Multivariate analysis of oral hygiene data from a representative sample

This analysis was based on an oral health survey with dental examination and interview by questionnaire.

The data set comprised a representative random sample of the over 14-year-old residents of the German Federal State of Saxony (n=714), in which a multivariate logistic regression analysis was conducted. The dependent variable was the debris index (DI) score of the oral hygiene index scoring system. The subjects were split into 2 groups by dividing DI scores into tertiles: those with acceptable oral hygiene (1st tertile, DI 0.00< or =1.20) and unacceptable oral hygiene (2nd and 3rd tertile, DI>1.20). The most important explanatory variable was the community periodontal index of treatment need (CPITN).

The higher the CPITN score, the lower the probability of acceptable oral hygiene. In relation to the reference category score 0, higher CPITN scores were associated with acceptable oral hygiene, with an odds ratio 0.05 (95% confidence interval (CI)=0.01-0.23) for CPITN score 1, and an odds ratio 0.02 (95% CI: 0.01-0.08) for score 4. Further significant variables were: self-evaluation of tooth condition, gender, and the number of missing teeth.