Systemic Inflammatory Biomarkers and Non-Suicidal Self-Injury in Adolescents with Mental Disorders: A Threshold Effect Analysis

Background Non-suicidal self-injury (NSSI) is a prevalent and serious psychiatric concern among adolescents. Emerging evidence implicates systemic inflammation in the pathophysiology of mood disorders and impulsivebehaviors; however, the specific association between peripheral inflammatory biomarkers and NSSI in adolescents remains poorly characterized. This study aimed to evaluate the predictive value of six systemic inflammatory indices for NSSI and to explore potential non-linear threshold effects in a clinical adolescent population. Methods This cross-sectional study enrolled 500 adolescents hospitalized with psychiatric disorders (316 with NSSI, 184 without NSSI) between June 2024 and June 2025. Six systemic inflammatory markers—neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio (MLR), Systemic Immune-Inflammation Index (SII), Systemic Inflammatory Response Index (SIRI), and Aggregate Index of Systemic Inflammation (AISI)—were calculated from routine blood tests. Multivariable logistic regression, generalized additive models (GAM), and two-piecewise linear regression were employed to assess both linear and non-linear associations after adjusting for potential confounders. Results Adolescents with NSSI exhibited significantly elevated levels of NLR, PLR, and SII compared to the non-NSSI group (allp < 0.05). After full adjustment, the highest tertiles of PLR (OR = 2.05), SII (OR = 2.30), SIRI (OR = 1.97), and AISI (OR = 2.39) were independently associated with increased NSSI risk. GAM analysis revealed significant non-linear relationships, and threshold effect analysis identified critical breakpoints for PLR (K = 73.26), SIRI (K = 1.72), and AISI (K = 616.92), below which NSSI risk increased sharply. Among all markers, SII demonstrated the highest predictive performance (AUC = 0.584). Conclusions Composite inflammatory indices, particularly SII and AISI, are robust independent predictors of NSSI in adolescents with mental disorders. The identification of non-linear threshold effects provides novel insights into the inflammation-NSSI relationship and supports the clinical utility of these cost-effective biomarkers for early risk stratification.