Psychosocial Predictors of Spinal Cord Injury Pain: The Canadian Spinal Cord Injury Pain Survey

Arch Phys Med Rehabil. 2026 Apr 8:S0003-9993(26)00643-X. doi: 10.1016/j.apmr.2026.04.003. Online ahead of print.

ABSTRACT

OBJECTIVE: The objective of this study was to examine the relationship between sociodemographics, injury characteristics, personality traits, and emotions with pain severity and interference using data from the Canadian Spinal Cord Injury (SCI) Pain Survey.

DESIGN: Cross-sectional community-based survey.

SETTING AND PARTICIPANTS: An online survey completed by 200 English-speaking adults (≥18 years old), with non-congenital SCI living in Canada (age=51±14 years, 51% men), recruited through SCI community and research organizations.

MAIN OUTCOME MEASURES: Pain severity and interference were measured using the Brief Pain Inventory-Short Form subscales. Hierarchical blocks of predictor variables included: 1) sociodemographics; 2) injury characteristics; and psychosocial variables comprising 3) personality traits; and 4) emotions.

RESULTS: For pain severity, age was the only significant predictor in the final model (B=.04, 95% CI [.008, .072], p=.015). Psychosocial variables (blocks 3 and 4 combined) contributed additional variance (ΔR² = .02-.05) but did not significantly improve model fit (p>.05), indicating that psychosocial factors did not meaningfully predict pain severity. The final model explained 19.7% of the variance in pain severity (adjusted R²=.09, p=.03), corresponding to a small to medium effect size (Cohen's f² ≈.10). For pain interference, age (B=.05, 95% CI [.018, .081], p=.002), anger (B=.05, 95% CI [.005, .099], p=.029) and depression (B=.07, 95% CI [.010, .123], p=.022) were statistically significant predictors. Psychosocial variables contributed additional variance (ΔR²≈.11, p<.05), supporting the hypothesis that psychosocial factors contribute meaningfully to pain interference. The final model explained 38.2% of the variance in pain interference (adjusted R²=.30, p<.001), reflecting a very large effect size (Cohen's f² ≈.62).

CONCLUSIONS: Psychosocial variables, particularly emotions, are associated with pain interference but not severity in people with SCI. These factors appear more important than injury-related characteristics and highlight potentially modifiable targets for intervention and self-management and support the integration of psychosocial treatment in SCI pain care.

PMID:41962647 | DOI:10.1016/j.apmr.2026.04.003