Evaluation of the effects of thyroid functions on frailty in geriatric patients using the Edmonton, SOF and FRAIL Scales

Abstract

Background and rationale

Thyroid dysfunction in older adults often mimics the signs of aging, impacting metabolism and overall physiological balance. While age-related chronic conditions have been extensively studied, the relationship between thyroid function and frailty remains underexplored.

Objective

This study aimed to evaluate the effects of thyroid dysfunction on frailty among individuals aged 65 years and older. Thyroid-stimulating hormone (TSH), free triiodothyronine (fT3), free thyroxine (fT4), and thyroid peroxidase antibody (anti-TPO) levels were analyzed. The study further examined the correlation between thyroid dysfunction, chronic diseases, sociodemographic factors, and optimal TSH levels in relation to frailty, using the Study of Osteoporotic Fractures (SOF), Edmonton Frail Scale (EFS), and FRAIL scales.

Methods

This cross-sectional study included 220 older adults with either treated or untreated thyroid dysfunction. Comprehensive geriatric assessments were conducted, including detailed medical histories, sociodemographic data collection, and thyroid function tests. Frailty was assessed using the SOF, EFS, and FRAIL scales. Multivariate logistic regression analyses were performed to identify significant associations between thyroid parameters and frailty.

Results

The median age of participants was 73 years, and 68.2% (n = 150) were women. Frailty prevalence was significantly higher in individuals with abnormal TSH levels (outside the 0.5–6 mIU/L range). Lower fT3 levels and the fT3/fT4 ratio were significantly associated with frailty, particularly as assessed by the SOF and EFS scales. In contrast, the FRAIL scale revealed a significant association between increased frailty and lower fT3 levels only. Subgroup analysis indicated that in individuals aged ≥ 80 years, a lower fT3/fT4 ratio was consistently associated with frailty across all frailty scales, whereas in those aged < 80 years, lower TSH levels showed a strong association with frailty as assessed by the FRAIL scale. These findings underscore age-specific variations in the relationship between thyroid function and frailty.

Conclusion

This study highlights the significant impact of thyroid dysfunction on frailty in older adults. Lower fT3 levels and the fT3/fT4 ratio emerged as key indicators of increased frailty, particularly on the SOF and EFS scales. Subgroup analysis further emphasized the importance of age-specific assessments, with a lower fT3/fT4 ratio being a critical indicator of frailty in individuals aged ≥ 80 years, while lower TSH levels were significant in those aged < 80 years. Abnormal TSH levels were strongly associated with frailty on the SOF scale, suggesting the need to consider thyroid dysfunction as a modifiable risk factor. Additionally, factors such as age, sex, education, thyroid medication use, and comorbidities influenced frailty status. Incorporating thyroid function tests into frailty assessments could enhance early identification and targeted interventions for at-risk older adults, particularly when age-specific thresholds are applied.

Key Summary Points

AbstractSection Aim

To investigate the impact of thyroid function on frailty in geriatric patients using the Edmonton Frail Scale (EFS), Study of Osteoporotic Fractures (SOF) scale, and FRAIL scale.

AbstractSection Findings

The study identified significant correlations between altered thyroid function parameters and increased frailty scores across the EFS, SOF, and FRAIL scales, highlighting the substantial influence of thyroid dysfunction on geriatric health.

AbstractSection Message

Thyroid function plays a critical role in determining frailty status among older adults, emphasizing the need for targeted monitoring and management to improve health outcomes in this population.

Introduction

Aging can be biologically characterized as a period of cellular differentiation, molecular change, and functional decline. During this phase, geriatric syndromes such as frailty, falls, immobility, malnutrition, pain, osteoporosis, and dementia become prevalent health issues [1, 2]. Frailty is defined as a deterioration in adaptive capacity caused by cumulative and multisystem regression of body systems, which results in reduced functional reserves. This leads to functional limitations, falls, hospitalizations, or even death among older adults. Affecting 10–25% of the older population, frailty significantly increases risks such as falls, depression, and malnutrition, thereby severely impacting the quality of life and daily activities of older individuals [3,4,5,6].

Although data on frailty prevalence in Turkey is limited, existing studies highlight frailty as a significant public health concern among older adults. Reported frailty prevalence in older populations ranges from 27 to 45%, with higher rates observed in women aged 65 and older compared to men. Additionally, frailty prevalence is notably higher among hospitalized older patients. These findings underline the profound individual and societal implications of frailty in Turkey. However, there is a need for more extensive, nationally representative studies to better understand the prevalence and impact of frailty in this population [7,8,9].

Diagnosing and assessing frailty is complex due to the multifactorial influences on its progression [10,11,12]. As a result, various frailty scales have been developed to provide a more objective assessment method. Prominent scales such as the Study of Osteoporotic Fractures (SOF) frailty scale, Edmonton Frail Scale (EFS), and FRAIL frailty scale are instrumental in diagnosing, treating, and monitoring older individuals who are frail or at risk of frailty [10, 13]. With aging, the endocrine system undergoes substantial changes, which significantly affect older adults [1, 10, 14]. Understanding the factors that influence the progression of frailty remains an area of ongoing research [15].

While studies have investigated the relationship between frailty and various conditions in the geriatric population—such as cardiovascular diseases, chronic kidney disease, and cancer—research on the association between thyroid dysfunction and frailty is limited [16,17,18]. Due to the widespread receptor distribution of thyroid hormones and their effects on multiple systems, thyroid hormone levels are hypothesized to influence frailty [18].

This study aimed to evaluate the effects of different thyroid function states—including overt and subclinical hypothyroidism, euthyroidism, and overt and subclinical hyperthyroidism—and hormone levels such as thyroid-stimulating hormone (TSH), free thyroxine (fT4), and free triiodothyronine (fT3), as well as thyroid peroxidase antibody (anti-TPO) positivity, on frailty in patients aged 65 and older. Furthermore, the study examined the relationships between frailty and comorbidities, sociodemographic factors, and optimal TSH levels for older adults using the SOF, EFS, and FRAIL scales. Subgroup analyses were also performed to explore age-specific variations in the relationships between thyroid function parameters and frailty.

Study population and method

The study population comprised patients aged 65 and older who visited the Internal Medicine, Endocrinology, and Geriatrics outpatient clinics at Ankara Training and Research Hospital between December 2019 and December 2020. This cross-sectional observational study was designed to evaluate the association between thyroid functions and frailty in geriatric patients using specific frailty assessment tools, namely the SOF, EFS, and FRAIL scales.

All research was conducted in accordance with the Helsinki Declaration and Good Clinical Practice (GCP) guidelines. Informed consent was obtained from all participants after they were thoroughly informed about the details of the study. The study was ethically approved by the Clinical Research Ethics Committee of the Health Sciences University Ankara Training and Research Hospital on December 5, 2019 (Committee No. E-19; Study No. 151).

A total of 220 patients were included in the study. These included individuals with overt or subclinical hypothyroidism, individuals with overt or subclinical hyperthyroidism who were receiving medical treatment, and treatment-naive euthyroid patients. Patients were excluded if they were under 65 years of age, unwilling to participate, immobile, had cognitive disorders, were taking amiodarone or beta-blockers, had language barriers, or did not meet the inclusion criteria.

After providing consent, patients underwent a thorough evaluation in the outpatient clinics. Their medical history, medication use, comorbid conditions (e.g., diabetes, hypertension), physical examination findings, and demographic data were recorded. Urinary incontinence was assessed using a three-question test (3 Incontinence Questions) or by evaluating the use of antimuscarinic or β3 agonist medications [19]. Comprehensive geriatric assessments, including the tools described in the Assessment Tools section [20], were conducted. Biochemical tests for thyroid function (TSH, fT3, fT4, and anti-TPO) were also performed.

Assessment tools

Various validated tools were utilized to comprehensively assess the general health and functional status of the participants. Activities of Daily Living (ADL) were evaluated using the Katz Index of Independence in Activities of Daily Living (score range: 0–6, where higher scores indicate greater independence) and the Lawton-Brody Instrumental Activities of Daily Living (IADL) Scale (score range: 0–8, where higher scores reflect greater functional autonomy) [21, 22]. Nutritional status was assessed using the Mini Nutritional Assessment-Short Form (MNA-SF), with scores interpreted as follows: 0–7 indicating malnutrition, 8–11 indicating a risk of malnutrition, and 12–14 indicating normal nutritional status [23]. Body Mass Index (BMI) was calculated using height and weight measurements obtained according to World Health Organization (WHO) protocols to ensure consistency [24, 25]. Depression symptoms were evaluated using the Geriatric Depression Scale-Short Form (GDS-SF), with scores interpreted as follows: 0–4 indicating no depression, 5–9 indicating mild depression, and 10–15 indicating severe depression [26], administered in a face-to-face interview format. Balance and mobility were measured using the Timed Up and Go (TUG) test, where completion times below 10 s indicated normal mobility, 10–20 s suggested mild limitations, and times above 20 s reflected significant mobility impairments [27]. Cognitive status was assessed using the Mini-Mental State Examination (MMSE), where scores of 24–30 indicated normal cognitive function, 18–23 indicated mild cognitive impairment, and 0–17 indicated severe cognitive impairment [27, 28].

Assessment of comorbidities using the modified charlson comorbidity index

The comorbidity burden of the participants was evaluated using the Modified Charlson Comorbidity Index (mCCI). The mCCI was calculated by assigning one point for every decade of age above 50 years. Additional points were allocated for specific comorbid conditions, including myocardial infarction, heart failure, chronic obstructive pulmonary disease, stroke, chronic kidney disease, diabetes (complicated or uncomplicated), and malignancies, based on their clinical significance [29].

Assessment of frailty

Frailty was assessed using three validated scales: the SOF, EFS, and FRAIL scales. The SOF scale is a rapid diagnostic tool consisting of three questions evaluating over 5% weight loss in the last year, the ability to rise from a chair five times without using arms, and feeling full of energy. Scores range from 0 to 3, where 0 indicates robustness, 1 suggests vulnerability, and 2–3 indicates frailty [30].

The EFS assesses multiple domains, including functional independence, nutrition, incontinence, polypharmacy, mood, social support, emotional state, quality of life, and two performance-based tasks: the clock drawing test for cognitive assessment and the timed up-and-go (TUG) test for mobility and balance [31].

The FRAIL scale evaluates fatigue, mobility, resilience, weight loss, and comorbidities. Each response is scored as 0 or 1, with total scores of 0, 1–2, and > 2 indicating non-frailty, pre-frailty, and frailty, respectively [32].

Assessment of thyroid function

Thyroid function tests were conducted using standard laboratory protocols. Blood samples for TSH, fT3, fT4, and anti-TPO were collected in 5 ml BD vacutainer plastic SST gel tubes. Samples were centrifuged (4000 rpm for 5 min) to obtain plasma, which was stored at −80 °C until analysis. Measurements were performed using chemiluminescence on a Roche Cobas 6000 Immunoassay System at the Biochemistry Laboratory of Ankara Training and Research Hospital. The following laboratory reference ranges were used: TSH, 0.5–6 mIU/L; fT4, 0.93–1.6 ng/dL; fT3, 2.5–4.40 pg/mL; and anti-TPO < 34 IU/mL. Optimal TSH values for older adults were determined based on recommendations from the literature [33,34,35,36]. Blood samples were collected by trained nursing staff between 08:00 and 11:00 am, with 99% of participants fasting at the time of collection. All procedures adhered to institutional ethical guidelines to ensure participant safety and comfort.

Statistical analysis

The required sample size was calculated using PASS version 11 software. The logistic regression formula proposed by Hsieh et al. was applied, considering a significance level (α) of 0.05, a power (1–β) of 0.80, an odds ratio of 1.5, and a frailty prevalence of approximately 40% in similar populations. A minimum of 210 participants was deemed necessary, and 220 participants were included to account for potential missing data or dropouts [37].

Data analysis was performed using IBM SPSS Statistics for Windows, version 15.0 (IBM Corp., Armonk, NY, USA). Continuous variables were evaluated for normality using the Kolmogorov–Smirnov test, histograms, detrended plots, and measures of central tendency (mean, median, standard deviation). Categorical variables were presented as numbers and percentages, while non-normally distributed continuous variables were summarized as medians [minimum (min) – maximum (max)].

For bivariate comparisons, categorical variables were analyzed using the chi-square or Fisher’s exact test, while ordinal and non-normally distributed continuous variables were compared using the Jonckheere–Terpstra test.

Logistic regression analysis was conducted to identify independent predictors of frailty. Two models were constructed: the first included variables with p < 0.05 in bivariate analyses, and the second incorporated additional variables with p < 0.2 to minimize the risk of type I errors. Model fit was assessed using the Hosmer–Lemeshow test, and multicollinearity was evaluated through a correlation matrix.

Odds ratios (OR) were used to quantify the strength of associations, with OR > 1 indicating a positive association and OR < 1 indicating a protective effect. The 95% confidence intervals (CI) represent the range within which the true OR is expected to lie with 95% certainty. Statistical significance was inferred if the CI did not include 1 and the p-value was < 0.05.

For multivariate analyses, frailty test results were dichotomized into frail and non-frail, and TSH levels were categorized as normal (0.5–6 mIU/L) or abnormal (< 0.5 mIU/L or > 6 mIU/L).

Results

A total of 220 patients were included in the study, of whom 68.2% (n = 150) were female. The median age of the study population was 73 years (min: 65, max: 96). Most of the participants (71.4%, n = 157) had a primary school education or less, and 60.5% (n = 133) were living with their spouses. The most common comorbidities were hypertension (64.1%, n = 141) and diabetes mellitus (46.4%, n = 102). To assess the overall comorbidity burden, the mCCI was calculated, yielding a median score of 4 (min: 1, max: 10). Symptoms of urinary incontinence were reported by 42.3% (n = 93) of the participants. Regarding thyroid function parameters, the median TSH level was 1.84 mIU/L (min: 0.01, max: 107), the median fT4 level was 1.21 ng/dL (min: 0.01, max: 7.77), and the median fT3 level was 2.67 pg/mL (min: 0.88, max: 21.57). Anti-TPO positivity was observed in 28.6% (n = 63) of the participants. Thyroid medication was being used by 37.3% (n = 82) of patients, 5.0% (n = 11) had a history of radioactive iodine (RAI) treatment, and 11.4% (n = 25) had undergone thyroid surgery (Table 1).

Table 1 Demographic Data of the Study Group (n = 220)

Anthropometric and geriatric assessment results further highlighted the participants’ characteristics. The median weight was 72.1 kg (min: 40.9, max: 122.5), and the median BMI was 26.5 kg/m² (min: 18.2, max: 37.8). Functional assessments revealed a median ADL score of 5 (min: 2, max: 6) and a median IADL score of 6 (min: 3, max: 8), indicating varying levels of independence. Nutritional evaluation using the MNA-SF showed a median score of 11 (min: 5, max: 14), suggesting that some participants were at risk of malnutrition. The median GDS-SF score was 3 (min: 0, max: 9), indicating depressive symptoms in certain individuals, while cognitive assessment using the MMSE revealed a median score of 25 (min: 20, max: 30). Mobility, as measured by the TUG test, showed a median time of 12.5 s (min: 7.4, max: 20.0), reflecting varying degrees of functional mobility within the population (Table 1).

Frailty prevalence according to the scales was as follows: 44.5% (n = 98) of patients were classified as frail according to the SOF scale, 30.4% (n = 67) according to the EFS scale, and 36.4% (n = 80) according to the FRAIL scale (Table 2). The relationship between TSH categories and frailty scale scores was statistically significant. Patients were grouped into five TSH categories: Category 1 (TSH < 0.50 mIU/L), Category 2 (TSH 0.50–3 mIU/L), Category 3 (TSH 3–6 mIU/L), Category 4 (TSH 6–10 mIU/L), and Category 5 (TSH > 10 mIU/L). According to the SOF scale, frailty prevalence was 64.6% (n = 31) in Category 1, 28.4% (n = 29) in Category 2, 30.3% (n = 10) in Category 3, 87.5% (n = 14) in Category 4, and 66.7% (n = 14) in Category 5. According to the FRAIL scale, frailty prevalence was 52.1% (n = 25) in Category 1, 20.6% (n = 21) in Category 2, 36.4% (n = 12) in Category 3, 43.8% (n = 7) in Category 4, and 71.4% (n = 15) in Category 5. According to the EFS scale, frailty prevalence was 41.7% (n = 20) in Category 1, 23.5% (n = 24) in Category 2, 24.3% (n = 8) in Category 3, 31.3% (n = 5) in Category 4, and 47.6% (n = 10) in Category 5. The observed differences in frailty prevalence across TSH categories were statistically significant for all three scales (SOF: p < 0.001; FRAIL: p < 0.001; EFS: p < 0.001) (Table 3).

Table 2 Distribution of patients according to frailty tests

Table 3 Parameters identified as significantly different in univariate analysis among TSH categories

Diagnostic concordance among the frailty scales was evaluated using the kappa test. The concordance between EFS and FRAIL was moderate (kappa = 0.53, p < 0.001), between EFS and SOF was low to moderate (kappa = 0.45, p < 0.001), and between FRAIL and SOF was moderate (kappa = 0.51, p < 0.001).

Significant differences were observed among frailty categories in thyroid function parameters and clinical characteristics. For the EFS categories, significant differences were found in low fT3 levels (p < 0.001), low fT3/fT4 ratios (p = 0.010), advanced age (p < 0.001), low educational attainment (p < 0.001), the presence of urinary incontinence (p < 0.001), the presence of major depressive disorder (p < 0.001), stroke history (p = 0.001), and the presence of Alzheimer’s disease (p = 0.001). For the FRAIL categories, significant differences were observed in low fT3 levels (p = 0.002), advanced age (p < 0.001), low educational attainment (p = 0.006), alcohol use (p = 0.020), duration of thyroid medication use (p = 0.008), the presence of urinary incontinence (p < 0.001), the presence of Alzheimer’s disease (p = 0.008), and stroke history (p = 0.001). For the SOF frailty scale categories, significant differences were noted in low fT3 levels (p = 0.003), low fT3/fT4 ratios (p = 0.030), advanced age (p = 0.006), low educational attainment (p = 0.003), the presence of hypothyroidism (p = 0.012), the presence of Alzheimer’s disease (p = 0.002), duration of thyroid medication use (p < 0.001), thyroid surgery history (p = 0.043), and the presence of urinary incontinence (p < 0.001) (Table 4).

Table 4 Parameters Identified as Significantly Different in the Univariate Analysis Among the Edmonton, FRAIL and SOF Frailty Scale Categories

Parameters demonstrating significant differences among TSH categories were included in the multivariate logistic regression analysis. The presence of frailty according to the SOF frailty scale [OR: 7.71; 95% CI = (3.66–16.23); p < 0.001], positivity for anti-TPO [OR: 5.01; 95% CI = (2.29–10.97); p < 0.001], duration of thyroid medication use [OR: 1.01; 95% CI = (1.03–1.17); p = 0.004], and the presence of Graves’ disease [OR: 22.17; 95% CI = (4.13–118.98); p < 0.001] were found to be significantly different. Parameters that differed significantly among the EFS categories were also included in the multivariate logistic regression analysis. Among these were having an education level above primary school [OR: 0.27; 95% CI = (0.10–0.77); p = 0.014], a history of cancer [OR: 5.49; 95% CI = (1.57–19.19); p = 0.008], a diagnosis of stroke [OR: 8.83; 95% CI = (2.42–32.18); p = 0.001], the presence of Alzheimer’s disease [OR: 10.15; 95% CI = (1.99–51.76); p = 0.005], the presence of major depressive disorder [OR: 3.19; 95% CI = (1.31–7.74); p = 0.010], and the presence of urinary incontinence [OR: 3.22; 95% CI = (1.55–6.70); p = 0.002]. Parameters showing significant differences among the FRAIL categories were also included in the multivariate logistic regression analysis. These factors included having an education level above primary school [OR: 0.41; 95% CI = (0.18–0.95); p = 0.037], a history of alcohol use [OR: 0.14; 95% CI = (0.03–0.76); p = 0.022], a diagnosis of heart failure [OR: 5.08; 95% CI = (1.36–19.00); p = 0.016], a diagnosis of stroke [OR: 17.13; 95% CI = (3.68–79.73); p < 0.001], the presence of Alzheimer’s disease [OR: 4.95; 95% CI = (1.39–17.60); p = 0.013], the presence of atrial fibrillation [OR: 5.10; 95% CI = (1.01–25.65); p = 0.048], and current use of thyroid medication [OR: 2.61; 95% CI = (1.36–5.02); p = 0.004] (Table 5).

Table 5 Results of Multivariate Logistic Regression Analysis for Abnormal TSH Level Categories According to the Edmonton, FRAIL and SOF Frailty Scales

In the subgroup analysis, significant differences were observed between individuals aged ≥ 80 years and those aged < 80 years regarding demographics, thyroid function parameters, and frailty measures. Among individuals aged ≥ 80 years, 37 (67.3%) were female, and 18 (32.7%) were male, whereas in the < 80 years group, 113 (68.5%) were female. According to the SOF frailty scale, 25 individuals (45.5%) aged ≥ 80 years were classified as frail, compared to 68 individuals (41.2%) in the < 80 years group, although this difference was not statistically significant (p = 0.085). However, the FRAIL scale revealed a significant difference, with 25 individuals (45.5%) aged ≥ 80 years classified as frail compared to 50 individuals (30.3%) in the < 80 years group (p = 0.001). Similarly, the EFS frailty scale identified 28 individuals (50.9%) aged ≥ 80 years as frail, compared to 39 individuals (23.6%) in the < 80 years group, showing a strong association with age ≥ 80 years (p < 0.001). Among thyroid function parameters, individuals aged ≥ 80 years had a median TSH level of 1.58 mIU/L (min: 0.01, max: 16.54), a median fT4 level of 1.18 ng/dL (min: 0.53, max: 2.29), and a median fT3 level of 2.44 pg/mL (min: 0.88, max: 5.07), with anti-TPO positivity observed in 19 individuals (34.5%). In contrast, the < 80 years group had a median TSH level of 1.91 mIU/L (min: 0.01, max: 107.00), a median fT4 level of 1.22 ng/dL (min: 0.10, max: 7.77), and a median fT3 level of 2.71 pg/mL (min: 0.88, max: 21.57), with anti-TPO positivity in 44 individuals (26.7%). These differences in anti-TPO positivity were not statistically significant (p = 0.263), nor were the gender distributions among individuals aged ≥ 80 years (p = 0.867). A lower fT3/fT4 ratio was significantly associated with frailty in individuals aged ≥ 80 years according to the FRAIL (p = 0.038), SOF (p = 0.015), and EFS (p = 0.020) scales. However, neither fT3 nor fT4 levels were significantly associated with frailty in this age group according to the FRAIL (p = 0.068 and p = 0.879), SOF (p = 0.056 and p = 0.478), or EFS (p = 0.057 and p = 0.973) scales. In individuals aged < 80 years, TSH levels in the lower category were significantly associated with frailty according to the FRAIL scale (p = 0.038), while no significant associations were found for SOF (p = 0.354) or EFS (p = 0.539). Similarly, fT3 and fT4 levels were not significantly associated with frailty in individuals aged < 80 years across the FRAIL (p = 0.056 and p = 0.527), SOF (p = 0.098 and p = 0.726), or EFS (p = 0.176 and p = 0.468) scales. These findings highlight the higher prevalence of frailty in individuals aged ≥ 80 years, particularly as assessed by the FRAIL and EFS scales, underscoring the need for tailored assessments and interventions in this vulnerable population.

Discussion

Frailty is a multidimensional and dynamic syndrome that increases morbidity and mortality risks in older adults. With the rise in life expectancy, the prevalence and incidence of frailty are steadily increasing in the population [28, 38]. Our study is a novel investigation that comprehensively examines the relationship between thyroid dysfunction and frailty using the FRAIL, SOF, and EFS scales. Our findings demonstrate that low fT3 levels and a low fT3/fT4 ratio are significantly associated with increased frailty. In subgroup analyses, a lower fT3/fT4 ratio was significantly associated with frailty in individuals aged ≥ 80 years across all scales (FRAIL, SOF, and EFS), while in individuals aged < 80 years, lower TSH levels were significantly associated with frailty only on the FRAIL scale. These results highlight the importance of age-specific evaluations of thyroid function in frailty assessment. The SOF and EFS scales more effectively revealed the impact of thyroid dysfunction on frailty, while the FRAIL scale showed a significant association only between low fT3 levels and frailty. These discrepancies may stem from the heterogeneity in the scales’ capacities to evaluate different subdimensions of frailty. Our findings underscore the importance of considering thyroid function in the assessment of frailty.

Our subgroup analysis revealed that the effects of thyroid function parameters on frailty vary significantly with age. In individuals aged ≥ 80 years, a low fT3/fT4 ratio was significantly associated with frailty across all scales, while in individuals aged < 80 years, low TSH levels were significantly associated with frailty according to the FRAIL scale. Studies in the literature examining the relationship between TSH levels, frailty, and mortality support our findings [34, 35, 39,40,41,42,43]. For example, researchers such as Zhai and Wang have suggested that TSH reference ranges may broaden with advancing age and that this should be considered in clinical evaluations and treatment planning [34, 39]. A retrospective study identified a relationship between subclinical hypothyroidism and increased mortality in the geriatric population and demonstrated that maintaining TSH levels within the range of 5–10 mIU/L reduces the risk of frailty and mortality in older individuals [44,45,46,47,48,49]. In alignment with these findings, our study highlights that age-specific TSH reference ranges, particularly in individuals aged ≥ 80 years, could serve as a crucial guide for managing frailty and providing personalized treatment for thyroid dysfunctions. Furthermore, age-specific assessments not only improve frailty management but also contribute to strategies aimed at reducing morbidity and mortality in older adults. These findings emphasize the multidimensional effects of thyroid function during aging and underscore the necessity of individualized clinical approaches.

The studies by Veronese and Virgini also support our findings. Veronese et al. conducted cross-sectional and prospective cohort studies that evaluated the relationship between TSH levels and frailty using the parameters of the Fried Frailty Index, demonstrating that abnormal TSH levels are independently associated with frailty. However, unlike Veronese's study, our research utilized the FRAIL, SOF, and EFS scales, providing a broader perspective on this relationship [50]. Similarly, the study by Virgini et al., which examined the impact of subclinical thyroid dysfunction on frailty, also identified an independent association between abnormal TSH levels and frailty [51]. In line with these findings, our study underscores the necessity of considering thyroid function as a critical component in the assessment of frailty.

Compared to the literature, our findings align with studies by Pasqualetti, Bertoli, and Valentini, which emphasize the critical role of fT3 levels and the fT3/fT4 ratio in the assessment of frailty [52,53,54]. Pasqualetti and Bertoli reported that low fT3 levels and a reduced fT3/fT4 ratio are associated with increased frailty, prolonged hospital stays, and higher mortality [52, 53]. Valentini further highlighted the significant contribution of low fT3 levels and the fT3/fT4 ratio to the development of frailty [54]. Additionally, Bano’s U-shaped relationship hypothesis, which suggests that both low and high thyroid function increase the risk of frailty, partially aligns with our findings. Bano specifically noted that both low and high fT4 levels may elevate the risk of frailty, underscoring the importance of maintaining thyroid function within an optimal range [55]. However, our study did not find a significant association between fT4 levels and frailty. In the literature, Yeap has suggested that elevated fT4 levels contribute to an increased risk of frailty, but our findings did not confirm this association [56]. Moreover, a longitudinal study by Correia et al. in euthyroid older adults demonstrated that higher fT4 levels significantly increased the risk of functional decline, whereas lower fT4 levels had a protective effect [57]. Similarly, van den Beld et al. reported that thyroid hormone concentrations influence physical function and mortality in older men, further supporting the complexity of these relationships [58]. These discrepancies may be attributed to differences in population characteristics, methodological variations, and the specific scales used [43, 59]. While our findings support the critical role of fT3 levels and the fT3/fT4 ratio in frailty, the complex associations between fT4 levels and frailty in the literature highlight the need for further research to clarify these relationships.

The relationship between chronic diseases and frailty has been widely emphasized in the literature. Conditions such as Alzheimer’s disease, a history of stroke, urinary incontinence, cardiovascular diseases, and depression are strongly associated with frailty and mortality. Our study corroborates these associations and highlights the critical role of thyroid function in understanding the multidimensional effects of frailty [41, 60,61,62,63,64]. Additionally, the literature reports a strong link between hyperthyroidism and sarcopenia. Although our study did not directly evaluate sarcopenia, it addressed the broader impact of thyroid dysfunction on overall health through its relationship with frailty [65].

The cross-sectional design of this study limits the ability to establish causal relationships between thyroid function and frailty. Additionally, the single-center nature of the data and the relatively small sample size, despite being meticulously calculated using PASS software, may reduce the generalizability of the findings. Although the mCCI and CGA were calculated, the relationships between frailty and the total mCCI score or CGA parameters were not specifically evaluated. Instead, the associations between individual comorbidities and frailty were analyzed separately. Furthermore, our study did not comprehensively analyze other geriatric syndromes such as malnutrition, falls, sarcopenia, muscle strength, polypharmacy, and total medication use, as the primary focus was to investigate the relationship between thyroid function and frailty. Additionally, thyroid function tests in this study were based on a single measurement, which may not account for short-term physiological or pathological fluctuations in hormone levels. Future studies should consider multiple measurements over time to provide a more comprehensive assessment of thyroid function and its relationship with frailty. The exclusion of other inflammatory and hormonal parameters, which are known to influence frailty, may also limit the scope of our findings in capturing the broader biological mechanisms underlying frailty [66]. Future studies integrating these variables and utilizing longitudinal designs are essential to provide a more comprehensive understanding of frailty in older adults. Despite these limitations, our study has several notable strengths. First, frailty was comprehensively assessed using three validated frailty scales (FRAIL, SOF, and EFS), which enabled a detailed evaluation of the biological, psychosocial, and clinical dimensions of frailty. This multi-dimensional approach enhanced the methodological robustness of the study and allowed us to explore frailty from a broader perspective. Second, our findings identified fT3 levels and the fT3/fT4 ratio as potential biomarkers for frailty, highlighting their potential utility in frailty assessments. These findings contribute to a growing body of evidence supporting the integration of thyroid function parameters into frailty screening protocols. Third, the evaluation of age-specific thyroid parameters and the identification of distinct associations between abnormal TSH levels and frailty provided critical insights into the nuanced relationship between thyroid dysfunction and frailty. This age-specific analysis addressed an important gap in the literature and has implications for developing individualized clinical approaches tailored to older adults. Finally, by focusing on thyroid dysfunction as a modifiable risk factor, this study emphasizes the importance of early identification and targeted interventions, which could ultimately improve outcomes and quality of life for older adults at risk of frailty.

This study provides a comprehensive evaluation of the relationship between thyroid function parameters and frailty using multiple frailty scales, addressing a critical gap in the literature. Our findings suggest that fT3 levels and the fT3/fT4 ratio are significant biomarkers for frailty and underline the necessity of incorporating thyroid function assessments into frailty management protocols, particularly in older adults. Subgroup analyses revealed that in individuals aged ≥ 80 years, a lower fT3/fT4 ratio was significantly associated with frailty across all frailty scales, while in individuals aged < 80 years, lower TSH levels were significantly associated with frailty as assessed by the FRAIL scale. These results reinforce the importance of integrating thyroid function tests into routine frailty screenings, facilitating earlier detection and personalized interventions. Furthermore, the study highlights the potential for age-specific thyroid function reference ranges to enhance clinical decision-making and management strategies. Future research with larger, more diverse populations and longitudinal designs is essential to confirm these associations and establish causality, ultimately contributing to the development of targeted strategies to improve outcomes for older adults.