Functional Impairment and Post-Stroke Depression: Potential Roles of Sleep Quality and Fatigue in a Cross-Sectional...

To examine the association between functional impairment and depressive symptoms in hospitalized patients undergoing post-stroke rehabilitation, and to explore whether sleep quality and fatigue were statistically linked to this association within a structural equation modeling framework. We also compared the relative contributions of the Barthel Index (BI) and modified Rankin Scale (mRS).

In this single-center retrospective cross-sectional study, 137 patients with stroke admitted for rehabilitation were included. Depressive symptoms were assessed using the Patient Health Questionnaire-9 (PHQ-9), sleep quality using the Pittsburgh Sleep Quality Index (PSQI), and fatigue using the Fatigue Assessment Scale (FAS). Functional status was evaluated with the BI (reverse-transformed to Barthel_rev) and mRS. Structural equation modeling with maximum likelihood estimation and bootstrap resampling (2,000 samples) was used to estimate direct and indirect associations.

In the primary latent-variable model, greater functional impairment was associated with poorer sleep quality (β=0.407, p=0.012) and greater fatigue (β=0.692, p=0.010). Poorer sleep quality was associated with higher PHQ-9 scores (β=0.296, p=0.003), whereas fatigue was not. The indirect association through sleep quality was significant (b=0.068, 95% CI 0.018-0.131), but the indirect association through fatigue was not. In a supplementary mutually adjusted model, mRS showed significant indirect associations with depressive symptoms through both sleep quality and fatigue, whereas Barthel_rev was linked primarily through fatigue. Because the supplementary model showed suboptimal fit, these findings should be interpreted cautiously.

These findings suggest that sleep quality may represent a more consistent statistical pathway linking functional impairment and depressive symptoms in post-stroke rehabilitation, whereas the role of fatigue appears to vary by functional measure and model specification. Early screening of sleep disturbance and fatigue may still be clinically relevant, but causal inferences cannot be drawn from this cross-sectional study.