Social disparities in Periodontal Disease: The NHANES 1999-2004

Objectives: Disparities in the prevalence of periodontal disease by race/ethnicity and socioeconomic position in the United States have been pervasive for years. However, few studies have investigated these disparities independent of each other in the same population. Methods: This study uses the 1999-2004 National Health and Nutrition Examination Survey (NHANES) to investigate whether race/ethnicity, income and education are independently associated with periodontal disease, before and after controlling for selected characteristics among non-Hispanic black, non-Hispanic white or Mexican-American adults 18 years and older with a complete periodontal examination during NHANES 1999-2004 (n=10,648). Since NHANES 1999-2000 and NHANES 2001-2004 used different periodontal assessment, definitions accounting for this difference were developed. Specifically, periodontitis was defined for NHANES 1999-2000, as a person who had at least three sites with CAL > 4mm and at least two sites with PD > 3mm, while for NHANES 2001-2003, as a person who had at least three sites with CAL > 4mm and at least three sites with PD > 3mm. The main predictors were race/ethnicity, education and income. All analyses were performed in SUDAAN to account for the complex sampling design of NHANES. Results: Low income (<$5,000/ year), low educational attainment (<12 years) and being Non-Hispanic black are significant independent predictors of periodontitis. After adjustment for age, sex, marital status, income, health insurance, smoking status, and diabetes, persons with low education (OR: 2.56; 95%CI: 1.76-3.72) and non-Hispanic blacks (OR: 1.82; 95%CI: 1.28-2.60) exhibited higher odds of periodontitis than highly educated and non-Hispanic whites. The models for race/ethnicity and education were mutually adjusted for each other. Conclusions: Race/ethnicity and education were independent predictors of periodontitis. Thus, although race/ethnicity and education may be related, it is possible they act in different ways to affect periodontal disease. Supported by NIDCR, RWJ and CHUM.