Factor Analysis to Determine Subgroups of Gastrointestinal Symptoms in Systemic Sclerosis.

Abstract

Objective: Systemic sclerosis (SSc) gastrointestinal (GI) disease is heterogeneous in presentation, with individual symptoms lacking specificity for specific anatomical and functional abnormalities. We used factor analysis to investigate whether latent subgroups of SSc-GI symptoms can be detected.

Methods: Using the University of California, Los Angeles Scleroderma Clinical Trials Consortium Gastrointestinal 2.0 (GIT 2.0) questionnaire data from 773 Australian Scleroderma Cohort Study participants, we performed a factor analysis of, firstly, GIT 2.0 domains scores, and then individual GIT 2.0 question responses to identify latent factors. A subsequent cluster analysis was performed to explore whether clinically definable SSc phenotypes were associated with specific GI symptoms.

Results: SSc-GI symptoms were highly correlated. Factor analysis of individual GIT 2.0 question responses revealed 4 latent factors within the dataset that could be clinically described as upper GI tract symptoms, bloating, diarrhea and incontinence, and constipation. Cluster analysis revealed 2 patient clusters, distinguished by disease duration and severity of GI manifestations. Anticentromere antibodies and pulmonary arterial hypertension were more common in participants with severe GI disease.

Conclusion: Despite the high correlation between GI manifestations, it is possible to detect subgroups of SSc-GI symptoms. Improved understanding of these subgroups of SSc-GI disease may advance the discovery of targeted interventions to improve the daily function and quality of life of those living with SSc.