PURPOSE To determine if estrogen replacement therapy (ERT) is associated with improved tooth retention and lower risk of edentulism (no natural teeth remaining) in a cohort of elderly women.

PATIENTS AND METHODS Subjects were 488 women, aged 72 to 95, who participated in the 23rd examination cycle (1994 to 1995) of the Framingham Heart Study, a population-based study begun in 1948. The number of teeth remaining and their location were recorded by a trained observer. History and duration of ERT were obtained from records kept since cycle 10 (1960 to 1963). Third molars were excluded from all analyses.

RESULTS Women who ever used ERT were younger than nonusers by 1 year (80 ± 4 years, n = 184, versus 81 ± 4 years, n = 304, P = 0.019). Estrogen users had more teeth remaining than nonusers (12.5 ± 0.8 versus 10.7 ± 0.8 versus 10.7 ± 0.6 teeth, P = 0.046, mean ± SE) after controlling for age, smoking status, and education. Duration of estrogen use was an independent predictor of the number of teeth remaining (P = 0.015) such that each 4.2-year interval of use was associated with an increased mean retention of 1 tooth. Long-term estrogen users (more than 8 years, n = 48) had an average of 3.6 more teeth than women who never used estrogen (14.3 ± 1.5 versus 10.7 ± 0.6 teeth, P <0.02). The association with duration of use was present among different types of teeth (incisors, canines, and premolars) but less strong for molars. The odds of being edentulous were reduced by 6% for each 1-year increase in duration of estrogen use (odds ratio = 0.94, P = 0.038, 95% confidence interval = 0.90 to 0.99).

CONCLUSIONS These data suggest that ERT protects against tooth loss and reduces the risk of edentulism. The associations of estrogen use and tooth retention are evident for all but the molars.