The trajectory of depression-related symptom clusters following stroke: A network and latent class analysis

Post-stroke depression (PSD) frequently co-occurs with other symptoms, such as anxiety, fatigue, sleep disturbances, and stigma, forming complex symptom clusters. While research has identified associations between PSD and specific symptoms, the intricate interrelationships within these clusters remain largely unexplored. This study aimed to examine the longitudinal trajectory of these symptoms, classify distinct symptom cluster profiles, uncover symptom interconnections, identify core symptoms, and explore associated factors.

Depressive symptoms, anxiety, fatigue, sleep disturbance, and stigma were assessed in 195 stroke survivors at baseline (T1), one (T2), three (T3), and six (T4) months post-stroke. Network analysis examined symptom relationships and identified core symptoms. Latent profile analysis was used to classify distinct symptom clusters.

Network analysis revealed positive correlations among the five symptoms across four time points. "Anxiety" was the most central symptom within the symptom network at T1 (Strength = 2.134, EI = 2.134), while "depressive symptoms" held the central position at T2 (Strength = 2.595, EI = 2.595), T3 (Strength = 2.689, EI = 2.689), and T4 (Strength = 2.789, EI = 2.789). At T1, four clusters emerged: moderately affected, anxiety-insomnia, resilient, and severely affected. At T2, T3, and T4, a two-cluster solution was identified: resilient and symptomatic.

This study revealed a dynamic interplay of psychological symptoms following stroke. Depressive symptoms, anxiety, fatigue, sleep disturbance, and stigma exhibited a U-shaped trajectory, initially improving but subsequently worsening. Network analysis demonstrated a stable symptom cluster structure, with depressive symptoms and anxiety as core components. Symptom cluster profiles varied across time points.