Abstract
Background
Self-reported oral health questions (OHQs) are commonly used for epidemiological surveillance of periodontal disease (PD). Our objective is to investigate how OHQs are associated with well-established systemic comorbidities of PD and their impact on all-cause mortality. We hypothesize that OHQs exhibit similar associations with systemic comorbidities as PD.
Methods
Two independent datasets were utilized to achieve these objectives: the Women’s Health Study (WHS), a prospective cohort of middle-aged women with self-reported information on PD, OHQs, cardiovascular disease (CVD), diabetes, and osteoporosis in various timeframes (1992-present) and the United States National Health and Nutrition Examination Survey (NHANES), with data on OHQs and linked mortality (1999-2018). We applied multivariate logistic regression models and Cox proportional hazard regression survival analyses to test our hypotheses.
Results
WHS women who reported having PD until 2006 were more likely to later report deteriorating oral health, bone loss around their teeth, or periodontal treatment in 2018. Self-rated fair or poor oral health was independently associated with increased risk of CVD [odds ratio, OR(95%CI)=1.39(1.14-1.69), p<.001], diabetes [1.21(1.02-1.43), p=0.028], and osteoporosis [1.60(1.38-1.84), p<.001]. NHANES participants who self-rated fair or poor oral health had higher risks of all-cause mortality [hazard ratio HR(95%CI)=1.18(1.02-1.37), p=0.027].
Conclusions
Self-reported oral health demonstrated similar magnitude of associations with systemic comorbidities previously established with PD. Moreover, self-rated fair or poor oral health, suboptimal dental visits, or infrequent flossing were associated with increased all-cause mortality.
Practical implications
Our results support the utility of OHQs in assessing systemic connections, especially when clinical dental access is limited.
Keywords: oral health, surveys and questionnaires, periodontal disease, cardiovascular disease, diabetes mellitus, osteoporosis, mortality
INTRODUCTION
Periodontal disease (PD) affects around 42% of US adult, with 7.8% in its severe form.1, 2 PD is a microbially induced chronic inflammatory disease with severity influenced by factors such as smoking and genetics.3 Previous studies have established associations between PD and systemic conditions including cardiovascular disease (CVD),4, 5 type 2 diabetes,6–8 and osteoporosis.9 Furthermore, previous reports from large epidemiological cohorts have employed self-reported PD diagnosis to explore associations with cardiovascular outcomes.4, 10–13
Plain language oral health questions (OHQs) were developed by the Centers for Disease Control and Prevention and the American Academy of Periodontology for assessing periodontal status.14, 15 These OHQs have been validated and utilized in the National Health and Nutritional Examination Survey (NHANES),15 and also adopted in the UK Biobank16 and the Million Veterans Program.17 Self-reported OHQs provide new opportunities for exploring associations between oral health and systemic health in the absence of comprehensive clinical examinations. Importantly, survey approaches utilizing OHQs may facilitate accessing oral health among disadvantaged populations who lack dental care.18 However, unlike earlier reports based on self-reported PD diagnosis, there has been limited research on the association between OHQs and systemic health outcomes. In this report, we aim to investigate how OHQs are related to well-established systemic comorbidities of PD and their impact on all-cause mortality. We hypothesize that OHQs demonstrate comparable associations with systemic comorbidities as PD, providing valuable insights in situations where clinical assessment is unavailable.
The Women’s Health Study (WHS) is a prospective cohort study in the U.S., investigating initially healthy middle-aged women with multiple systemic health outcomes.19–21 Uniquely, WHS participants self-reported PD diagnosis from 1992 to 2006, and in 2018, validated OHQs was administered to the WHS participants. A previous report demonstrated that there was a greater likelihood of developing PD among WHS women with a family history of myocardial infarction.22 Therefore, in this report, we first aim to compare the recently obtained self-reported OHQs (2018) with earlier self-reported PD data (1992-2006) among the WHS women. Secondly, analyses were conducted to assess connections between 2018 OHQs and established systemic comorbidities of PD, namely CVD,4, 5 diabetes,6–8 and osteoporosis.9
Lastly, we re-examine such associations between OHQs and systemic health outcomes in the NHANES study by linking longitudinal mortality data. Researchers previously reported that the number of remaining natural teeth in NHANES data indicates risks of all-cause and disease-specific mortality.23 Since NHANES data includes dental visit history, self-rated teeth and gum health, and dental flossing data linked to mortality until 12/31/2019, our investigation independently examined whether self-reported oral health correlated with all-cause mortality, which complements the WHS study.
METHODS
Study design and population
Figure 1 study timeline displays the chronological data collections in the WHS and NHANES datasets.
Figure 1. Flow chart of study timeline.
This report analyzes two independent studies. First, the Women’s Health Study (WHS) is a prospective cohort study that originally began as a clinical trial in 1992-1994 to test the benefits and risks of low-dose aspirin and vitamin E in the primary prevention of cardiovascular disease and cancer. Annual follow-up questionnaires included self-reported periodontal disease (PD) status from the study inception until 2006. In 2018, oral health questions (OHQs) were administered to gather more recent information on oral health conditions. Systemic outcomes such as cardiovascular events and incidence of diabetes were assessed annually throughout the study period (until 2018). Additionally, osteoporosis was evaluated in 2009 using bone density reports. The WHS previously reported that women with baseline or incident PD had a higher risk of cardiovascular events. The second study used in this report is the National Health and Nutrition Examination Survey (NHANES). As of December 31, 2019, there were 59,062 participants eligible for mortality data linkage. We analyzed NHANES participants who had complete information on the number of teeth, demographics, education, income, smoking, body mass index (BMI), existing health conditions (cardiovascular diseases, diabetes, hypertension, and stroke), and physical activities. Among them, 29,510 NHANES participants provided information on their last dental visit (collected during 1999-2004 and 2011-2018), 21,488 responded to the question regarding ‘self-rated oral health (teeth and gums)’, and 18,392 answered the question regarding the frequency of dental flossing. Both the ‘self-rated oral health (teeth and gums)’ and ‘dental flossing frequencies’ were collected during 2009-2018. Detailed characteristics of NHANES participants can be found in Supplementary Table 1–4.
WHS.
The WHS was initially designed as a 10-year randomized clinical trial among 39,876 initially healthy female health professionals aged 45 years or older at baseline (1992-1994) evaluating low-dose aspirin and vitamin E for cancer and CVD prevention.19, 20 After the trial ended (2005), observational follow-up continued for an average of 27 years. Health and lifestyle data were collected at baseline and annually by follow-up questionnaires. Written consents were obtained for all WHS participants and the WHS study was approved by the Institutional Review Board. This report conforms to the STROBE guidelines.24
NHANES.
The NHANES program utilizes a stratified, multistage, cluster sampling design to gather representative health and nutrition data in the U.S. population. Data were collected via home interviews and health/dental examinations conducted in the mobile centers.25 Data from 1999 onwards are publicly available, released biennially with survey and examination manuals. The NHANES program is approved by the United States National Center for Health Statistics Research Ethics Review Board and conforms to the STROBE guideline.24 Among the 59,062 participants with mortality data (1999-2018), 29,554 had complete information on traditional risk factors, dental examination, and time span since their last dental visits. For this investigation, 21,490 participants self-rated their teeth and gum health, and 18,393 reported their dental flossing frequencies. Please see Supplementary Tables 1–4 for detailed participant characteristics.
Self-reported PD and OHQs in the WHS
WHS participants were asked at study entry (1992-1994) whether they had ever been diagnosed with a number of health outcomes, explicitly including PD.4 Newly diagnosed PD were also inquired at 36, 48, 60, 72, 84, 96, 108, and120 months during the trial, and in 2006.4 The question stems were ‘Since you started the trial (around 3 years ago)/In the past year, were you newly diagnosed with/have you had any of the following’, with ‘Periodontal disease’ as one possible response.16
In 2018, complementing the initial self-report of PD, additional oral health information was collected through the following questions:14, 15 Q1. Do you think you might have gum disease? (yes/no/don’t know); Q2. Overall, how would you rate the health of your teeth and gums? (excellent/very good/good/fair/poor/don’t know); Q3. Have you ever had treatment for gum disease such as scaling and root planing, sometimes called “deep cleaning”? (yes/no/don’t know); Q4. Have you ever been told by a dental professional that you lost bone around your teeth? (yes/no/don’t know); Q5. Aside from brushing your teeth with a toothbrush, in the LAST 7 DAYS, on how many DAYS did you use dental floss or any other device to clean between your teeth? (0/1/2/3/4/5/6/7); Q6. In the past 12 months, have you visited a dentist or dental hygienist? (yes/no/don’t know);26 Q7. How often do you usually visit the dental office for routine check-ups or cleanings? (more than once per year/once per year/less than once per year/don’t know).26
The analyses presented are based on responses from 14,663 women who completed self-reported PD prior to 2006, the 2018 OHQs, and all other systemic conditions ascertained, consented to blood sampling at baseline, and were of European ancestry. Please see Supplementary Table 5 for details of the administered WHS OHQs.
Covariates and systemic outcomes in the WHS.
In the WHS, covariates encompassed age at the OHQ distribution as well as education, smoking, body mass index (BMI), history of hypercholesterolemia or hypertension, and physical activity, assessed from the WHS baseline questionnaire.20 Endpoints of incident myocardial infarction and stroke during annual follow-up questionnaires were confirmed by physician or committee reviewing medical records.20 Two composite cardiovascular endpoints were created: major CVD, a composite of non-fatal myocardial infarction, non-fatal ischemic stroke, or death from cardiovascular disease, and total CVD, defined as major CVD, bypass surgery, or percutaneous coronary angioplasty. Self-reported incident type-2 diabetes was ascertained either by a telephone interview with a physician, or a self-administered questionnaire inquiring about symptoms, diagnostic testing, and use of diabetes medication.20, 27, 28 The 2009 self-reported diagnosis of osteoporosis was confirmed by a bone density scan.21
Covariates, existing systemic conditions, mortality, and number of remaining teeth in the NHANES.
Age, sex, race and ethnicity, education levels, income-to-poverty ratios, physical activities, and smoking status were obtained from the self-reported NHANES surveys.23, 25, 29 Physical activities were obtained from questionnaires during the home interview and categorized as sedentary (5 hours or more sedentary activities per day), insufficient, moderate (moderate intensity activities 150 minute or more per week), or vigorous (75 minutes or more vigorous-intensity aerobic activities or a combination of moderate and vigorous-intensity aerobic activities per week) as suggested by a recent guideline.29 BMI was from the NHANES body measures examination. Mortality was ascertained by the National Center for Health Statistics through record matching using the National Death Index death certificates and converted for public-use.23, 30 Existing systemic comorbidities, such as total cardiovascular disease (including congestive heart failure, coronary heart disease, angina and heart attack), and stroke, were summarized from the self-reported medical questionnaires.25 Hypertension was defined with either self-report, the use of antihypertensive medications, or averaged blood pressure greater than 140/90 mmHg. Diabetes mellitus was defined with either self-report, the presence of a fasting plasma glucose level greater than 140 mg/dL, plasma glucose level greater than 200 mg/dL, or the use of diabetic medications.25 The NHANES dental tooth count is charted as primary, permanent, implant, missing or residual roots, excluding third molars. We used the sum of permanent dentition (excluding residual roots) as the number of remaining natural teeth.23
Statistical analysis
WHS.
Group means and proportions were compared using t-tests for continuous variables and chi-square tests for categorical variables. Multivariate logistic regression was performed to examine associations between 2018 OHQs, self-reported PD until 2006, and systemic outcomes. Model 1 adjusted for periodontal disease-related risk factors: age at OHQs, baseline education, and smoking.3, 31 Model 2 included additional systemic outcome risk factors, such as BMI (kg/m2), physical activity, and history of hypertension and hypercholesteremia. Reference groups for the designated analyses were women who responded ‘yes’ to Q6 (visited a dentist within the past year) but answered ‘no’ to Q1 (do you think you might have gum disease), Q3 (ever had treatment for gum disease), and Q4 (been told you lost bone around teeth). Responses to Q2 (self-rated teeth and gum health) were grouped as binary, with ‘excellent/very good/good’ as the reference group compared to ‘fair/poor’. Similarly, women who flossed twice or less per week (Q5, ‘0/1/2’) were compared to the reference group of flossing three times a week or more (‘3/4/5/6/7’), and responses to dental check-ups or cleaning frequencies (Q7) were grouped as ‘less than once’ versus ‘once or more’ (reference group) per year. For subgroup analyses of self-rated teeth and gum health (Q2), each response was tested as an independent dummy variable with ‘excellent’ as the reference group. Trend for self-rated teeth and gum health responses was tested using linear regression, treating levels of responses as an ordinal categorical variable.
NHANES.
Group means and proportions were compared using t-tests or analysis of variance tests for continuous variables and chi-square tests for categorical variables. Cox-proportional hazard regression survival analyses were used to assess risks of all-cause mortality accounting for multiple risk factors. Mortality risks were evaluated among those who had last seen a dentist more than a year ago compared to those who visited a dentist within a year (reference group), among those who self-rated their teeth and gum health as worse (fair/poor) compared to those who reported ‘excellent/very good/good’ (reference group), and among those who flossed twice or less per week (‘0/1/2’) compared to those who flossed three times a week or more (‘3/4/5/6/7’). Covariates were adjusted for age, gender, race, smoking as in Model 1. Model 2 further included education, income-to-poverty ratio, BMI, physical activities, and pre-existing medical conditions (hypertension, cardiovascular diseases, diabetes, and stroke). In Model 3, all covariates in Model 2 plus the number of remaining teeth were adjusted for. Additional details of subgroups survival analyses as specified are provided in Supplementary Table 7–9 and Figure 2.
Figure 2. Kaplan-Meier survival curves by ‘last dental visit’, ‘self-rated oral health (teeth and gums)’, and ‘dental flossing frequencies’.
We plotted different survival curves for all-cause mortality based on binary or multiple groups of responses for NHANES participants’ ‘last dental visit’, ‘self-rated oral health (teeth and gums)’, and ‘dental flossing frequencies’. Hazard ratio estimates and p-values derived from the Cox proportional hazard regression survival analyses are provided in the upper left for each panel. Detailed estimates can be found in Supplementary Table 7–9.
All analyses were conducted using the R statistical software, and statistical significance was set at P<0.05.
RESULTS
Characteristics of WHS women based on their self-reported PD until 2006
A total of 14,663 women from the WHS were analyzed based on their self-reported PD until 2006, their 2018 OHQs, and their systemic health outcomes (Table 1). Compared to those without PD, women with PD diagnosis by 2006 were slightly older (mean age 77.87 vs. 76.79, p<.001), more likely to be current smokers (14.8% vs. 6.8%, p<.001), had an increased prevalence of hypercholesterolemia (28.5% vs. 25.7%), p=0.002, and lower levels of physical activity at baseline (rarely/never, 35.5% vs. 33.9%, p=0.045).
Table 1.
Characteristics of Women’s Health Study participants who responded to the 2018 oral health questions, by their self-reported periodontal disease (PD) status until 2006
| Self-reported Ever-having PD until 2006 | ||||
|---|---|---|---|---|
|
| ||||
| Total | No | Yes | P b | |
| N | 14663 | 11260 (77%) | 3403 (23%) | |
|
| ||||
| Age at OHQsa | 77.0 (5.6) | 76.79 (5.5) | 77.87 (5.8) | <.001 |
| Baseline information at the beginning of the clinical trial | ||||
| Education | 0.072 | |||
| < Bachelor’s degree | 7609 (51.9) | 5890 (52.3) | 1719 (50.5) | |
| ≥ Bachelor’s degree | 7054 (48.1) | 5370 (47.7) | 1684 (49.5) | |
| Smoking | ||||
| Current | 1269 (9) | 764 (6.8) | 505 (14.8) | <.001 |
| Past | 5510 (38) | 4049 (36.0) | 1461 (43.0) | |
| Never | 7976 (54) | 6441 (57.2) | 1435 (42.2) | |
| Body Mass Index (kg/m2) | 25.6 (4.7) | 25.55 (4.7) | 25.74 (4.7) | 0.049 |
| Hypertension | 2873 (20) | 2173 (19.3) | 700 (20.6) | 0.11 |
| Hypercholesterolemia | 3859 (26) | 2891 (25.7) | 968 (28.5) | 0.002 |
| Physical activity | ||||
| Rarely / Never | 5019 (34) | 3813 (33.9) | 1206 (35.5) | 0.045 |
| < 1 time per week | 2930 (20) | 2224 (19.8) | 706 (20.8) | |
| 1-3 times per week | 4930 (34) | 3820 (33.9) | 1110 (32.6) | |
| 4+ times per week | 1779 (12) | 1400 (12.4) | 379 (11.1) | |
|
| ||||
| Responses to OHQs in 2018 | ||||
| Last dental visit > 1 year ago | 1501 (10) | 1143 (10.2) | 358 (10.6) | 0.53 |
| Self-rated OH (fair/poor) | 1615 (11) | 954 (8.6) | 661 (19.9) | <.001 |
| Floss ≤2 per week | 2671 (18) | 2065 (18.5) | 606 (18.0) | 0.5 |
| Bone loss around teeth | 2884 (20) | 1382 (12.8) | 1502 (47.4) | <.001 |
| Check-up <1 per year | 1314 (9) | 987 (8.9) | 327 (9.8) | 0.13 |
| Had periodontal treatment | 3426 (23) | 1361 (12.3) | 2065 (62.9) | <.001 |
| Gum disease in 2018 | 1072 (7.3) | 418 (3.8) | 654 (20.1) | <.001 |
|
| ||||
| Systemic health outcomes across the observational study period | ||||
| Total CVD (2018) | 845 (6) | 593 (5.3) | 252 (7.4) | <.001 |
| Major CVD (2018) | 474 (3) | 325 (2.9) | 149 (4.4) | <.001 |
| Myocardial infarction (2018) | 235 (2) | 156 (1.4) | 79 (2.3) | <.001 |
| Stroke (2018) | 286 (2) | 206 (1.8) | 80 (2.4) | 0.06 |
| Diabetes (2017) | 1392 (9) | 1054 (9.4) | 338 (9.9) | 0.34 |
| Osteoporosis (2009) | 2266 (15) | 1665 (17.4) | 601 (20.9) | <.001 |
Data are presented as mean (standard deviation) for continuous variables; and n (%) for categorical variables.
P-values are from chi-square test for categorical variables or t-test for continuous variables comparing groups.
Abbreviations: Self-rated OH, self-rated oral health (teeth and gums); CVD, cardiovascular disease.
Regarding dental behaviors, no significant differences emerged in dental visit frequency, weekly flossing, or yearly dental check-ups between those with or without 2006 self-reported PD. Notably, a greater percentage of women reporting PD until 2006 rated their 2018 teeth and gum health as fair or poor (19.9% vs. 8.6%, p<.001). Similarly, women with 2006 self-reported PD were more likely to report bone loss around their teeth (47.4% vs. 12.8%, p<.001), gum disease diagnosis (20.1% vs. 3.8%, p<.001), and underwent periodontal treatment in 2018 (62.9% vs. 12.3%, p<.001). In terms of systemic health outcomes, women with 2006 self-reported PD exhibited a significantly higher likelihood of later reporting CVD events and osteoporosis (20.9% vs. 17.4%, p<.001). However, there was no significant difference in the prevalence of diabetes between women with or without 2006 self-reported PD.
2018 OHQs and systemic health outcomes in the WHS
Associations between the 2018 OHQs and systemic conditions are outlined in Table 2. In Model 1, adjustment accounted for established periodontal disease risk factors: age, education, and smoking.3, 31 In Model 2, we adjusted for additional recognized systemic outcome risk factors, namely, BMI, physical activity, and histories of hypertension and hypercholesteremia. Results in both models revealed significant associations with multiple systemic health outcomes, particularly among women who flossed twice or less weekly or who rated their teeth and gum health as fair or poor. Specifically, women with fair/poor self-rated teeth and gum health were more likely to report a history of total CVD [OR(95%CI), 1.39 (1.14-1.69), p<.001], major CVD [1.36(1.05-1.74), p=0.018], stroke [1.48(1.06-2.01), p=0.017], incident diabetes [1.21(1.02-1.43), p=0.028], and osteoporosis [1.60(1.38-1.84), p<.001]. All OHQ responses, except periodontal treatment, were significantly associated with osteoporosis. Lastly, women with infrequent oral health care behaviors (flossing two times or less per week, no dental visit or check-up within the past year) or those with fair or poor self-rated teeth and gum health had higher odds of incident diabetes. It is also worth noting that, comparable to self-reported PD until 2006, participants who self-rated their teeth and gum health as fair or poor in 2018 exhibited similar effect sizes for associations with total and major CVD, slightly higher for stroke, diabetes, and osteoporosis. Please see Supplementary Table 6 for detailed total CVD output (Model2).
Table 2.
Associations of WHS 2018 oral health questions with systemic health outcomes
| PD in 2006 | Floss <=2 /week | Last dental visit > 1 year ago | Check-up <1 per year | Bone loss around teeth | Self-OH (binary)* | Periodontal treatment | Gum disease in 2018 | |
|---|---|---|---|---|---|---|---|---|
| OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | OR (95%CI) | |
|
|
||||||||
| Model 1 | ||||||||
| Total CVD | 1.26 (1.08-1.47)** | 1.35 (1.14-1.59)*** | 1.28 (1.04-1.56)* | 1.33 (1.07-1.64)** | 0.98 (0.82-1.17) | 1.54 (1.27-1.86)*** | 1.03 (0.87-1.22) | 1.02 (0.77-1.33) |
| Major CVD | 1.34 (1.09-1.64)** | 1.32 (1.06-1.63)* | 1.06 (0.8-1.39) | 1.23 (0.92-1.62) | 0.88 (0.69-1.12) | 1.47 (1.14-1.87)** | 1.03 (0.82-1.28) | 1.08 (0.75-1.51) |
| Myocardial infarction | 1.42 (1.07-1.87)* | 1 (0.71-1.37) | 1.58 (1.1-2.21)** | 1.67 (1.15-2.36)** | 1.06 (0.76-1.45) | 1.31 (0.9-1.84) | 1.13 (0.83-1.52) | 1.17 (0.71-1.82) |
| Stroke | 1.16 (0.89-1.51) | 1.53 (1.16-1.99)** | 0.71 (0.46-1.05) | 0.8 (0.51-1.19) | 0.75 (0.53-1.04) | 1.55 (1.13-2.1)** | 0.93 (0.69-1.24) | 1.07 (0.66-1.64) |
| Incident diabetes | 1.06 (0.93-1.2) | 1.82 (1.6-2.07)*** | 1.76 (1.5-2.05)*** | 1.77 (1.5-2.09)*** | 0.92 (0.8-1.06) | 1.64 (1.4-1.91)*** | 0.89 (0.78-1.02) | 1.12 (0.9-1.37) |
| Osteoporosis (2009) | 1.17 (1.05-1.3)** | 1.12 (0.99-1.26) | 1.22 (1.05-1.42)* | 1.14 (0.97-1.34) | 1.36 (1.21-1.52)*** | 1.46 (1.27-1.68)*** | 1.04 (0.93-1.16) | 1.36 (1.15-1.61)*** |
| Model 2 | ||||||||
| Total CVD | 1.27 (1.08-1.49)** | 1.23 (1.03-1.46)* | 1.1 (0.89-1.35) | 1.18 (0.94-1.47) | 1.02 (0.84-1.22) | 1.39 (1.14-1.69)*** | 1.04 (0.88-1.24) | 1 (0.75-1.31) |
| Major CVD | 1.31 (1.07-1.61)** | 1.21 (0.96-1.5) | 0.96 (0.71-1.26) | 1.14 (0.85-1.51) | 0.91 (0.71-1.16) | 1.36 (1.05-1.74)* | 1.03 (0.82-1.28) | 1.06 (0.73-1.49) |
| Myocardial infarction | 1.42 (1.07-1.88)* | 0.87 (0.62-1.21) | 1.41 (0.97-1.99) | 1.54 (1.05-2.2)* | 1.09 (0.78-1.5) | 1.17 (0.8-1.67) | 1.11 (0.81-1.5) | 1.18 (0.72-1.85) |
| Stroke | 1.12 (0.85-1.46) | 1.45 (1.1-1.91)** | 0.61 (0.38-0.92)* | 0.69 (0.43-1.05) | 0.76 (0.54-1.05) | 1.48 (1.06-2.01)* | 0.96 (0.71-1.28) | 1.04 (0.63-1.61) |
| Incident diabetes | 1 (0.86-1.15) | 1.46 (1.27-1.68)*** | 1.23 (1.03-1.46)* | 1.33 (1.11-1.6)** | 0.96 (0.82-1.12) | 1.21 (1.02-1.43)* | 0.85 (0.73-0.99)* | 1.04 (0.82-1.29) |
| Osteoporosis (2009) | 1.2 (1.07-1.33)** | 1.2 (1.06-1.35)** | 1.34 (1.14-1.57)*** | 1.24 (1.05-1.47)* | 1.32 (1.17-1.47)*** | 1.6 (1.38-1.84)*** | 1.03 (0.92-1.15) | 1.38 (1.16-1.63)*** |
Models: The associations of individual oral health questions with systemic health outcomes were assessed by multivariate logistic regression adjusting for (Model 1) age at the time of OHQs, baseline information on educational and smoking, and in Model 2, variables in Model 1 plus body mass index (kg/m2), physical activities, history of hypertension and hypercholesteremia.
Abbreviations: Self-OH, self-rated oral health (teeth and gums), with binary grouping of levels of responses as excellent/very good/good (reference) vs. fair/poor; CVD, cardiovascular diseases.
Annotation for P-values:
P<0.05;
P<0.01,
P<0.001.
Ordinal self-rated oral health responses and systemic health outcomes
Associations between different self-rated teeth and gum health levels and systemic health outcomes are detailed in Table 3. Deteriorating oral health ratings (from excellent to poor) corresponded to elevated odds of experiencing various systemic comorbidities. For instance, odds of total CVD events increased from ‘excellent’(reference) to ‘very good’ [OR(95% CI), 1.04(0.83-1.31), p=0.72], ‘good’ [1.27(1.02-1.59), p=0.037], ‘fair’ [1.51(1.15-1.98), p=0.003], and ‘poor’ [1.92(1.22-2.94), p=0.003]. Importantly, the increasing trends of odds ratios associated with total CVD, diabetes, and osteoporosis across self-reported oral health responses were significant in both Models 1 and 2 (p<.001).
Table 3.
Associations with systemic health outcomes by levels of responses to the WHS self-rated oral health (teeth and gums) question
| Self-Rated Oral Health of Teeth and Gums | ||||||
|---|---|---|---|---|---|---|
|
| ||||||
| Groups | Excellent | Very good | Good | Fair | Poor | Ptrend |
|
|
||||||
| N (%) | 2651 (18) | 5683 (39) | 4512 (31) | 1376 (10) | 239 (2) | |
|
| ||||||
| OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | OR (95% CI) | ||
|
| ||||||
| Model 1 | ||||||
|
| ||||||
| Total CVD | Ref. | 1.04 (0.83-1.31) | 1.38 (1.11-1.72)** | 1.71 (1.31-2.22)*** | 2.26 (1.44-3.44)*** | <.001 |
| Major CVD | Ref. | 0.97 (0.73-1.3) | 1.37 (1.04-1.83)* | 1.63 (1.16-2.3)** | 1.84 (0.99-3.21) | <.001 |
| Myocardial infarction | Ref. | 0.89 (0.6-1.33) | 1.28 (0.88-1.91) | 1.22 (0.74-2.0) | 2.23 (1.03-4.42)* | 0.006 |
| Stroke | Ref. | 1.16 (0.8-1.71) | 1.36 (0.94-2.0) | 2.06 (1.34-3.19)** | 1.11 (0.38-2.62) | 0.008 |
| Incident diabetes | Ref. | 1.14 (0.95-1.36) | 1.79 (1.51-2.14)*** | 2.02 (1.63-2.51)*** | 3.1 (2.16-4.39)** | <.001 |
| Osteoporosis (2009) | Ref. | 0.98 (0.85-1.11) | 1.17 (1.02-1.34)* | 1.49 (1.24-1.78)*** | 1.85 (1.30-2.61)*** | <.001 |
|
| ||||||
| Model 2 | ||||||
|
| ||||||
| Total CVD | Ref. | 1.04 (0.83-1.31) | 1.27 (1.02-1.59)* | 1.51 (1.15-1.98)** | 1.92 (1.22-2.94)** | <.001 |
| Major CVD | Ref. | 0.97 (0.73-1.3) | 1.28 (0.96-1.71) | 1.47 (1.04-2.09)* | 1.65 (0.89-2.89) | 0.001 |
| Myocardial infarction | Ref. | 0.91 (0.61-1.37) | 1.18 (0.8-1.77) | 1.06 (0.63-1.75) | 1.96 (0.9-3.92) | 0.05 |
| Stroke | Ref. | 1.16 (0.8-1.71) | 1.33 (0.92-1.96) | 1.95 (1.26-3.04)** | 1.06 (0.36-2.51) | 0.02 |
| Incident diabetes | Ref. | 1.11 (0.92-1.34) | 1.48 (1.23-1.78)*** | 1.45 (1.15-1.83)** | 1.69 (1.13-2.5)** | <.001 |
| Osteoporosis (2009) | Ref. | 0.99 (0.87-1.13) | 1.26 (1.1-1.45)*** | 1.66 (1.38-1.99)*** | 2.3 (1.6-3.28)*** | <.001 |
Models: The associations of each response of self-assessed oral health as an indicator variable and the trend for levels of responses as an ordinal categorical variable to systemic outcomes were assessed using multivariate logistic regression adjusting for (Model 1) age at the time of self-rated oral health responses, baseline information of educational and smoking (3 groups), and additional in Model 2 with body mass index (kg/m2), physical activities, history of hypertension and hypercholesteremia.
Abbreviations: CVD, cardiovascular diseases.
Annotation for P-values:
P<0.05;
P<0.01,
P<0.001.
All-cause mortality risks among NHANES participants
Having identified significant associations between self-reported oral health and systemic health outcomes in the WHS, we conducted an independent analysis using the NHANES data to examine whether OHQs exhibit similar associations with mortality (Table 4). Detailed NHANES participant characteristics are provided in Supplementary Tables 1–4. The risks of all-cause mortality were estimated based on the NHANES participants’ OHQ responses. Participants who reported no dental visit within the past year had higher all-cause mortality risks compared to those who visited a dentist within the past year [HR(95%CI), 1.17(1.07-1.27), p<.001]. Individuals self-rating teeth and gum health as fair or poor also had elevated risks of all-cause mortality compared to those rating it as excellent/very good/good [HR(95%CI), 1.18(1.02-1.37), p=0.027]. Furthermore, individuals flossing two times or less weekly exhibited higher mortality risks compared to those flossing three times or more per week [HR(95%CI), 1.28(1.07-1.52), p=0.006]. Please refer to Figure 2 for survival curves and Supplementary Tables 7–9 for detailed estimates from these analyses. It is noteworthy that the subgroup analyses, further dividing dental visit, flossing frequencies, and self-rated oral health into three or more groups, consistently showed increasing mortality risks compared to the reference group.
Table 4.
Risk of all-cause mortality in the National Health and Nutrition Examination Survey based on last dental visit, self-rated oral health (teeth and gums), or frequency of dental flossing
| All-cause Mortality | Model 1 | Model 2 | Model 3 |
|---|---|---|---|
| N (event) | 29510 (2937) | 29510 (2937) | 29510 (2937) |
|
|
|||
| Last dental visit | HR (95% CI) | HR (95% CI) | HR (95% CI) |
|
|
|||
| ≤ 1 year | Ref. | Ref. | Ref. |
| > 1 year | 1.52 (1.41-1.64)*** | 1.26 (1.16-1.36)*** | 1.17 (1.08-1.27)*** |
|
| |||
| N (event) | 21488 (791) | 21488 (791) | 21488 (791) |
|
|
|||
| Self OH | HR (95% CI) | HR (95% CI) | HR (95% CI) |
|
|
|||
| Excellent ~ Good | Ref. | Ref. | Ref. |
| Fair / Poor | 1.39 (1.20-1.61)*** | 1.18 (1.02-1.38)* | 1.18 (1.02-1.37)* |
|
| |||
| N (event) | 18392 (775) | 18392 (775) | 18392 (775) |
|
|
|||
| Flossing frequency | HR (95% CI) | HR (95% CI) | HR (95% CI) |
|
|
|||
| ≥ 3 per week | Ref. | Ref. | Ref. |
| ≤ 2 per week | 1.74 (1.49-2.03)*** | 1.46 (1.25-1.71)*** | 1.27 (1.07-1.51)** |
Cox proportional hazard regression survival analyses were adjusted for the following covariates: Model 1: age, gender, race (4 groups) and smoking (3 groups); Model 2: covariates in Model 1, education (3 groups), income-to-poverty ratio, body mass index, physical activities (4 groups), and pre-existing medical conditions of hypertension, total cardiovascular diseases, diabetes, and stroke; Model 3: covariates in Model 2 and the total number of remaining teeth.
Abbreviations: Self OH, self-rated oral health (teeth and gums)
Annotation for P-values:
P<0.05;
P<0.01,
P<0.001.
In summary, our investigations, spanning two independent large studies, underscore robust connections between self-reported oral health, systemic health outcomes, and heightened risks of all-cause mortality.
DISCUSSION
The link between poor oral health, periodontal disease and heightened mortality and frailty is well-documented.32–34 In alignment with prior findings,4–9 our analysis of self-reported OHQs from WHS women reinforces robust associations with systemic health outcomes. Similarly, NHANES participants displaying infrequent dental visits, limited flossing, or self-rated fair or poor oral health faced elevated risks of all-cause mortality, independent of their remaining teeth count. Therefore, our present results emphasize the significance of self-reported OHQs in delineating oral health’s connection with systemic health outcomes across large-scale studies. Additionally, our findings of infrequent dental flossing among WHS women significantly linking to higher odds of stroke, diabetes and osteoporosis have not been reported as of today.
Existing evidence indicates that oral health behaviors, such as the frequency of toothbrushing, are linked to cardiovascular events35 as well as changes markers like C-reactive protein,36, 37 fibrinogen,36 and vascular function.37 Recent analysis utilizing a composite score based on self-reported oral health identified heightened incident hypertension risks tied to poor oral health conditions.38 Our results based on OHQ responses are consistent with previous studies that established epidemiological associations of PD or tooth loss with CVD,4, 39 diabetes,6–8 and osteoporosis.9, 40, 41 Thus, combining with prior evidence,35–37 toothbrushing and dental flossing warrant further exploration as preventive interventions for systemic health outcomes.
Remarkably, our data elucidates that WHS women reporting PD until 2006 were likelier to self-assess their teeth and gum health as fair or poor, report undergoing periodontal treatment, bone loss around their teeth, or consider themselves as still having gum disease in 2018. These findings reveal strong correlations between the responses across two distinct time points. Our study has several limitations. There is no clinical data related to the WHS self-reported OHQs, except a limited subset of women with retrieved dental records. Among those 63 women, fewer teeth were noted among those with worse oral health conditions (previously reported)42 Secondly, cross-sectional nature of the presented WHS analyses and different timing of PD and OHQ information limit causal interpretation, necessitating further research for additional exploration and validation.
To reinforce the associations between self-reported OHQs and systemic health outcomes, we independently conducted survival analyses for all-cause mortality in the NHANES. This was separate from the WHS observations but similarly relied on self-reported oral health and systemic disease data based on a representative population. Results indicated higher risks of all-cause mortality for participants without recent dental visits within the past year, those self-rating teeth and gum health as fair or poor, or individuals flossing two times or less weekly. Hence, the WHS and NHANES derived results together underscore the oral-systemic connection and expand the rationale for employing self-reported OHQs in population studies.
In conclusion, our study elucidates the value of self-reported oral health indicators, especially the duration since the last dental visit, self-assessed teeth and gum health, and dental flossing frequencies. These metrics prove invaluable for assessing oral health in extensive studies or data-rich databases where thorough clinical examinations or dental records might be lacking. Moreover, our investigations, spanning two independent large studies, underscore significant relationship between self-reported oral health, systemic health outcomes, and heightened risks of all-cause mortality. The significant oral-systemic correlations we have identified not only reinforced the well-established links between periodontal disease and systemic comorbidities,4–9, 43 as previously reported, but also highlight the need for future investigations. Furthermore, our finding of increased all-cause mortality among those having suboptimal dental visits emphasizes the urgency of addressing barriers to accessing dental care.
Supplementary Material
Acknowledgements
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The WHS study is supported by HL043851, HL080467, HL099355, from the National Heart, Lung, and Blood Institute, and by CA047988 and CA182913 from the National Cancer Institute. Dr. Yu is supported by NIH/NIDCR K23DE026804 and the NIH Loan Repayment Programs to conduct this research work.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of interest
The authors have declared no conflict of interests.
Clinical trial registry information
The Women’s Health Study is registered under NCT00000479 in the clinicaltrial.gov.
REFERENCES:
- 1.Eke PI, Thornton-Evans GO, Wei L, et al. Periodontitis in US Adults: National Health and Nutrition Examination Survey 2009-2014. J Am Dent Assoc 2018;149(7):576–88 e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Eke PI, Borgnakke WS, Genco RJ. Recent epidemiologic trends in periodontitis in the USA. Periodontol 2000 2020;82(1):257–67. [DOI] [PubMed] [Google Scholar]
- 3.Meyle J, Chapple I. Molecular aspects of the pathogenesis of periodontitis. Periodontol 2000 2015;69(1):7–17. [DOI] [PubMed] [Google Scholar]
- 4.Yu YH, Chasman DI, Buring JE, Rose L, Ridker PM. Cardiovascular risks associated with incident and prevalent periodontal disease. J Clin Periodontol 2015;42(1):21–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Tonetti MS, Van Dyke TE, working group 1 of the joint EFPAAPw. Periodontitis and atherosclerotic cardiovascular disease: consensus report of the Joint EFP/AAP Workshop on Periodontitis and Systemic Diseases. J Periodontol 2013;84(4 Suppl):S24–9. [DOI] [PubMed] [Google Scholar]
- 6.Genco RJ, Borgnakke WS. Diabetes as a potential risk for periodontitis: association studies. Periodontol 2000 2020;83(1):40–45. [DOI] [PubMed] [Google Scholar]
- 7.Demmer RT, Holtfreter B, Desvarieux M, et al. The influence of type 1 and type 2 diabetes on periodontal disease progression: prospective results from the Study of Health in Pomerania (SHIP). Diabetes Care 2012;35(10):2036–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Chapple IL, Genco R, working group 2 of the joint EFPAAPw. Diabetes and periodontal diseases: consensus report of the Joint EFP/AAP Workshop on Periodontitis and Systemic Diseases. J Periodontol 2013;84(4 Suppl):S106–12. [DOI] [PubMed] [Google Scholar]
- 9.Penoni DC, Fidalgo TK, Torres SR, et al. Bone Density and Clinical Periodontal Attachment in Postmenopausal Women: A Systematic Review and Meta-Analysis. J Dent Res 2017;96(3):261–69. [DOI] [PubMed] [Google Scholar]
- 10.Joshipura KJ, Rimm EB, Douglass CW, et al. Poor oral health and coronary heart disease. J Dent Res 1996;75(9):1631–6. [DOI] [PubMed] [Google Scholar]
- 11.Momen-Heravi F, Babic A, Tworoger SS, et al. Periodontal disease, tooth loss and colorectal cancer risk: Results from the Nurses’ Health Study. Int J Cancer 2017;140(3):646–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Joshipura KJ, Hung HC, Rimm EB, Willett WC, Ascherio A. Periodontal disease, tooth loss, and incidence of ischemic stroke. Stroke 2003;34(1):47–52. [DOI] [PubMed] [Google Scholar]
- 13.Hung HC, Willett W, Merchant A, et al. Oral health and peripheral arterial disease. Circulation 2003;107(8):1152–7. [DOI] [PubMed] [Google Scholar]
- 14.Eke PI, Genco RJ. CDC Periodontal Disease Surveillance Project: background, objectives, and progress report. J Periodontol 2007;78(7 Suppl):1366–71. [DOI] [PubMed] [Google Scholar]
- 15.Eke PI, Dye BA, Wei L, et al. Self-reported measures for surveillance of periodontitis. J Dent Res 2013;92(11):1041–7. [DOI] [PubMed] [Google Scholar]
- 16.Shungin D, Haworth S, Divaris K, et al. Genome-wide analysis of dental caries and periodontitis combining clinical and self-reported data. Nat Commun 2019;10(1):2773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Gaziano JM, Concato J, Brophy M, et al. Million Veteran Program: A mega-biobank to study genetic influences on health and disease. J Clin Epidemiol 2016;70:214–23. [DOI] [PubMed] [Google Scholar]
- 18. .Oral Health in America: Advances and Challenges. Bethesda (MD); 2021. [PubMed] [Google Scholar]
- 19.Ridker PM, Chasman DI, Zee RY, et al. Rationale, design, and methodology of the Women’s Genome Health Study: a genome-wide association study of more than 25,000 initially healthy american women. Clin Chem 2008;54(2):249–55. [DOI] [PubMed] [Google Scholar]
- 20.Ridker PM, Cook NR, Lee IM, et al. A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women. N Engl J Med 2005;352(13):1293–304. [DOI] [PubMed] [Google Scholar]
- 21.Estrada K, Styrkarsdottir U, Evangelou E, et al. Genome-wide meta-analysis identifies 56 bone mineral density loci and reveals 14 loci associated with risk of fracture. Nat Genet 2012;44(5):491–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Yu YH, Doucette-Stamm L, Rogus J, et al. Family History of MI, Smoking, and Risk of Periodontal Disease. J Dent Res 2018;97(10):1106–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Yu YH, Cheung WS, Steffensen B, Miller DR. Number of teeth is associated with all-cause and disease-specific mortality. BMC Oral Health 2021;21(1):568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007;370(9596):1453–7. [DOI] [PubMed] [Google Scholar]
- 25.Dye BA, Thornton-Evans G. A brief history of national surveillance efforts for periodontal disease in the United States. J Periodontol 2007;78(7 Suppl):1373–9. [DOI] [PubMed] [Google Scholar]
- 26.Del Pinto R, Monaco A, Ortu E, et al. Access to dental care and blood pressure profiles in adults with high socioeconomic status. J Periodontol 2022;93(7):1060–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Pradhan AD, Cook NR, Manson JE, Ridker PM, Buring JE. A randomized trial of low-dose aspirin in the prevention of clinical type 2 diabetes in women. Diabetes Care 2009;32(1):3–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Song Y, Manson JE, Buring JE, Liu S. A prospective study of red meat consumption and type 2 diabetes in middle-aged and elderly women: the women’s health study. Diabetes Care 2004;27(9):2108–15. [DOI] [PubMed] [Google Scholar]
- 29.Piercy KL, Troiano RP, Ballard RM, et al. The Physical Activity Guidelines for Americans. JAMA 2018;320(19):2020–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Statistics NCfH. NCHS Public-use Linked Mortality Files: National Center for Health Statistics; 2022. [Google Scholar]
- 31.Tomar SL, Asma S. Smoking-attributable periodontitis in the United States: findings from NHANES III. National Health and Nutrition Examination Survey. J Periodontol 2000;71(5):743–51. [DOI] [PubMed] [Google Scholar]
- 32.Tanaka T, Takahashi K, Hirano H, et al. Oral Frailty as a Risk Factor for Physical Frailty and Mortality in Community-Dwelling Elderly. J Gerontol A Biol Sci Med Sci 2018;73(12):1661–67. [DOI] [PubMed] [Google Scholar]
- 33.Gupta A, Felton DA, Jemt T, Koka S. Rehabilitation of Edentulism and Mortality: A Systematic Review. J Prosthodont 2019;28(5):526–35. [DOI] [PubMed] [Google Scholar]
- 34.Sabbah W, Slade GD, Sanders AE, Bernabe E. Denture wearing and mortality risk in edentulous American adults: A propensity score analysis. J Dent 2020;100:103360. [DOI] [PubMed] [Google Scholar]
- 35.Janket SJ, Lee C, Surakka M, et al. Oral hygiene, mouthwash usage and cardiovascular mortality during 18.8 years of follow-up. Br Dent J 2023:1–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.de Oliveira C, Watt R, Hamer M. Toothbrushing, inflammation, and risk of cardiovascular disease: results from Scottish Health Survey. BMJ 2010;340:c2451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Matsui S, Maruhashi T, Kishimoto S, et al. Poor tooth brushing behavior is associated with high risk of cardiovascular events: A prospective observational study. Int J Cardiol 2022;350:111–17. [DOI] [PubMed] [Google Scholar]
- 38.Carra MC, Fessi S, Detzen L, et al. Self-reported periodontal health and incident hypertension: longitudinal evidence from the NutriNet-Sante e-cohort. J Hypertens 2021;39(12):2422–30. [DOI] [PubMed] [Google Scholar]
- 39.Lee HJ, Choi EK, Park JB, Han KD, Oh S. Tooth Loss Predicts Myocardial Infarction, Heart Failure, Stroke, and Death. J Dent Res 2019;98(2):164–70. [DOI] [PubMed] [Google Scholar]
- 40.Wakai K, Naito M, Naito T, et al. Tooth loss and risk of hip fracture: a prospective study of male Japanese dentists. Community Dent Oral Epidemiol 2013;41(1):48–54. [DOI] [PubMed] [Google Scholar]
- 41.Yu YH, Cheung WS, Miller DR, Steffensen B. Number of teeth is associated with hip fracture and femoral neck bone mineral density in the NHANES. Arch Osteoporos 2021;16(1):105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Yu YH, Steffensen B, Ridker PM, Buring JE, Chasman DI. Candidate loci shared among periodontal disease, diabetes and bone density. Front Endocrinol (Lausanne) 2022;13:1016373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Lamster IB, Lalla E, Borgnakke WS, Taylor GW. The relationship between oral health and diabetes mellitus. J Am Dent Assoc 2008;139 Suppl:19S–24S. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.