest that the vast quantities of longitudinal data accumulating in electronic health information systems present a largely untapped opportunity for improving medical screening and diagnosis," the researchers wrote. "Beyond the direct implications for prediction of suicide risk, this general approach has far-reaching implications for the automated screening of a wide range of clinical conditions for which longitudinal historical information may be beneficial for eliminating clinical risk."
More articles on analytics:
Carilion Clinic, PeraHealth partner to deploy predictive analytics solutions
University of Iowa Hospitals uses predictive analytics to curb SSIs
Are today's hospitals underinvesting in their ability to transform data into actionable information?