A machine learning algorithm that predicts a suicide attempt has recently undergone a prospective trial at the institution where it was developed, Vanderbilt University Medical Center. The results of the trial have been published in the JAMA Network Open.
The algorithm, called the Vanderbilt Suicide Attempt and Ideation Likelihood (VSAIL) model, uses routine information from electronic medical records (EHRs) to calculate the risk of 30-day visits per suicide attempt and, by extension , suicidal ideation.
During the 11-month trial, some 78,000 adult patients were seen at the VUMC hospital, emergency room and surgical clinics.
When stratifying adult patients into eight groups based on their suicide risk scores according to the algorithm, the upper stratum alone accounted for more than a third of all documented suicide attempts in the study and about half of all cases suicidal ideation.
As documented in the EHR, one in 23 people in this high-risk group reported suicidal thoughts and one in 271 attempted suicide.
Currently, suicide has increased in the new generations in many first world countries. But, even without taking such peaks into account, in countries like Spain, where there are comparatively few suicides, there are ten times more suicides than homicides. In 2017, more than 47,000 Americans died by suicide and an estimated 1.4 million suicide attempts .
There is an average of 129 suicides in the United States per day, and Tennessee represents a higher than average rate of one suicide every eight hours. Suicide is the 10th leading cause of death in the United States.