Prediction of Mean Caries Experience of Samples of the Australian Population

Publisher Website
Google Scholar
Abstract
A multiple linear regression model for the prediction of mean caries experience scores of Australian adolescents was studied. A model incorporating eight explanatory variables was specified through forward stepwise multiple linear regression. The explantory variables were age, mean proportion of the sample subjects’ lives exposed to optimally fluoridated water supplies, time of the study, use of fluoride supplements in nonfluoridated areas, biases toward either high or low socio-economic status groups, extent of coverage for dental care by the Australian School Dental Scheme, and an interaction variable between the School Dental Scheme variable and age and time. The model had a multiple correlation coefficient of 0.96, explaining 92% of the variance in the caries experience scores used for its development. Actual mean DMFT scores used for the model’s development were well predicted by the specified model, but the validity of the model for predicting scores requires further research.