Prevalence and psychosocial interventions of mental disorders following traumatic spinal cord injury: a systematic review and meta-analysis based on knowledge graphs.

BACKGROUND

Evaluating and addressing the mental and behavioral health issues of patients with spinal cord injury (SCI) during the rehabilitation process is essential. This systematic review and meta-analysis investigated the prevalence of mental disorders post-traumatic SCI and evaluated psychosocial interventions, leveraging a Knowledge Graph of Spinal Cord Injury (SCIKG) to overcome limitations of traditional literature searches.

METHODS

The SCIKG integrated structured ontologies (SNOMED CT, UMLS, MeSH) to enable hierarchical semantic reasoning across mental disorder subconcepts. SPARQL queries identified literature from PubMed, Embase, PsycINFO, Web of Science, and Cochrane Central up to July 2025, supplemented by citation tracing. Two researchers independently screened studies against predefined criteria. Prevalence was pooled using DerSimonian-Laird random-effects models (RevMan 5.4; Stata metaprop) to account for heterogeneity (I4 > 50%). SCIKG quality was assessed across Coverage, Accuracy, Freshness, and Consistency dimensions

RESULTS

From 16,861 screened records, pooled prevalence estimates of anxiety, depression, post-traumatic stress disorder (PTSD), cognitive impairments, suicide ideation, drug abuse, and sleep dysregulation post-SCI were 23% (95% confidence interval (CI), 21%-26%), 24% (95% CI, 22%-26%), 25% (95% CI, 21%-29%), 55% (95% CI, 27%-82%), 18% (95% CI, 12%-24%), 36% (95% CI, 30%-42%), and 41% (95% CI, 8%-73%), respectively. Subgroup analyses revealed significant variations: Western cohorts reported 30% higher anxiety than Asian populations; self-report tools overestimated depression by 40% versus clinical interviews; violence-related SCI increased PTSD risk by 75%. Psychosocial interventions mainly focused on specific subgroups in SCI patients with increased levels of pain, psychological distress, or pressure ulcers. Studies indicated inconsistent enhancements in participants' psychosocial adaptation, cognitive appraisal, and mental health. However, no significant effect was observed on their coping abilities.

CONCLUSIONS

Mental disorders are prevalent and heterogeneously distributed post-SCI, influenced by cultural, methodological, and injury-phase factors. Knowledge graphs enhance evidence synthesis precision, revealing critical gaps in assessment standardization. Routine screening with validated, culturally adapted tools is warranted. Future research must prioritize phase-stratified designs, SCI-specific diagnostic thresholds, and large-scale trials of tailored psychosocial interventions.

CLINICAL TRIAL NUMBER: Not applicable.