Abstract
Even before dementia becomes apparent, cognitive decline may contribute to deterioration in oral health. This cohort study of middle-aged adults evaluated associations of six-year change in cognitive function with oral health behaviors and conditions in the Atherosclerosis Risk in Communities (ARIC) study. Cognitive function was measured at study visits in 1990-1992 and 1996-1998 with three tests: (a) Delayed Word Recall (DWR), (b) Digit Symbol Substitution (DSS), and (c) Word Fluency (WF). Cognitive decline scores were computed as ‘studentized’ residuals of 1996-1998 scores regressed against 1990-1992 scores. In 1996-1998, 10,050 participants answered dental screening questions, and 5,878 of 8,782 dentate participants received a comprehensive oral examination. Multiple regression models used cognitive change to predict oral health behaviors and conditions with adjustment for covariates. In the fully adjusted models, greater decline in all three measures of cognitive function was associated with increased odds of complete tooth loss. Greater decline in DSS and WF scores was associated with infrequent toothbrushing. Decline in WF scores was also associated with higher plaque levels. In these middle-aged adults, six-year cognitive decline was modestly associated with less frequent toothbrushing, plaque deposit, and greater odds of edentulism, but not with other oral behaviors or diseases.
Keywords: oral hygiene, tooth loss, periodontitis, cognition, executive function, memory
Introduction
Cognition refers to mental processes, including attention, memory, producing and understanding language, solving problems, and decision-making. Cognitive ability decreases with age (Drag and Bieliauskas, 2010), while low cognitive ability, even within a normal range, is predictive of adverse health outcomes (St John et al., 2002; Pavlik et al., 2003). We posit that perceptions and behaviors regarding oral health can be adversely affected if cognitive ability is impaired, and may be accompanied by potential harm to oral health.
Various neuropsychological tests have been used to measure cognitive ability in studies of cognition and oral health. In studies of people who had already developed cognitive impairment or dementia (Chalmers et al., 2003, 2005; Avlund et al., 2004; Yu and Kuo, 2008; Ellefsen et al., 2009; Syrjälä et al., 2012), participants had more dental caries, fewer teeth, and poorer periodontal health than adults without dementia. Poor oral health may be attributable to functional decline, poor oral hygiene care, and less use of dental services (Avlund et al., 2004; Wu et al., 2007). Ideally, interventions to prevent oral disease would begin prior to manifestation of cognitive impairment, when change in cognitive function could be assessed. However, little is known about middle-aged adults when cognitive decline is at an early stage, or about the effects of cognitive change on dental behaviors and oral health. A clearer understanding of this relationship at different points in the life span could advance our knowledge of biological and environmental determinants of worsening oral health.
We used data from the Atherosclerosis Risk in Communities (ARIC) study, wherein cognitive function was measured at two visits separated by six years, and oral health was evaluated at the latter visit. The aim was to evaluate associations between a six-year change in cognitive function and a wide range of oral health measures and behaviors. We hypothesized that initial signs of cognitive decline would be associated with a greater likelihood of unfavorable dental health behaviors, plaque accumulation, and gingivitis, but not with indicators of chronic oral diseases, such as periodontitis and tooth loss.
Methods
Data and Participants
This study included a secondary analysis of data from the ARIC study. Details regarding study design and participant eligibility requirements have been reported previously (The ARIC Investigators, 1989; Cerhan et al., 1998). In brief, ARIC is a prospective, community-based study of vascular disease in a biracial cohort of middle-aged adults (45-64 yrs of age at inception) followed since 1987-1989 (examination 1). Participants were residents of four U.S. study sites: (a) Forsyth County, NC; (b) Minneapolis, MN; (c) Washington County, MD; and (d) Jackson, MS. The Dental ARIC, an ancillary study of the ARIC, was conducted from 1996-1998 (examination 4), when 11,375 participants answered dental screening questions. Of the 9,726 dentate respondents, 6,676 attended their local ARIC study center for a comprehensive oral examination.
Cognitive Function
Cognitive function assessment in the ARIC consisted of the Delayed Word Recall (DWR) Test, the Word Fluency (WF) Test, and the Digit Symbol Substitution (DSS) Test. The DWR is a test of verbal learning and recent memory (Knopman and Ryberg, 1989). The DSS measures executive function and processing speed (Wechsler, 1981). The WF is a test of executive function and expressive language (Benton et al., 1981). Cognitive tests were administered by trained examiners using standard protocols (Cerhan et al., 1998).
Oral Health Measures
Outcomes of interest included: (a) complete tooth loss, (b) number of teeth, (c) periodontitis, (d) plaque deposits, (e) gingival inflammation, (f) oral hygiene care, and (g) dental utilization. Self-reported data about complete tooth loss were available for all dentally screened study participants. If participants had dental prostheses, they were asked when the prostheses were first used. For dentally examined people with no contraindications to periodontal probing, examiners recorded the number of teeth present and measured probing pocket depth and gingival recession at 6 sites for all teeth present. For each tooth, the examiners also assessed the extent of gingivitis and plaque deposits. During data analysis, the severity of periodontitis was classified as none/mild, moderate, or severe based on CDC/AAP case definitions (Page and Eke, 2007). Variables reflecting oral hygiene care and dental utilization were from dental interview questions among dentate participants (Appendix 1). Edentulous individuals were excluded from the analysis of oral health behaviors, because their patterns of oral hygiene care and dental visits differed substantially from those of dentate individuals (Appendix Table 4).
Covariates
Potential confounding variables were ARIC study sites, demographic variables, socio-economic status, and health history. Health status measures were comprised of diabetes mellitus, hypertension, stroke, coronary heart disease (CHD), smoking, and alcohol consumption (Appendix 1).
Data Analyses
The summary scores for change in cognitive function were computed from race- and gender-specific linear regression models that regressed examination 4 scores against examination 2 scores. The residual was divided by its standard error, creating a ‘studentized’ residual (hereafter labeled ‘cognitive decline score’) that was the main predictor variable (Appendix Fig. 1).
Four multiple regression models were created to evaluate each oral health outcome: (a) model 1, adjusted for age, gender, and race-center; (b) model 2, adjusted for model 1 covariates and education; (c) model 3, adjusted for model 2 covariates, smoking, and alcohol use; and (d) model 4, adjusted for model 3 covariates and medical history. Income was not used as a covariate because it was correlated with education, producing variance inflation greater than 10. Appropriate regression methods were used according to the scale of the following dependent variables: (a) logistic regression for binary outcomes (e.g., severe periodontitis vs. moderate/none); (b) multinomial logistic regression for categorical outcomes (e.g., toothbrushing); and (c) linear regression for continuous outcomes (e.g., number of teeth). The final sample included participants who completed cognitive function assessments during Examinations 2 and 4 as well as a dental screening (n = 10,050) or an additional periodontal examination (n = 5,878) (Appendix Fig. 2).
Review by the IRB determined that this study was limited to secondary data analysis of de-identified data. The manuscript was prepared in accordance with the STROBE statement.
Results
At examination 2, about 90% of participants were aged 46-65 yrs, with 44.5% males and 81.5% whites. Overall, African Americans had poorer socio-economic status, general and oral health status, and worse dental behaviors when compared with whites. African American participants also had a higher prevalence of complete tooth loss (18.2% vs. 11.3%), fewer teeth, poorer oral hygiene care, and fewer dental visits. About one-third of African Americans, compared with 80% of whites, visited dentists on a regular basis (Table 1). Most participants with no teeth (98%) and approximately half (46%) of dentate participants had dentures (Appendix Table 5).
Table 1.
Race- and Gender-specific Characteristics of Study Participants at ARIC Examination 2
| African American |
White |
||||
|---|---|---|---|---|---|
| Characteristics | Female n = 1,214 | Malen = 647 | Femalen = 4,368 | Malen = 3,821 | Alln = 10,050 |
| Age at examination 2, mean ± SD | 55.6 ± 5.6 | 55.7 ± 5.7 | 56.7 ± 5.6 | 57.5 ± 5.7 | 56.8 ± 5.7 |
| Study sites,% | |||||
| Forsyth | 12.4 | 12.1 | 28.8 | 28.7 | 25.7 |
| Jackson | 87.6 | 87.9 | 0 | 0 | 16.3 |
| Minneapolis | 0 | 0 | 36.7 | 37.4 | 30.1 |
| Washington | 0 | 0 | 34.5 | 33.9 | 27.9 |
| Education,% | |||||
| Basic | 30.6 | 30.0 | 13.2 | 14.5 | 16.9 |
| Intermediate | 32.0 | 29.7 | 51.5 | 39.7 | 43.2 |
| Advanced | 37.4 | 40.3 | 35.3 | 45.8 | 39.9 |
| Income,% | |||||
| < $25,000 | 62.8 | 43.4 | 27.9 | 17.6 | 29.3 |
| $25,000 < $50,000 | 23.6 | 27.8 | 37.5 | 38.0 | 35.4 |
| $50,000 or more | 11.3 | 25.0 | 32.5 | 42.3 | 33.2 |
| Refused to provide | 2.3 | 2.8 | 2.1 | 2.1 | 2.2 |
| Cigarette use,% | |||||
| Current | 18.5 | 29.1 | 18.4 | 18.3 | 19.1 |
| Former | 23.2 | 41.7 | 30.5 | 53.3 | 39.0 |
| Never | 58.3 | 29.2 | 51.1 | 28.3 | 41.9 |
| Alcohol use,% | |||||
| Current | 25.7 | 49.9 | 62.0 | 70.1 | 59.9 |
| Former | 27.5 | 34.2 | 13.2 | 19.6 | 18.7 |
| Never | 46.8 | 15.9 | 24.8 | 10.3 | 21.4 |
| Diabetes mellitus,% | 22.2 | 19.0 | 8.6 | 12.5 | 12.4 |
| Hypertension,% | 53.4 | 45.6 | 26.3 | 30.1 | 32.3 |
| Coronary heart disease,% | 2.5 | 4.9 | 1.7 | 8.4 | 4.6 |
| Stroke,% | 1.6 | 1.7 | 0.9 | 1.3 | 1.2 |
| Oral health conditions and behaviors | |||||
| Edentulous,% | 20.8 | 13.3 | 10.4 | 12.4 | 12.6 |
| Number of teeth1, mean ± SD | 17.1 ± 7.5 | 18.1 ± 7.9 | 23.0 ± 6.4 | 22.7 ± 6.8 | 22.0 ± 7.1 |
| Extent of plaque deposits1,2, mean ± SD | 66.6 ± 44.8 | 79.4 ± 36.9 | 28.0 ± 30.7 | 42.3 ± 34.2 | 41.1 ± 37.6 |
| Extent of gingival inflammation3, mean ± SD | 79.6 ± 36.6 | 86.8 ± 30.4 | 17.3 ± 27.5 | 25.7 ± 33.1 | 32.2 ± 39.1 |
| Severe periodontal disease1,% | 13.2 | 34.5 | 10.5 | 21.5 | 16.7 |
| Brushing4,% | |||||
| Not at all | 1.5 | 2.7 | 0.4 | 2.7 | 1.6 |
| Once a day | 25.0 | 42.2 | 18.0 | 38.3 | 28.1 |
| Twice a day or more | 73.5 | 55.1 | 81.6 | 59.0 | 70.3 |
| Flossing4,% | |||||
| Not at all | 44.2 | 61.8 | 21.9 | 44.2 | 35.4 |
| Once a week | 8.3 | 8.9 | 6.7 | 9.9 | 8.3 |
| Twice a week or more | 47.5 | 29.2 | 71.4 | 45.9 | 56.3 |
| Last time visited a dentist4,% | |||||
| > 36 mos | 24.2 | 25.7 | 4.8 | 7.3 | 9.2 |
| 12-36 mos | 26.3 | 24.6 | 8.5 | 11.7 | 12.7 |
| < 12 mos | 49.5 | 49.7 | 86.7 | 81.0 | 78.1 |
| Reasons to visit a dentist 4,% | |||||
| Do not visit a dentist | 3.8 | 3.9 | 0.6 | 0.5 | 1.1 |
| When problems occur | 59.7 | 64.7 | 13.5 | 22.4 | 25.2 |
| A regular basis | 36.5 | 31.4 | 85.9 | 77.1 | 73.7 |
| Change in cognitive scores | |||||
| Delayed word recall, mean ± SD | −0.2 ± 1.7 | −0.3 ± 1.7 | −0.1 ± 1.5 | −0.2 ± 1.5 | −0.1 ± 1.5 |
| Digit symbol substitution, mean ± SD | −2.3 ± 9.1 | −1.4 ± 8.1 | −2.8 ± 6.8 | −2.7 ± 6.1 | −2.6 ± 7.0 |
| Word fluency, mean ± SD | −0.7 ± 8.3 | −1.8 ± 8.5 | −0.1 ± 7.8 | −0.8 ± 7.9 | −0.6 ± 8.0 |
n, total number in study group; SD, standard deviation.
Of dentate participants, 5,878 completed periodontal examinations.
Extent of plaque deposits was a percentage of buccal surfaces with visible plaque.
Extent of gingival inflammation was available for 5,638 participants, calculated as a percentage of teeth with Löe & Silness Gingival Index (GI) score ≥ 1.
Only dentate participants (n = 8,782).
Change in cognitive function over the course of 6 yrs was small for all tests. In general, older participants with low education, low income, and more medical co-morbidities had greater cognitive decline (Appendix Tables 2, 3).
A one-unit increase in all three cognitive decline scores was associated with fewer teeth and greater odds of complete tooth loss. However, in the fully adjusted models, the association between cognitive decline and the number of teeth was not evident (Table 2). The greater decline in DSS was associated with not brushing the teeth and gingivitis, but only the association with not brushing teeth remained robust after adjustment for all covariates. The greater decline in the WF was associated with infrequent toothbrushing and higher plaque deposits in both the unadjusted and adjusted models (Tables 2, 3). Even when associations were statistically significant, odds ratios were no greater than 1.3. We did not observe a statistically significant effect of cognitive decline scores on periodontitis, dental visits, and the use of dental floss (Tables 2-4).
Table 2.
Association between a Six-year Cognitive Decline1 and Oral Health Conditions
| β (upper, lower 95% CI) |
OR (upper, lower 95% CI) |
||||
|---|---|---|---|---|---|
| Cognitive Function | Mean Number of Remaining Teethn = 5,878 | Extent of Dental Plaque2n = 5,878 | Extent of Gingival Inflammation3n = 5,638 | Severe Periodontitis4n = 5,878 | Complete Tooth Loss5n = 10,050 |
| Delayed word recall | |||||
| Model 16 | −0.20 (−0.37, −0.02) | 0.97 (0.13, 1.81) | 0.29 (−0.47, 1.06) | 1.01 (0.94, 1.09) | 1.19 (1.12, 1.26) |
| Model 27 | −0.12 (−0.29, 0.05) | 0.71 (−0.13, 1.54) | 0.12 (−0.63, 0.88) | 1.00 (0.93, 1.08) | 1.13 (1.07, 1.21) |
| Model 38 | −0.08 (−0.25, 0.09) | 0.64 (−0.19, 1.47) | 0.09 (−0.67, 0.85) | 0.99 (0.92, 1.07) | 1.12 (1.05, 1.19) |
| Model 49 | −0.07 (−0.24, 0.10) | 0.54 (−0.28, 1.37) | 0.03 (−0.73, 0.78) | 0.99 (0.92, 1.06) | 1.11 (1.04, 1.18) |
| Digit symbol substitution | |||||
| Model 16 | −0.35 (−0.53, −0.17) | 1.50 (0.65, 2.35) | 1.22 (0.45, 1.99) | 1.08 (1.01, 1.17) | 1.17 (1.11, 1.25) |
| Model 27 | −0.24 (−0.42, −0.07) | 1.15 (0.30, 1.99) | 1.01 (0.24, 1.78) | 1.07 (0.99, 1.15) | 1.12 (1.05, 1.19) |
| Model 38 | −0.18 (−0.35, −0.01) | 0.96 (0.12, 1.79) | 0.91 (0.14, 1.68) | 1.06 (0.98, 1.14) | 1.10 (1.03, 1.06) |
| Model 49 | −0.16 (−0.33, 0.01) | 0.76 (−0.08, 1.59) | 0.76 (−0.01, 1.53) | 1.05 (0.97, 1.13) | 1.08 (1.01, 1.15) |
| Word fluency | |||||
| Model 16 | −0.27 (−0.44, −0.10) | 1.64 (0.82, 2.46) | 0.66 (−0.09, 1.41) | 1.00 (0.93, 1.07) | 1.21 (1.13, 1.28) |
| Model 27 | −0.13 (−0.29, 0.04) | 1.19 (0.37, 2.00) | 0.37 (−0.38, 1.12) | 0.98 (0.91, 1.05) | 1.12 (1.05, 1.20) |
| Model 38 | −0.12 (−0.29, 0.04) | 1.09 (0.28, 1.90) | 0.32 (−0.42, 1.07) | 0.98 (0.91, 1.05) | 1.12 (1.05, 1.20) |
| Model 49 | −0.11 (−0.27, 0.05) | 0.97 (0.16, 1.77) | 0.24 (−0.51, 0.98) | 0.98 (0.91, 1.05) | 1.10 (1.03, 1.18) |
OR, Odds ratios; CI, confidence interval; β, regression coefficient.
A six-year change in cognitive scores was expressed as race- and gender-specific ‘studentized’ residuals.
Extent of plaque deposits was a percentage of buccal surfaces with visible plaque.
Extent of gingival inflammation was a percentage of teeth with Löe & Silness Gingival Index (GI) score ≥ 1.
OR and 95% CI from binary logistic regression models estimated the associations between cognitive decline, and odds of severe periodontitis vs. none/mild/moderate periodontitis [The Centers for Disease Control and Prevention/The American Academy of Periodontology (CDC/AAP) classification].
OR and 95% CI from binary logistic regression models estimated the associations between cognitive decline, and odds of complete tooth loss vs. dentate.
Adjusting for age, gender, and race-center.
Adjusting for covariates in model 1 and education.
Adjusting for covariates in model 2, cigarette use, and alcohol use.
Adjusting for covariates in model 3 and health history.
Table 3.
Association between a Six-year Change in Cognitive Decline1 and Oral Hygiene Care
| OR (upper, lower 95% CI) |
||||
|---|---|---|---|---|
| Toothbrushing2 (n = 8,782) |
Dental Flossing3 (n = 8,782) |
|||
| Cognitive Function | No brushing | Brushing once a day | No flossing | Flossing once a week |
| Delayed word recall | ||||
| Model 14 | 1.00 (0.84, 1.19) | 1.08 (1.02, 1.13) | 1.07 (1.02, 1.13) | 1.03 (0.95, 1.12) |
| Model 25 | 0.96 (0.80, 1.15) | 1.06 (1.01, 1.11) | 1.05 (1.00, 1.11) | 1.03 (0.95, 1.11) |
| Model 36 | 0.96 (0.80, 1.15) | 1.06 (1.00, 1.11) | 1.05 (1.00, 1.10) | 1.03 (0.95, 1.12) |
| Model 47 | 0.95 (0.78, 1.13) | 1.06 (1.00, 1.11) | 1.04 (0.99, 1.09) | 1.03 (0.95, 1.12) |
| Digit symbol substitution | ||||
| Model 14 | 1.33 (1.14, 1.56) | 1.02 (0.97, 1.07) | 1.04 (1.00, 1.10) | 1.05 (0.97, 1.14) |
| Model 25 | 1.31 (1.11, 1.55) | 1.00 (0.95, 1.05) | 1.02 (0.96, 1.07) | 1.04 (0.96, 1.13) |
| Model 36 | 1.31 (1.11, 1.55) | 1.00 (0.95, 1.05) | 1.01 (0.96, 1.06) | 1.04 (0.96, 1.13) |
| Model 47 | 1.29 (1.08, 1.53) | 0.99 (0.94, 1.04) | 1.01 (0.96, 1.06) | 1.04 (0.96, 1.13) |
| Word fluency | ||||
| Model 14 | 1.31 (1.10, 1.55) | 1.09 (1.04, 1.14) | 1.08 (1.03, 1.13) | 1.07 (0.99, 1.16) |
| Model 25 | 1.23 (1.02, 1.47) | 1.06 (1.01, 1.11) | 1.04 (0.99, 1.09) | 1.06 (0.98, 1.14) |
| Model 36 | 1.22 (1.02, 1.46) | 1.06 (1.01, 1.11) | 1.04 (0.99, 1.09) | 1.05 (0.97, 1.14) |
| Model 47 | 1.21 (1.01, 1.45) | 1.06 (1.01, 1.11) | 1.03 (0.99, 1.08) | 1.05 (0.97, 1.14) |
OR, Odds ratios; CI, confidence interval.
A six-year change in cognitive scores was expressed as race- and gender-specific ‘studentized’ residuals.
OR and 95% CI from multinomial logistic regression models estimated the associations between cognitive decline, and odds of no toothbrushing and brushing once per day vs. brushing twice per day.
OR and 95% CI from multinomial logistic regression models estimated the associations between cognitive decline, and odds of no flossing and flossing once per week vs. brushing twice per week.
Adjusting for age, gender, and race-center.
Adjusting for covariates in model 1 and education.
Adjusting for covariates in model 2, cigarette use, and alcohol use.
Adjusting for covariates in model 3 and health history.
Table 4.
Association between a Six-year Cognitive Decline1 and Dental Utilization
| OR (upper, lower 95% CI) |
||||
|---|---|---|---|---|
| Last Dental Visit2 (n = 8,782) |
Reasons to Visit a Dentist3 (n = 8,782) |
|||
| Cognitive Function | >36 mos | 12- < 36 mos | Do not visit a dentist | When problems occur |
| Delayed word recall | ||||
| Model 14 | 1.07 (0.99, 1.16) | 1.04 (0.98, 1.12) | 1.13 (0.92, 1.39) | 1.11 (1.05, 1.17) |
| Model 25 | 1.03 (0.95, 1.11) | 1.02 (0.96, 1.09) | 1.09 (0.88, 1.34) | 1.07 (1.01, 1.13) |
| Model 36 | 1.02 (0.94, 1.11) | 1.02 (0.95, 1.09) | 1.08 (0.87, 1.33) | 1.06 (1.01, 1.13) |
| Model 47 | 1.01 (0.93, 1.09) | 1.01 (0.94, 1.08) | 1.05 (0.85, 1.30) | 1.05 (0.99, 1.11) |
| Digit symbol substitution | ||||
| Model 14 | 1.06 (0.98, 1.14) | 1.06 (0.99, 1.14) | 1.21 (0.99, 1.48) | 1.12 (1.06, 1.18) |
| Model 25 | 1.01 (0.93, 1.09) | 1.04 (0.97, 1.11) | 1.14 (0.93, 1.41) | 1.08 (1.02, 1.14) |
| Model 36 | 1.00 (0.92, 1.08) | 1.03 (0.96, 1.10) | 1.13 (0.92, 1.40) | 1.06 (1.01, 1.13) |
| Model 47 | 0.97 (0.89, 1.06) | 1.01 (0.94, 1.08) | 1.09 (0.88, 1.34) | 1.04 (0.98, 1.10) |
| Word fluency | ||||
| Model 14 | 1.10 (1.02, 1.18) | 1.05 (0.98, 1.12) | 1.23 (1.01, 1.50) | 1.11 (1.05, 1.17) |
| Model 25 | 1.03 (0.95, 1.11) | 1.01 (0.95, 1.08) | 1.15 (0.93, 1.41) | 1.05 (0.99, 1.11) |
| Model 36 | 1.02 (0.94, 1.10) | 1.01 (0.95, 1.08) | 1.15 (0.93, 1.41) | 1.05 (0.99, 1.10) |
| Model 47 | 1.00 (0.92, 1.08) | 0.99 (0.93, 1.06) | 1.10 (0.89, 1.36) | 1.02 (0.97, 1.08) |
OR, Odds ratios; CI, confidence interval.
A six-year change in cognitive scores was expressed as race- and gender-specific ‘studentized’ residuals.
OR and 95% CI from multinomial logistic regression models estimated the associations between cognitive decline, and odds of last dental visit > 36 mos and 12- < 36 mos vs. < 12 mos.
OR and 95% CI from multinomial logistic regression models estimated the associations between cognitive decline, and odds of not visiting a dentist and visiting a dentist only when problems occur vs. visiting a dentist on a regular basis.
Adjusting for age, gender, and race-center.
Adjusting for covariates in model 1 and education.
Adjusting for covariates in model 2, cigarette use, and alcohol use.
Adjusting for covariates in model 3, and health history.
Discussion
In these middle-aged adults, declines in two cognitive measures were modestly associated with lack of toothbrushing, but not with other dental behaviors or clinical gingivitis or periodontitis. Cognitive decline in all tests was associated with greater levels of tooth loss. Associations with tooth loss were attributable, in part, to confounding effects of education and general health status, which tended to be poorer in people who experienced cognitive decline. The result regarding toothbrushing is consistent with the study hypothesis, although it was not corroborated by the findings regarding gingivitis or dental plaque for DSS. The observed association with tooth loss was counter to expectations and suggests instead that this indicator of treatment for past dental disease may be a marker of elevated risk for cognitive decline.
One limitation of this study was the absence of oral health measures at baseline, which meant that changes in cognition could be related only to oral health at follow-up instead of changes in oral health between baseline and follow-up. If cognition truly affects oral health, and if change in cognition is associated with the level of cognition at baseline, then we would expect baseline levels of oral health to differ according to levels of change in cognition. If that were the case, observed associations between changes in cognition and levels of oral health at follow-up might simply reflect poorer oral health at baseline for people who subsequently experienced a decline in cognition. To avoid this problem, we computed the change in cognition using residuals, which meant that the change in cognition was not related to baseline levels of cognition.
Self-reported behaviors probably become less reliable in people who experience cognitive decline. People who report infrequent toothbrushing might simply have forgotten that they truly did brush their teeth. Maybe this is why we observed an effect of cognitive decline on reported toothbrushing, but not on plaque or gingivitis. An effect between cognitive decline and dental flossing was not observed, but this may be explained by a ‘floor effect’, meaning that, although people are cognitively intact, they are less likely to floss their teeth routinely. Therefore, a relatively small change in cognitive function cannot make a difference in the frequency of flossing. For dental visits and periodontitis, the non-significant associations may be partly due to a threshold effect, whereby oral health measures will be affected only when there are very large reductions in cognition.
The decline in executive function as measured by DSS and WF was associated with infrequent toothbrushing. These findings add to the body of knowledge regarding the contributions of the early stages of cognitive decline to oral health. Unlike in older adults with severely impaired cognition, failure to perform routine oral hygiene care in middle-aged adults is less likely to be attributable to functional impairment (Chalmers et al., 2003, 2005; Yu and Kuo, 2008; Ellefsen et al., 2009; Syrjälä et al., 2012). Instead, non-compliance with routine oral hygiene care may be an initial sign of altered cognitive processes (e.g., planning, reasoning, initiation, or decision-making). Associations between impaired executive function and greater risks of adverse health outcomes (Pavlik et al., 2003; Rosano et al., 2008), health behaviors (Dong et al., 2010) and oral health (Stewart et al., 2008; Yu and Kuo, 2008) have also been shown.
The effect sizes observed in this study are modest. For example, never brushing the teeth was reported as only 1.6% (n = 137), and, thus, the observed odds ratio of 1.3 is equivalent to a net increase of 0.45%, which is associated with one unit (i.e., one standard deviation) of the cognitive decline scores (Appendix 2). Although executive functioning is a high-order cognitive skill that controls goal-directed behaviors (Drag and Bieliauskas, 2010), it is only one factor contributing to successful oral health care. Other factors, such as lifestyle or personality, may also be important. The effect of cognitive decline on self-neglect behavior and deteriorating oral health may become more apparent after a longer follow-up period.
Interpretations of the study results should be made with caution due to the possibility of reverse causation that arises because several study measurements are permanent markers of oral disease history (e.g., tooth loss and attachment level) and baseline oral health status was not available, as mentioned previously. Dentate participants who received periodontal examinations were more likely to be healthier and to have better cognitive ability than those who did not, which could lead to underestimation of the association between cognitive decline and periodontitis (Appendix Table 6). Further, despite the extensive characterization of this cohort, uncontrolled confounding cannot be ruled out.
Previous studies among the elderly have raised the issue of bidirectional associations between poor oral health and impaired cognition (Stewart and Hirani, 2007; Starr et al., 2008; Stewart et al., 2008; Kaye et al., 2010; Batty et al., 2011). That is, periodontitis and tooth loss may be not only consequences of cognitive impairment, but also risk factors for cognitive impairment. In the present study, most edentulous participants tended to lose many teeth early in life (i.e., they received their first false teeth at an average age of 36.2 yrs), which was about 20 yrs before the first cognitive assessment (average age of 58.3 yrs). The observed association with cognitive decline likely supports the hypothesis that tooth loss in early life is a risk indicator for cognitive decline in later life, which has been explained by mechanisms involving inflammatory load in early life (Gatz et al., 2006), malnutrition (Kim et al., 2007), and loss of neurosensory function in the animal model (Onozuka et al., 1999). An alternative interpretation derives from the observation that poor oral health and cognitive decline share common risk factors, such as cardiovascular diseases and low socio-economic status (Singh-Manoux et al., 2008; Matthews et al., 2011; Xu and Lu, 2011). It has been proposed that low socio-economic status in adulthood associated with poor oral health at an older age may reflect low cognitive ability in childhood (Sabbah et al., 2009).
It is difficult to compare these results regarding change in cognitive function with results from studies that measured cognitive function at one time (Stewart et al., 2008; Sabbah et al., 2009; Matthews et al., 2011). There is also a lack of uniformity in cognition domains measured, instruments used, or handling of cognitive scores for statistical analysis. Our results cannot be extrapolated to the U.S population, since study participants were recruited from only four areas in the United States. Strengths of this study include the large sample of community-dwelling middle-aged adults, repeated assessments of cognitive function, and comprehensive examination of periodontal health (i.e., full-mouth examination).
Despite the intuitive plausibility of a relationship between cognitive decline and oral health, no study has examined this relationship in an adult population without dementia. These findings shed new light on this likely complex relationship. Specifically, we show that even small reductions in cognitive function were associated modestly with some oral health behaviors, although not with periodontal disease. However, we caution that the association might not be causal, and other relationships might represent reverse-causation: Similar levels of cognitive decline were associated with greater odds of edentulism, a condition that usually occurred decades before the cognitive decline measured here.
Supplementary Material
Footnotes
The ARIC study was carried out as a collaborative study supported by National Heart, Lung, and Blood Institute contracts (HHSN268201100005C, HHSN268201100006C, HHSN26820 1100007C, HHSN268201100008C, HHSN268201100009C, HHSN2682011000010C, HHSN2682011000011C, and HHSN 2682011000012C). The Dental ARIC study was supported by the National Institute of Dental and Craniofacial Research (DE13807-01A1 and DE 11551).
The authors declare no potential conflicts of interest with respect to the authorship and/or publication of this article.
A supplemental appendix to this article is published electronically only at http://jdr.sagepub.com/supplemental.
References
- The ARIC investigators. (1989). The Atherosclerosis Risk in Communities (ARIC) Study: design and objectives. Am J Epidemiol 129:687-702. [PubMed] [Google Scholar]
- Avlund K, Holm-Pedersen P, Morse DE, Viitanen M, Winblad B. (2004). Tooth loss and caries prevalence in very old Swedish people: the relationship to cognitive function and functional ability. Gerodontology 21:17-26. [DOI] [PubMed] [Google Scholar]
- Batty GD, Li Q, Huxley R, Zoungas S, Taylor BA, Neal B, et al. (2011). Oral disease in relation to future risk of dementia and cognitive decline: prospective cohort study based on the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified-Release Controlled Evaluation (ADVANCE) trial. Eur Psychiatry 28:49-52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Benton AL, Eslinger PJ, Damasio AR. (1981). Normative observations on neuropsychological test performances in old age. J Clin Neuropsychol 3:33-42. [DOI] [PubMed] [Google Scholar]
- Cerhan JR, Folsom AR, Mortimer JA, Shahar E, Knopman DS, McGovern PG, et al. (1998). Correlates of cognitive function in middle-aged adults. Atherosclerosis Risk in Communities (ARIC) Study Investigators. Gerontology 44:95-105. [DOI] [PubMed] [Google Scholar]
- Chalmers JM, Carter KD, Spencer AJ. (2003). Oral diseases and conditions in community-living older adults with and without dementia. Spec Care Dentist 23:7-17. [DOI] [PubMed] [Google Scholar]
- Chalmers JM, Carter KD, Spencer AJ. (2005). Caries incidence and increments in Adelaide nursing home residents. Spec Care Dentist 25:96-105. [DOI] [PubMed] [Google Scholar]
- Dong X, Simon MA, Wilson RS, Mendes de Leon CF, Rajan KB, Evans DA. (2010). Decline in cognitive function and risk of elder self-neglect: finding from the Chicago Health Aging Project. J Am Geriatr Soc 58:2292-2299. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Drag LL, Bieliauskas LA. (2010). Contemporary review 2009: cognitive aging. J Geriatr Psychiatry Neurol 23:75-93. [DOI] [PubMed] [Google Scholar]
- Ellefsen B, Holm-Pedersen P, Morse DE, Schroll M, Andersen BB, Waldemar G. (2009). Assessing caries increments in elderly patients with and without dementia: a one-year follow-up study. J Am Dent Assoc 140:1392-1400. [DOI] [PubMed] [Google Scholar]
- Gatz M, Mortimer JA, Fratiglioni L, Johansson B, Berg S, Reynolds CA, et al. (2006). Potentially modifiable risk factors for dementia in identical twins. Alzheimers Dement 2:110-117. [DOI] [PubMed] [Google Scholar]
- Kaye EK, Valencia A, Baba N, Spiro A, Dietrich T, Garcia RI. (2010). Tooth loss and periodontal disease predict poor cognitive function in older men. J Am Geriatr Soc 58:713-718. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kim JM, Stewart R, Prince M, Kim SW, Yang SJ, Shin IS, et al. (2007). Dental health, nutritional status and recent-onset dementia in a Korean community population. Int J Geriatr Psychiatry 22:850-855. [DOI] [PubMed] [Google Scholar]
- Knopman DS, Ryberg S. (1989). A verbal memory test with high predictive accuracy for dementia of the Alzheimer type. Arch Neurol 46:141-145. [DOI] [PubMed] [Google Scholar]
- Matthews JC, You Z, Wadley VG, Cushman M, Howard G. (2011). The association between self-reported tooth loss and cognitive function in the REasons for Geographic And Racial Differences in Stroke study: an assessment of potential pathways. J Am Dent Assoc 142:379-390. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Onozuka M, Watanabe K, Mirbod SM, Ozono S, Nishiyama K, Karasawa N, et al. (1999). Reduced mastication stimulates impairment of spatial memory and degeneration of hippocampal neurons in aged SAMP8 mice. Brain Res 826:148-153. [DOI] [PubMed] [Google Scholar]
- Page RC, Eke PI. (2007). Case definitions for use in population-based surveillance of periodontitis. J Periodontol 78(7 Suppl):1387S-1399S. [DOI] [PubMed] [Google Scholar]
- Pavlik VN, de Moraes SA, Szklo M, Knopman DS, Mosley TH, Hyman DJ. (2003). Relation between cognitive function and mortality in middle-aged adults: the Atherosclerosis Risk in Communities study. Am J Epidemiol 157:327-334. [DOI] [PubMed] [Google Scholar]
- Rosano C, Newman AB, Katz R, Hirsch CH, Kuller LH. (2008). Association between lower digit symbol substitution test score and slower gait and greater risk of mortality and of developing incident disability in well-functioning older adults. J Am Geriatr Soc 56:1618-1625. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Sabbah W, Watt RG, Sheiham A, Tsakos G. (2009). The role of cognitive ability in socio-economic inequalities in oral health. J Dent Res 88:351-355. [DOI] [PubMed] [Google Scholar]
- Singh-Manoux A, Sabia S, Lajnef M, Ferrie JE, Nabi H, Britton AR, et al. (2008). History of coronary heart disease and cognitive performance in midlife: the Whitehall II study. Eur Heart J 29:2100-2107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- St John PD, Montgomery PR, Kristjansson B, McDowell I. (2002). Cognitive scores, even within the normal range, predict death and institutionalization. Age Ageing 31:373-378. [DOI] [PubMed] [Google Scholar]
- Starr JM, Hall RJ, Macintyre S, Deary IJ, Whalley LJ. (2008). Predictors and correlates of edentulism in the healthy old people in Edinburgh (HOPE) study. Gerodontology 25:199-204. [DOI] [PubMed] [Google Scholar]
- Stewart R, Hirani V. (2007). Dental health and cognitive impairment in an English national survey population. J Am Geriatr Soc 55:1410-1414. [DOI] [PubMed] [Google Scholar]
- Stewart R, Sabbah W, Tsakos G, D’Aiuto F, Watt RG. (2008). Oral health and cognitive function in the Third National Health and Nutrition Examination Survey (NHANES III). Psychosom Med 70:936-941. [DOI] [PubMed] [Google Scholar]
- Syrjälä AM, Ylöstalo P, Ruoppi P, Komulainen K, Hartikainen S, Sulkava R, et al. (2012). Dementia and oral health among subjects aged 75 years or older. Gerodontology 29:36-42. [DOI] [PubMed] [Google Scholar]
- Wechsler D. (1981). Manual for the Wechsler Adult Intelligence Scale-Revised. New York, NY: The Psychological Corporation. [Google Scholar]
- Wu B, Plassman BL, Liang J, Wei L. (2007). Cognitive function and dental care utilization among community-dwelling older adults. Am J Public Health 97:2216-2221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Xu F, Lu B. (2011). Prospective association of periodontal disease with cardiovascular and all-cause mortality: NHANES III follow-up study. Atherosclerosis 218:536–542. [DOI] [PubMed] [Google Scholar]
- Yu YH, Kuo HK. (2008). Association between cognitive function and periodontal disease in older adults. J Am Geriatr Soc 56:1693-1697. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.