Methods: Data were analyzed from subjects ages 13 years and older in the National Heath and Nutrition Examination Survey (1999-2004) using cluster analysis to group subjects into beverage consumption patterns based on the proportions of total fluid intake across 8 beverage categories. Eight clusters were formed: plain water; milk; sweetened carbonated beverages; unsweetened carbonated beverages; fruit juice and fruit drinks; low calorie fruit drinks; coffee and tea; and water in food. The relationships between beverage cluster and caries susceptibility and experience as measured by DFS (decayed and filled surfaces) were analyzed using logistic regression to model caries susceptibility and negative binomial regression to model caries experience. Associations controlled for age, race, sex, poverty income ratio, and education. Survey design elements were included.
Results: Relative to people who consumed a high level of tap water, people who consumed high levels of sweetened carbonated beverages (t = -2.60, p < .02) or people who obtained most of their fluid from food (t = -2.54, p < .02) were more likely to be susceptible to caries. Furthermore, relative to people who consumed a high level of tap water, people who consumed a high level of unsweetened carbonated beverages had greater caries experience (t = 2.76, p < .01).
Conclusions: Results show the need to distinguish between sweetened and unsweetened carbonated beverages when analyzing caries susceptibility and caries experience. The small sample sizes in the unsweetened carbonated beverage and low calorie fruit drink clusters require follow up to determine the differences in caries susceptibility and experience.
Keywords: Caries, Diet, Epidemiology, Nutrition and beverage