Associations of Sleep Disturbances in Systemic Lupus Erythematosus 1 With Physical and Psychological Outcomes: A Cross-Sectional Latent 2 ProfileAnalysis

Ling MaLi YanhongXin GuoYing Wang^*

• Department of Rheumatology and Immunology, West China Hospital, Sichuan University, Chengdu, China

Purpose: Patients with systemic lupus erythematosus (SLE) frequently experience poor sleep 11 quality. This cross-sectional study aimed to identify distinct sleep disturbance profiles in SLE 12 patients and examine their associations with demographic, disease-related, and psychosocial 13 factors. 14 Methods: A total of 331 patients with SLE were included. Latent profile analysis (LPA) was 15 conducted using the tidyLPA package. Logistic regression models were constructed to assess 16 associations between the identified sleep disturbance clusters and physical and psychological 17 outcomes, based on factors significantly influencing the LPA results. The physical and 18 psychological outcomes were estimated using the Hospital Anxiety and Depression Scale 19 (HADS) and the Fatigue Severity Scale (FSS). Sleep clusters were analyzed through multivariate 20 logistic regression. 21 Results: Three distinct sleep disturbance profiles were identified: Cluster 1 (severe sleep 22 disturbance) (n = 42), Cluster 2 (moderate sleep disturbance) (n = 174), and Cluster 3 (mild sleep 23 disturbance) (n = 115). LPA yielded an entropy value of 0.996 for the three-cluster model. The 24 mean total Pittsburgh Sleep Quality Index (PSQI) score for the SLE samples was 7.59 ± 3.44. 25 Among the various sleep quality domains, sleep latency and subjective sleep quality were the 26 most significantly affected in SLE patients. The analysis revealed that disease duration, severity 27 of fatigue, use of calcium supplements, impaired renal function, anxiety, and depression were all 28 significant factors influencing cluster membership. 29 Conclusion: This study identified three distinct patterns of sleep disturbance among SLE 30 patients. Cluster 1 (severe sleep disturbance) was characterized by prolonged sleep latency 31 despite high sleep efficiency and subjective sleep quality scores. Cluster 2 (moderate sleep 32 disturbance) exhibited longer sleep duration than Cluster 1, while Cluster 3 (mild sleep 33 disturbance) had the lowest scores across all sleep quality domains. These findings suggest that 34 sleep disturbance profiling may facilitate personalized sleep management strategies for patients 35 with SLE.