*2.3. Questionnaire*

#### 2.3.1. Quality of Life (QOL)

Quality of life was evaluated by the short form 36(SF-36). The Sf-36 consists of 36 items. These items are classified into 8 subscales: physical functioning (PF), role physical (RP), bodily pain (BP), general health (GH), vitality (VT), social functioning (SF), role emotional (RE), and mental health (MH).

The values of these subscales were standardized and calculated by a program provided by iHope International (Kyoto, Japan) [33–35].

#### 2.3.2. Activities of Daily Living (ADLs)

Instrumental activity of daily living was assessed by The Tokyo Metropolitan Institute of Gerontology index of competence (TMIG index) [36]. TIMG index consists of three subscales/dimensions: self-maintenance (S.M), intellectual activity (I.A.), and social role (S.R.). These subscales consist of 5, 4, and 4 items, respectively. If subjects answered yes or able, one point was given for each item. A low IADL (≤4 points), IA (≤2 points), or SR (≤2 points) score is regarded as declining function [39,40]. The TMIG index has been widely used in epidemiological surveys [41–45].

The items included in these subscales are

S.M.: Using public transportation, shopping, preparing, meals, paying bills, managing deposits I.A.: Filling out pension forms, reading the newspaper, reading books, becoming interested in a new story or program about health.

S.R.: Visiting friends, being called on for advice, visiting sick friends, talking to young people.

#### 2.3.3. Self-Assessed Chewing Ability

Self-assessed chewing ability was investigated using the following question about 15 different foods: Can you chew the following 15 foods? The response was a simple dichotomous choice (yes/no). Several epidemiological studies have applied this questionnaire for the evaluation of chewing ability [37,46].

#### *2.4. Statistical Analysis*

#### 2.4.1. Item Response Theory (IRT)

To calculate the summary score for chewing ability, a three-parameter logistic model of the item response theory (IRT) was applied. In addition, factor analysis by the major factor method with varimax rotation was carried out. Summary scores were calculated within each factor [42–49]. IRT analysis was performed using R ver3.50 with the LTR and irtoys packages.

#### 2.4.2. Structural Equation Modeling (SEM)

Before performing structural equation modeling (SEM), factor analysis by the major factor method with varimax rotation was carried out. Based on the results of the factor analysis, latent variables were constructed. The models were modified through a comparison with the correction index to improve the fitness of the data. For the evaluation of the fitness, the root-mean-square error of approximation (RMSEA) was used for the goodness of fit index [50]. Factor analysis was carried out using SPSS Statistics ver24.0 (IBM, Tokyo, Japan) and SEM was carried out busing AMOS ver24.0 (IBM, Tokyo, Japan).

#### 2.4.3. Generalized Linear Model

To assess the subscales and items of QOL and IADL, the generalized linear model was applied. The distribution of response and link functions was selected using Akaike's Information Criterion (AIC). The generalized linear model analysis was carried out using SPSS Statistics ver24.0 (IBM, Tokyo, Japan)

#### 2.4.4. Survival Analysis

Survival rates were calculated using the Kaplan–Meier analysis. A log rank test was used to compare significant di fferences in survival curves. A Cox proportional hazards model was applied to calculate the hazard ratios. Survival analysis was carried out using SPSS Statistics ver24.0 (IBM, Tokyo, Japan)

#### *2.5. Ethics Approval and CONSENT to Participate*

Informed written consent was obtained from all of the participants at the baseline survey visit. This study was approved by the Ethics Committee of Tsurumi University School of Dental Medicine (Approval Number: 1515).
