Is modelling dental caries a 'normal' thing to do?

Objective To introduce and encourage the use of generalised linear models ( GLMs) in analysing caries data that do not require the response to be treat ed necessarily as a sample from a normal distribution. Basic research desig n At the present time, it is most likely that the sampling distribution of dmf/DMF in industrialised countries will not approximate normality. General ised linear modelling can be conducted assuming many underlying distributio ns which, in fact, includes the normal distribution. In this paper three GL Ms are employed (normal, Poisson, negative binomial) for modelling an examp le caries data set. In addition, a binomial model is used to model the dich otomous outcome of caries-free/caries-present. Clinical setting The data co mprised 871 Old Trafford, Manchester primary school children aged between 4 years 0 months and 5 years 11 months. Results The effect of one study cova riate was prominent in a normal model applied to all available dmf data but not in two non-normal models which used dmf > 0 data only. Furthermore, th e same covariate was significant at the 5% level in a binomial model indica ting that it influenced whether or not caries was present and not the level of dmf. Conclusion A suitable modelling approach for caries data is to emp loy a Poisson or a negative binomial model for the dmf/DMF response and a b inomial model for the caries-free/caries-present outcome. This allows separ ate estimation of those factors which influence the magnitude of caries and those factors which influence whether caries is actually present or not.