An example of the application of Markov models in dentistry--a prediction of the fate of molar teeth in 12-year-old Scottish children.

Markov analysis is a mathematical modelling technique which is particularly useful for predicting prognoses and probabilities of chronic diseases within populations. This paper applies Markov analysis to data concerning the condition of upper left permanent first molars in a group of individuals who were surveyed at age 12 and 15 years in Scotland by the Office of Population Censuses and Surveys. The Markov analysis predicted that in 1998, when the group of individuals would be 27 years old, 3.6 per cent of upper left first permanent molar teeth would be sound, 2.6 per cent would be decayed, 58.4 per cent would be filled and 35.4 per cent would be missing. A comparison of the Markov prediction for 18 year-olds in Scotland in 1989 with United Kingdom data for 16-24 year-olds published in 1988 indicated that the observed health status of the teeth was considerably better than that predicted by the Markov technique. The applicability of the Markov method to tooth status may therefore lie less in its ability to predict events as in being able to provide 'baselines' for future dental states in particular groups. Such an application of the Markov method could be used to determine whether dental deterioration among a group is occurring at a faster or slower rate than in the past.