Nutritional Status, Fat-Free Mass Index, and Functional Dependency in Older Adults: A Cross-Sectional Path Analysis Authors: Hacer Alataş, Nurgül Arslan, Ezgi Nur Güneş, Harun Kurt, Serkan Bentli Published: 2026 ABSTRACT AbstractObjectiveThis study aimed to investigate the cross-sectional associations among nutritional status, body composition, and functional dependency in older adults and to evaluate whether fat-free mass index (FFMI) statistically contributed to the indirect association between nutritional and functional status using observed-variable path analysis.MethodsThis retrospective cross-sectional study included 165 adults aged ≥65 years residing in long-term care centers in Istanbul, Türkiye, of whom 51.5% were women. Nutritional status was assessed using the Mini Nutritional Assessment (MNA), functional status using the Barthel Index, and body composition using multifrequency bioelectrical impedance analysis (InBody S10). Handgrip strength, gait speed, dysphagia, and sarcopenia were also evaluated. Multivariable regression, receiver operating characteristic (ROC), and observed-variable path analyses were performed.ResultsOverall, 44.2% of participants were malnourished, 32.7% were at risk of malnutrition, and 70.3% were functionally dependent. Higher MNA scores (OR=0.66, 95% CI: 0.57–0.77), FFMI (OR=0.84, 95% CI: 0.72–0.98), handgrip strength (OR=0.73, 95% CI: 0.61–0.88), and gait speed (OR=0.64, 95% CI: 0.45–0.91) were associated with lower odds of functional dependency, whereas dysphagia (OR=8.27, 95% CI: 2.23–30.65) and sarcopenia (OR=3.90, 95% CI: 1.43–10.63) were associated with higher odds. The MNA score showed excellent discriminatory ability for functional dependency (AUC=0.894). Path analysis showed a strong positive direct association between MNA and Barthel Index scores (β=0.882, p<0.001) and a small negative indirect association through FFMI (β=−0.023, bootstrap 95% CI: −0.053 to −0.002). The opposing directions of the direct and indirect associations were consistent with a statistical suppression pattern rather than a conventional mediating relationship.ConclusionPoor nutritional status, low FFMI, reduced muscle strength, slower gait speed, dysphagia, and sarcopenia were associated with functional dependency in older adults. Although FFMI contributed to the statistical decomposition of the association between nutritional and functional status, the observed indirect association was small and opposite in direction to the direct association. These findings should be interpreted as concurrent cross-sectional relationships rather than evidence of mediation, temporal ordering, or causality. Longitudinal studies are needed to clarify their temporal and clinical significance.Keywords: malnutrition, functional dependency, fat-free mass index, older adults, path analysis Introduction Population aging has become one of the most important global public health issues due to the increasing prevalence of chronic diseases, disability, frailty, and loss of independence among older adults.1 Functional dependency, which refers to the inability to independently perform activities of daily living, is associated with reduced quality of life, increased healthcare utilization, hospitalization, institutionalization, and mortality in geriatric populations.2,3 Therefore, identifying modifiable factors contributing to functional decline has become increasingly important in geriatric healthcare. Malnutrition is one of the most common geriatric syndromes and is highly prevalent among older adults receiving long-term care and institutional support.4 Physiological aging, chronic diseases, reduced appetite, swallowing difficulties, cognitive decline, polypharmacy, and decreased mobility may contribute to inadequate nutritional intake and deterioration in nutritional status in elderly individuals.4,5 Studies conducted in institutionalized and hospitalized older adults have reported high rates of malnutrition or risk of malnutrition; for example, Chatindiara et al (2018) reported that almost three-quarters of older adults admitted to hospital were either malnourished or at risk of malnutrition. The same study also demonstrated strong associations between malnutrition risk, cognitive decline, dysphagia risk, low muscle strength, and unintentional weight loss, highlighting the multifactorial nature of nutritional deterioration in older adults.6 Previous studies have demonstrated that malnutrition is closely associated with frailty, sarcopenia, impaired physical performance, prolonged hospitalization, and increased mortality risk.4,7,8 In addition, inadequate nutritional status has been identified as an important predictor of reduced functional capacity and dependency in older adults.8 The relationship between malnutrition and functional dependency may partly be explained by age-related alterations in body composition, particularly reductions in lean body mass. Although body mass index (BMI) is commonly used in clinical practice, it may not adequately reflect changes in skeletal muscle mass and body composition in geriatric individuals.9 In contrast, fat-free mass index (FFMI), which represents lean body mass adjusted for height, has emerged as a more sensitive indicator of nutritional and muscle status in older adults.10,11 Reduced FFMI has been associated with decreased muscle strength, impaired mobility, sarcopenia, and poor functional outcomes.11,12 Therefore, body composition assessment may provide important additional information beyond conventional anthropometric measurements. Previous studies have demonstrated that malnutrition is closely associated with frailty, sarcopenia, impaired physical performance, prolonged hospitalization, and increased mortality risk.4,7,8 In addition, inadequate nutritional status has been identified as an important predictor of reduced functional capacity and dependency in older adults.8 The relationship between malnutrition and functional dependency may partly be explained by age-related alterations in body composition, particularly reductions in lean body mass. Although body mass index (BMI) is commonly used in clinical practice, it may not adequately reflect changes in skeletal muscle mass and body composition in geriatric individuals.9 In contrast, fat-free mass index (FFMI), which represents lean body mass adjusted for height, has emerged as a more sensitive indicator of nutritional and muscle status in older adults.10,11 Reduced FFMI has been associated with decreased muscle strength, impaired mobility, sarcopenia, and poor functional outcomes.11,12 Therefore, body composition assessment may provide important additional information beyond conventional anthropometric measurements. Nevertheless, BIA-derived estimates should be interpreted with consideration of hydration status and fluid distribution, as age-related and clinical fluid shifts may influence impedance-based body composition measurements in institutionalized older adults. Muscle strength, physical performance, and swallowing function are important determinants of functional independence in older adults. Reduced handgrip strength, slow gait speed, and dysphagia have been associated with frailty, disability, hospitalization, mortality, and impaired nutritional intake.12 Their coexistence with malnutrition and sarcopenia may further increase the risk of functional dependency. FFMI may be particularly relevant when examining the relationship between nutritional and functional status in older adults because nutritional deterioration frequently coexists with reductions in lean tissue, muscle strength, and physical performance.12–14 Unlike BMI, which does not distinguish lean tissue from adipose tissue, FFMI provides a height-adjusted estimate of the fat-free component of body weight and may therefore offer complementary information regarding body composition and functional vulnerability. Importantly, FFMI should not be regarded a priori as a confirmed biological mediator of the nutrition–function relationship. Rather, its inclusion in the present study was intended to evaluate whether variation in fat-free mass statistically contributes to the cross-sectional association between nutritional and functional status. From a clinical perspective, determining whether FFMI provides information beyond conventional anthropometric measures may help clarify the potential value of incorporating body-composition assessment into comprehensive geriatric nutritional and functional evaluation.15,16 Despite growing evidence regarding the associations between nutritional status, body composition, and functional outcomes, the statistical pathways underlying these relationships remain insufficiently understood. In particular, limited studies have examined whether FFMI contributes to the indirect association between nutritional and functional status. Observed-variable path analysis enables the simultaneous estimation of direct and indirect associations among measured variables and may provide additional insight into these relationships.17 Therefore, the present study aimed to investigate the cross-sectional associations among nutritional status, body composition, and functional dependency in older adults and to evaluate the direct and indirect statistical associations among nutritional status, FFMI, and functional status using observed-variable path analysis. We hypothesized that nutritional status would be associated with functional status and that FFMI would contribute to the observed association. Materials and Methods Study Design and Participants This single-center, retrospective cross-sectional study included older adults residing in long-term care centers affiliated with a single institution in Istanbul, Türkiye. Data were obtained from individuals who underwent routine comprehensive geriatric assessment between January 2024 and January 2026. Clinical, nutritional, anthropometric, functional, and body composition data were retrieved from institutional health records, archived patient files, and routine assessment forms. Individuals aged 65 years or older were eligible for inclusion. Participants were excluded if they were younger than 65 years, had severe cognitive impairment that prevented reliable assessment, had a terminal illness, had an acute infection or inflammatory condition at the time of evaluation, or had incomplete data on nutritional status, body composition, or functional assessment. A total of 195 older adults were screened. Of these, 30 were excluded: eight were younger than 65 years, seven had severe cognitive impairment, five had a terminal illness, four had an acute infection or inflammatory condition, and six had missing or incomplete data (Figure 1). The final analysis therefore included 165 participants. Participants were classified as functionally independent or functionally dependent according to their Barthel Index scores. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Malatya Turgut Özal University Health Sciences Scientific Research Ethics Committee (Approval No. E-30785963-020-383857). Owing to the retrospective nature of the study and the use of routinely collected clinical data, the requirement for written informed consent was waived by the ethics committee. All patient information was handled confidentially, and data were anonymized prior to analysis to protect participant privacy. Flowchart of the study design and analytical process. Sample Size Calculation The minimum sample size was estimated for multiple linear regression analysis using G*Power software, version 3.1. Assuming a small-to-moderate effect size of f2=0.10f2=0.10f2=0.10, a two-sided significance level of 0.05, a statistical power of 80%, and a maximum of eight predictors in the final model, the minimum required sample size was calculated as 159 participants. The final study sample comprised 165 individuals and therefore exceeded the minimum sample size required for the planned multivariable analyses. The final sample size also provided an adequate number of outcome events for the multivariable logistic regression analysis. In addition, the path model was parsimonious, comprising three observed variables, and indirect effects were estimated using 5,000 bootstrap resamples. Nutritional Assessment Nutritional status was evaluated using the Mini Nutritional Assessment (MNA), a validated screening tool specifically developed for older adults.18 The MNA assesses multiple domains including anthropometric measurements, dietary intake, mobility, neuropsychological problems, and self-perceived health and nutritional status. The total score ranges from 0 to 30 points, with higher scores indicating better nutritional status. According to established cut-off values, participants were classified as having normal nutritional status, being at risk of malnutrition, or being malnourished.18 In the present study, MNA score was used both as a continuous variable and as a categorical nutritional status indicator. Dietary intake information was retrospectively obtained from hospital nutrition follow-up records documented by clinical dietitians during routine geriatric evaluation.4 Detailed definitions and coding procedures for the study variables are provided in Supplementary Table S1. Functional Dependency Assessment Functional dependency was evaluated using the Barthel Index, a widely validated and frequently used instrument for assessing activities of daily living in geriatric populations.19 The Barthel Index measures an individual’s ability to independently perform basic daily self-care and mobility-related activities and is considered an important indicator of functional status and independence in older adults. The scale assesses multiple domains including feeding, bathing, grooming, dressing, bowel and bladder continence, toilet use, chair and bed transfers, mobility, and stair climbing. Each item is scored according to the degree of assistance required, and the total score ranges from 0 to 100 points. Higher scores indicate greater functional independence, whereas lower scores reflect increased functional dependency and poorer physical functioning.19 Barthel Index assessments included in the present study were retrospectively obtained from routinely documented geriatric evaluation records. In addition to continuous score analyses, participants were categorized according to functional dependency status based on Barthel Index classification as functionally independent or functionally dependent. For categorical analyses, participants with a Barthel Index score ≥90 were classified as having higher functional independence, whereas those with a score <90 were classified as functionally dependent. Anthropometric Measurements Anthropometric measurements routinely performed during geriatric assessment were obtained from hospital records. Body weight and height were measured using standardized clinical procedures while participants wore light clothing and no shoes. Body mass index (BMI) was calculated as body weight divided by height squared (kg/m2).20 Calf circumference measurements were obtained using a non-elastic measuring tape from the widest part of the calf while participants were in a seated position. Calf circumference was included as an indicator of muscle mass and nutritional status, as anthropometric and muscle-related parameters are clinically relevant in the assessment of malnutrition and functional decline among older adults.4,21 Body Composition Analysis Body composition was assessed using a multifrequency bioelectrical impedance analyzer (InBody S10, InBody Co., Ltd., Seoul, Republic of Korea). Measurements were performed according to the routine institutional BIA protocol, which aimed to standardize premeasurement conditions as far as clinically feasible. Participants were assessed in a stable clinical condition, preferably before meals and exercise, after bladder emptying, and following a period of rest. Measurements were performed in the supine position using the device-specific electrode configuration. The InBody S10 has been validated for body-composition assessment in older adults,22 and BIA-derived parameters have been widely used in older populations to evaluate lean mass, fat mass, and hydration-related body-composition compartments.23 The body composition parameters recorded for the study included fat-free mass, fat mass, total body water (TBW), extracellular water (ECW), and intracellular water (ICW). Fat-free mass was considered an indicator of lean tissue compartments, including skeletal muscle mass, whereas fat mass reflected total adipose tissue. TBW, ECW, and ICW were used to characterize hydration-related body composition parameters. Fat-free mass index (FFMI) was calculated by dividing fat-free mass in kilograms by height squared in meters: FFMI = fat-free mass (kg) / height2 (m2). For categorical analyses, low FFMI was defined as <18 kg/m2 in men and <15 kg/m2 in women.24 Muscle Strength and Physical Performance Assessment Handgrip strength was assessed using a calibrated hand dynamometer according to standardized procedures during routine geriatric evaluation.21,25 The highest recorded value was used in the statistical analyses. Low muscle strength was defined as a handgrip strength of <28 kg in men and <18 kg in women, in accordance with the 2019 Asian Working Group for Sarcopenia (AWGS) criteria.26 Physical performance was evaluated using usual gait speed, expressed in meters per second. A gait speed of <1.0 m/s was considered indicative of low physical performance.27 Appendicular skeletal muscle mass was assessed using bioelectrical impedance analysis and indexed to height squared (ASM/height2). Low muscle mass was defined as an ASM/height2 value of <7.0 kg/m2 in men and <5.7 kg/m2 in women. Sarcopenia was diagnosed when low muscle mass was accompanied by low muscle strength. Severe sarcopenia was defined as the concurrent presence of low muscle mass, low muscle strength, and low physical performance, in accordance with the AWGS 2019 diagnostic algorithm.12,21 Sarcopenic obesity was evaluated separately from the AWGS sarcopenia classification. It was defined as the coexistence of low appendicular skeletal muscle mass and elevated body fat percentage. Low absolute appendicular skeletal muscle mass was defined as <19.75 kg in men and <15.02 kg in women, while elevated body fat percentage was defined as >25% in men and >35% in women.26 Dysphagia Assessment Swallowing function and dysphagia severity were assessed using the Gugging Swallowing Screen (GUSS), a validated bedside screening instrument designed to evaluate swallowing impairment and aspiration risk. The GUSS follows a stepwise protocol comprising an indirect swallowing assessment and, when appropriate, a direct swallowing assessment.28 The indirect assessment evaluates level of consciousness, voluntary cough, saliva swallowing, voice quality, and the presence of drooling or abnormal oral secretions. Participants who successfully completed this stage proceeded to the direct swallowing assessment, which includes semisolid, liquid, and solid swallowing trials. During these trials, delayed swallowing, coughing, choking, voice changes, and other clinical signs suggestive of aspiration were monitored. The total GUSS score ranges from 0 to 20, with higher scores indicating better swallowing function and a lower risk of aspiration. Lower scores indicate more severe swallowing impairment and greater aspiration risk. In the present study, the total GUSS score, indirect swallowing assessment score, direct swallowing assessment score, and categorical dysphagia status were included in the analyses. Dysphagia-related variables were examined in relation to nutritional status, body composition, muscle strength, physical performance, and functional dependency.28 Statistical Analysis All statistical analyses were performed using IBM SPSS Statistics, version 29.0 (IBM Corp., Armonk, NY, USA). Observed-variable path analysis was conducted using IBM SPSS Amos, version 29.0. Detailed operational definitions and coding procedures for all study variables are provided in Supplementary Table S1. The distribution of continuous variables was evaluated using histograms and Q–Q plots, together with the Kolmogorov–Smirnov and Shapiro–Wilk tests. Continuous variables are presented as mean ± standard deviation, and categorical variables as frequency and percentage. Between-group comparisons according to functional dependency status were performed using the independent-samples t-test for continuous variables and the chi-square test or Fisher’s exact test for categorical variables, as appropriate. Additional comparisons according to nutritional status are presented in Supplementary Table S2. For exploratory analyses, participants were categorized into tertiles according to their MNA scores. Trends across tertiles were evaluated using one-way analysis of variance for continuous variables and the linear-by-linear association test for categorical variables. Associations among nutritional, anthropometric, body composition, functional, muscle-related, and dysphagia-related variables were examined using Spearman’s rank correlation analysis. Multivariable linear regression analysis was performed to identify factors independently associated with Barthel Index score, which was entered as a continuous dependent variable. Higher Barthel Index scores indicated greater functional independence. Variables considered clinically relevant or associated with the outcome in preliminary analyses were entered into the multivariable model. Results are reported as unstandardized regression coefficients (B), standardized coefficients (β), standard errors (SE), 95% confidence intervals (CI), and p-values. Multicollinearity was evaluated using variance inflation factor values; detailed diagnostics for the final model are provided in Supplementary Table S4. Multivariable binary logistic regression analysis was performed to identify factors independently associated with functional dependency status. Functional dependency was entered as the binary dependent variable. Results are reported as odds ratios (OR) with corresponding 95% CIs and p-values. Because of the cross-sectional design, regression findings were interpreted as associations rather than causal or prospective risk estimates. Receiver operating characteristic (ROC) curve analysis was used to evaluate the discriminative performance of nutritional and muscle-related parameters for identifying functional dependency. The area under the curve (AUC), 95% CI, sensitivity, specificity, and optimal cut-off values were calculated. Optimal thresholds were determined using the Youden index.29 Detailed ROC estimates are presented in Supplementary Table S5. Observed-variable path analysis was performed to examine the indirect association between nutritional status and functional independence through fat-free mass index (FFMI). MNA score was specified as the exposure variable, FFMI as the intermediate variable, and Barthel Index score as the outcome variable. Direct, indirect, and total effects were estimated using 5,000 bootstrap resamples. Standardized path coefficients (β), percentile-based bootstrap 95% CIs, and p-values were reported.30 Because the model included three observed variables and was saturated, global model fit indices were not considered informative. The indirect effect was interpreted according to both its magnitude and direction relative to the direct effect. Opposing signs between the direct and indirect effects were considered indicative of an inconsistent mediation or suppression pattern rather than conventional partial mediation. Given the retrospective cross-sectional design, the path model was interpreted as a statistical decomposition of concurrent associations and not as evidence of a temporal, mechanistic, or causal pathway. Detailed estimates of the direct, indirect, and total effects are provided in Supplementary Table S3. All tests were two-tailed, and a p-value <0.05 was considered statistically significant. Results Table 1 summarizes participant characteristics according to functional dependency status. Functionally dependent participants were significantly older than those with higher functional independence (p<0.001). Body weight, BMI, sex distribution, mean FFMI, and inadequate dietary intake did not differ significantly between groups. The functionally dependent group had significantly lower MNA scores and a higher prevalence of malnutrition risk or malnutrition (both p<0.001). Low FFMI was more common among functionally dependent participants (57.8% vs 34.7%, p=0.011). Barthel Index scores, handgrip strength, and gait speed were significantly lower in the functionally dependent group, while dysphagia, sarcopenia, and polypharmacy were significantly more prevalent (all p<0.001). Additional comparisons according to nutritional status are presented in Supplementary Table S2. Participants with malnutrition or risk of malnutrition had significantly lower BMI, FFMI, Barthel Index scores, and handgrip strength, together with higher prevalences of dysphagia, sarcopenia, low FFMI, and functional dependency. Table 1Participant Characteristics According to Functional Dependency StatusCharacteristicsHigher FunctionalIndependence (n=49)FunctionalDependency (n=116)p-ValueDemographic and anthropometric characteristicsAge (years)77.24 ± 9.6883.05 ± 8.41<0.001Body weight (kg)70.38 ± 16.8368.95 ± 14.910.589Height (cm)163.24 ± 6.84167.59 ± 10.360.008Body mass index (kg/m2)25.46 ± 3.4024.50 ± 5.060.226Female sex, n (%)30 (61.2)55 (47.4)0.147Nutritional and body composition parametersMini Nutritional Assessment score23.63 ± 3.0910.78 ± 9.66<0.001Fat-free mass index (kg/m2)16.64 ± 2.8316.22 ± 3.450.454Low fat-free mass index, n (%)17 (34.7)67 (57.8)0.011Inadequate dietary intake, n (%)31 (63.3)66 (56.9)0.558Functional and physical performance parametersBarthel Index score (0–100)83.93 ± 12.1024.83 ± 24.37<0.001Handgrip strength (kg)5.84 ± 4.791.74 ± 2.50<0.001Gait speed (m/s)1.78 ± 2.580.43 ± 1.37<0.001Clinical characteristicsAt risk of malnutrition or malnourished, n (%)2 (4.1)71 (61.2)<0.001Dysphagia, n (%)17 (34.7)111 (95.7)<0.001Sarcopenia, n (%)13 (26.5)80 (69.0)<0.001Polypharmacy, n (%)39 (79.6)113 (97.4)<0.001Notes: Data are presented as mean ± standard deviation or number (percentage). Continuous and categorical variables were compared using the independent-samples t-test and chi-square or Fisher’s exact test, respectively. Higher Barthel Index scores indicate greater functional independence.Abbreviations: MNA, Mini Nutritional Assessment; FFMI, fat-free mass index. Participant Characteristics According to Functional Dependency Status Notes: Data are presented as mean ± standard deviation or number (percentage). Continuous and categorical variables were compared using the independent-samples t-test and chi-square or Fisher’s exact test, respectively. Higher Barthel Index scores indicate greater functional independence. Abbreviations: MNA, Mini Nutritional Assessment; FFMI, fat-free mass index. Table 2 presents the multivariable linear regression analysis of factors associated with Barthel Index score. Higher MNA scores, greater handgrip strength, and faster gait speed were independently associated with higher Barthel Index scores. In contrast, older age and dysphagia were independently associated with lower Barthel Index scores. MNA score showed the strongest association with functional status (β=0.67, p<0.001). Sarcopenia and polypharmacy were not significantly associated with Barthel Index score after adjustment for the other variables. The model explained 82.1% of the variance in Barthel Index score (adjusted R2=0.813). Table 2Multivariable Linear Regression Analysis of Factors Associated with Barthel Index ScorePredictorBSEStandardized βt95% CI for Bp-ValueVIFAge (years)−0.090.04−0.08−2.21−0.18 to −0.010.0281.20MNA score0.680.050.6713.960.58 to 0.78<0.0011.99Handgrip strength (kg)0.460.120.173.870.22 to 0.69<0.0011.66Gait speed (m/s)0.550.200.102.820.17 to 0.940.0051.15Dysphagia, yes−4.101.07−0.17−3.81−6.22 to −1.97<0.0011.65Sarcopenia, yes0.900.860.041.05−0.80 to 2.600.2961.49Polypharmacy, yes−1.441.40−0.04−1.03−4.21 to 1.320.3041.17Notes: Dependent variable: Barthel Index score. Higher scores indicate greater functional independence. Model fit: R2=0.821; adjusted R2=0.813; F=102.80; p<0.001.Abbreviations: B, unstandardized coefficient; SE, standard error; β, standardized coefficient; CI, confidence interval; VIF, variance inflation factor; MNA, Mini Nutritional Assessment. Multivariable Linear Regression Analysis of Factors Associated with Barthel Index Score Notes: Dependent variable: Barthel Index score. Higher scores indicate greater functional independence. Model fit: R2=0.821; adjusted R2=0.813; F=102.80; p<0.001. Abbreviations: B, unstandardized coefficient; SE, standard error; β, standardized coefficient; CI, confidence interval; VIF, variance inflation factor; MNA, Mini Nutritional Assessment. Table 3 presents the multivariable logistic regression analysis of factors associated with functional dependency. Each one-year increase in age was associated with an 8% increase in the odds of functional dependency (OR=1.08, 95% CI: 1.02–1.15, p=0.008). Dysphagia was associated with more than eightfold higher odds of functional dependency (OR=8.27, 95% CI: 2.23–30.65, p=0.002), while sarcopenia was associated with nearly fourfold higher odds (OR=3.90, 95% CI: 1.43–10.63, p=0.008). Higher MNA scores were associated with lower odds of functional dependency (OR=0.66, 95% CI: 0.57–0.77, p<0.001). Similarly, higher FFMI (OR=0.84, 95% CI: 0.72–0.98, p=0.023), greater handgrip strength (OR=0.73, 95% CI: 0.61–0.88, p=0.001), and faster gait speed (OR=0.64, 95% CI: 0.45–0.91, p=0.015) were independently associated with lower odds of dependency. Polypharmacy was not significantly associated with functional dependency after adjustment for the other variables (OR=2.10, 95% CI: 0.54–8.08, p=0.283). Table 3Multivariable Logistic Regression Analysis of Factors Associated with Functional DependencyPredictorBSEWald χ2OR95% CIp-ValueAge (years)0.080.036.941.081.02–1.150.008MNA score−0.420.0733.810.660.57–0.77<0.001FFMI (kg/m2)−0.180.085.140.840.72–0.980.023Handgrip strength (kg)−0.310.0911.670.730.61–0.880.001Gait speed (m/s)−0.440.185.970.640.45–0.910.015Dysphagia, yes2.110.679.928.272.23–30.650.002Sarcopenia, yes1.360.517.063.901.43–10.630.008Polypharmacy, yes0.740.691.152.100.54–8.080.283Notes: Dependent variable: Functional dependency. Reference categories: No dysphagia, no sarcopenia, and no polypharmacy.Abbreviations: B, regression coefficient; SE, standard error; OR, odds ratio; CI, confidence interval; MNA, Mini Nutritional Assessment; FFMI, fat-free mass index. Multivariable Logistic Regression Analysis of Factors Associated with Functional Dependency Notes: Dependent variable: Functional dependency. Reference categories: No dysphagia, no sarcopenia, and no polypharmacy. Abbreviations: B, regression coefficient; SE, standard error; OR, odds ratio; CI, confidence interval; MNA, Mini Nutritional Assessment; FFMI, fat-free mass index. The Figure 1 flowchart illustrates participant screening and selection, nutritional, body-composition, functional, and clinical assessments, grouping according to functional dependency status, and the statistical analyses performed. Nutritional status was assessed using the Mini Nutritional Assessment (MNA), body composition was evaluated using bioelectrical impedance analysis and fat-free mass index (FFMI), functional status was assessed using the Barthel Index (BI), handgrip strength, and gait speed, and clinical assessment included dysphagia, sarcopenia, and polypharmacy. Participants were classified as having higher functional independence (BI ≥90) or functional dependency (BI <90). The analytical approach included descriptive, correlation, multivariable regression, receiver operating characteristic (ROC), and observed-variable path analyses. This heatmap presents Spearman correlation coefficients among Mini Nutritional Assessment (MNA) score, fat-free mass index (FFMI), Barthel Index score, handgrip strength, gait speed, sarcopenia, and dysphagia. Blue tones indicate positive correlations, whereas red tones indicate negative correlations. Values shown within the cells are Spearman correlation coefficients (ρ). p<0.05, *p<0.01, and **p<0.001 (Figure 2). Heatmap of correlations among nutritional, functional, and muscle-related parameters. Values are Spearman correlation coefficients (p). *p<0.05, **p<0.01, ***p<0.001. Receiver operating characteristic curves showing the discriminative performance of the Mini Nutritional Assessment (MNA) score and handgrip strength for identifying functional dependency in older adults. The MNA score showed an AUC of 0.894 (95% CI: 0.847–0.936), while handgrip strength showed an AUC of 0.772 (95% CI: 0.687–0.848). Optimal cut-off values determined using the Youden index were ≤20.0 for the MNA score and ≤4.0 kg for handgrip strength (Figure 3). ROC Analysis for Identifying Functional Dependency. This figure compares clinical and functional parameters across tertiles of the Mini Nutritional Assessment (MNA) score. Lower MNA tertiles were characterized by lower Barthel Index scores, handgrip strength, gait speed, and FFMI values, together with higher prevalences of dysphagia and sarcopenia. Trend p-values are shown for each parameter (Figure 4). Clinical and Functional Parameters According to MNA Tertiles. Data are presented as mean ± SD or percentage. ***p<0.001. The Figure 5 path diagram illustrates the direct and indirect associations between nutritional status, fat-free mass index (FFMI), and functional independence. Nutritional status was assessed using the Mini Nutritional Assessment (MNA), and functional independence was assessed using the Barthel Index. Standardized path coefficients (β) and corresponding p-values are shown for each pathway. Higher MNA scores were positively associated with FFMI (β=0.242, p=0.002), whereas FFMI was inversely associated with Barthel Index score after adjustment for MNA score (β=−0.095, p=0.021). The direct association between MNA and Barthel Index scores remained strongly positive (β=0.882, p<0.001). The standardized indirect effect through FFMI was small and negative (β=−0.023, bootstrap 95% CI: −0.053 to −0.002, p=0.027). Because the direct and indirect effects were in opposite directions, the model indicated an inconsistent mediation or suppression pattern rather than conventional partial mediation. Direct, indirect, and total effects were estimated using 5,000 bootstrap resamples. Given the cross-sectional design, the findings were interpreted as a statistical decomposition of concurrent associations rather than evidence of a causal pathway (Figure 5). Path Analysis of the Association Between Nutritional Status and Functional Independence Through Fat-Free Mass Index. Discussion The present study investigated the associations among nutritional status, body composition, and functional dependency in older adults and evaluated the indirect association between nutritional status and functional status through fat-free mass index (FFMI) using observed-variable path analysis. Poorer nutritional status was associated with greater functional dependency, lower handgrip strength, slower gait speed, and higher prevalences of dysphagia and sarcopenia. Low FFMI was also more common among functionally dependent participants. Multivariable linear regression showed that MNA score, handgrip strength, gait speed, age, and dysphagia were independently associated with Barthel Index scores, whereas logistic regression additionally identified FFMI and sarcopenia as factors associated with functional dependency. Path analysis further demonstrated a small but statistically significant negative indirect association between nutritional status and functional status through FFMI. Because the direct and indirect associations were opposite in direction, the observed pattern was most consistent with statistical suppression. Given the cross-sectional design, this pattern should be interpreted as a feature of the concurrent association structure rather than evidence of a temporal or causal mediating mechanism. Malnutrition is a common but frequently underrecognized geriatric syndrome that may adversely affect physical performance, independence, quality of life, healthcare use, and survival.31,32 In the present study, participants who were at risk of malnutrition or malnourished had substantially lower Barthel Index scores and markedly higher rates of functional dependency than those with normal nutritional status. MNA score showed the strongest independent association with functional status in the linear regression analysis and was also associated with lower odds of dependency in the logistic regression model. These findings are consistent with previous studies reporting that poor nutritional status is associated with impaired activities of daily living, frailty, disability, prolonged hospitalization, and mortality in older adults.4,18,33,34 Inadequate energy and protein intake, reduced appetite, chronic disease burden, and age-related physiological changes may contribute to losses in muscle mass and physical performance, thereby limiting independence. The strong association between MNA and Barthel Index scores should nevertheless be interpreted cautiously. The MNA includes mobility, neuropsychological problems, and self-perceived health in addition to dietary and anthropometric components.18 These domains may partly overlap with functional characteristics assessed by the Barthel Index. Therefore, the magnitude of the observed association may reflect both a genuine nutritional–functional relationship and some shared construct content. This overlap may also partly explain the high proportion of variance in Barthel Index scores accounted for by the linear model. However, the independent associations of handgrip strength, gait speed, age, and dysphagia indicate that functional dependency was not explained by nutritional status alone. Another important finding was the association between FFMI and functional outcomes. Although mean FFMI did not differ significantly between dependency groups, low FFMI was considerably more prevalent among functionally dependent participants, and higher FFMI was independently associated with lower odds of dependency. In contrast, BMI did not differ significantly according to functional status. This supports previous evidence that BMI may not adequately reflect age-related changes in body composition because it does not distinguish between fat mass and fat-free mass. FFMI may therefore provide more clinically relevant information regarding lean tissue depletion and physiological reserve in older adults.32,35 Muscle strength and physical performance were also strongly associated with functional status. Functionally dependent participants had markedly lower handgrip strength and gait speed, and both variables remained independently associated with Barthel Index scores and the odds of dependency. These findings are consistent with previous studies showing that low handgrip strength and slow gait speed are markers of frailty, disability, falls, hospitalization, loss of independence, and mortality.36–38 Handgrip strength also showed good discriminatory ability for functional dependency in the ROC analysis, although its performance was lower than that of the MNA score. Because the identified cut-off values were derived from a single institutionalized population, they require external validation before wider clinical use. Sarcopenia was substantially more prevalent among functionally dependent participants and was independently associated with nearly fourfold higher odds of dependency. This finding reinforces the importance of assessing muscle mass together with muscle strength and physical performance. Sarcopenia may result from the combined effects of aging, inactivity, inadequate nutritional intake, chronic disease, inflammation, and prolonged institutional care.12,21 The coexistence of low FFMI, reduced handgrip strength, slow gait speed, and sarcopenia highlights the close relationship between muscle deterioration and loss of independence. Another notable finding was the strong association between dysphagia and functional dependency. Dysphagia was present in 95.7% of functionally dependent participants compared with 34.7% of those with higher functional independence. It was also independently associated with lower Barthel Index scores and more than eightfold higher odds of dependency. Swallowing impairment may reduce dietary intake, impair hydration, and increase the risk of aspiration, hospitalization, weight loss, and muscle depletion. Conversely, advanced frailty, neurological disease, sarcopenia, and severe disability may increase the likelihood of dysphagia, suggesting a bidirectional relationship. The very high prevalence of dysphagia in the dependent group should be interpreted in the context of the institutionalized, advanced-age population with a high burden of dependency, sarcopenia, and polypharmacy. Because dysphagia and functional status were assessed during the same retrospective cross-sectional period, it is not possible to determine whether dysphagia preceded dependency, resulted from advanced disability, or developed concurrently as part of a broader frailty syndrome. Nevertheless, the coexistence of dysphagia, malnutrition, sarcopenia, and dependency emphasizes the need for multidisciplinary assessment including nutritional, swallowing, muscle, and functional evaluation.39,40 A central finding was the significant indirect association between nutritional status and functional status through FFMI. Observed-variable path analysis showed that higher MNA scores were associated with higher FFMI, whereas higher FFMI was weakly associated with lower Barthel Index scores after adjustment for MNA score. The direct association between MNA and Barthel Index scores remained strongly positive. Consequently, the indirect effect through FFMI was small and negative, while the direct effect was positive. This opposing pattern indicates inconsistent mediation or statistical suppression rather than conventional partial mediation. This pattern suggests that variation in lean body mass may contribute to the observed statistical relationship between nutritional and functional status; however, it does not establish FFMI as a biological or temporal intermediary. Poorer nutritional status may coexist with lower fat-free mass, while reduced lean tissue may be associated with impaired strength, mobility, and daily functioning. This interpretation is consistent with previous studies showing that FFMI and related body composition measures may provide additional clinical information beyond BMI when evaluating physical function, sarcopenia, and functional vulnerability.35,38,41 However, this indirect association should not be interpreted as evidence of a confirmed biological or temporal mechanism. Nutritional status, FFMI, and functional status were measured concurrently, and the retrospective cross-sectional design does not establish temporal ordering. Reverse or reciprocal relationships are plausible: functional dependency may reduce mobility, contribute to muscle loss, and impair food intake, whereas low FFMI may reflect chronic disease, inflammation, inactivity, or frailty in addition to poor nutrition. Therefore, the path analysis findings should be interpreted as a statistical decomposition of cross-sectional associations rather than proof that reduced FFMI causally links malnutrition to functional dependency. In addition, because the direct and indirect effects were in opposite directions, the observed pattern should not be described as conventional partial mediation. Longitudinal studies with repeated measurements are needed to clarify the temporal sequence. The findings support the clinical value of assessing body composition alongside conventional nutritional and functional measures. FFMI, handgrip strength, gait speed, dysphagia, sarcopenia, and functional performance may provide complementary information not captured by body weight or BMI alone. Current geriatric nutrition guidelines similarly emphasize routine malnutrition screening, individualized nutritional care, and multidisciplinary intervention.4,25,41 However, the present findings do not establish that improving these parameters will prevent dependency; interventional studies are needed to evaluate whether nutritional support, resistance exercise, swallowing rehabilitation, or combined approaches can preserve lean mass and independence. Strengths and Limitations The study’s strengths include the simultaneous evaluation of nutritional status, body composition, muscle strength, gait performance, dysphagia, sarcopenia, and functional status, together with complementary linear regression, logistic regression, ROC, and observed-variable path analyses. Several limitations should also be considered. First, the retrospective cross-sectional design precludes conclusions regarding causality or temporal sequence and does not exclude reverse or bidirectional relationships. Accordingly, the direct and indirect associations identified in the path analysis should be interpreted as statistical relationships among concurrently measured variables rather than evidence of a temporal or causal pathway. Second, the study was conducted in long-term care centers affiliated with a single institution, limiting the generalizability of the findings to other older populations. Third, the use of routine clinical records may have introduced information bias and variation in measurement procedures. Fourth, although BIA measurements were performed according to routine institutional procedures designed to minimize variation in hydration and other premeasurement conditions, complete adherence to each condition could not be retrospectively verified for every participant. As BIA-derived estimates are sensitive to hydration status and fluid distribution, some residual measurement variability cannot be excluded. Fifth, conceptual overlap between the MNA and Barthel Index domains may have strengthened the observed association between nutritional and functional status. Finally, residual confounding cannot be excluded, and the ROC analyses were performed in the same sample without external validation. Conclusions Poor nutritional status, low FFMI, reduced muscle strength, slow gait speed, dysphagia, and sarcopenia were associated with functional dependency in institutionalized older adults. Path analysis identified a small negative indirect association involving FFMI alongside a strong positive direct association between nutritional and functional status. Because these variables were measured concurrently, the observed pattern should be interpreted as a statistical suppression pattern within the cross-sectional data rather than evidence of mediation, temporal ordering, or causality. FFMI may nevertheless provide complementary information when considered together with nutritional screening, muscle function, mobility, and swallowing assessment. Prospective multicenter studies with repeated measurements are needed to clarify temporal relationships and the clinical significance of these associations. Disclosure The authors report no conflicts of interest in this work. References