Abstract
The association between periodontitis and lifestyle factors has been widely investigated. However, an association between periodontitis and dietary patterns has not been explored. Therefore, this study investigated the association between periodontitis and food consumption among a Southern Brazil population. Data from the 1982 Pelotas Birth Cohort were used (n = 537). The exposure, periodontitis, was clinically measured and classified using the AAP/CDC system, then two latent variables were defined: 'initial' and 'moderate/severe' periodontitis. The consumption of in natura, processed, and ultra-processed foods (NOVA classification) was the outcome and measured in calories using the food frequency questionnaire (FFQ). Confounders were sex, maternal education, smoking status, xerostomia, and halitosis. Data were analyzed by structural equation modeling. 'Initial' periodontitis was associated with a higher consumption of in natura food (standardized coefficient (SC) 0.102; p-value = 0.040), versus processed (SC 0.078; p-value = 0.129) and ultra-processed (SC 0.043; p-value = 0.400) foods. 'Moderate/severe' periodontitis was associated with higher consumption of ultra-processed foods (SC 0.108; p-value = 0.024), versus processed (SC 0.093; p-value = 0.053) and in natura (SC 0.014; p-value = 0.762) foods. 'Moderate/severe' periodontitis appears to be associated with the consumption of processed and ultra-processed foods.