Cultural Adaptation and Validation of the Oral Frailty Index‐8 for Turkish Older Adults

ABSTRACT

Introduction

Oral frailty, a multidimensional decline in oral function involving chewing, swallowing, and oral behaviours, is a recognised precursor of dysphagia, malnutrition, and physical frailty. The Oral Frailty Index‐8 (OFI‐8) is a brief self‐report screening tool for assessing oral frailty. This study's aim is to adapt the OFI‐8 into Turkish (OFI‐8‐TR) and evaluate its psychometric properties among older adults.

Methods

A cross‐sectional study was conducted in 341 adults aged ≥ 65 years attending a geriatrics outpatient clinic. Internal consistency (KR‐20), test–retest reliability (ICC), and construct validity (convergent, known‐groups, and confirmatory factor analyses) were evaluated. Comparator measures included the FRAIL (Fatigue, Resistance, Ambulation, Illnesses, and Loss of weight) scale, Clinical Frailty Scale (CFS), Eating Assessment Tool‐10 (EAT‐10), Functional Oral Intake Scale (FOIS), and Mini Nutritional Assessment–Short Form (MNA‐SF).

Results

The OFI‐8‐TR showed acceptable internal consistency (KR‐20 = 0.62) and excellent test–retest reliability (ICC = 0.898). Confirmatory factor analysis supported a one‐factor model (Comparative Fit Index (CFI) = 0.93, Goodness‐of‐Fit Index (GFI) = 0.95, adjusted Goodness‐of‐Fit Index (AGFI) = 0.91). Convergent validity was confirmed through strong correlations with the EAT‐10 (r = 0.63) and FOIS (r = −0.62), and moderate correlations with FRAIL, CFS, and MNA‐SF in the expected directions. Known‐groups validity showed significantly higher scores among participants with malnutrition or risk of malnutrition.

Conclusions

The OFI‐8‐TR is a valid and reliable instrument for assessing oral frailty and identifying swallowing‐related and nutritional vulnerability in older adults. Its brevity and multidimensional structure make it a practical screening tool for routine geriatric assessments to support early identification of swallowing‐ and nutrition‐related vulnerability.

1. Introduction

Oral frailty, defined as a decline in oral health and functional capacity, has emerged as an important geriatric health concept and a growing public health priority [1]. Early conceptual work emphasized its multidimensional, age‐related nature, linking impairments in oral structure and function with declines in physical, cognitive and social domains [2]. The 2024 Japanese national consensus statement further formalized oral frailty as a progressive weakening of oral functions, including chewing, swallowing, and articulation, which has been associated with increased frailty, disability and mortality in older adults [3]. Later theoretical refinements emphasized oral frailty as a dynamic biopsychosocial process linking oral health to overall resilience [4, 5]. Global estimates suggest that approximately one‐quarter to one‐third of older adults experience oral frailty, and nearly half are at risk [6, 7, 8]. Strong associations have been reported with advancing age, physical frailty, malnutrition, dysphagia, social isolation and chronic conditions such as polypharmacy or depression, highlighting oral frailty as a multifactorial syndrome shaped by biological, functional and behavioral factors [6, 9, 10].

Evidence increasingly supports oral frailty as both a marker and driver of systemic vulnerability. Recent meta‐analyses have shown that oral frailty is associated with an increased likelihood of sarcopenia, malnutrition, falls, and functional decline, underscoring its value as a modifiable determinant of late‐life health trajectories [11, 12, 13]. However, despite rising global interest, validated tools for assessing oral frailty remain limited outside East Asia, restricting international comparability and early risk identification.

To address this gap, Tanaka et al. developed the Oral Frailty Index‐8 (OFI‐8), an eight‐item self‐administered screening instrument designed to identify early oral frailty among community‐dwelling older adults [14]. Among available oral frailty screening approaches, the OFI‐8 was specifically designed as a brief, self‐reported instrument capturing not only chewing and swallowing function but also oral health–related behaviours relevant to ageing. Its multidimensional structure and ease of administration make it particularly suitable for use in non‐dental, geriatric outpatient and community‐based settings, where rapid screening without clinical oral examination is required.

Better oral function is a key component of healthy ageing, as the primary goal in later life is not merely to prolong survival but to enhance quality of life and maintain functional independence. Therefore, validating the OFI‐8 for the Turkish population is an important step toward increasing oral frailty awareness and providing a culturally relevant instrument for both research and clinical practice. The aim of this study was to evaluate the validity and reliability of the Oral Frailty Index‐8 in older adults.

2. Methods

2.1. Study Design and Participants

The source population consisted of community‐dwelling older adults attending a geriatrics outpatient clinic, from whom participants were consecutively recruited during routine outpatient visits. The Geriatrics Outpatient Clinic is a university‐affiliated tertiary care clinic providing comprehensive geriatric assessment for community‐dwelling older adults referred for multimorbidity, frailty, nutritional problems, cognitive complaints, and functional decline. Patients are referred from primary care or other hospital departments and are routinely evaluated using a multidisciplinary geriatric assessment approach. This cross‐sectional study was conducted at the Geriatrics Outpatient Clinic between April and August 2025. Inclusion criteria were being aged 65 years or older, able to provide reliable information, and having given informed consent. Individuals residing in long‐term care facilities were not included.

A total of 400 older adults were evaluated, and 59 (15%) were excluded due to incomplete data, acute medical conditions affecting nutritional or oral status, severe sensory impairments, or cognitive impairment that precluded meaningful communication. When assistance from caregivers was required due to visual or literacy limitations, caregivers were limited to reading or clarifying questionnaire items only. All responses reflected the participant's own answers, and no proxy responses were accepted. The attending geriatrician supervised the interview process to ensure comprehension and confirm that responses were provided by the participant themselves. Severe sensory impairments were defined as uncorrected hearing or visual deficits that prevented understanding the study procedures or questionnaire items, as judged by the attending geriatrician during the clinical assessment. Cognitive impairment precluding meaningful communication was determined based on clinical judgement supported by Mini‐Mental State Examination (MMSE) scores and the participant's inability to reliably understand or respond to study questions during the interview. The OFI‐8‐TR was administered as a structured face‐to‐face interview conducted by geriatricians, rather than as a self‐completed questionnaire, to minimise missing data and ensure comprehension among older participants. Administration of the OFI‐8‐TR required approximately 2–3 min per participant. Sociodemographic variables (age, sex, living status, marital status, education) were recorded.

2.2. Sample Size Calculation

Sample size estimation for factor analysis was based on the recommendation by Hogarty et al. of a 20:1 participant‐to‐item ratio, requiring a minimum of 170 participants for the 8‐item scale [15]. Based on expected exclusion rates of approximately 10%–20% in geriatric outpatient studies, over recruitment was planned to ensure an adequate final sample size for psychometric analyses. A larger sample of 341 participants was recruited to strengthen factor‐analytic and reliability estimates.

Formal a priori sample size calculations for reliability analyses were not performed. However, for internal consistency assessment, sample sizes above 100 are generally considered adequate for short scales, and the full study sample was used. For test–retest reliability, a subsample of 25 participants was included, which is consistent with methodological recommendations indicating that 20–30 participants are sufficient to estimate intraclass correlation coefficients in reliability studies.

2.3. Measurement Tools

The OFI‐8 is an eight‐item self‐reported screening questionnaire developed to identify oral frailty. It assesses oral frailty–related symptoms and behaviours, including perceived difficulties eating tough foods, choking on liquids or soup, denture use, dry mouth, reduced frequency of going out, ability to eat hard foods, tooth brushing frequency, and regular dental visits. The total score ranges from 0 to 11, with higher scores indicating greater oral frailty. Questions 1–3 are scored as 2 points for “Yes,” questions 4–5 as 1 point for “Yes,” and questions 6–8 (reverse coded) as 1 point for “No” [14, 16]. In this study, a score of ≥ 4 was used to indicate a high oral frailty category, in line with the original instrument.

The OFI‐8‐TR was compared with medical and geriatric clinical measures rather than objective oral clinical variables because it was originally developed as a self‐reported screening instrument intended for use in non‐dental settings. Its primary purpose is to capture perceived oral functional decline and related behaviors within the broader context of functional vulnerability and aging. Therefore, convergent and clinical validity were examined using established geriatric assessment tools that reflect functional status, frailty, nutrition, swallowing, and oral intake, aligning with the conceptual framework of oral frailty as an integrative geriatric construct rather than a purely dental diagnosis.

Comparator instruments were selected based on their conceptual relevance to oral frailty and their routine use in comprehensive geriatric assessment. These included the Clinical Frailty Scale (CFS) and the FRAIL Scale (Fatigue, Resistance, Ambulation, Illnesses, and Loss of weight) to assess general frailty; the MMSE to assess cognitive status; the Mini Nutritional Assessment–Short Form (MNA‐SF) to evaluate nutritional status; the Eating Assessment Tool (EAT‐10) to capture swallowing‐related symptoms; and the Functional Oral Intake Scale (FOIS) to assess oral intake level. All comparator instruments were used in their validated Turkish versions [17, 18, 19, 20, 21, 22, 23, 24, 25] and were applied to examine convergent, known‐groups, and clinical validity of the OFI‐8‐TR.

2.4. Translation and Cross‐Cultural Adaptation Process

Because the original developers could not be reached, the English version of the OFI‐8 was translated using standard forward–backward translation procedures following Beaton et al.'s guideline [26]. Two bilingual geriatricians produced independent translations, which were synthesised and back‐translated by a bilingual translator. An expert committee reviewed all versions for semantic and conceptual equivalence. Pretesting was conducted in 15 older adults, leading to minor wording adjustments (e.g., replacing culturally unfamiliar food examples with “nuts” and “raw radish”). The final OFI‐8‐TR was approved by the expert panel.

2.5. Reliability

Internal consistency and test–retest reliability of the OFI‐8‐TR were evaluated to assess the stability and coherence of the scale. Measurement error indices and item‐level agreement were also examined.

Test–retest reliability was assessed in a subsample of 25 participants who were consecutively selected from the overall study population. These participants were community‐dwelling older adults attending the geriatrics outpatient clinic who were clinically stable, had no acute medical conditions, and were able to return for reassessment within a 2‐week interval. The subsample was comparable to the full sample in terms of age, sex, and functional status.

2.6. Validity

Content validity was ensured during expert review and cognitive debriefing. Convergent, known‐groups, clinical, and construct validity were examined to evaluate the performance of the OFI‐8‐TR. Criterion validity was not assessed, as no gold‐standard oral frailty measure exists.

2.7. Statistical Analysis

Data analysis was performed using SPSS 25.0 and Jamovi. Continuous variables included age, OFI‐8‐TR total score, FRAIL score, CFS score, EAT‐10 score, FOIS score, MNA‐SF score, ADL, IADL, and MMSE scores. Categorical variables included sex (female/male), living status (alone/not alone), marital status (married/widowed/divorced/single), education level (illiterate/literate/no formal education/primary school or higher), frailty categories (robust/prefrail/frail), nutritional status categories defined by the MNA‐SF (normal nutrition, at risk of malnutrition, malnutrition), and oral frailty category (OFI‐8‐TR < 4 vs. ≥ 4). Normality was tested by the Shapiro–Wilk test. Continuous variables were expressed as mean ± SD or median (min–max), categorical variables as frequencies and percentages. Chi‐square or Fisher's exact tests were used for categorical comparisons, and t‐test or Mann–Whitney U test for continuous variables. A p‐value < 0.05 was considered statistically significant. Permissions for all comparator scales were obtained.

Internal consistency was evaluated using the Kuder–Richardson 20 (KR‐20) coefficient, appropriate for dichotomous items. Values > 0.50 were considered acceptable given the short scale length. Test–retest reliability was evaluated in 25 participants over a 2‐week interval using the intraclass correlation coefficient (ICC, two‐way random‐effects, absolute agreement). ICC values between 0.75 and 0.90 were considered good, and > 0.90 excellent [27]. The standard error of measurement (SEM) and minimal detectable change at the 95% confidence level (MDC₉₅) were calculated. Item‐level agreement was assessed with Cohen's κ coefficients and the McNemar test [28].

The OFI‐8‐TR total score was treated as a continuous variable for convergent, known‐groups, and clinical validity analyses. Categorization of the OFI‐8‐TR score using the established cut‐off value (≥ 4 points) was applied only for descriptive purposes and for test–retest reliability analyses.

Convergent validity was assessed using Spearman's correlation coefficients between OFI‐8‐TR and FRAIL, CFS, EAT‐10, FOIS, and MNA‐SF scores. Correlations > 0.5 were considered strong, 0.35–0.5 moderate, and < 0.35 weak.

Known‐groups validity refers to the ability of an instrument to discriminate between groups that are known, a priori, to differ in the construct being measured. In this study, known‐groups validity of the OFI‐8‐TR was examined by comparing scores across nutritional status categories defined by the Mini Nutritional Assessment–Short Form (MNA‐SF) (normal nutrition (scores = 12–14), at risk of malnutrition [8, 9, 10, 11], and malnutrition (≤ 7)), under the hypothesis that individuals with poorer nutritional status would demonstrate higher oral frailty scores using the Kruskal–Wallis test with Dunn–Bonferroni post hoc comparisons. In this study, the term nutritional risk was used to refer to participants classified as either malnourished or at risk of malnutrition according to the MNA‐SF.

Clinical validity was examined within the framework of the Comprehensive Geriatric Assessment (CGA), a multidimensional and interdisciplinary diagnostic process used in geriatric practice to evaluate medical, functional, cognitive, and psychosocial domains in older adults. For clinical validity analyses, participants were grouped according to components of the Comprehensive Geriatric Assessment. Independence in Activities of Daily Living (ADL; scores 5–6 vs. ≤ 4) and Instrumental Activities of Daily Living (IADL; 8 vs. < 8), frailty status assessed by the Clinical Frailty Scale (CFS; ≤ 4 vs. > 4), and cognitive status assessed by the MMSE (MMSE; normal cognition vs. ≤ 24) were used for group comparisons. Lower ADL/IADL scores, higher CFS scores, and lower MMSE scores indicated greater dependency, frailty, and cognitive impairment, respectively. OFI‐8‐TR scores were compared between groups using Mann–Whitney U tests.

Construct validity was examined using confirmatory factor analysis (CFA) with a diagonally weighted least squares (DWLS) estimator, appropriate for ordinal/dichotomous data. Model fit was considered acceptable when χ ²/df < 3, Root Mean Square Error of Approximation (RMSEA) < 0.08, and Comparative Fit Index (CFI), Goodness‐of‐Fit Index (GFI), Adjusted Goodness‐of‐Fit Index (AGFI), Incremental Fit Index (IFI), and Normed Fit Index (NFI) ≥ 0.90.

3. Results

3.1. Participant Characteristics

After exclusions, 341 participants were included in the final analysis. The study sample had a mean age of 75.9 years; 65.7% (n = 224) were women and 17.6% (n = 60) lived alone. Based on the FRAIL scale, 27.9% (n = 95) of participants were classified as robust, 41.6% (n = 142) as prefrail, and 30.5% (n = 104) as frail. The median OFI‐8‐TR total score was 4.0 (interquartile range: 3.0; range: 0–11). Normality testing indicated a non‐normal distribution of OFI‐8‐TR scores (Shapiro–Wilk test, p < 0.001). Therefore, non‐parametric statistical methods were applied.

Table 1 summarises the clinical and functional characteristics of the study population. Participants demonstrated predominantly mild‐to‐moderate levels of frailty, with preserved basic and instrumental activities of daily living and cognitive performance. Nutritional status and swallowing‐related measures covered a wide range of values, reflecting heterogeneity in oral and nutrition‐related functional profiles.

TABLE 1.

Characteristics of the participants.

Variables	All (n = 341)
	Mean	SD	Median	IQR	Min–Max
Age, years	76.0	7.0	75	70–81	65–94
Clinical Frailty Scale	3.7	1.1	4	3–4	2–7
Mini Nutritional Assessment–Short Form	12.4	2.0	13	12–14	3–14
Activities of Daily Living	5.5	0.9	6	5–6	0–6
Instrumental Activities of Daily Living	7.3	1.5	8	7–8	0–8
Mini‐Mental State Examination	25.2	4.0	26	23–28	10–30
Oral Frailty Index–8 Turkish version	4.7	2.4	4	3–6	0–11
Eating Assessment Tool–10	1.8	4.0	0	0–2	0–33
FRAIL	1.6	1.3	1	0–3	0–5
Functional Oral Intake Scale	6.7	0.6	7	6–7	4–7

Note: Data are presented as mean ± standard deviation (SD), median (interquartile range [IQR]), or minimum–maximum, as appropriate.

Abbreviations: FRAIL, fatigue, resistance, ambulation, illnesses, and loss of weight scale; IQR, interquartile range; SD, standard deviation.

3.2. Translation and Cross‐Cultural Adaptation

The translation and adaptation process was completed without major linguistic or conceptual difficulties. During cognitive debriefing with 15 older adults of varying educational and socioeconomic backgrounds, all items were reported to be clear, comprehensible, and culturally appropriate. Minor wording adjustments were made to improve readability and local relevance. Specifically, in the item referring to chewing hard foods, the original examples (“dried squid” and “pickled radish”) were replaced with culturally familiar foods (“nuts” and “raw radish”) while maintaining the original conceptual meaning. No other modifications were required, and the expert committee approved the final Turkish version of the OFI‐8‐TR (see Supporting Information S1).

3.3. Reliability Analysis

3.3.1. Internal Consistency

The internal consistency of the OFI‐8‐TR, calculated in the full sample (n = 341), was acceptable, with a Kuder–Richardson 20 (KR‐20) coefficient of 0.621 (0.648 when based on standardised items). Item‐level analyses showed that all items contributed to the overall scale reliability, and deletion of any single item did not result in an improvement of the reliability coefficient, supporting the internal coherence of the instrument (Table 2).

TABLE 2.

Internal consistency of the Oral Frailty Index‐8‐TR items.

	Internal consistency (n = 341)
	Corrected item‐ total correlation	Cronbach's alpha if item deleted
Q1. Having any difficulties eating tough foods compared 6 months ago	0.503	0.525
Q2. Choking on your tea or soup recently	0.472	0.537
Q3. Using denture	0.147	0.681
Q4. Often experience having a dry mouth	0.362	0.579
Q5. Going out less frequently than you did last year	0.410	0.569
Q6. Eating hard foods like squid jerky or pickled radish	0.506	0.545
Q7. Brush teeth per day	0.115	0.633
Q8. Visiting dental clinic at least annually	0.188	0.619

3.3.2. Test–Retest Reliability

Among 25 participants who completed the scale twice within a 2‐week interval, the intraclass correlation coefficient (ICC) for total OFI‐8‐TR score was 0.898 (95% CI: 0.756–0.960, p < 0.001), demonstrating good to excellent reliability. The average‐measures ICC was 0.946 (95% CI = 0.861–0.979). The SEM was 0.64 points, corresponding to a MDC₉₅ of 1.77 points. Thus, a change of approximately 2 points in the total score can be considered a true clinical change beyond measurement error. At the categorical level, agreement for high‐risk oral frailty classification between test and retest assessments was perfect. Of 25 participants, 12 remained in the low‐risk group and 13 remained in the high‐risk group, with no discordant classifications. The McNemar test confirmed no significant change between assessments (exact p = 1.000), indicating excellent stability of categorical risk classification.

3.4. Validity Analysis

3.4.1. Convergent Validity

The OFI‐8‐TR total score showed significant correlations with conceptually related measures, supporting its convergent validity. Strong positive correlations were observed with the FRAIL scale (r _s = 0.578) and the EAT‐10 (r _s = 0.627). Moderate positive correlations were found with the Clinical Frailty Scale (r _s = 0.430). Significant negative correlations were identified with the FOIS (r _s = −0.622) and the MNA‐SF (r _s = −0.408).

3.4.2. Known‐Groups Validity

The OFI‐8‐TR scores differed significantly across nutritional status groups defined by the MNA‐SF (Kruskal–Wallis H = 57.809, df = 2, p < 0.001). Median OFI‐8‐TR scores were highest among those with malnutrition (median = 6, IQR = 4–10) and those at risk (median = 6, IQR = 5–9), and lowest among those with normal nutrition (median = 4, IQR = 3–5). Corresponding mean ranks were 149.5, 247.4, and 221.3, respectively.

3.4.3. Clinical Validity

The OFI‐8‐TR scores significantly differed according to functional, frailty, and cognitive status.

Participants classified as frail had higher scores than robust participants (median rank = 227.3 vs. 154.9 in robust; Z = −5.74, p < 0.001). Those dependent in ADL (262.9 vs. 165.0; Z = −4.48, p < 0.001), or IADL (220.2 vs. 152.0; Z = −5.82, p < 0.001) also had significantly higher scores. Similarly, participants with cognitive impairment scored higher than those with normal cognition (222.0 vs. 152.7; Z = −5.81, p < 0.001).

3.4.4. Construct Validity

CFA using the DWLS estimator supported a one‐factor structure for the OFI‐8‐TR. Overall model fit indices indicated an acceptable fit, with most indices meeting conventional criteria, as summarised in Table 3. CFA using the DWLS estimator (N = 341) supported a one‐factor structure. Model fit indices were: χ ²/df = 4.53, RMSEA = 0.10 (90% CI [0.08–0.12]), CFI = 0.93, GFI = 0.95, and AGFI = 0.91, as presented in Table 3.

TABLE 3.

Model fit indices of confirmatory factor analysis for the Oral Frailty Index‐8‐TR.

Fit index	Model results	Acceptable criteria	Fit assessment
χ 2/df	4.53	< 3.0	Moderate fit
RMSEA (90% CI)	0.10 (0.008–0.12)	≤ 0.08	Marginal fit
CFI	0.93	≥ 0.90	Acceptable fit
GFI	0.95	≥ 0.90	Good fit
AGFI	0.91	≥ 0.90	Good fit

Note: Model fit was evaluated using the diagonally weighted least squares (DWLS) estimator in Jamovi due to the ordinal nature of the items. Fit indices were interpreted according to conventional cutoff values: χ ²/df < 3 indicates good fit; RMSEA ≤ 0.08 acceptable fit; and CFI, GFI, and AGFI ≥ 0.90 acceptable or good fit. Although RMSEA was slightly above the optimal threshold, the remaining indices supported an acceptable one‐factor model consistent with the original OFI‐8‐TR structure.

Abbreviations: χ ²/df, Chi‐square divided by degrees of freedom; AGFI, Adjusted Goodness‐of‐Fit Index; CFA, Confirmatory Factor Analysis; CFI, Comparative Fit Index; GFI, Goodness‐of‐Fit Index; RMSEA, Root Mean Square Error of Approximation.

Standardised factor loadings ranged from 0.26 to 0.88, with several items demonstrating higher loadings and others showing lower values. The standardised loadings and the single‐factor structure of the OFI‐8‐TR are illustrated in Figure 1.

FIGURE 1.

Path diagram of the OFI‐8‐TR confirmatory factor analysis model. Standardised factor loadings (β) are shown on the paths. The full wording of each OFI‐8 item is provided in Table 2. The one‐factor model of the Oral Frailty Index–8 Turkish version (OFI‐8‐TR) demonstrated acceptable fit (χ ²/df = 4.53, RMSEA = 0.10, CFI = 0.93, GFI = 0.95, AGFI = 0.91). Standardised factor loadings (β) are displayed on the paths from the latent variable (OFI) to the eight observed indicators (OFI1–OFI8). Solid lines indicate statistically significant loadings (p < 0.001).

4. Discussion

This study provides the first psychometric validation of the OFI‐8‐TR in Turkish older adults. The findings demonstrated acceptable internal consistency, excellent test–retest reliability, and a unidimensional factor structure, supporting the reliability and construct validity of the OFI‐8‐TR as a culturally appropriate tool for assessing oral frailty.

Several limitations should be considered when interpreting these findings. Although we attempted to contact the original developers of the OFI‐8 prior to initiating the translation process, no response was received. Contact was attempted via the corresponding author's institutional email address listed in the original publication, with two follow‐up emails sent over a 3‐month period. In the absence of a response, the adaptation proceeded in accordance with established cross‐cultural translation and validation guidelines, without modification of the original scale structure or scoring. Test–retest reliability was assessed in a relatively small subsample, which may have widened confidence intervals, although temporal stability remained good to excellent. In addition, the moderate internal consistency observed is typical for short, multidimensional dichotomous instruments and has been reported in other OFI‐8 validation studies. Another limitation relates to the need for occasional caregiver assistance during questionnaire administration. Although caregivers were restricted to reading or clarifying items and all responses reflected the participant's own answers, the possibility that proxy involvement may have influenced response interpretation in some cases cannot be entirely excluded. This limitation is inherent to studies involving older adults with sensory or literacy limitations and should be considered when interpreting the findings. Furthermore, oral health–related quality of life instruments such as the Oral Health Impact Profile (OHIP) or the Geriatric Oral Health Assessment Index (GOHAI) were not included. This was a deliberate methodological choice, as the OFI‐8 was originally developed as a self‐reported screening tool for use in non‐dental geriatric settings, focusing on perceived oral functional decline and behaviours rather than dental status or oral health–related quality of life. Accordingly, validity analyses were conducted using geriatric clinical measures reflecting functional, nutritional, and swallowing vulnerability, in line with the conceptual framework of oral frailty. Finally, as the study included only community‐dwelling older adults, generalisability to institutionalised or hospitalised populations may be limited. Given the self‐administered nature of the OFI‐8, its applicability in settings involving more advanced oral functional decline, acute illness, or limited cognitive or communicative capacity requires cautious consideration. In such populations, proxy‐assisted administration or complementary objective oral assessments may be necessary. Accordingly, any extension of validation to these settings should be undertaken within the conceptual boundaries of oral frailty as an early‐stage, potentially reversible condition.

The findings are consistent with previous international validations of the OFI‐8. The KR‐20 coefficient of 0.62 indicates moderate internal consistency, which is acceptable for an eight‐item dichotomous scale and consistent with previous validations of the OFI‐8. Tanaka et al. reported a Cronbach's alpha of 0.69, while the Portuguese adaptation showed a KR‐20 value of 0.63, supporting similar reliability across languages [14, 16]. In short, for multidimensional screening instruments with binary response formats, lower inter‐item correlations are expected because items intentionally reflect heterogeneous but related domains. Therefore, the observed internal consistency likely reflects construct breadth rather than measurement weakness, particularly when considered alongside the excellent test–retest reliability and coherent factor structure observed in this study.

The CFA replicated the single‐factor model observed in the Japanese and Portuguese validations, indicating that all items reflect a unified construct encompassing chewing, swallowing, saliva secretion, and oral behaviour [14, 16]. Although the RMSEA value slightly exceeded the optimal threshold (0.10 vs. 0.08), other indices (CFI = 0.93, GFI = 0.95, AGFI = 0.91) demonstrated overall acceptable fit. This pattern is common in short binary instruments, where item heterogeneity modestly elevates error estimates. As in previous studies, chewing‐ and swallowing‐related items showed the strongest loadings, while behavioural items contributed more modestly but meaningfully.

Convergent and clinical validity analyses showed significant correlations between the OFI‐8‐TR and measures of frailty, nutritional status, swallowing difficulty, and oral intake, indicating close associations between oral frailty, physical frailty, and nutritional vulnerability. Prior research has reported associations between oral frailty and adverse functional outcomes. In Tanaka et al.'s cohort study, higher OFI‐8 scores were associated with incident functional disability [14]. Hironaka et al. reported associations between oral and social frailty and physical frailty [29], while population‐based studies demonstrated indirect effects through reduced dietary variety and social isolation [30]. Watanabe et al. further reported an association between oral frailty and all‐cause mortality, independent of physical and psychological frailty [31]. Together, these studies suggest that oral frailty represents an integrative construct spanning oral, social, nutritional, and physical domains of ageing.

It should be noted that several comparator measures used in this study, including the FRAIL scale, CFS, and MNA‐SF, assess constructs that are conceptually related to oral frailty. Therefore, the observed correlations should be interpreted with appropriate caution, as part of the shared variance likely reflects overlapping domains of functional vulnerability rather than entirely distinct constructs. Nevertheless, the consistent pattern of associations observed across multiple functional, nutritional, and swallowing‐related measures supports the internal coherence of the OFI‐8‐TR within a comprehensive geriatric assessment framework.

Dental‐specific quality‐of‐life instruments such as the Oral Health Impact Profile (OHIP) or the Geriatric Oral Health Assessment Index (GOHAI) were not included in this study. The OFI‐8 was specifically designed as a brief self‐reported screening tool focusing on functional oral decline rather than oral health–related quality of life and was intended for use in non‐dental, geriatric settings. As such, convergent validity was examined using geriatric and functional measures more commonly applied in medical outpatient care. Future studies integrating dental examinations and oral health–related quality‐of‐life instruments would be valuable to further explore discriminant validity across dental and geriatric domains.

Known‐groups validity supported the sensitivity of the OFI‐8‐TR to nutritional differences, with higher scores among individuals with malnutrition or at risk of malnutrition as defined by the MNA‐SF. Similar associations have been reported in community and clinical settings; oral frailty increases the odds of malnutrition [32], predicts nutritional deterioration and sarcopenia progression in haemodialysis patients [33], and contributes to poorer oral intake through chewing and swallowing impairment as well as social withdrawal [34]. Chewing and swallowing difficulties are consistently central to oral frailty [35], and high OFI‐8 scores have been associated with dysphagia and poorer nutritional indices among hospitalised older adults [36]. These findings reinforce the relevance of the OFI‐8‐TR for identifying older adults at heightened nutritional risk.

Taken together, these findings highlight the potential clinical utility of the OFI‐8‐TR. As a brief and self‐reported instrument, it may facilitate early identification of oral frailty in outpatient settings and support timely preventive and rehabilitative interventions. Future studies should examine measurement invariance, responsiveness to change, and predictive validity for adverse outcomes across diverse clinical populations.

5. Conclusion

The findings of this study indicate that the Turkish version of the OFI‐8‐TR demonstrates acceptable reliability, construct validity, and cultural adaptability for the assessment of oral frailty in community‐dwelling older adults. Comparable to other language versions, it represents a brief and feasible screening instrument for use within comprehensive geriatric assessment. However, as this study did not evaluate sensitivity, predictive validity, or responsiveness to change, conclusions regarding its clinical prognostic utility should be interpreted with caution. Future longitudinal and interventional studies are warranted to further examine these properties and to clarify the role of the OFI‐8‐TR as a screening and risk stratification tool for early identification and monitoring of oral frailty within geriatric care.

Author Contributions

F.O.K.K. conceived and designed the study, supervised data collection, performed statistical analyses, and drafted the manuscript. All authors contributed to data interpretation, critically revised the manuscript for important intellectual content, and approved the final version for submission.

Funding

The authors have nothing to report.

Ethics Statement

The study was approved by the Clinical Research Ethics Committee of Tepecik Training and Research Hospital (Date: 5 February 2025; Approval No: 2025/01‐15) and conducted in accordance with the ethical standards of the Declaration of Helsinki.

Consent

Written informed consent was obtained from all participants or, when applicable, from their caregivers.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

File S1: Turkish version of the Oral Frailty Index‐8 (OFI‐8).

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.