MKVPCI: a computer program for Markov models with piecewise constant intensities and covariates

Area of Science:

• Biostatistics
• Computational Statistics
• Epidemiology

Background:

• Markov models are essential for analyzing time-to-event data.
• Estimating time-dependent transition intensities in Markov models presents computational challenges.
• Understanding covariate effects on these intensities is crucial for accurate modeling.

Purpose of the Study:

• To present a novel computer program for fitting Markov models with piecewise constant intensities.
• To enable the estimation of covariate effects on transition intensities.
• To facilitate statistical hypothesis testing within these models.

Main Methods:

• Development of a computer program implementing a modified time-homogeneous Markov model.
• Introduction of artificial time-dependent covariates to represent time-varying transition intensities.
• Utilizing maximum likelihood estimation for parameter and standard error estimation.

Main Results:

• The program successfully estimates baseline transition intensities and regression coefficients.
• It provides maximum likelihood estimates and their standard errors.
• The program allows for the testing of various statistical hypotheses.

Conclusions:

• The developed program offers a robust method for analyzing Markov models with piecewise constant intensities.
• It effectively estimates the influence of covariates on transition intensities.
• The program is applicable to real-world epidemiological studies, such as analyzing smoking habits in children.