Comparison of oral health data from self-administered questionnaire and clinical examination.

In order to assess the validity of self-reported oral health data 319 subjects randomly selected from the respondents to an oral health survey were examined clinically. The level of agreement between the two methods of examination was estimated by the kappa statistic. Substantial agreement (kappa = 0.65) was found between the reported and observed number of remaining teeth and the difference between underreporting and overreporting was small. The results indicate that valid data for planning purposes and for monitoring changes in such oral health indicators as the proportion of individuals having 28 or more teeth, the proportion of subjects having a minimum of 20 functional teeth, the level of edentulousness and the presence of complete and partial dentures, can be obtained from postal questionnaires.