Differences in Symptom Experience Among Patients with Systemic Sclerosis: a Cluster Analytic Approach to Identifying Subgroups

Objectives: Symptoms of people who have SSc are heterogeneous and difficult to address clinically. Because diverse symptoms often co-occur and may share common underlying mechanisms, identifying symptoms that cluster together may better target treatment approaches. We sought to identify and characterize patient subgroups based on symptom experience.

Methods: An exploratory hierarchical agglomerative cluster analysis was conducted to identify subgroups from a large SSc cohort from a single US academic medical centre. Patient-reported symptoms of pain interference, fatigue, sleep disturbance, dyspnoea, depression and anxiety were used for clustering. A multivariate analysis of variance (MANOVA) was used to examine the relative contribution of each variable across subgroups. Analyses of variance were performed to determine participant characteristics based on subgroup assignment. Presence of symptom clusters were tallied within subgroup.

Results: Participants (n = 587; 84% female, 41% diffuse cutaneous subtype, 59% early disease) divided into three subgroups via cluster analysis based on symptom severity: (i) no/minimal, (ii) mild, and (iii) moderate. Participants in mild and moderate symptoms subgroups had similar disease severity, but different symptom presentation. In the mild symptoms subgroup, pain, fatigue and sleep disturbance was the main symptom cluster. Participants in the moderate symptoms subgroup were characterized by co-occurring pain, fatigue, sleep disturbance, depression and anxiety.

Conclusion: Identification of distinct symptom clusters, particularly among SSc patients who experience mild and moderate symptoms, suggests potential differences in treatment approach and in mechanisms underlying symptom experience that require further study.