Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
February 28, 2025BMC Medical Research MethodologyOpen Access

The proposed Discrete-Time Split-State Framework achieved 94% coverage for Markov processes and 93% for non-Markov processes in simulations, and successfully modeled age-specific transition rates between heart disease states in the ARIC cohort.

View Full Paper
Ask AI
Bookmark
Share

Why the study?

The course of chronic disease is understudied, and existing Markov and semi-Markov models are unsuitable for chronic diseases that are largely non-memoryless.

Population

Individuals in a large cohort study

Design

Methodological framework proposal with simulation study and cohort application

Authors

MDMing DingHCHaiyi ChenFLFeng‐Chang Lin

Discussion

Loading...

Member takes

Overview

Enables history-dependent modeling of heart disease trajectories in cohorts; extends multi-state methods but leaves open clinical utility pending validation.

Structured PICO

P
Population
15,027 participants without coronary heart disease (CHD) or heart failure at baseline from the Atherosclerosis Risk in Communities (ARIC) study.
O
Outcome
Age-specific transition rates and probabilities between multi-states of heart disease (healthy, at-metabolic-risk, CHD, heart failure, and mortality)

The proposed discrete-time split-state framework provides a robust methodological approach for modeling the complex, non-Markovian progression of chronic cardiovascular diseases.

Limitations

  • Under-coverage (72%) for the semi-Markov process in simulation, likely due to the approximation of transition rates in discrete time
  • Assumption of a forward model of heart disease progression required classifying reverse transitions to a forward path, which may affect study power

Cite This Study

Ding et al. (2025) studied this question.

synapsesocial.com/papers/6a7628d63ddd128930fbe251https://doi.org/10.1186/s12874-025-02512-6
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A Multi-state Non-Markov Framework to Estimate Progression of Chronic Disease2024
  2. 2Summary Estimates Derived from a Multi-state Non-Markov Framework to Characterize the Course of Heart Disease2024 · 1 citations
  3. 3Summary measures derived from a multi-state modeling framework to characterize the course of heart disease2026
  4. 4Multistate Markov models for disease progression with classification error2003 · 282 citations
  5. 5Abstract TH877: A multi-state Causal Framework to Estimate the Effects of Treatment Regimes over the Course of Heart Disease2026