Clinical Report: Development of a Machine Learning Nomogram for NSSI Risk
Overview
This study developed and validated a nomogram to estimate the risk of non-suicidal self-injury (NSSI) in adolescents with depression using data from a multicenter cohort. The nomogram incorporates eight key variables.
Background
Non-suicidal self-injury (NSSI) is a significant concern among adolescents with depression, often linked to severe mental health issues and increased suicide risk. Current tools for assessing NSSI risk are limited. This study creates a machine learning-based nomogram to aid clinicians in evaluating NSSI risk.
Data Highlights
Variable
Importance
Depression Score
High
Sleep Medication Use
Moderate
Difficulty Identifying Feelings
High
Age
Moderate
Perceived Family Support
Moderate
Gender (Female)
High
Hallucination
High
Externally Oriented Thinking
Moderate
Key Findings
The nomogram integrates eight clinical and psychosocial variables to estimate NSSI risk.
AUC values for the nomogram were 0.754 in the training cohort and 0.748 in the validation cohort.
Variables identified include depression score, sleep medication use, and perceived family support.
The model was validated using calibration curves and decision curve analysis.
Over 80% of individuals engaging in NSSI have comorbid mental disorders.
Clinical Implications
The developed nomogram provides a structured approach for clinicians to assess NSSI risk in adolescents with depression.
Conclusion
The study presents a validated nomogram that can assist healthcare professionals in estimating NSSI risk among depressed adolescents.
