Identifying symptom-based subgroups in Sjögren's disease: a latent class analysis of a patient co-designed survey

To identify symptom-based subgroups in Sjögren's disease using patient-reported data and explore differences in symptom burden, functional impact, psychological distress and healthcare priorities.

This secondary analysis used cross-sectional data from a patient co-designed online survey conducted between June and September 2022. Individuals self-identifying as living with Sjögren's disease (diagnosed or symptomatic) were eligible to participate. Latent class analysis was applied to 24 patient-reported symptom variables. European Alliance of Associations for Rheumatology (EULAR) Sjögren's Syndrome Patient Reported Index and the Hospital Anxiety and Depression Scale were used in post hoc analyses to characterise and compare classes.

A total of 296 respondents were included. Four symptom-based subgroups were identified: a sicca-predominant low-burden group (Class 1), a fatigue-dryness dominant group (Class 2), a musculoskeletal-cognitive group (Class 3) and a multisystem high-severity group (Class 4). Symptom burden, psychological distress and functional impact differed significantly across classes (all p<0.001). Inability to work increased from 3.4% in Class 1 to 33.9% in Class 4 (p<0.001). Higher-burden classes also reported greater unmet healthcare needs. Urgent need for effective treatments increased from 56% in Class 1 to 86% in Class 4 (p=0.0031), while urgent need for diagnostic testing increased from 38% to 68% (p<0.0001). Improved education of doctors and other healthcare professionals was consistently identified as a priority for improving care across all classes.

Patient-reported data identified clinically interpretable symptom-based subgroups in Sjögren's disease that differed in symptom burden, psychological distress, functional impact and perceived unmet healthcare need. These findings support multidimensional patient-reported assessment for characterising disease heterogeneity and informing patient-centred care. Limitations include the cross-sectional design and reliance on self-reported data.