In the mental health world, predictive analytics are being applied to help prevent suicide. According to Kaiser Permanente, utilizing the data stored in EHRs provides a powerful “treasure trove to support risk detection.” They conducted a study with Mental Health Research Network, utilizing EHR data. They discovered that: “In the 90 days following an office visit, suicide attempts and deaths among patients in the highest 1 percent of predicted risk were 200 times more common than among those in the bottom half of predicted risk.”
To find this, the research team incorporated data points like prior suicide attempts, diagnoses, prescriptions, inpatient or emergency room care, and scores from a standardized depression questionnaire. Other models that do not rely on predictive analytics and use fewer data points are less accurate. First author on the report, Gregory E. Simon, MD, MPH said that “We demonstrated that we can use electronic health record data in combination with other tools to identify people at high risk for suicide attempt or suicide death.”
While it may not be surprising that those who were considered “high risk” were indeed at a higher risk of self-harm, it proves how powerful predictive analytics can be. If mental health and substance use health providers are able to tell when a patient is more likely to hurt themselves, they can adjust their sessions and medications accordingly.