Cluster analysis of clinical data identifies fibromyalgia subgroups

Area of Science:

• Rheumatology
• Clinical Data Analysis
• Patient Stratification

Background:

• Fibromyalgia (FM) presents with widespread pain and diverse symptoms, complicating assessment and management.
• Reducing FM heterogeneity is crucial for effective clinical strategies.
• Classifying clinical data into simplified dimensions aids in defining FM subgroups.

Purpose of the Study:

• To classify clinical data into simplified dimensions for defining fibromyalgia subgroups.
• To reduce heterogeneity in fibromyalgia patient data.
• To identify distinct fibromyalgia patient clusters.

Main Methods:

• Evaluated 48 variables in 1,446 Spanish FM patients meeting 1990 ACR criteria.
• Performed partitioning analysis to group similar variables into dimensions.
• Utilized clustering on composite indexes to define FM subgroups, cross-validated on a large cohort.

Main Results:

• Variables clustered into three dimensions: "symptomatology", "comorbidities", and "clinical scales".
• FM subgroups were defined using symptomatology and comorbidity dimensions.
• Three clusters emerged: low symptoms/comorbidities, high symptoms/comorbidities, and high symptoms/low comorbidities, showing varied disease severity.

Conclusions:

• Identified three fibromyalgia subgroups within a large patient cohort using clinical data clustering.
• Highlighted the significance of family and personal history of comorbidities in FM.
• The resulting patient clusters suggest potentially different disease forms, relevant for future research and clinical assessment.