Oral health and cognitive function in the Third National Health and Nutrition Examination Survey (NHANES III)

• Published in: Psychosomatic medicine 2008-10, Vol.70 (8), p.936-941
• Main Authors: Stewart, Robert, Sabbah, Wael, Tsakos, Georgios, D'Aiuto, Francesco, Watt, Richard G
• Format: Article
• Language: English
• Subjects: Adjustment; Adult; Age Factors; Aged; Attachment; Cognition & reasoning; Cognition Disorders - epidemiology; Cognition Disorders - psychology; Cognitive functioning; Comorbidity; Dental health; Female; Gingival Hemorrhage - epidemiology; Gingival Hemorrhage - psychology; Health education; Humans; Male; Memory; Mental Recall; Middle Aged; Neuropsychological Tests - statistics & numerical data; Nutrition; Nutrition Surveys; Oral Health; Periodontal Attachment Loss - epidemiology; Periodontal Attachment Loss - psychology; Psychometrics; Regression analysis; Statistics as Topic; Tooth Loss - epidemiology; Tooth Loss - psychology; Young Adult
• ISSN: 0033-3174; 1534-7796
• DOI: 10.1097/psy.0b013e3181870aec

To investigate the association between oral health and cognitive function in early-, mid-, and late-adult life. A secondary analysis was carried out of a large, well-characterized community sample (NHANES III). Analyzed variables included three measures of oral health (gingival bleeding, loss of periodontal attachment, loss of teeth) and three measures of cognitive function: the Symbol Digit Substitution Test (SDST), the Serial Digit Learning Test (SDLT) (both in 5138 participants aged 20-59 years), and a Story Recall test (in 1555 participants aged >or=70 years). Other covariates in linear regression models included age, gender, ethnicity, education and poverty, and cardiovascular risk factors. Worse scores on all three measures of oral health status were significantly associated with poorer performance on all three measures of cognitive function after adjustment for age. Education was an important confounding factor. However, after full adjustment for all other covariates, gingival bleeding (%) and loss of periodontal attachment (%) remained associated with relative impairment on SDST score (B coefficients both = 0.003), and gingival bleeding was associated with relative impairment on SDLT (B = 0.017). No effect modification by age was observed. Poor oral health is associated with worse cognitive function throughout adult life. This may, in part, be accounted for by early life education and social status. However, the possibility of direct causal pathways requires further investigation.