These Early Childhood Risk Factors Can Help Predict Self-Harm in Teens

Nearly one in five adolescents is thought to self-harm at least once in their lifetime. But despite its pervasiveness, relatively little is known about how to accurately predict which teens are at risk.

Now, researchers from the University of Cambridge are trying to change that by pinpointing early life risk factors that predict self-harm during adolescence. And some of these risk factors were apparent in kids as early as age 5.

In analyzing data on youth in the U.K., researchers identified two subgroups of adolescents who self-harm. Each group showed childhood risk factors that could have predicted self-harm a decade earlier.

The method they used, which involved machine learning, could pave way for earlier intervention. Study author Stepheni Uh, a PhD candidate at the University of Cambridge, tells Verywell that the models clinicians currently use to treat self-harm are primarily reactive.

"We wait for problems to escalate, and this really overwhelms the system," Uh says. "We recognize that there needs to be some sort of shift towards more preventive or proactive models to treat self-harm, and that was the main motivator of this study."

The study was published in the Journal of the American Academy of Child & Adolescent Psychiatry in early May.

What Are the Risk Factors for Self-Harm?

Researchers used data from the Millennium Cohort Study, a large, ongoing developmental study of young people throughout the U.K. They analyzed behavioral, socioemotional, and physical data on participants beginning when they were 9 months old.

In narrowing down the data, Uh and colleagues included 10,827 participants. By the age of 14, all participants had completed responses to two different questionnaires used to evaluate risk factors for self-harm and depression.

Of the almost 11,000 respondents, 1,580 (73% female) reported self-harm. However, researchers ended up limiting their analysis to 1,284 participants (74% female) because these individuals fit into the two distinct behavioral clusters.

They also compared these participants to a random subsample of 900 participants who did not self-harm as reference.

Two Major Self-Harm Pathways

Researchers identified two distinct developmental pathways of young people who self-harm by the age of 14:

1. The "psychopathology" pathway: A smaller group with a long history of emotional and behavioral difficulties.
2. The "adolescent risky behavior” pathway: A larger group without a long history of emotional and behavioral difficulties, but with different risk factors that showed up later in life.

Individuals in the first subgroup more commonly reported being bullied. They also shared they were more likely to struggle with emotion regulation and that their caregivers struggled with emotional difficulties as well.

Uh says that individuals in this group might fit more into society's expectations of people who self-harm.

But this first group was much smaller than the second: Only 379 individuals, or about a third of all those who self-harmed, showed these patterns.

In the second, larger subgroup, individuals did not show predictable patterns in childhood. However, over time, they showed a greater willingness to take part in risky behavior, as well as reported feeling less security with friends and family at age 14.

They also performed lower on the Mood and Feelings Questionnaire compared to the control group, indicating poorer mental health. At age 11 they reported greater concern about the feelings of others, compared to other groups.

Given these different groups, Uh says, "this is not one homogenous group of young people who self-harm." However, across all participants who self-harmed, many reported sleep difficulties and low self-esteem at age 14.

Limitations

In analyzing the data, researchers still acknowledge various limitations. First, self-harming behavior was a yes or no question, and the research did not capture nuance or motive.

Also, the sample was overwhelmingly White (90%) and based out of the U.K. "This limits our capability to investigate potential intersectionality between ethnicity and self-harm risk profiles, and highlights the importance of replicating our study with other nationally representative samples," the authors write.

The Importance of Identifying Risk

Studies like this, Uh says, can shift the mental health conversation from treatment to prevention.

"I think that's where there's this big push for machine learning and AI," she says, noting the algorithms can make it easier to identify patterns and subgroups. "We can say that we have at least two, but there are likely to be more that are at-risk."

This type of study can help fill a gap since most clinical services currently available for self-harm are "mostly responsive."

Preventive Measures

Overall, Uh and colleagues identified several specific interventions to prevent self-harm in adolescents, including:

• Anti-bullying interventions in schools
• Emotion regulation training
• Screening adolescents for self-harm on a regular basis
• Sleep training

Bullying, the authors write, "casts a shadow over adult as well as childhood mental health." They mention that there are many evidence-based anti-bullying interventions that can, and should, be deployed at a school level. After all, this study found that being bullied by age 11 made self-harm 30% more likely.

"Anti-bullying interventions at a school level, that are properly resourced, can be really effective at an assortment or different things, but also for self-harm," Uh adds.

Teachers, too, should be trained in how to respond when they learn a student is self-harming. This is important, the authors write, considering that teachers "are often the first people to hear about self-harm but may have difficulty responding."

For emotion regulation, Uh says that creating more of a supportive environment from an early stage, in the school and home, can help. The study found that individuals were between 30 and 50% more likely to self-harm if they had experienced emotion regulation difficulties.

Additionally, adolescents could be asked about self-harming behavior at annual physicals, "with careful consideration of non-stigmatizing language," the authors write. That is, clinicians should be trained in how to ask about it without making it sound like the individual's fault.

And since sleep difficulties were a strong predictor of self-harm across the board, sleep training could prove useful.

"A lot of researchers are becoming more interested in the effects of sleep," Uh says. "This could be really good to look into in terms of finding interventions to support good sleep habits even at a young age. It could be very important for future development."

The bottom line, Uh says, is that identifying patterns can help to prevent self-harm in the difficult period of adolescence. "Many of the risk factors we identified are tractable," she says. "They are apparent years before self-harming behavior is reported."