Open Access

ARTICLE

Longitudinal Relationship between Emotional Disorder and Non-Suicidal Self-Injury in Chinese University Students: A Random Intercept Cross-Lagged Panel Model

Xin Xu¹, Shaobo Liang^2,3,*

1 School of Journalism and Communication, Shanghai University, Shanghai, China
2 Faculty of Education, University of Macau, Macau, China
3 School of Humanities, Dalian University of Technology, Dalian, China

* Corresponding Author: Shaobo Liang. Email:

International Journal of Mental Health Promotion 2026, 28(9), 10 https://doi.org/10.32604/ijmhp.2026.082789

Received 23 March 2026; Accepted 28 May 2026; Issue published 22 September 2026

Abstract

Objectives: Whether emotional disorder (ED) and non-suicidal self-injury (NSSI) influence each other over time remains underexplored, particularly at the within-person level. This study examined the reciprocal longitudinal associations between ED and NSSI among mainland Chinese university students. Methods: Three-wave panel data were collected at six-month intervals from 574 Chinese university students at Wave 1 (51.22% female; mean age = 22.58 years, SD = 2.89). ED was assessed using the Multidimensional Emotional Disorder Inventory (MEDI), and NSSI frequency was measured using a validated six-item scale. A Random Intercept Cross-Lagged Panel Model (RI-CLPM) was used to separate stable between-person differences from within-person longitudinal effects. Missing data were handled using full-information maximum likelihood. Longitudinal measurement invariance was tested before examining cross-lagged associations. Results: At the between-person level, students with persistently higher ED also reported higher overall NSSI frequency (β = 0.487, p < 0.001). At the within-person level, ED at Wave 2 significantly predicted increased NSSI at Wave 3 (β = 0.351, p < 0.05), whereas the Wave 1 to Wave 2 path was not significant. Conversely, NSSI at Wave 1 predicted higher ED at Wave 2 (β = 0.183, p < 0.05), but this effect was not replicated from Wave 2 to Wave 3. Longitudinal measurement invariance supported the comparability of constructs across waves. Conclusions: ED and NSSI showed evidence of reciprocal within-person associations, although the effects were interval-specific rather than consistently bidirectional across all waves. These findings suggest that university counselling services should address both transdiagnostic emotional distress and self-injury behaviors in order to target the interval-specific associations identified here. Culturally sensitive interventions may be particularly important in the Chinese university context.

---

Keywords

---