Oral frailty and gait performance in community-dwelling older adults: findings from the Takashimadaira study

Purpose: This cross-sectional study compared gait performance between community-dwelling older adults with and without accumulated deficits in oral health, defined as oral frailty. Methods: A total of 1,082 individuals (439 men and 643 women; mean age, 77.1 years) from the Takashimadaira study were included in the current analysis. Based on a multifaceted oral health assessment, oral frailty was defined as having three or more of the following six components: (i) fewer teeth, (ii) low masticatory performance, (iii) low articulatory oral motor skills, (iv) low tongue pressure, (v) difficulties in eating, and (vi) swallowing. Eight gait parameters were assessed using an electronic walkway. Gait characteristics comparison between groups with and without oral frailty was performed using multiple linear regression models. Models were adjusted for age, sex, educational status, income, smoking, drinking, physical activity level, height, body mass index, comorbidities, and the presence of chronic pain. Results: Oral frailty was observed in 227 (21.0%) participants. After adjusting for potential confounders, the participants with oral frailty had slower gait speed, shorter stride and step length, wider step width, and longer double support duration as well as higher variability of stride length and step length. Conclusions: Oral frailty was associated with poor gait performance among community-dwelling older adults.


Introduction
Gait performance is an important aspect of geriatric health. Poor gait performance, such as slower speed and greater gait variability, significantly increases the risk of falling among older adults [1][2][3]. Falls can result in fractures, hospitalization, mobility reduction, and mortality [4]. In addition, poor gait performance and its adverse sequelae lead to inactivity and vulnerability to secondary impairments, such as frailty and dementia [5].
There have been several reports on the association between poor oral health status, including fewer teeth, lack of posterior occlusal support, and low oral function, with impairment of balance and gait as well as reduced lower limb strength [6][7][8][9][10][11]. Several potential explanations for these relationships include (i) the impact of occlusal interference and change of mandibular position on the body posture control/stability and knee muscular performance [6,[11][12][13][14] as well as (ii) the effect of insufficient dietary intake caused by masticatory and swallowing disorders on the muscle strength and gait performance [6,[15][16][17].
However, to date, comprehensive gait characteristics in association with oral function have not been fully examined. Furthermore, only an individual aspect of oral health has been investigated. As the underlying mechanisms between oral health and gait performance involve multiple pathways [11][12][13][14]18,19], a global measure of oral function is necessary to capture the association between oral health and gait parameters. To date, no standardized global measure of oral health conditions is available.
Recently, the concept of oral frailty, defined as a series of phenomena characterized by ignored and neglected vulnerability in multiple oral health conditions such as reduced oral motor skills, chewing disability, and swallowing disorders, ultimately leading to physical and mental disorders [20], has been introduced in Japan. Importantly, this concept has been operationally defined as accumulated deficits in oral health [21], which enable researchers to test the validity of this concept. The operationalized definition of oral frailty was validated with regard to undernutrition, physical frailty, sarcopenia, disability, and mortality [21,22]. However, the association between oral frailty and poor gait performance, which is the established factor for disability [4], has not been addressed.
This cross-sectional study investigated various quantitative gait parameters in relation to comprehensive assessments of oral health to clarify whether oral frailty is associated with gait in community-dwelling older adults. We hypothesized that individuals with oral frailty would have poorer gait performance compared to those without oral deficits. Oral frailty has the potential to be a standardized global measure of oral health.
We believe that the results of the present study can contribute to further understanding of the concept of oral frailty as well as the association between oral health and physical functions.

Study population
The study population comprised participants of an ongoing cohort study, the Takashimadaira study [23]. The Takashimadaira study was initiated in 2016 and targeted adults aged ≥70 years living at Takashimadaira, Itabashi ward, Tokyo, Japan. The questionnaire on demographic and socioeconomic status, health behavior, and health status were mailed to these targeted adults. Thereafter, the investigators obtained completed questionnaires through home visits. During the home visits, the investigators asked the targeted adults about their willingness to participate in a comprehensive health assessment to be conducted at a healthcare facility. Those who agreed to undergo a comprehensive health assessment, including the oral health examination, gait parameter measurement, anthropometric measurement, and comorbidity assessment, through a medical interview were included in the study.
For the purpose of the current investigation, we used data from 2016. Individuals with missing data and those with medical histories of Parkinson's disease, dementia, or depression were excluded. The Takashimadaira study was conducted in full accordance with the ethical principles of the Declaration of Helsinki and was approved by the Ethics Committee of the Tokyo Metropolitan Institute of Gerontology (Approval number: 9 and 31 in 2016). All participants provided written informed consent prior to participation.
The Takashimadaira study aimed to develop a model of dementiafriendly communities in a metropolitan area. Hence, considering that our current investigation was based on the secondary use of data from this population-based study, no priori sample size calculation was carried out.

Questionnaire survey
Data on participants' age, sex, educational status (i.e., years of schooling), income, smoking status, alcohol consumption, physical activity level, vision status, and presence of chronic pain in the lower back or the knee were obtained through a self-administered questionnaire that was mailed before the health examination. The contents of the questionnaire survey are summarized in Table 1.

Assessment of oral health status
Oral health status, including the number of teeth and denture use, was assessed by a trained dentist. Masticatory performance was assessed using a fully automatic measuring device with β-carotene-containing gummy jelly (UHA Mikakuto Co., Ltd., Osaka, Japan) following Nokubi et al. (2013) [24]. The device was specially designed to measure the increase in the surface area of the ground gummy jelly in mm 2 , which was used as a parameter of masticatory performance. The repetitive articulatory rate (oral diadochokinesis) was measured to evaluate articulatory oral motor skill by using an oral function measuring device (KENKOU-KUN handy, Takei Scientific Instruments Co. Ltd., Niigata, Japan) [25]. The number of repetitions of the monosyllable "ta" per second was recorded. The force produced by the contact between the anterior part of the hard palate and tongue, called the tongue pressure (TP), was measured using the JMS Tongue Pressure Device (JMS Co., Ltd., Hiroshima, Japan) [26]. TP was recorded on three occasions, and the mean value of these measurements was used. In addition, the questions "Do you have any difficulties eating tough foods compared to six months ago?" and "Have you choked on your tea or soup recently?" on the Kihon Checklist (developed by the Japanese Ministry of Health, Labour and Welfare [27]) were asked to assess selfperceived oral function.
Following Tanaka et al. [21], oral frailty was defined as the presence of three or more of the following components: (i) fewer teeth, (ii) low masticatory performance, (iii) low articulatory oral motor skill, (iv) low TP, and subjective difficulties in (v) eating and (vi) swallowing. Detailed criteria for each component are presented in Table 2. This operationally defined oral frailty variable was specified as the main exposure variable.

Gait parameters measurement
Gait parameters were assessed using a modular platform system, P-WALK (BTS Bioengineering Corp., Milano, Italy), measuring gait variables in dynamic phases. A detailed description of the measurement gait parameters was provided by Kawai et al. [28].
Study participants were asked to walk at their normal pace in the straight 5-meter P-WALK sensor area. Measurements were performed twice to obtain the following gait parameters: gait speed, stride length, step length, step width, double support duration, single support duration, and stride time. The coefficient of variation (CV) for all parameters (except for gait speed) was also calculated. The reliability of the measured parameters has been indicated in the previous studies [29][30][31]. The aforementioned comprehensive gait parameters were specified as the outcomes.

Anthropometric measurement and comorbidity assessment
Anthropometric measures included weight (kg) and height (cm). The individual body mass index (BMI) was calculated by dividing the weight in kilograms by the squared height in meters. Comorbidity factors, such as diabetes, respiratory diseases, and stroke, were identified through medical interviews.

Statistical analyses
Analyses were performed using the statistical software package STATA version 16.1 (StataCorp, College Station, TX, USA). Descriptive statistics were performed to characterize the study population and compare the groups with and without oral frailty. T-test, the Mann-Whitney U-test, or the Chi square test were used when appropriate. The level of significance (two-tailed test) was set at 0.05.
A comparison of gait characteristics between groups with and without oral frailty was performed using simple and multiple linear regression models. The models included gait characteristics as outcome variables and oral frailty as an exposure variable. In line with previous studies [10,11,23,32], the following variables were included in the multivariable models as potential confounders: denture use, age, sex, educational status, income, smoking status, alcohol consumption, physical activity level, height, BMI, diabetes, respiratory diseases, stroke, vision impairment, chronic lower back pain, and chronic knee pain. The statistical threshold was adjusted using the Bonferroni method.
In addition, another set of multiple linear regression analyses was conducted to evaluate the association between the number of oral frailty components and gait parameters. Similar to the aforementioned primary analyses, the statistical threshold was adjusted using the Bonferroni method.

Results
In 2016, a survey questionnaire was sent to 7,614 adults aged ≥70 years, as listed in the basic resident register of Takashimadaira. Of these, 5,430 (71.3%) returned their completed questionnaires. Additionally, they were asked whether they were willing to participate in the comprehensive health assessment, wherein 1,248 individuals (23.0%) responded positively and underwent the assessment. Among them, 38 did not complete the oral health examinations, while 59 did not complete other examinations. Furthermore, 69 had medical histories of Parkinson's disease, dementia, or depression and were thus excluded. The remaining 1,082 individuals (439 men and 643 women; mean [standard deviation] age: 77.1 [4.7] years) were included in the present analysis.
Among the 1,082 individuals examined, oral frailty was observed in 227 (21.0%). The median (interquartile range) of the number of components of oral frailty was 1 (0-2). The prevalence of each component ranged from 20.1% (low masticatory performance) to 38.2% (few remaining teeth), as presented in Table 2. Table 3 presents characteristics according to presence of oral frailty. Participants with oral frailty had fewer teeth, had poorer oral function, and were more likely to use dentures. In addition, those with oral frailty were older and had fewer years of education, lower income, higher frequency of current smoking, were less physically active, and were likely to have vision impairment, and chronic pain in the lower back and knee. Table 4 shows gait characteristics in relation to the presence of oral frailty; descriptive statistics of gait characteristics and estimated differences (with 95% confidence intervals) in gait parameters between groups with and without oral frailty are presented. In comparison to participants without oral frailty, those with oral frailty had slower gait speed, shorter stride and step length, wider step width, and longer double support duration, as well as higher CVs of stride length and step length.
Essentially similar results were obtained from the multiple linear regression analysis, including the number of components of oral frailty as the exposure variable (Table 5). Negative linear relationships of the number of components of oral frailty were observed with gait speed, stride length, and step length. In contrast, the number of components of oral frailty had positive linear relationships with step width and double support duration. In addition, a positive linear relationship was observed between the number of components of oral frailty, and CVs of stride length and step length.

Discussion
This study demonstrated that community-dwelling older Japanese adults with oral frailty, defined as accumulated deficits in oral health, showed poor gait performance, mirrored by slow gait speed, accompanied by short stride and step lengths and long double support duration. In addition, poor gait stability among those with oral frailty was mirrored by the high variability of various gait parameters. Furthermore, the accumulation of oral health deficits had a dose-response relationship with a degree of decline in gait performance. For instance, one component of oral frailty increases as gait speed, step length, and stride length decrease; step width and double support duration as well as variability of gait parameters increase. To our knowledge, this is the first study to comprehensively demonstrate the gait characteristics of older adults with poor oral health and function. Our results highlight the importance of conducting a comprehensive assessment of oral health status among older adults. Although further studies are needed to verify these conclusions, oral assessments can be effective tools in detecting individuals who are at a higher risk of falling in the community setting, as gait impairment is a proven risk factor for falls in older adults [1][2][3].
A previous epidemiological study involving 749 older adults demonstrated a significant association between poor dentition status and low leg extensor power, low stepping rate, and short one-leg standing time [11]. Song-Yu et al. (2012) observed that low occlusal force was significantly associated with impairment of postural stability among 87 older adults [33]. Takata et al. (2004) studied the 80-year-old population and found that poor masticatory performance based on self-report was associated with low isokinetic leg extensor power and shorter standing time on one leg [9]. Edentulous individuals were observed to have slower gait speed and exhibited larger decline in gait velocity among participants in the English Longitudinal Study of Ageing [10]. Furthermore, Hatta et al. (2019) revealed that lack of posterior occlusal support predicted the incidence of gait speed reduction among participants in the Septuagenarians, Octogenarians, Nonagenarians Investigation with Centenarians study [6]. The current findings corroborate previous studies showing that poor oral health status is associated with gait performance, impaired physical function, and posture regulation.
According to previous studies [1][2][3], gait characteristics of older individuals who are at risk of falling include slow gait speed, shorter stride and step lengths, wide step width, long double support duration, and high variability of gait parameters. In the current study, these gait characteristics were observed in participants with oral frailty. Therefore, it may be inferred that gait deficits have a connection to oral health and falling and fall-related adverse health events, including disability and death. In fact, previous epidemiological studies reported that poor oral health was associated with an increased incidence of falls among community-based older adults [32] as well as among older adults with dementia [34]. Other studies indicate that poor oral health is associated with a risk of functional disability [21,35,36] and even mortality [21].
The observed associations between oral frailty and gait impairment appear to be biologically plausible. First, oral diseases cause occlusal interference and change of mandibular position, which lead to changes in the head and neck posture, ultimately leading to disturbances in body posture control and equilibrium [12]. Second, malocclusion owing to oral diseases may decrease the sensory afferent input of the masticatory   [45] reporting the average annual income of aged households was 3 million Japanese Yen.  muscular and dentoalveolar ligaments, hindering proprioception and posture stability [13]. A previous study involving 34 edentulous individuals (mean age=75.6 years) reported that the use of complete dentures was associated with the improvement of gait velocity and posture stability [19]. Third, dental occlusion can have negative impacts on the muscle tone, which is related to the knee muscular performance, eventually influencing gait [11,14]. Fourth, abnormal habituation, such as one-sided masticatory predominance, can be preceded by oral diseases and deterioration of dentition. Abnormal habituation ultimately leads to disequilibrium of the muscle balance, affecting systemic equilibrium [11]. Finally, a pathway involving nutrition has also been advocated. Poor oral health, including masticatory and swallowing disorders, leads to avoidance of foods that are difficult to chew, such as meat, vegetables, and fruits, which are generally rich in nutrients [15][16][17]. We have previously shown that older adults with oral health problems typically have less protein, fewer vitamins, and less dietary fiber compared to those without oral problems [37][38][39]. Thus, poor nutritional status caused by insufficient dietary intake may lead to decreased muscle strength and impaired gait performance [18]. Importantly, a large-scale epidemiological study of older adults showed a significant association between masticatory performance and gait speed and found that this association was mediated by protein intake [7]. Overall, although the evidence is growing, there are not enough studies to fully explain the association between oral health and gait performance, indicating the need for future research in this field.
The strength of the current study is its large sample size and the availability of data. This enabled us to define oral frailty based on the comprehensive measurement of oral health. As aforementioned, while several different pathways can be considered in the observed association between oral health and gait performance, it is plausible to assume that we found a dose-response relationship between components of oral frailty and degree of gait deficits. This finding distinguishes our study from previous studies that examined the associations of single oral health condition with gait performance [6,8,9,11]. Our findings, based on the comprehensive assessments of oral health, suggest that accumulated deficits in oral health have adverse effects on gait performance in a dose-response manner. Since oral health problems, such as the tooth loss, masticatory and swallowing function, and tongue pressure were interrelated [21], assessing a single oral health condition could not capture these phenomena. Oral frailty, which operationally defined accumulated deficits in oral health, has the potential to be a global measure of oral health.
Tanaka et al. exhibited that oral frailty was associated with increased risks of physical frailty, sarcopenia, disability, and mortality [21]. Their findings indicated that among the monosyllables "pa," "ta," and "ka," only "ta" was significantly associated with incidence of adverse events; thus, it was subsequently included as the oral frailty component. In their study, the prevalence rate of oral frailty was 15.9%, which was lower than that in the current study. This discrepancy can be partly explained by the fact that participants in our study were, relatively older. Furthermore, in a study conducted by Tanaka et al. [21], masticatory performance was objectively evaluated by color-changing chewing gum (Xylitol; LOTTE Co. Ltd., Tokyo, Japan). Results were obtained as the chewed gum's red photogenesis (a* values) was assessed using a colorimeter (Color Reader CR-13; Konica Minolta, Inc., Tokyo, Japan). A lower a* value indicated a lower performance, and a sex-specified lowest quintile of the a*value was used for categorizing low masticatory performance in both men and women. Since we did not use color-changing chewing gum, results of Table 3. Characteristics of the study population according to the presence of oral frailty.
The present study has several limitations. First, the cross-sectional design prevented us from assessing the temporal association of oral frailty with gait performance. However, we can partially transform the current data into a longitudinal analysis when the follow-up dataset is available. Evaluating longitudinal associations between oral frailty and adverse consequences of gait deficits such as falls and fractures could be ascertained. In this study, we expected that oral frailty would affect gait performance; however, the opposite was also a possibility. Older adults with functional disabilities might have limited ability to maintain their oral hygiene and restricted access to necessary dental care. In addition, systemic low muscle mass and function mirrored by impaired gait performance might have adverse effects on muscles involved in mastication and swallowing, which ultimately leads to a decline in multiple aspects of oral health. Second, the individuals were from a convenience cohort of those who voluntarily participated in the survey. Additionally, the present study population included only older Japanese adults from one specific region of the country; therefore the results may not be generalizable to other groups. Finally, although we controlled for various potential confounders, residual ones owing to unexpected confounding variables may exist, as with any multivariable analysis.
Prosthetic rehabilitation can improve masticatory performance [41]. Furthermore, prosthetic rehabilitation with nutritional counseling has beneficial effects on patients' dietary intake [42][43][44]. These findings suggest that oral frailty is potentially reversible after prosthetic rehabilitation. The current study demonstrating the association between oral frailty and gait performance can be used as a basis for future studies to assess whether prosthetic treatment could reverse oral frailty and have beneficial effects on gait performance in older adults.

Conclusion
To conclude, in this study, poorer gait performance was observed among adults with oral frailty compared to those without oral health deficits.