Disease progression in chronic heart failure is linear: Insights from multistate modelling

Area of Science:

• Cardiology
• Biostatistics
• Public Health

Background:

• Chronic heart failure (HF) poses significant challenges to patient care and healthcare systems.
• Understanding HF disease progression patterns is crucial for effective management and resource allocation.

Purpose of the Study:

• To describe the disease trajectory of patients with chronic heart failure (HF) using a four-state Markov model.
• To assess the predictability of HF progression at a population level.

Main Methods:

• A multistate Markov model was developed using data from 4918 patients followed for 2 years.
• Model predictions for transitions (e.g., hospitalization to death, rehospitalization) were compared with observed data up to 5 years.
• Covariate inclusion in the model did not enhance predictive accuracy.

Main Results:

• The Markov model accurately predicted HF population behavior in subsequent years.
• Predicted transition probabilities closely matched observed rates, demonstrating model validity.
• The model estimated average life expectancy around 10-13 years, aligning with observed 5-year mortality.

Conclusions:

• A Markov chain model indicates a constant annual transition proportion between states for chronic HF patients.
• HF disease progression at a population level appears more linear and predictable than commonly believed.
• Current patient status is a strong predictor of future HF course.