Machine learning tool helps identify suicide risk factors: Study

In a first, researchers have used machine learning (ML), and health data from the entire Danish population to create sex-specific suicide risk profiles, an advance that may help predict the chances of someone taking their own life.

The researchers, including those from Boston University in the US, used data from the whole population of Denmark, and analysed it with a machine-learning system to identify suicide risk factors.

The study, published in the journal JAMA Psychiatry, found that physical illness and injuries raised the risk of suicide in men, but not in women.

The research also revealed a range of other insights into the complex factors that may increase a person's risk of committing suicide.

"Suicide is incredibly challenging to predict, because every suicide death is the result of multiple interacting risk factors in one's life," said study lead author Jaimie Gradus, an associate professor of epidemiology at Boston University.

The study noted that Denmark has a national health care system with the entire population's clinical information compiled in government registries.

Using this, Gradus and his team looked at thousands of factors in the health histories of all 14,103 individuals who died from suicide in the country from 1995 through 2015.

They also assessed the health histories of 2,65,183 other Danes in the same period, using a machine-learning system to look for patterns.

The findings confirmed previously-identified risk factors, such as psychiatric disorders and related prescriptions.

The research team also found new potential risk patterns, including that diagnoses and prescriptions four years before a suicide were more important to prediction than those six months before.

The scientists added that physical health diagnoses were particularly important to men's suicide prediction but not women's.

According to the researchers, this study do not create a model for perfectly predicting suicide in part because medical records rarely include the more immediate experiences like the loss of a job or relationship, which combine with these longer-term factors to precipitate suicide.

Another limitation of the study, they said, is that the risk factors and patterns may also be different outside of Denmark.

Gradus said the findings may, however, point to new factors that can be examined to prevent this persistent public health issue.