Latent profile analysis of resilience and its relationship with depression-anxiety-stress among Chinese nursing students: a cross-sectional study

Resilience is an important psychological resource for nursing students, but little is known about how item-level resilience profiles relate to depression, anxiety and stress among Chinese nursing students. We identified resilience profiles and examined their associations with these symptoms. A cross-sectional study included 2,378 nursing students from four medical colleges in Hubei Province, China. Data were collected between January 2025 and March 2026. Resilience was assessed with the 25-item Connor-Davidson Resilience Scale and Depression-Anxiety-Stress with the 21-item Depression Anxiety Stress Scales. Latent profile analysis was used to identify resilience profiles. Chi-square tests, multinomial logistic regression, the Bolck-Croon-Hagenaars procedure and hierarchical linear regression were used to examine profile characteristics and Depression-Anxiety-Stress outcomes. The three-profile solution (entropy = 0.958) comprised a lower adversity-adaptation profile (52.2%), an intermediate profile (40.6%) and a higher profile (7.2%). Depression, anxiety, stress and total DASS-21 scores followed a graded pattern across the profiles (all p < 0.001). Relative to the higher profile, the lower and intermediate profiles were associated with 18.661- and 5.775-point higher total DASS-21 scores, respectively. The model with six demographic, family and relational covariates explained 8.7% of the variance (R² = 0.087); adding profile membership increased explained variance to 18.1% (R² = 0.181; ΔR² = 0.094). Three relative adversity-adaptation profiles were identified among Chinese nursing students, with differences in sociodemographic and relational characteristics. Depression, anxiety and stress scores were highest in the lower profile and lowest in the higher profile. After adjustment, both the lower and intermediate profiles remained associated with higher total DASS-21 scores than the higher profile. These findings may help nursing schools combine universal support with direct symptom assessment and mentoring based on students’ needs. Longitudinal and intervention studies are needed to determine whether profile-informed support improves outcomes.