The profiles were developed from their study of children and adolescents aged 5-18 years who had been admitted with a neuropsychiatric event to two children’s hospitals.
The researchers used Bayesian regression to identify the profiles developed from 32 covariates: age, sex, and 30 mental health diagnostic groups from April 2016 to March 2020. The profiles include low-, moderate-, high- and very-high-risk categories.
The study, led by Mert Sekmen with the division of hospital medicine at Monroe Carell Jr. Children’s Hospital, and a student at Vanderbilt University Medical Center in Nashville, Tenn., included 1,098 children, average age 14. Of those, 406 (37%) were diagnosed with a self-harm event.
Traditionally, single diagnoses have been linked with risk of self-harm, independent of other comorbidities, but this study gauges risk for a set of diagnoses.
Findings were published online in Pediatrics.
The risk groups were described as follows:
- Low risk. (45% of the study population; median risk of 0.04 (interquartile range, 0.03-0.04; odds ratio, 0.08). The group included children aged 5-9 years with a non–mental health diagnosis, and without mood, behavioral, psychotic, developmental, trauma, or substance-related disorders.
- Moderate risk. (8% of the study group). This group had the same risk as the baseline risk for the entire cohort (37%) and served as the reference group, with a median risk of 0.30 (IQR, 0.27-0.33). This profile was characterized by several mood disorders and behavioral disorders but without depressive disorders.
- High risk. (36%) This group had an average risk of 0.69 (IQR, 0.67-0.71; OR, 5.09). This profile included female adolescents ages 14-17 with depression and anxiety in conjunction with substance- and trauma-related disorders. Personality and eating disorders were significant in this group. Importantly, the authors wrote, the high-risk group did not include behavioral and developmental disorders.
- Very high risk. (11%) The very-high-risk profile had the highest average risk of 0.79 (IQR, 0.73-0.79; OR, 7.21) and included male children aged 10-13. This profile, like the high-risk profile, included anxiety and depressive disorders. The very-high-risk profile differed from the high-risk with its inclusion of bipolar disorder; attention-deficit/hyperactivity disorder; and trauma-related and developmental disorders such as autism spectrum disorder or intellectual disability, along with conduct disorders. Neither the high- nor the very-high-risk profiles included a concurrent non–mental health diagnosis.
Differences by sex
The authors explained some of the differences by sex. They noted that in a study of children aged 5-11, deaths by suicide were more prevalent among boys. A mental health diagnosis was identified in 31%, the most common being ADHD, depression, and other unspecified co-occurring disorders.
“The very-high-risk group also reflects a concerning rise in death by suicide among (males) aged 10-13, who have seen rates nearly triple from 2007 to 2017,” the authors wrote.
The authors pointed out that, although incidence of anxiety and depressive disorders between male and female children is much the same before adolescence, “female adolescents are twice as likely to be diagnosed with either disorder during adolescence. Girls also have higher rates of suicidal ideation and attempts after puberty.”
Eating disorders were also included in the high-risk profile. A study showed that emergency department visits for adolescent girls attempting suicide were 51% higher from February to March 2021, compared with the same period in the pre-COVID-19 year 2019.
Jason Lewis, PhD, psychologist and section director of mood, anxiety and trauma disorders in the department of child and adolescent psychiatry and behavioral sciences at Children’s Hospital of Philadelphia, who was not part of the research team, said the “constellations of risk factors put into acuity levels” helps to better project risk than knowing the risk associated with a particular diagnosis.