Evaluating cardiovascular mortality in type 2 diabetes patients: an analysis based on competing risks Markov chains...

Area of Science:

• Epidemiology
• Biostatistics
• Diabetology

Background:

• Type 2 diabetes is strongly linked to increased cardiovascular mortality.
• Accurate prediction of mortality causes is crucial for effective patient management.

Purpose of the Study:

• To estimate cause-specific mortality using Cox and Aalen models.
• To assess how covariate patterns influence mortality predictions.
• To compare the performance of different statistical methods.

Main Methods:

• Cox and Aalen additive regression models with Markov chain approach.
• Analysis of a cohort of 2865 type 2 diabetic patients (no insulin).
• Comparison of model accuracy for varying patient severity.

Main Results:

• Models showed similar cumulative incidence functions for patients with favorable profiles.
• Significant differences emerged for patients with severe profiles; Cox model: 0.26 (95% CI 0.21-0.31) vs. Aalen model: 0.14 (95% CI 0.09-0.18) for cardiovascular mortality.
• Aalen model demonstrated superior risk estimation for severe cases.

Conclusions:

• Both Cox and Aalen models effectively capture risk for average profiles.
• Aalen model is superior for identifying cause-specific mortality risk in severe type 2 diabetes cases.
• Findings support the development of advanced analytical tools for diabetes care research and resource allocation.