Given the rise in reported suicide in the context of the national pediatric mental health emergency that our patients continue to endure, an article being early released this week in Pediatrics entitled, “Serious Self-Harm Events in Children and Adolescents” by Sekmen et al (10.1542/peds.2022-059817) provides a detailed view of factors associated with serious self-harm among children and adolescents. Having more detailed/granular suicide risk profiles can, according to the authors, help providers better predict the likelihood of serious self-harm.
Sekmen et al delineated four a priori psychiatric comorbid profiles (low, moderate, high, and very high risk), using a Bayesian profile regression among nearly 1,098 children ages 5-18 who were hospitalized with a neuropsychiatric event during a 4-year period that concluded just prior to the declaration of the COVID-19 pandemic. The authors incorporated age, sex [sic], medical, and psychiatric [sic] diagnoses together (as opposed to associating risk of self-harm with individual and independent factors such as prior psychiatric diagnosis) into their formula for estimating risk of suicide. It is interesting to note that serious self-harm events were defined by the authors as any acute suicidal ideation or intentional self-harm event given that the former (ideation) doesn’t necessarily equate with a true suicide attempt.
Each profile had the following characteristics:
- Low-risk: Ages 5-9 without trauma/substance-related disorders nor major mood, behavioral or developmental disorders
- Moderate-risk: Children with mood disorders and/or behavioral disorders without depression and represented the baseline/reference group
- High-risk: Females 14–17-years-old with depression and anxiety in conjunction with substance- and trauma- related disorders
- Highest-risk: 10–13-year-old male children who also had bipolar, ADHD, and or trauma-related and developmental disorders
While knowing these self-harm risk profiles can be instructive, their use must be tempered by the fact the categories were drawn from a retrospective (secondary analysis of a previous cross-sectional study) sample of patients who were already hospitalized for a neuropsychiatric event. Being mindful of the stigma that already exists with regard to mental health, we must be cautious in using the data to make prospective determinations of suicide risk. However, this study does give insight about certain individuals who may be at increased risk of self-harm and can help guide future research. Looking ahead, it would be helpful to determine how accurate the model is among patients admitted on/after the declaration of the COVID-19 pandemic as the pandemic itself could be a factor in the model.