Community-dwelling Older Individuals using Day Services: Prevalence and Associations Between Walking, Toe-grip Strength, and Foot Problems -Cross-sectional Study-Kashiko

Background: Foot disorders in older individuals contribute to balance, posture, and gait instability, causing a decrease in activity of daily living and quality of life. However, the existing foot problems of older individuals in the community have not been studied well in Japan. Method: This cross-sectional study analyzed the foot-related data of 176 frail, old-aged participants who attended day service centers in Japan. Foot conditions were assessed and measured from various aspects, as was weight distribution to the oor. Multiple regression analysis was used to nd foot item variables that correlated with toe-grip strength and walking speed. Results: Prevalence of skin dryness and suspected and existing nail fungal infection was high in both sexes regardless of care level. Prevalence rate of toe deformity and arch deformity was signicantly higher in men and women requiring long-term care, respectively. The multiple regression analysis using stepwise method indicated the right toe-grip strength had a statistically signicant association with the arch deformity, suspected or existing nail fungal infection, and nail thickness of the right sided foot (p-value, 0·012, 0·034, 0·040 respectively). Coecients of all independent variables were negative. Any independent variables were not selected from the left foot. The corn, callus and toe deformity of right sided foot was signicantly associated with walking test (p-value, 0·026, 0·033 respectively). Conclusions: Our study supported that individuals requiring more long-term care tend to have more foot problems. Some specic conditions were associated with toe-grip force and walking speed. This nding contributes to future strategies to protect foot health for community-dwelling older individuals, particularly the frail population.

Our target participants were older people aged > 65 years who attended day service centers; day service centers or day care centers offering professional rehabilitation programs under long-term care insurance. We assumed that participants could walk or stand even using medical appliances; therefore, majority of the participants would be categorized into support leve1 1 to care level 3.
In Japan, there are two types of insurance for healthcare of older people: medical insurance and long-term care insurance. To use long-term health insurance system, older people in Japan have to apply and obtain certi cation by going through certain processes, including screening. Certi ed older people are categorized into seven levels (support level 1 or 2, or care level 1 to 5) [21]. The long-term insurance service is divided into home-based, community-based, or facility based. Under the home-based services, day service and day care centers provide daily cares, such as hygiene, meal, recreation, and physical rehabilitation, depending on the client's plan. In general, older people using one-day service centers are frailer than those who do not use such services. The author previously described this in another study [22].
According to an estimate based on the 2015 population census, there are 36,019,000 individuals > 65 years old in Japan as of March 2020 [23].
As of March 2019, the Ministry of Health, Labour and Welfare (MHLW) conducted the survey, which revealed that nearly 18·3% of the population > 65 years of age in Japan is eligible for using long-term care insurance [24].
For those who were categorized between care level 1 and 5 1,159,200 people used the day service, whereas 436,600 used the day care service offering rehabilitation [25].
According to the sample size calculator [26], 385 was the required number per population, with 95% con dence interval and 5% margin error. The sample size was also calculated using G*power. According to Cohen's parameter for the effect size for linear multiple regression [27] xed mode, and R2 deviation from zero, we input the effect size for f2 (medium = 0·15), a = 0·05, power = 0·08 and dependent variables 3 and nally obtained a total sample size of 119. Therefore, although we aimed to recruit 385 individuals, a total sample size of 119 satis es the requirement of the number of participants.
To achieve the initial target number of 385, we assumed that 500 would be the ideal invitation number of clients (400 facilities × 5 participants per facility × 0·25) by considering that nearly 23% would be rejected or drop out during the study.
The MHLW provided a list of in-home service providers in the target city of prefecture A, Japan for this study. We randomly selected 400 one-day service centers via the Excel random sheet. In addition, we selected a few facilities that were introduced through the author's connection as candidates for the study.
We sent an invitation mail and a postcard with a checkbox indicating the level of willingness to participate in the survey to the randomly selected facilities. Out of 400 facilities, 27 facilities responded. The author contacted the facilities with an active answer via telephone and visited majority of them to explain the study contents. Ultimately, 20 facilities participated in this study. Some of them requested to conduct the study a few times for different clients using separate dates due to a high demand from clients. Some facilities were initially eager to participate, but they declined because of staff shortage or unreadiness of preparing for the study.
The manager or staff members of the facility were willing to explain to clients and their family beforehand. Because the survey time was limited for each facility, selection of study participants was conducted by the facilities. The study included clients who understood the study contents and for those who did not fully understand, their family agreed their participation in the study. The mini-mental state exam (MMSE) score was set at ≥ 10. However, clients who obtained a score of < 10 were included if a family agreed and signed the agreement letter. Clients with foot infections, such as ulcer or wound, were excluded.
Although there were many obstacles because of limited time allocation for the eld survey, 6-9 clients from each facility were able to participate the study.
Each facility assisted to obtain written study permission from the family member instead of the author. Moreover, we explained and obtained written consents from study participants right before the survey. Unexpectedly, several clients could not write the letter using a pen because of hand tremor and low sight vision. In those cases, the examiner or staff signed his/her name instead of the clients themselves with the agreement. The majority of participants of both sexes were support level 1 to care level 3. There were only two participants for each sex with care level 4. There were no people with care level 5.

Study instruments
The main author created a two-page assessment sheet for data gathering for another study (submission for other paper) and this study. Appendix 1 presents the de nition, measurement unit and methods of foot-related items for assessment and measurement of the item selection of the assessment sheet. Previous studies used not only the foot assessment scores but also the cumulative methods of multiple foot problems or dichotomous scales, based on the presence or absence of foot disorders [4,28]. In this study, we employed the dichotomous method for some items. An assessment sheet was developed by the lead researcher and evaluated by a doctor with abundant foot knowledge and practical skills in this domain. With approval from a foreign researcher via mail, some of their assessment items were introduced into the assessment sheet used in this study.
We used two types of machines for measurements. One of the machines was Foot Look (Foot Look Inc.), which precisely, timely and visually measures various items, including foot length, foot width, and toe angle from the sole using a computer and scanner. It provides a visual image of soles including the hallux valgus and oating toe, and weight distribution of the body to the oor (Fig. 1). We measured toe-grip strength by using a toe-grip dynamometer (T.K.K.3364 Takei Scienti c Instrument) (Fig. 2). Participants were asked to sit upright, with both the hip and knee joints at 90 degree and the ankle joints in a neutral position. While xing the heel by a stopper, the grip bar was positioned using the rst proximal phalanx. Thereafter, the participants gripped the bar with the greatest possible force for approximately 3 seconds using their toes [29,30], and then the machine displayed the force meter. We also used a thermometer to check skin temperature of the lower extremity.
Foot-related items were classi ed as either items requiring clients to perform active action or do passive action at the time of foot measurement, depending on the level of participation of the clients during the examination. The items were considered passive when the participants' feet were assessed by the examiner while merely sitting down. When the participants needed to move their toes or feet or walk or step on the measurement board of the machine, the item was considered active.

Study preparation
The rst author established a research team from a blueprint. First, research assistants were recruited from the university students through application indicated in a poster. Nine students participated in this research during the summer. They were educated and trained to gain knowledge of foot and practice skills for measuring foot items.
The author arranged the visiting schedule to each facility. The research team, which consisted of the main author, 3-5 research assistants, and a co-author, visited each facility during the summer of 2019.

Procedure
Each facility provided a space where we could set up the machine and chairs. Research assistants mainly measured the walking speed, oating toe, hallux valgus angles (HVAs), and weight distribution using the machine and collected the data accordingly. With the use of the assessment sheet, foot assessment, which required extensive training, was conducted by the main author.
Through this survey, safety, such as fall and injury prevention, was the primary concern. In particular, when clients participated in the walking test and in stepping up to the table of the Foot Look machine, we carefully paid attention to their movement.
The rst author obtained consultation and audit from a foot care specialist with > 20 years of experience based on anonymized foot images .

Ethics
The research conformed to the Declaration of Helsinki 2000. We obtained informed consents from all participants, with the approval of facility providers and some of the families. We selected the STROBE checklist to describe the study. The study was further approved by the ethics committee of Nagoya University (approval number: 2019 − 0150).

Analysis
The demographic characteristics and foot-related assessment and measurement data were rst calculated using descriptive statistics. The items were categorized as continuous variables and categorical variables. Majority of the items were analyzed by categorizing into right and left foot. Sequentially, we simply observed the histogram, calculated the mean, SD, median, maximum, and minimum values, and employed the paired t-test, Fisher's exact test, and Mann-Whitney U test. We compared the data by four categories, namely, men, women, support level, and care level under the long-term insurance .
We initially counted the number of toes that had certain problems, such as thickened nail and ingrown nail. However, most foot-related items for assessment were coded as 1 (problem) or 0 (no problem), In this study, "problem" refers to the detected condition of the foot, deviating from the normal known condition.
For prevalence analysis, some items that were de ned in numbers were changed into dichotomous scales.
In stepwise multiple regression analysis, the dependent variables included the right and left toe-grip strength and walking speed, whereas the independent variables comprised the width of opened toes, oating toe, toe deformities, corn and callus, arch deformities, nail thickness, ingrown nail, suspected or existing nail fungal infection, HVA(HVA ≤ 15°, 15 < HVA ≤ 20°, 20° < HVA ≤ 40°, and HVA > 40°), and weight distribution. The independent variables were selected by predicting their association with toe-grip strength and walking.
For regression analysis, the continuous variable (number of toes, etc.) was used for the width of opened toes, whereas the dischotomoous variables (problem "1" or no problem "0" data) were used for the remaining items. Data were analyzed using the SPSS Statistics 26 (IBM Corp., USA)

Results
A ow diagram of participants is shown in Fig. 3. The total number of participants was 176. However, the nal number of analyzed clients on foot prevalence was 160 due to the exclusion of some missing data. The association of right and left toe-grip force with foot-related items was 136 and 135 respectively. The association between walking and foot-related items was 119. Some clients were not able to step up to the table of the Foot Look Machine because of physical limitation or were unable to grip the bar on the toe-grip dynamometer, or some other data were missing due to one-sided paralysis caused by cerebrovascular or hip problems or other issues.. Furthermore, some clients were not able to participate for measurements that required active movement, such as opening toes. Therefore, participant number for the analysis was varied as per items. The sample number was valid for multiple regression analysis, as previously mentioned.
Tables 1-3 enumerate the participants' demographic data (age, 4 m walking speed, skin temperature, degree of hallux valgus, toe-grip force, Barthel Index, widths of opened toes, and medical history). Paired t-test, *p < 0.05; **p < 0.01; ***p < 0.001 Those insured under the long-term insurance system include category 1 (people aged 65 years or older), and 2 (people aged 40-64 years who are covered by a health insurance program). Applicants for care services must be screened. Accepted applicants are categorized into support level 1 and 2 and care level 1-5.
Applicants in support level can perform most of the basic activities in daily life by themselves, but they need nursing care up to certain extent. Applicants in care level need some kind of assistance to do certain activities in daily life because they cannot perform them by themselves, although it is possible for them to complete most of the basic activities in daily life.  Mann-Whitney U Test, *p < 0.05; **p < 0.01; ***p < 0.001; Data are median, 25%, and 75% percentile The Barthel Index is a simple assessment and evaluation method to measure the 10 types of movement required for daily life. Widths of opening toes were measured with changes of the length of the midpoint between the 2nd toe and the 5th toe. The examiner asked participants to open the toes to the maximum Note: Some participants were not able to answer the questions of Barthel Index and test of widths of opening toes. Considering the time constraint, answering all the MMSE questions was a tremendous emotional burden for the clients and the facilities. Therefore, on reaching 10 points, we stopped them from answering the remaining questions.
Falling information was initially included but was eventually eliminated because the information provided by the clients and the staff was uncertain. Meanwhile, physical activity was measured by short physical-performance battery tests including balance, chair standing, and walking tests, which were also subsequently declined. At rst visit, performing balance and chair standing tests made the clients, as well as the staff, extremely anxious because of mechanical body changes, such as heart rate elevation and instability during standing. Therefore, we only selected the 4-meter walking test. The rst selected geriatric depression test was also declined because of complaints from several older clients.
The mean scores of walking speed were higher in care level for both men and women (6.0 ± 2.2 vs 8.6 ± 4·3 sec for men and 5.6 ± 1.9 vs 7.9 ± 3.8 sec for women with support level and care level, respectively), and right and left widths of opened toes were signi cantly narrow in the support level as compared to that in the care level for both men and women.
Both the right and left toe-grip strength was signi cantly higher for men with support level than care level. Skin temperature on lower extremity was 33·6 °C to 34·2 °C. Table 4 presents the foot assessment results. Signi cant differences were only observed in arch deformity between women with care level and those with support level (p-value 0·037). Skin dryness and suspected or existing nail fungal infection were highly prevalent in both men and women in general (skin dryness for men above 85·2% for right and left side, for women above 77·8% for right side and left side; suspected or existing nail fungal infection for men above 92·6% and 95·2% for women). The prevalence of nail color change (above 69·8%), arch deformity (63%) was relatively high in both sexes. However, the prevalence of nail color change was higher in men, whereas that of arch deformity was higher in women for both sexes. Meanwhile, corn and callus were relatively low for both sexes. The distribution rate of the right-side weight to the oor was higher than that of left-side weight. The HVA was categorized into HVA ≤ 15°, 15 < HVA ≤ 20°, 20° < HVA ≤ 40°, and HVA > 40°(see Appendix1). The prevalence of hallux valgus above 15° was higher in women than in men (8·3%-26·7% and 28·8%-43·1%, respectively). Fisher's exact test, *p < 0.05; **p < 0.01; ***p < 0.001 N (%) represents the number of participants whose foot conditions deviated from normal. Hallux valgus angle ≤ 15 was considered normal. Number of oating toe, hallux valgus angle (HVA), and weight distribution were measured using Foot Look (Foot Look Inc.) Table 5 shows the sensitivity and vascular condition of the feet. Over 92.6% of both men and women had touch sensation, although some of them had diabetes. The rate of the palpable posterior tibial arteries was 29·6%-58·8% for men and 47·2%-57·1% for women with support level or care level. N (%) represents the number of participants who answered "I feel it" when an examiner touched the their toe (sensitivity) N (%) represents the number of participants whose pulse could be palpated by the examiner Table 6 illustrates the multiple regression analysis results. Figure 4 illustrates the logical tree obtained from the regression analysis. Selected items which may associate with toe-grip strength and walking speed were input for analysis. The multiple regression analysis using stepwise method indicated the right toe-grip strength had a statistically signi cant association with the arch deformity, suspected or existing nail fungal infection, and nail thickness of the right sided foot (p-value, 0·012, 0·034, 0·040 respectively). Coe cients of all independent variables were negative. The highest standardized partial regression coe cient was right arch deformity. Adjusted R 2 was 0.099. Any independent variables were not selected from the left foot. The corn, callus and toe deformity of right sided foot was signi cantly associated with walking test (p-value, 0·026, 0·033 respectively). No signi cant association was found between left sided toe-grip strength and left sided foot-related items. Weight distribution Table 7 represents the result of association between right and left side of toe-grip strength and walking speed.

Discussion
To the best of our knowledge, this is the rst study to analyze the foot problem prevalence, the association between foot problems and walking, and toe-grip strength in older people attending day service and day care centers in Japan. Our target participants were frailer older people who use services under long-term care insurance but are not in a bedridden condition.
The strength of the study is that we obtained the data based on clinical assessment instead of self-reporting. In addition, the current foot problems were identi ed by data collection using multiple items so that the whole picture of the foot was captured. Three research questions were answered.
The prevalence of foot condition in older people Our study target was frail older people in the community who use long-term care insurance. As frailty advances, various factors of the body may negatively affect the foot health of older people. Our investigation on foot items indicated that the rate of incidence of oating toes, toe deformities, nail color change, arch deformities, nail length, thickened nail, skin dryness, edema, skin color change, suspected or existing nail fungal infection was relatively high, whereas the rate of corn and callus were pretty low. Various factors including aging may trigger these foot problems. Physical condition varies among community-dwelling older people; hence, foot conditions also differ among them. The high prevalence of hypertension, diabetes, and brain diseases may be associated with foot conditions, leading to lower immune system, circulation impediment, and increase of muscle generation with aging.
The reason of lower incidence of corn and callus may indicate that walking time is less for the frail older population; therefore, the repeated pressure on the sole is minimum. The rate of incidence of palpable tibial arteries may indicate the changes in the circulatory system based on aging. The change in the condition of peripheral blood vessels with aging is related to nail color and nail growth [31,32]. However, as we did not use Doppler, the pressure level of the examiner may be associated with the detection of palpating the pulse. The temperature of the lower extremities indicated the condition of foot circulation because of aging, sedentary behavior, and lack of foot movement. This indicates that temperature is associated with the coldness of the feet of many older people using the service, which is observed when the examiners touch the skin and palpate the pulse of the posterior tibial artery.
The other study indicated that awake temperature of 30·6 °C (SD 2·6) at awake time and 34·0 °C (SD1.8) at sleep time also varies from 15·9 °C to 37·5 °C depending on seasons [33]. For our future study, the consistency of the targeted site of temperature measurement and time with larger number of participants, a larger number of samples, and a comparison of the data with that from younger people would provide more accurate data.
The author's previous research demonstrated that foot skin dryness was found in 89·5-100% of frail older people who used homecare service [34]. A few studies reported that 44-45% of older people in-home care [10] and 52·6% in residents of nursing home had dry skin [35]. Both the caregiver and older people should understand the mechanism of skin barrier and apply the appropriate skin-protecting ointment.
This study assessed for possible nail fungal infection based on observation and the clinical hallmarks of onychomycosis were followed for some participants by a dermatologist. The nail with fungal infection becomes friable and appears yellowish [36]. The nail condition can be detected through mycological test, but the false-negative culture rate of this test is at least 30% [37]. Fungal infection of the foot is an indicator of lower limb cellulitis development [38].
Additionally, it may lead to psychological disorders due to its appearance [39].
Toe and arch deformity are interlinked with muscle degeneration, weakening of tendons and ligament with aging, lifestyle, walking, or the selection of shoes, or less walking. These deformities cause oating toes, leading to balance instability.
As we expected, walking speed was slower and width of opening toes was narrow and toe-grip strength was weak for people with more care. Previous studies indicated that toe-grip strength has declined with aging [40,41].
Uritani et al [41] classi ed the data of toe-grip strength into six age groups and reported that women in their 70 s (men: 10·4, women: 7·3) had a signi cantly weaker toe-grip strength than those in their 20s-60 s. The mean age of our study participants was 83 years, which was lower than their report (70 years, [mean: 4·026, median: 3·3]). Thus, building strategies to maintain or enhance toe-grip strength in this frail population is essential.
The foot item variables correlate with toe-grip strength and walking speed The right toe-grip strength had a statistically signi cant association with the arch deformity, suspected or existing nail fungal infection, and nail thickness of the right sided foot although no independent variables were selected on left foot. The corn, callus, and toe deformity of the right sided foot was signi cantly associated with walking test (p-value, 0·026, 0·033 respectively). The nding may be associated with a higher rate of participants represented by right distribution to the oor. However, more studies are needed to explore the nding. Although the adjusted R 2 / was 0.099 for right toe grip force and 0.067 for walking speed were week, signi cant P values indicated that relationship between predictors and response variables. The individual variability of foot may associate with the lower of R 2 scores. Further study will be needed to explore the generation of combination of higher adjusted R2 and lower P value.
Several reasons can be considered for the association between toe-grip strength and nail problems. The association between nail problems and toe force was previously reported [42][43][44]; however, the number of studies was limited in Japan and overseas. Nail abnormality including thickened, deformed or friable nails caused by fungal infection may trigger the decline of toe-grip strength because nail has the power to rmly step on the oor. Thus, nail care signi cantly affects the toe-grip strength. In the present study, abnormalities of the right foot arch led to a decrease in the right toe-grip strength. A few studies investigated the changes in foot shape with aging [45][46][47]. Deformity of arches may lead to the decline of range of motion, change in plantar loading pattern, and postural stability [48,49], causing instability of balance.
The foot arch is known to be associated with the plantar fascia, which is also known as plantar aponeurosis [50]. In one study, Erdemir et al described the role of plantar fascia during walking [51]. The study, which evaluated the association between plantar fascia and toe force, was conducted in 20 people and found that there was no relationship between the progression of the rst metatarsophalangeal (MTP) joint and the presence of plantar fascia. This study concluded that further investigation on this matter would be required [52].
Since the bar of the device measuring toe-grip strength is gripped at the MTP joint part of the foot, it is considered that an abnormality in the foot arch may lead to a decrease in the toe-grip strength. However, more data needs to be accumulated on the relationship between toe-grip strength and arch correction.
In a study, Digiovanni et al stated that adequate rehabilitation affects the stretching of the plantar fascia [53]. In another study, Mickel et al suggested "plantar fascia stretching, perception, and balance training, along with pain management and weight control" as components of rehabilitation program in order to prevent falling [54]. However, we suggest that the contents of an adequate rehabilitation program should further be assessed. For example, instead of just an exercise that moves the muscles, stretching of the foot before the exercise may prevent the generation of pain and could make the exercise more effective.
Additionally, there are various auxiliary tools and pads for arch correction, and thus, further data to verify the substantial effects on the strength of toe force should be gathered.
Interestingly, the walking speed was signi cantly associated with the existence of corn and callus as well as toe deformities. The formation of corn and callus itself and pain caused by them may be associated with plantar pressure, leading to decline of walking speed. The association between walking speed and plantar pressure has been widely studied [55][56][57][58].
Another study reported the effect of treatment to solve the pain due to the keratosis, such as corns and callus [59,60]. However, the association between corn and callus with plantar pressure remains unclear [61,62]. Another study revealed that plantar pressure is higher in older people with callused regions and an intervention study indicated the removal of callus reduces planter foot pressure [63]. The lower rate of having callus and corns indicated that the walking time of study participants was less due to frailty. Simply concluding the association between corn, callus, and toe-grip strength should be avoided because the generation of corn and callus are limited for this targeted population and rate of having corn and callus are low.
Toe deformities include the hallux valgus and lower-toe deformities. Interestingly, some patients have no toe deformity, whereas others have multiple toe deformities.
Mickle et al suggest that toe deformation of the lower toes changes the load distribution during walking because such a toe condition results in the toes being pulled back into extension, thus reducing the contact areas of the toes [55].
As the presence of corn and callus trigger pain, both toe deformities, including the hallux valgus and lower-toe deformities, may generate pain [59]. Therefore, pain may greatly affect the walking speed.
Hughes concluded that toes roles as the contact portion of three-quarter of the walking cycle and weight bearing function [64]. The balance transition from the medial to the lateral side of a patient with hallux valgus is signi cantly delayed [65]. The toe plays an important role in maintaining the body's balance and re ects the lower limb function. Therefore, appropriate efforts to strengthen the toe function, such as washing, nail care, stretching, and exercise are required.
In the present study, no signi cant relationship was observed between the toe-grip strength or walking speed and the hallux valgus and oating toes.
In the present study, no signi cant relationship was observed between toe-grip strength or walking speed and hallux valgus and oating toes.
When the foot arch deforms, occurrence of a bunion may be inevitable. When the foot spreads laterally due to muscle degeneration or other reasons such as arch deformation, it causes the adductor muscle of the toes to be pulled outward; thus leading to the formation of the hallux valgus. Many studies have majorly focused on hallux valgus among other toe deformities [66,67]. Further, a report also indicated an association between the degree of hallux valgus and toe-grip strength [68].
In this study, the analysis was conducted by dividing the degree of hallux valgus into three stages. Our study revealed that prevalence of hallux valgus (> 15°) in women was 29·9%-35·9%, which was higher than that in men; however, it did not reveal the angles of hallux valgus and walking speed. The small sample size may be associated with the results of undifferentiation of angles of hallux valgus.
Future studies with more samples may explore the association between the angles of hallux valgus and walking speeds. Arch deformity may prevent the toes from rmly attaching to the oor, resulting in oating toe complication, which has recently gained attention of the researchers [69,70]. However, the number of studies remain limited both in Japan as well as overseas. Uritani et al. concluded from their study that the oating toe is associated with walking [70]; however, further studies are required to con rm this.
The study showed that negative and signi cant correlation between walking speed and toe-grip strength. A previous study reported the association between the decreased strength of toe exor and slow walking [71]. However, correlation was not so strong. There is a great deal of individual variability, therefore, we were not able to conclude that individual with lower strength of toe exor have longer walking time. Further studies will explore this association in the strati ed aging group.
The foot data obtained by assessing and measuring the foot from various angles revealed the prevalence of foot disorder in frail old-aged people. Future strategies can be considered based on the foot deformities that have higher prevalence. For example, awareness regarding skin care should be spread among older people as well as care givers, which may lead to a higher usage of skin ointments and appropriate methods of skin care, leading to prevention of cracks in the skin.
The results of multiple regression analysis implicated the presence of an impaired area of foot condition, which may trigger lower walking speed and toe-grip force.
It provided insights into future strategies to prevent the foot condition from worsening as well as the possibility of effective nail care and prevention of skin lesion in maintaining the toe-grip strength or walking speed, which are key indicators of foot health. There are high chances of frail old-aged people falling and becoming bedridden in the future. According to the Ministry of Health, Labor and Welfare, fractures and falls are the third most common cause of need for long-term care for people who are certi ed as requiring care level 4 or 5 by long-term care insurance that requires severe long-term care [72]. Therefore, new approaches which incorporate appropriate foot care into daily practice among day service or day care would be desirable. Information on how to prevent the occurrence of nail thickness, nail fungal infection, toe and foot deformity, corn, and callus as well as information on how to appropriately provide foot care and promote the stretching of plantar fascia and exercise of foot would be key components for future strategies of preserving foot health for this age group as well as other age groups living in the community. In addition, the delivery of knowledge information and practice of foot care among the caregivers who provide care for people of this age group would be signi cantly helpful for preserving their foot health.
Our study has limitations with regard to the data collected. Because of the physical condition of older people, some of the participants were unable step on to the table of the Foot Look machine. Therefore, it was impossible to obtain data from all the participants. Although there are very few studies that describe the factors causing deformation of the arch and toes, there are various possible factors that need to be considered. This study, however, does not mention to them. In addition, the increased association between the decrease in skeletal muscle mass with arch and toe deformity because of aging was not evaluated in this study, although this is something that has gained much attention in the recent years.

Conclusions
In summary, the study exhibits that impaired foot area of older people by rate and key components may be associated with walking speed and toe-grip strength. This nding provides useful insight into the foot conditions of older people, in particular the frailer ones. Further studies should be conducted to target more people in this age group and to investigate strategies to prevent the development of falling and bedridden condition by incorporating foot care into routine practice at all facilities; however, we need to overcome many existing obstacles.

Declarations
Ethics approval and consent to participate: The research conformed to the Declaration of Helsinki 2000. We obtained informed consents from all participants, with the approval of facility providers and some of the families. The study was further approved by the ethics committee of Nagoya University (approval number: 2019-0150).

Consent for publication:
This paper does not reveal identity of any participant in any form. It is not applicable for this paper.
Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Figure 1
Foot look Using a computer and a special scanner, the foot is measured and the balance of the foot is visually analyzed to nd problems, such as loating toes and the degree of hallux valgus..

Figure 1
Foot look Using a computer and a special scanner, the foot is measured and the balance of the foot is visually analyzed to nd problems, such as loating toes and the degree of hallux valgus..

Figure 2
Toe-grip dyamometer measures toe force. Researchers developed the macine with assistantce from Takei Scienti c Insturment Inc. Participants gripped the bar for approximatly 3 seconds using their toes with the greatemet force as much as they can.

Figure 2
Toe-grip dyamometer measures toe force. Researchers developed the macine with assistantce from Takei Scienti c Insturment Inc. Participants gripped the bar for approximatly 3 seconds using their toes with the greatemet force as much as they can.

Figure 3
A ow of diagram of participants Initial study participants were 176. Due to the missing data, 160 were on analysis for assessing foot items. For the association between right or left grip force and selected foot items which may associate with the strenght, 136 and 135 parcitipants were on analysis. For the association between walking speed and seletected items were 119.

Figure 3
A ow of diagram of participants Initial study participants were 176. Due to the missing data, 160 were on analysis for assessing foot items. For the association between right or left grip force and selected foot items which may associate with the strenght, 136 and 135 parcitipants were on analysis. For the association between walking speed and seletected items were 119.

Figure 4
Logic tree obtained from the regression analysis Logic tree indicated foot items which is associated with dependent variables as right or left toe grip force, walking speed by breaking down the items. Logic tree obtained from the regression analysis Logic tree indicated foot items which is associated with dependent variables as right or left toe grip force, walking speed by breaking down the items.