Identifying subgroups in heart failure patients with multimorbidity by clustering and network analysis

Area of Science:

• Clinical Informatics
• Cardiology
• Public Health

Background:

• Multimorbidity, the presence of multiple chronic conditions, complicates patient care and healthcare systems.
• Understanding multimorbidity in Heart Failure (HF) patients is limited, hindering effective treatment and risk stratification.
• Electronic Health Records (EHR) offer a rich data source for characterizing complex patient populations.

Purpose of the Study:

• To develop and present a workflow for identifying and characterizing HF patients with multimorbidity using EHR data.
• To explore the heterogeneity within HF patient populations based on their multimorbidity profiles.
• To investigate the prognostic implications of different multimorbidity patterns on clinical outcomes like hospital admissions.

Main Methods:

• A clustering analysis was performed on a cohort of 3745 Heart Failure patients.
• Data utilized included demographics, comorbidities, laboratory values, and drug prescriptions from EHRs.
• The analysis aimed to identify distinct patient subgroups based on their multimorbidity profiles.

Main Results:

• Four distinct patient clusters were identified among HF patients.
• These clusters exhibited significant differences in their multimorbidity profiles.
• The identified clusters demonstrated differential prognostic implications, particularly concerning unplanned hospital admissions.

Conclusions:

• HF patient populations are highly heterogeneous regarding multimorbidity.
• EHR data can be effectively utilized to characterize patient subgroups within HF.
• Improved patient subgroup characterization holds potential for enhanced clinical risk stratification in Heart Failure management.