Abstract
BACKGROUND: Resilience plays a critical role in adaptation following spinal cord injury (SCI), yet there is no consensus on how resilience is defined or measured. Current unidimensional approaches may miss important patterns in resilience responses that could inform targeted interventions. OBJECTIVES: To identify distinct subgroups of individuals with SCI based on resilience patterns using latent class analysis (LCA) and examine their associations with demographic, injury-related, and psychosocial factors. METHODS: We conducted a cross-sectional analysis of 10,634 adults with SCI from the National SCI Model System Database who completed the Spinal Cord Injury-Quality of Life Resilience short form (SCI-QOL-R-SF) between 2016 and 2021. LCA identified response patterns across the 8 SCI-QOL-R-SF items. Polytomous logistic regression examined factors related to class membership. Linear regression assessed relationships between resilience classes and life satisfaction and depression symptoms. RESULTS: A 3-class solution provided optimal fit: high resilience (40.5%), medium resilience (38.4%), and low resilience (21.1%). Classes demonstrated a clear gradient pattern of responses from "always" to "sometimes" on the SCI-QOL items. Significant predictors of class membership included sex, age, employment status, self-care behaviors, health status, time since injury, and social support. Resilience class was the strongest predictor of both life satisfaction (accounting for 14% of variance) and depression symptoms (12.2% of variance), with large effect sizes between classes. CONCLUSION: Distinct resilience patterns exist among individuals with SCI and strongly predict psychosocial outcomes. These findings support tailored interventions targeting specific resilience profiles to improve life satisfaction and reduce depression in this population.