The relationship of aging, complete tooth loss, and having a dental visit in the last 12 months

Abstract Objectives To evaluate the extent to which dental health care visits in the past year differed among older adults with and without edentulism. Material and Methods We conducted a cross‐sectional study using the 2017 Medical Expenditure Panel Survey among participants aged ≥50 years (n = 10,480, weighted = 112,116,641). Two self‐reported outcome variables were used: loss of all teeth from upper and lower jaws (yes/no) and dental visit in the last 12 months (yes/no). Logistic models were used to estimate adjusted odds ratios (aOR) and 95% confidence intervals (CI). Results Overall, 11.4% of the non‐institutionalized U.S. population aged ≥50 years were edentulous; the prevalence was higher in those with advanced age. Adherence to annual oral health visits was 16% among those with edentulism, 52% among those without. The prevalence of dental care visits in the past year was higher among those with advanced age without edentulism, but for those with edentulism, the odds of visiting a dental care provider was lower in all age groups compared to those 50–59 years ((60–69 years): aOR: 0.58, CI:0.36–0.95; (70–79 years): aOR: 0.51, CI: 0.30–0.88; (≥ 80 years): aOR: 0.45, CI: 0.26–0.80)). Conclusion Although the prevalence of edentulism was higher in those with advanced age, oral health visits during the last 12 months were less frequent in older adults with edentulism. Interventions to improve adherence to dental care recommendations in the growing aging population are warranted.

Treatment for edentulism involves the provision of complete dentures to improve chewing and quality of life (Kroll et al., 2018;Krunic, Kostic, Petrovic, & Igic, 2015;Muller, Morais, & Feine, 2008). Annual visits with oral health care providers are recommended for persons with dental prostheses to evaluate condition and fit (American College of Prosthodontists: Position Statement, n.d.). Recent, population-based U.S. studies estimating adherence to the guidelines for annual evaluation of prostheses are lacking. Although historically dental insurance among retired people is low (Willink, Schoen, & Davis, 2017), increased enrollment in Medicare Advantage plans offering supplemental dental benefits has improved coverage; four in 10 Medicare Advantage enrollees had dental coverage in 2016 (Willink, 2019).

While supplemental dental insurance included in Medicare
Advantage plans may have improved adherence to annual oral health care visits, contemporary studies documenting routine care by age and edentulism status are lacking.

| AIMS
Using a U.S. population-based data resource, our study sought to provide contemporary estimates of the relationship between age and edentulism among older adults and to evaluate the extent to which dental health care visits in the past year differed among older adults with and without edentulism. We hypothesized that adherence to annual dental health care visits would decline with advanced age and would be greater among people with edentulism relative to those without edentulism.

| Ethics statement
Data were collected through a national survey that was approved by the Westat Institutional Review Board and the Office for Protection from Research Risk (Hill, Zuvekas, & Zodet, 2011). Participants provided informed consent. The data were de-identified and anonymized.
Data were released as open-source and available for public use and pose no risk to participants or individuals collecting the data.

| Study population
The MEPS 2017 household component included data from 31,880 participants. We excluded 21,400 participants <50 years of age and responses coded as "refused," "do not know," "not ascertained" on complete tooth loss of upper and lower jaws, born in the United States, education, and marital status. The final analytic sample included 10,480 respondents ≥50 years of age (weighted n = 112,116,641). Weighted respondents were cross-checked with U.S. census estimates for 2017, which totaled 114,217,553 (US Census Bureau: American Fact Finder, n.d.).

| Study outcomes
Teeth are fundamental, and pivotal in all aspects of individual and social function including the capacity to macerate food and quality of life (Tan, Peres, & Peres, 2016). As such, our primary study outcome was self-reported complete tooth loss of all upper and lower teeth (yes/no). The outcome was based on MEPS question: "Have you… lost all upper and lower teeth?" (Griffin et al., 2014).
It is imperative for edentulous persons to maintain an active relationship with an oral health care provider to ensure their prostheses, should they have them, are functioning optimally (American College of Prosthodontists: Position Statement, n.d.). As such, the American College of Prosthodontists recommends annual visits with oral health care providers (Felton et al., 2011). The outcome for the second aim was self-reported visit with an oral health care provider during 12 months prior to interview. Our outcome was based on MEPS question: "How many dental visits in the last 12 months?" (zero/one or more; Griffin et al., 2014;Meyerhoefer, Zukekas, Farkhad, Moeller, & Manski, 2019).

| Covariates
Individual and social characteristics were considered that may influence the ability to access dental services, be it through insurance or financial capacity. Personal characteristics included race/ethnicity, gender, education status (no degree, high school diploma/general education diploma, some college or beyond), born in the United States (yes, no), marital status (married, single, never married), family income as a percentage of the poverty line (poor/negative, near poor, low income, middle income, high income), dental insurance (yes/no), health insurance (private/public/none), dental visit in the last year (yes/no), active smoker in the last 12 months (yes/no). Education status was consolidated into three categories: no degree, high school diploma/general education diploma, and some college or greater. We categorized participants according to their race and ethnicity as Hispanic, non-Hispanic Black, non-Hispanic Asian, or non-Hispanic White. Mixed race/ethnicity persons were included as Hispanic if they identified as such (e.g., Asian-Hispanic, Black-Hispanic, White-Hispanic) or non-Hispanic mixed race.

| Data analysis
MEPS provided survey weights and approaches for handling single unit datapoints in the weighted measurement were followed (Wun, Ezzati-Rice, Diaz-Tena, & Greenblatt, 2007). Descriptive statistics were used to characterize the population according to edentulism. Analyses were stratified by age group. We calculated percentages for categorical variables. Bivariate associations were examined using Pearson Chi square tests for categorical variables. p-Values <.05 were considered statistically significant (two-sided tests). We then estimated the prevalence of edentulism by age (in years) and depicted this graphically ( Figure 1).
Logistic regression modeling was used to analyze the relationship between the primary determinant (four categories of age) and edentulism adjusting for potential confounders. We adjusted the partial odds ratio for sex, race/ethnicity, and marital status. We further adjusted the odds ratio for income level, smoking status, and dental insurance.
For the second aim, we first estimated the percent of participants who reported having an oral health care visit in the past 12 months, stratified by edentulism status and specific for each year of age ( Figure 2). We then conducted a stratified analysis by edentulism status using logistic regression modeling to examine the association between age and visiting an oral health care provider in the last 12 months. Partially adjusted odds ratios included sex, race/ethnicity, and marital status, and the fully adjusted model added family income and dental insurance. Smoking status was excluded from modeling since smoking status has lesser impact on edentulous persons visiting a dental care provider than other potential variables (Dolan, Gilbert, Duncan, & Foerster, 2008;Mittchell & Bennett, 2013). We used STATA version 15.1 (College Station, TX) for all analyses.

| RESULTS
Data from MEPS 2017 indicate that 11.4% of U.S. persons aged ≥50 years of age were edentulous (Table 1), and the prevalence of F I G U R E 1 Prevalence of edentulism by age among adults ≥50 years in the United States (2017) F I G U R E 2 Prevalence of dental visit in previous 12 months by age among adults aged ≥50 years, by edentulism status (2017

| DISCUSSION
There were two main findings from our study. First, using populationbased contemporary data, this study confirms the association between advanced age and edentulism. Overall 11.4% of adults aged T A B L E 4 Association between age and oral health care provider visit in the last 12 months, stratified by edentulism (2017) Age ( ≥50 years were edentulous; the prevalence increased in those with advanced age. While 6.2% of those aged 50-59 years were edentulous, 27.7% of those ≥80 years of age were edentulous. Second, adherence to guidelines regarding annual oral health provider visits was low with about half of those without edentulism and one in six of those with edentulism reporting a visit with an oral health provider in the past year. Furthermore, the relationship between age and use of oral health services in the past 12 months differed by edentulism status. Adherence to annual oral health care visits was less prevalent in older age groups among edentulous adults and was more prevalent in older age groups among non-edentulous adults. Using contemporary data, our population-based study confirmed the association between advanced age and edentulism. While there is some debate about the factors that contribute to complete tooth loss, people are more likely to lose their natural teeth as they age (Griffin et al., 2012;Hybels et al., 2016;Kanasi et al., 2016). America faces "a silent epidemic" of oral diseases and older adults are at greatest risk (Centers for Disease Control, n.d.). In the United States, older adults develop coronal caries at "approximately one new cavity per year" (Griffin, Griffin, Swann, & Zlobin, 2004;Griffin, Griffin, Swann, & Zlobin, 2005 (Mueller, Naharro, & Carlsson, 2007). Notably, a recent study of community-dwelling persons ages ≥65 in Italy found a 44% prevalence of edentulism among participants with some 17.5% of persons with edentulism using no protheses (Musacchio et al., 2007). Further research is needed to examine the potential financial expenditures of caring for aging persons oral health needs (Harford, 2009) as well as determining the availability of a qualified workforce.
Adherence to recommendations for annual oral health visits is poor among older adults. As such, population-level analyses examining the use of oral health services by age are important given the oral health care needs for this vulnerable population (Griffin et al., 2012). Edentulous persons require annual routine care from oral health providers (Felton et al., 2011). Individuals with edentulism require a complete denture to have a fully functional maceration capacity (Ekelund, 1989 Only 16% of the overall 13 million persons who have edentulism reported visiting an oral health care provider in the last 12 months.
That number in itself is troubling given the maintenance required for a complete denture. Unfortunately, the likelihood of a person visiting an oral health provider decreases with age, leaving persons who are more likely to have edentulism being the least likely to visit an oral health care provider. In our study, adjusting for dental insurance did not explain the decline in adherence to recommended annual oral health care visits. As such, lack of dental insurance may not be the rate limiting factor. Further research to understand factors associated with lack of adherence to routine oral health provider care among older adults is warranted.
The study strengths and limitations must be considered. Data were drawn from a nationally representative sample that provides vital insight into the oral health status of aging persons in the United States, and oral health utilization of a vulnerable group of persons (Christian et al., 2013). Our primary outcome variables from MEPS household data are self-reported and susceptible to response bias.
People may feel uncomfortable speaking about their oral health and concerned about social perceptions if they have edentulism (Lee, Shieh, Yang, Tsai, & Wang, 2007). MEPS interviews are conducted over the phone and persons are able to respond to the interviewers without fear of visual feedback (Hill et al., 2011). Further, studies support the validity of self-reported dentition in older adults (Douglass, Berlin, & Tennstedt, 1991).

| CONCLUSION
Edentulism is affecting a significant portion of our noninstitutionalized persons aged ≥50 years and has a profound impact on diet, overall health, and pre-existing conditions (Polzer, Schimmel, Mueller, & Biffar, 2011). People need teeth in order to chew and they require functional, well-cared for prostheses if they do not have a natural dentition. Our data show that persons are not receiving the annual care required to care for their complete denture, and that lack of dental insurance does not explain the age-related decrease in prevalence of adherence to annual oral health care provider visits.
Research is needed to understand how to better improve adherence to recommended annual oral health care provider visits for aging populations, particularly among older edentulous adults who have the greatest need for intervention.

ACKNOWLEDGMENTS
Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under award number UL1-TR001453. The content is solely