OBJECTIVE

Although global self-ratings of oral health are widely used in oral health research, the frames of reference on which older people's ratings are based are not known. This study used a quantitative approach in order to identify these referents.

METHODS

Data were collected from 498 dentate subjects aged 53 years and over who took part in the second stage of a three-phase longitudinal epidemiological and sociodental study. Data were obtained by means of a personal interview and clinical oral examination and a self-complete version of the 49-item Oral Health Impact Profile (OHIP). These data were used to construct measures of oral disorders, oral symptoms, the functional and psychosocial impacts of oral disorders, health behaviours and contextual variables such as general health status, socioeconomic status and sociodemographic characteristics. Bivariate and linear regression analyses were used to identify which of these variables predicted self-ratings of oral health.

RESULTS

One quarter of subjects stated that their oral health was only fair or poor. At the bivariate level most variables were associated with self-ratings of oral health. The regression model for all subjects indicated that the most important predictor of these self-ratings was the OHIP functional limitations sub-scale score. This explained 23% of the variation in the self-ratings. Six other variables entered the model and increased the R2 value to 0.36. There was some variation in the models and the influence of various factors by age and educational attainment.

CONCLUSIONS

The results suggest that the referents that inform older adults' ratings of oral health are broadly similar to those that have been reported to inform their ratings of general health and differ across groups.