|
Abstract:
|
We propose a semiparametric multi-state Markov frailty model for interval-censored time-to-event data in caries research. The proposed stochastic model is an attempt to describe the life history of dental caries at the tooth level, taking into account the multiplicity of the disease states and the intra-oral clustering of observations evaluated only at periodic time points. In particular, the model is used to investigate the intra-oral symmetries for caries formation at the quadrant level, and whether any of the symmetries vary with gender. We approximate the associated baseline intensities by linear splines. The estimation of the model is conducted using a penalized likelihood through a mixed-model representation. We develop a Bayesian method for predicting tooth-level caries transition probabilities, which can be used for tailoring tooth-level treatment plans. Intensive simulation studies indicate that the model performs well in realistic samples both in estimation and prediction. The practical utility of our model is illustrated using data generated from a unique longitudinal study on oral health among children from low-income families in the city of Detroit, Michigan.
|
