Frailty Models for Predicting Eruption Time and Sequence of Permanent Dentition in Sri Lankan Children

Abstract

The main goal of this study is to create a suitable model to predict the tooth eruption pattern of Sri Lankan children. Also, we identify the relationship between variables associated with eruption sequence, compare the eruption sequence between sexes, compare the eruption sequences between upper and lower jaws, identifying common polymorphisms of tooth eruption sequences of children and determine the frequencies of occurrence of emergence polymorphisms for different tooth pairs. This analysis was performed using the data of the extent of tooth eruption of all 28 teeth at 10 different time points in each year. Welch two-sample t-test was used to identify the relationship between variables associated with eruption sequence. Frailty models and Cox-Proportional Hazard models developed for each tooth type separately and the model selection procedures Akaike information criterion (AIC), Bayesian information criterion (BIC) and Root Mean Square Error (RMSE) values are measured for each model. Since Gamma Frailty models have the smallest AIC and BIC for seven types of tooth which divide according to the eruption stage of the each tooth, we choose Gamma Frailty models as the best predictor for the tooth eruption. There is a significant difference between the eruption pattern of gender and jaw associated with time. However, no significant difference between sides associated with the eruption sequence was observed.