Screening performance of a lifestyle-based questionnaire score for vitamin D deficiency and related functional and physiological characteristics in older women

Abstract

[Purpose]

This study aimed to determine if a lifestyle-based questionnaire on vitamin D can effectively screen for vitamin D deficiency in community-dwelling older women and whether such deficiency correlates with negative functional and physiological profiles.

[Methods]

Eighty older women completed a vitamin D questionnaire that assessed their dietary intake, sunlight exposure behaviors, and awareness of vitamin D. They also provided fasting blood samples, with serum 25-hydroxyvitamin D levels used to classify participants as deficient or non-deficient. Body composition and performance on the Senior Fitness Test (SFT) were evaluated, alongside measurements of circulating myostatin, follistatin, tumor necrosis factor-alpha, and interleukin-1 beta. Differences between groups and correlations among questionnaire scores, serum vitamin D levels, and assessed outcomes were analyzed.

[Results]

The group with vitamin D deficiency exhibited lower muscle mass, poorer muscle function, and reduced cardiorespiratory endurance compared to the non-deficient group. Additionally, the deficient group had higher levels of myostatin, lower levels of follistatin, a diminished follistatin-to-myostatin ratio, and elevated inflammatory cytokines. Positive associations were found between serum vitamin D levels and questionnaire total scores with muscle mass and functional measures, while inverse relationships were noted with myostatin and inflammatory markers.

[Conclusion]

The questionnaire-based approach proved effective for identifying older women at risk for vitamin D deficiency. Combined with serum vitamin D status, it was associated with functional impairment and an unfavorable myokine-inflammatory profile. This supports the implementation of community screening to identify high-risk individuals for further testing and targeted lifestyle interventions.

INTRODUCTION

Global population aging is accelerating, leading to an increasing number of older adults. Due to their longer life expectancy, women represent a larger proportion of this demographic [1]. This shift emphasizes the importance of preventing and early identifying late-life health risks, particularly in community settings, which require practical indicators to quickly assess nutritional status and vulnerability to functional decline [2]. In this context, vitamin D (VD) status is crucial in community-based care, as it is closely related to nutrition and regular physical activity patterns [2].

Older women are particularly at risk for VD deficiency. After menopause, declines in sex hormones often lead to reduced vitality and mood disturbances, which can decrease physical activity and outdoor engagement. These behavioral changes result in lower sunlight exposure and, consequently, reduced cutaneous VD synthesis [3]. Additionally, age-related declines in the skin’s capacity to produce VD further increase this risk, making older women a significant high-risk group for deficiency [4]. In Korea, high prevalence rates of VD deficiency among older women have been consistently reported, underscoring the need for systematic screening and management strategies to address functional decline and chronic disease burden [5].

Traditionally, VD has been primarily associated with bone metabolism [6]. However, emerging evidence suggests that vitamin D receptor (VDR) expression in skeletal muscle may connect VD status to the preservation of muscle mass and function [7]. Additionally, VD is implicated in immune and inflammatory regulation, indicating that a deficiency could worsen a chronic pro-inflammatory state [8]. Age-related declines in muscle mass and function increase the risk of falls, hinder independent daily activities, and diminish quality of life. Therefore, understanding how VD deficiency relates to muscle-related functional fitness and inflammatory physiological profiles holds significant clinical and public health relevance [9].

Serum 25-hydroxyvitamin D [25(OH)D] is the established biomarker for assessing VD status [10]. However, in community settings, the widespread use of blood testing is often limited by cost, accessibility, and the practicality of repeated measurements [11]. To overcome these challenges, questionnaire-based screening tools have been suggested to evaluate VD intake, sunlight exposure behaviors, and related awareness, facilitating the initial identification of individuals at higher risk before confirmatory testing [12]. Nonetheless, studies in Korean community-dwelling older women that simultaneously link questionnaire scores to serum 25(OH)D levels and characterize deficiency-related profiles—including body composition, functional fitness, muscle metabolism-related myokines, and inflammatory biomarkers—are still scarce. Previous research in postmenopausal women has often concentrated on bone health or specific outcomes [13-16], highlighting the necessity for a comprehensive framework that integrates field screening (questionnaire), functional status (fitness), and physiological vulnerability (myokines/inflammation).

This study aimed to determine whether a lifestyle-based questionnaire can effectively screen for vitamin D (VD) deficiency in community-dwelling older women prior to blood testing. Specifically, we sought to: (1) examine the associations among the total score of the VD questionnaire, serum 25(OH)D levels, body composition, functional fitness, and circulating myokines and inflammatory markers; (2) evaluate the discriminative performance of the questionnaire score in identifying VD deficiency; and (3) compare the functional and physiological profiles of VD-deficient and non-deficient groups to better characterize high-risk individuals and develop prioritization strategies for community-based management.

METHODS

Study Design and Ethical Approval

This study utilized a cross-sectional observational design to evaluate the associations among VD status, body composition, functional fitness, circulating myokines, inflammatory markers, and VD questionnaire scores in community-dwelling older women. All procedures were approved by the Institutional Review Board of Kangwon National University (KWNUIRB-2024-08-004-003) and conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Before participation, all participants received a detailed explanation of the study’s purpose and procedures, potential risks and benefits, and data protection measures, and they provided written informed consent.

Participants

A total of 80 community-dwelling older women were enrolled in the study. Participants were classified into two groups based on serum 25(OH)D concentration: a non-deficient group (NVDD, n = 46) and a deficient group (VDD, n = 34), with vitamin D deficiency defined as 25(OH)D < 20 ng/mL [17]. Individuals were excluded if they had medical conditions that could prevent safe participation in physical fitness testing, such as severe musculoskeletal disorders, cardiovascular or cardiopulmonary disease, stage ≥ 3 hypertension, or severe cognitive impairment. Additionally, individuals who had received a prescription for vitamin D or used vitamin D supplementation within the past 3 months, or who had other conditions that could potentially influence study outcomes, were also excluded. All participants voluntarily completed the questionnaires and assessments. Participant characteristics and body composition data are presented in Table 1.

Table 1.

Participant characteristics and body composition by vitamin D deficiency status

Variables	NVDD (n = 46)	VDD (n = 34)	t	p
Age (years)	72.78 ± 5.19	74.06 ± 4.51	-1.149	0.254
25(OH)D (ng/mL)	28.73 ± 5.00	18.03 ± 1.82	11.896	< .001
VD questionnaire total score (points)	44.96 ± 17.07	19.74 ± 13.96	7.046	< .001

Data are presented as mean ± SD. Participants were classified as non-vitamin D deficient (NVDD) or vitamin D deficient (VDD) based on serum 25-hydroxyvitamin D [25(OH)D] (deficiency: < 20 ng/mL). Between-group differences were assessed using an independent two-tailed t-test. Abbreviations: 25(OH)D, 25-hydroxyvitamin D.

Body Composition

Body composition was assessed using a multi-frequency bioelectrical impedance analyzer (InBody 720; Biospace, Seoul, Republic of Korea). Participants were measured barefoot and dressed in light clothing after removing shoes, socks, and heavy accessories. Body weight, fat mass, and skeletal muscle mass were recorded to the nearest 0.1 kg. Percent body fat was estimated based on impedance values obtained at multiple frequencies, following the manufacturer’s algorithm. Body mass index (BMI) was calculated by dividing weight (kg) by height squared (m²).

Senior Fitness Test

Functional fitness was evaluated using the Senior Fitness Test (SFT) battery, which includes assessments of muscular strength, muscular endurance, flexibility, balance, coordination, and cardiorespiratory endurance. Upper-body strength was measured with a digital handgrip dynamometer (BSHG; Biospace, Korea). Grip strength was recorded twice for each hand, and the highest value was retained; overall grip strength was calculated as the average of the peak values for the right and left hands. Relative grip strength was determined using the formula: [((peak right + peak left) / 2) × 100 / body weight (kg))].

Lower-body muscular endurance was assessed using the 30-s chair stand test on a 40-cm chair, counting the number of complete stands performed in 30 s. Flexibility was evaluated with a sit-and-reach test using a digital forward flexion meter (BS-FF; Biospace, Korea), with the best result from two trials recorded. Dynamic balance and mobility were assessed through the Timed Up and Go test, which measures the time taken to rise from a chair, walk 3 m, turn around a cone, return, and sit down. Coordination was evaluated using the Figure-of-8 Walk test, in which participants navigated a standardized figure-eight course. Cardiorespiratory endurance was measured with the 2-min step test, recording the total number of steps completed in place during that time. For strength and flexibility tests, two trials were conducted, and the best performance was used for analysis. All SFT assessments were administered by a certified health and exercise specialist.

Table 2.

Factor loadings and reliability of the vitamin D questionnaire

Domain	Item	Factor 1	Factor 2	Factor 3
VD awareness	Awareness 6	0.779	0.108	-0.012
	Awareness 7	0.762	0.121	0.066
	Awareness 3	0.710	0.339	0.226
	Awareness 2	0.702	0.285	0.173
	Awareness 5	0.613	0.040	0.020
	Awareness 4	0.534	0.493	0.212
VD intake	Diet 1	0.116	0.901	0.095
	Diet 2	0.173	0.881	0.109
	Awareness 1	0.417	0.603	0.259
Sunlight exposure behavior	Sunlight 7	0.020	0.051	0.769
	Sunlight 5	0.153	0.178	0.703
	Sunlight 1	0.060	0.083	0.541
Eigenvalue		3.090	2.460	1.597
Variance explained (%)		25.752	20.504	13.305
Cumulative variance (%)		25.752	46.256	59.561
Cronbach’s α		0.869	0.866	0.708
KMO=.804, Bartlett’s test of sphericity: χ2 = 2102.443, p < 0.001

Standardized factor loadings were derived from exploratory factor analysis (EFA) using principal component extraction with varimax rotation. Items loaded onto three factors: vitamin D awareness, vitamin D intake, and sunlight exposure behavior. Eigenvalues, variance explained (%), and cumulative variance (%) are reported for each factor. Internal consistency was assessed using Cronbach’s α. Sampling adequacy and factorability were evaluated using the Kaiser–Meyer–Olkin (KMO) measure and Bartlett’s test of sphericity.

Hematological Analysis

To standardize blood sampling and analysis, all procedures were conducted under identical conditions. Participants fasted for 12 h overnight after their evening meal, ensured adequate sleep, and avoided strenuous physical activity prior to sampling. Venous blood was collected at 08:00 from the antecubital vein. Samples were immediately centrifuged at 3,500 × g for 10 minutes at room temperature to separate the serum, which was then aliquoted and stored at −80°C until analysis. Serum 25(OH)D levels were quantified using a commercially available immunoassay kit (No. DE197; Demeditec Diagnostics, Kiel, Germany). Circulating myokines (myostatin and follistatin; DY788-05 and DY669; R&D Systems, Minneapolis, MN, USA) and inflammatory markers (tumor necrosis factor-α and interleukin-1β; DY210 and DY201; R&D Systems, Minneapolis, MN, USA) were measured using enzyme-linked immunosorbent assays (ELISA) following the manufacturers’ instructions. The follistatin-to-myostatin ratio (F/M ratio, %) was calculated as (follistatin / myostatin) × 100.

Vitamin D Questionnaire

A VD-related questionnaire was adapted from previous research to include items on sunlight exposure behaviors and VD intake [18]. The questionnaire consisted of three subdomains: VD awareness, VD intake, and sunlight exposure behavior. Each item was rated on a 5-point Likert scale, with a total score calculated (maximum of 60 points). To establish construct validity and reliability, exploratory factor analysis was conducted. Sampling adequacy and factorability were confirmed (KMO = 0.804; Bartlett’s test of sphericity: χ² = 2102.443, p < 0.001). Three factors were extracted, explaining 59.561% of the cumulative variance. Internal consistency was found to be acceptable to good (Cronbach’s α = 0.869 for awareness, 0.866 for intake, and 0.708 for sunlight exposure behavior).

Statistical Analysis

All statistical analyses were conducted using IBM SPSS Statistics (IBM Corp., Armonk, NY, USA). Data are presented as mean ± standard deviation (SD). The normality of the main variables was assessed using the Shapiro-Wilk test, and parametric procedures were applied as appropriate. Participants’ general characteristics and body composition were summarized using descriptive statistics. Linear associations among key variables were evaluated using Pearson’s correlation analysis.

To assess the discriminatory performance for VD deficiency, receiver operating characteristic (ROC) analyses were performed, using deficiency status (deficient = 1) as the reference outcome. Predictor variables included questionnaire scores, body composition and functional fitness indices, and blood biomarkers. The area under the curve (AUC), 95% confidence interval (CI), and p-values were reported. The optimal cutoff value was identified by maximizing the Youden index. For clarity and consistent ROC interpretation, variables indicating a higher likelihood of deficiency with lower values were reverse-coded prior to analysis.

Between-group differences based on vitamin D deficiency status were analyzed using independent-samples t-tests. Statistical significance was set at a two-tailed α level of 0.05.

RESULTS

Pearson’s Correlations Among Key Variables

Pearson’s correlation analysis revealed a positive association between the total VD questionnaire score and serum 25(OH)D (r = 0.664, p < 0.001) as well as skeletal muscle mass (r = 0.688, p < 0.001). The questionnaire score also showed positive correlations with relative handgrip strength (r = 0.827, p < 0.001), the 30-s chair stand test (r = 0.777, p < 0.001), and the 2-min step test (r = 0.544, p < 0.001). Serum 25(OH)D was positively correlated with skeletal muscle mass (r = 0.470, p < 0.001) and relative handgrip strength (r = 0.522, p < 0.001), and it exhibited smaller yet significant positive correlations with the 30-s chair stand test (r = 0.374, p < 0.001) and the 2-min step test (r = 0.224, p < 0.05).

Among myokines, myostatin was inversely correlated with the VD questionnaire score (r = −0.489, p < 0.001), serum 25(OH)D (r = −0.353, p < 0.01), skeletal muscle mass (r = −0.409, p < 0.001), and overall functional performance measures. In contrast, follistatin was positively associated with serum 25(OH)D (r = 0.291, p < 0.01) and skeletal muscle mass (r = 0.342, p < 0.01). The follistatin-to-myostatin (F/M) ratio was also positively correlated with the VD questionnaire score (r = 0.496, p < 0.001), serum 25(OH)D (r = 0.360, p < 0.01), and skeletal muscle mass (r = 0.530, p < 0.001). Regarding inflammatory markers, both tumor necrosis factor-alpha (TNF-α) and interleukin-1 beta (IL-1β) were inversely correlated with serum 25(OH)D (TNF-α: r = −0.413, p < 0.001; IL-1β: r = −0.390, p < 0.001) and with the VD questionnaire score (TNF-α: r = −0.364, p < 0.001; IL-1β: r = −0.312, p < 0.01).

Figure 1. Receiver Operating Characteristic (ROC) curves for predicting vitamin D deficiency.

ROC curves illustrate the ability of the vitamin D questionnaire total score (reversecoded), skeletal muscle mass (reverse-coded), interleukin- 1β, follistatin-to-myostatin ratio (F/M ratio, %, reversecoded), and tumor necrosis factor-α to discriminate vitamin D deficiency (serum 25-hydroxyvitamin D < 20 ng/mL). Variables for which lower values indicate a higher probability of deficiency were reverse-coded (_r) to maintain consistent directionality across predictors.

Table 3.

Pearson’s correlation coefficients among key study variables

Variables	VD score	25(OH)D	SMM	BF%	RGS	30s CST	2m ST	MSTN	FSTN	F/M Ratio	TNF-α	IL-1β
VD score	1
25(OH)D	.664***	1
SMM	.688***	.470***	1
BF%	-.160	.040	-.160	1
RGS	.827***	.522***	.562***	-.355**	1
30s CST	.777***	.374***	.505***	-.301**	.649***	1
2m ST	.544***	.224*	.392***	-.419***	.446***	.619***	1
MSTN	-.489***	-.353**	-.409***	.080	-.391***	-.386***	-.228*	1
FSTN	.265*	.291**	.342**	-.190	.170	.200	.284*	-.170	1
F/M Ratio	.496***	.360**	.530***	-.170	.355***	.405***	.369***	-.720***	.753***	1
TNF-α	-.364***	-.413***	-.295**	.120	-.269*	-.268*	-.363***	.234*	-.358**	-.388***	1
IL-1β	-.312**	-.390***	-.381***	-.090	-.130	-.190	-.060	.290**	-.120	-.273*	.170	1

Values are Pearson’s correlation coefficients (r) from two-tailed tests. Abbreviations: VD score, vitamin D questionnaire total score; 25(OH)D, serum 25-hydroxyvitamin D; SMM, skeletal muscle mass; BF%, body fat percentage; RGS, relative grip strength; 30s CST, 30-s chair stand test; 2m ST, 2-min step test; MSTN, myostatin; FSTN, follistatin; F/M ratio (%), follistatin-to-myostatin ratio calculated as (follistatin / myostatin) × 100; TNF-α, tumor necrosis factor-α; IL-1β, interleukin-1β. For functional fitness tests, higher values indicate better performance.

^*p < 0.05,

^**p < 0.01,

^***p < 0.001.

Table 4.

ROC analysis of key predictors for identifying vitamin D deficiency

Predictor	AUC	95% CI	p	Cut-off	Sensitivity	Specificity	Youden
VD score_r	0.858	0.775–0.940	< .001	38.50	0.853	0.761	0.614
SMM_r	0.852	0.768–0.936	< .001	26.74	0.794	0.804	0.598
IL-1β	0.843	0.757–0.928	< .001	1.50	0.912	0.696	0.607
F/M ratio_r	0.803	0.698–0.908	< .001	314.70	0.735	0.804	0.540
TNF-α	0.781	0.678–0.884	< .001	9.87	0.618	0.848	0.465
30s CST_r	0.760	0.652–0.868	< .001	24.50	0.853	0.609	0.462
MSTN	0.757	0.643–0.871	< .001	4.42	0.735	0.761	0.496
RGS_r	0.747	0.636–0.858	< .001	41.05	0.853	0.696	0.549
FSTN_r	0.745	0.629–0.861	< .001	12.95	0.647	0.826	0.473
2m ST_r	0.650	0.526–0.773	.018	85.34	0.588	0.696	0.284

Receiver operating characteristic (ROC) analysis was used to evaluate the ability of each predictor to discriminate vitamin D deficiency (serum 25-hydroxyvitamin D [25(OH)D] < 20 ng/mL; deficient = 1). Values are presented as area under the curve (AUC) with 95% confidence intervals (CI). Optimal cut-off values were determined using the Youden index (sensitivity + specificity − 1). Variables for which lower values indicate a higher likelihood of deficiency were reverse-coded and denoted with “_r” (higher values indicate a higher likelihood of deficiency). The follistatin-to-myostatin ratio (F/M ratio, %) was calculated as (follistatin / myostatin) × 100. P values test the null hypothesis of AUC = 0.50 (two-sided).

Key Predictors of Vitamin D Deficiency

Receiver operating characteristic (ROC) analyses identified the strongest discriminators of vitamin D deficiency as the reverse-coded vitamin D questionnaire total score and skeletal muscle mass (also reverse-coded), with areas under the curve (AUCs) of 0.858 (95% CI: 0.775-0.940, p < 0.001) and 0.852 (95% CI: 0.768-0.936, p < 0.001), respectively. Significant discriminatory performance was also observed for IL-1β (AUC = 0.843, 95% CI: 0.757-0.928, p < 0.001), the reverse-coded F/M ratio (AUC = 0.803, 95% CI: 0.698-0.908, p < 0.001), and TNF-α (AUC = 0.781, 95% CI: 0.678-0.884, p < 0.001). Optimal cut-off values, determined by maximizing the Youden index, were 38.50 points for the questionnaire score (sensitivity = 0.853, specificity = 0.761), 26.74 kg for skeletal muscle mass (sensitivity = 0.794, specificity = 0.804), 1.50 for IL-1β (sensitivity = 0.912, specificity = 0.696), 314.70 for the F/M ratio (sensitivity = 0.735, specificity = 0.804), and 9.87 for TNF-α (sensitivity = 0.618, specificity = 0.848).

Figure 2. Group differences in blood biomarkers by vitamin D deficiency status.

Serum 25-hydroxyvitamin D [25(OH)D], myostatin, follistatin, tumor necrosis factor-α (TNF-α), and interleukin-1β (IL-1β) are presented as mean ± SD (bar graphs). The follistatin-to-myostatin ratio (F/M ratio, %) was calculated as (follistatin / myostatin) × 100 and is presented as a box-and-whisker plot (median and interquartile range; whiskers indicate minimum to maximum). Between-group differences were assessed using an independent two-tailed t-test. **p < 0.01, ***p < 0.001.

Table 5.

Differences in body composition and SFT outcomes by vitamin D deficiency status

Variable	Groups	Means ± SD	t	p
Height (cm)	NVDD (n = 46)	155.18 ± 4.10	-0.398	0.692
	VDD (n = 34)	155.53 ± 3.67
Body weight (kg)	NVDD (n = 46)	58.67 ± 6.28	-1.401	0.165
	VDD (n = 34)	60.40 ± 4.08
Skeletal Muscle Mass (kg)	NVDD (n = 46)	24.16 ± 3.97	6.423	< .001
	VDD (n = 34)	18.80 ± 3.28
Body fat (%)	NVDD (n = 46)	34.28 ± 5.19	0.018	0.986
	VDD (n = 34)	34.26 ± 4.40
Body fat mass (kg)	NVDD (n = 46)	20.29 ± 4.76	-0.493	.623
	VDD (n = 34)	20.76 ± 3.46
BMI (kg/m2)	NVDD (n = 46)	24.37 ± 2.39	-1.233	0.221
	VDD (n = 34)	25.01 ± 2.13
Handgrip strength (kg)	NVDD (n = 46)	41.68 ± 10.03	4.191	< .001
	VDD (n = 34)	33.02 ± 7.77
30-s chair stand (reps)	NVDD (n = 46)	25.67 ± 5.19	4.462	< .001
	VDD (n = 34)	20.59 ± 4.82
2-min step test (reps)	NVDD (n = 46)	113.53 ± 10.26	2.531	0.013
	VDD (n = 34)	106.64 ± 14.09
Chair sit-and reach test (cm)	NVDD (n = 46)	11.98 ± 8.54	0.677	0.500
	VDD (n = 34)	10.79 ± 6.60
3-m timed up and-go test (s)	NVDD (n = 46)	6.47 ± 1.13	0.236	0.814
	VDD (n = 34)	6.41 ± 0.98
Figure-of-8 walk test (s)	NVDD (n = 46)	25.59 ± 4.72	-0.241	0.810
	VDD (n = 34)	25.84 ± 4.47

Data are presented as mean ± SD. Participants were classified as non–vitamin D deficient (NVDD) or vitamin D deficient (VDD) based on serum 25-hydroxyvitamin D [25(OH)D] concentration (deficiency: 25(OH)D < 20 ng/mL). Between-group differences were assessed using an independent t-test (two-tailed). Chair sit-and-reach is reported in cm; the 3-m Timed Up and Go and Figure-of-8 Walk tests are reported in seconds (lower values indicate better performance); the 30-s chair stand and 2-min step tests are reported as repetitions. Abbreviation: BMI, body mass index.

Group Differences in Body Composition and SFT by Vitamin D Status

Comparisons based on VD status revealed that the deficient group had significantly lower skeletal muscle mass than the non-deficient group (18.80 ± 3.28 kg vs. 24.16 ± 3.97 kg, p < 0.001). Additionally, the deficient group exhibited poorer functional fitness, including lower relative handgrip strength (33.02 ± 7.77 kg vs. 41.68 ± 10.03 kg, p < 0.001), fewer repetitions in the 30-s chair stand test (20.59 ± 4.82 vs. 25.67 ± 5.19, p < 0.001), and lower performance in the 2-minute step test (106.64 ± 14.09 vs. 113.53 ± 10.26, p = 0.013). In contrast, no significant differences were observed between the groups regarding body weight, body mass index, or body fat percentage (all p > 0.05). Similarly, there were no differences in flexibility (sit-and-reach), dynamic balance/mobility (Timed Up and Go), or coordination (Figure-of-8 Walk test) between the groups (all p > 0.05). The total score on the vitamin D questionnaire was significantly lower in the deficient group compared to the non-deficient group (19.74 ± 13.96 points vs. 44.96 ± 17.07 points, p < 0.001).

Group Differences in Blood Biomarkers by Vitamin D Status

In the blood biomarker analysis, the VD-deficient group showed significantly higher circulating myostatin concentrations than the non-deficient group (5.36 ± 1.42 vs. 4.17 ± 0.99 ng/mL, p < 0.001), whereas follistatin concentrations were significantly lower in the deficient group (7.21 ± 2.24 vs. 8.85 ± 2.17 ng/mL, p = 0.002). Accordingly, the F/M ratio was markedly reduced in the deficient group (147.72 ± 69.01 vs. 220.30 ± 64.72, p < 0.001). Inflammatory cytokines were also elevated in the deficient group, with higher TNF-α (11.27 ± 3.19 vs. 8.14 ± 2.15 pg/mL) and IL-1β (1.88 ± 0.43 vs. 1.39 ± 0.38 pg/mL) compared with the non-deficient group (both p < 0.001).

DISCUSSION

This cross-sectional study suggests that a questionnaire-based approach may help identify older women at higher risk for vitamin D (VD) deficiency in community settings. Additionally, VD deficiency may be associated with less favorable functional and physiological profiles. Compared to the non-deficient group, the VD-deficient group exhibited lower skeletal muscle mass and poorer performance on selected functional fitness tests, including handgrip strength, the 30-second chair stand, and the 2-minute step test. The VD-deficient group also had higher myostatin levels, lower follistatin levels, a reduced F/M ratio, and elevated TNF-α and IL-1β levels. Furthermore, serum 25(OH)D levels and the VD questionnaire score were positively correlated with indices of muscle mass and function, while inversely correlated with myostatin and inflammatory markers. These patterns suggest that VD status may reflect not only nutritional exposure but also functional capacity and physiological vulnerability in older adults. These observations align with previous reports that associate lower levels of 25(OH)D with reduced muscle strength and functional performance in older populations [15,16]. This underscores the potential clinical relevance of VD status in community-based care.

A key implication of this study is the potential practicality of using a questionnaire as a preliminary screening tool before blood testing. While serum 25(OH)D is the gold standard for assessing VD status, widespread blood-based screening in community settings can be limited by cost and feasibility [10,11]. In this study, the total score from the questionnaire was associated with serum 25(OH)D levels and effectively distinguished individuals with VD deficiency, indicating its potential as an initial screening tool. This finding aligns with previous questionnaire-based methods developed to estimate VD deficiency risk across various populations [17-19]. For instance, the EVIDENCe-Q, designed for Italian adults, included lifestyle factors relevant to VD metabolism and demonstrated effective discrimination for deficiency based on serum 25(OH)D, achieving an AUC of approximately 0.77 with a proposed cut-off of 25(OH)D < 20 ng/mL [19].

From an implementation perspective, the 38.50-point threshold is best viewed as a triage cut-off rather than a standalone diagnostic criterion because PPV varies with the underlying prevalence. Based on the observed test characteristics, the expected PPV is ~0.61 at 30% prevalence and ~0.78 at 50% prevalence (NPV ~0.92 and ~0.84), and would increase in higher-prevalence settings. Accordingly, this cut-off may help prioritize individuals for confirmatory serum testing and targeted lifestyle guidance within a stepwise community strategy.

The lower muscle mass and reduced muscle-related functional performance in the VD-deficient group align with mechanistic literature suggesting that VD signaling through the VDR plays a role in skeletal muscle metabolism [20,21]. VD-related pathways have been linked to calcium handling, mitochondrial function, and energy metabolism, all of which may impact muscle performance [15]. Experimental studies indicate that diminished VDR signaling can lead to abnormal muscle phenotypes and decreased strength [22]. In this context, the functional differences observed here may reflect a broader vulnerability profile associated with low VD status, although causality and directionality cannot be established. The lower performance in the 2-minute step test further suggests that these differences may extend beyond strength-related outcomes and could relate to overall functional capacity relevant to daily activities [24,25]. In contrast, the lack of clear between-group differences in flexibility, balance, and coordination may indicate that VD status is more directly associated with muscle mass and strength, while other components may be influenced by a variety of sensory, neural, and motor factors.

The myokine profile observed in the VD-deficient group offers further physiological insight into the VD-muscle relationship. Myostatin acts as a negative regulator of muscle growth and is linked to pathways associated with atrophy, while follistatin counteracts myostatin and promotes a more anabolic environment. The balance between these two factors is crucial for maintaining muscle homeostasis [22,26]. Previous studies have indicated that active VD may correlate with lower myostatin levels and higher follistatin expression in skeletal muscle [26]. Chang et al. (2020) found that VD treatment decreased myostatin secretion and increased markers of myogenic differentiation in a cell-based model, suggesting that VD-related signaling may prioritize anabolic processes over catabolic ones in muscle [27]. Mechanistically, 1,25(OH)2D activates VDR-RXR-mediated transcription and has been linked to the attenuation of pro-inflammatory signaling (e.g., NF-κB), which is implicated in muscle wasting and myostatin-related atrophy programs [28,29]. Given that myostatin primarily signals through the ActRIIB-SMAD2/3 axis, VD-VDR signaling could plausibly counteract downstream atrophic signaling and promote a more anabolic balance [30]. Additionally, VD-supported myogenic differentiation may increase follistatin availability, thereby antagonizing myostatin activity [31]. Taken together, the higher myostatin levels, lower follistatin levels, and reduced F/M ratio observed in the VD-deficient group suggest a shift toward a less favorable muscle protein metabolic environment. However, these mechanistic links remain speculative and require confirmation through longitudinal and interventional studies.

The higher concentrations of TNF-α and IL-1β in the VD-deficient group warrant attention. VD exerts immunomodulatory effects through the VDR expressed in immune cells and has been linked to anti-inflammatory actions in previous studies [32,33]. In contrast, lower levels of 25(OH) D have been associated with elevated inflammatory cytokines and diminished negative feedback on inflammatory signaling [34-36]. In this context, the current findings support the notion that lower VD status may coincide with a more pro-inflammatory environment in older women. Given that chronic low-grade inflammation is associated with muscle protein breakdown and impaired regeneration [7], the combination of lower VD status, higher inflammatory markers, and poorer muscle-related outcomes may represent a plausible pathway contributing to functional vulnerability with aging, while recognizing that causality cannot be established.

Several limitations should be acknowledged. First, the cross-sectional design prevents the establishment of causal relationships between VD deficiency and functional or physiological outcomes, and residual confounding factors, such as physical activity and overall nutritional status, cannot be completely ruled out. Although we excluded participants with conditions that could compromise safe SFT participation, as well as those who had received a prescription for vitamin D or used vitamin D supplementation within the past three months, we did not systematically collect data on chronic diseases and socioeconomic indicators. This limitation restricts our ability to adjust for covariates.

Second, participants were limited to older women from a single community, which may constrain generalizability. Additionally, a single assessment may not adequately capture seasonal variations in VD status. Third, due to inter-individual variability in biomarkers, larger studies with repeated measurements are necessary.

In summary, lower VD status in community-dwelling older women was associated with reduced skeletal muscle mass, poorer performance in selected functional outcomes, and a biomarker profile characterized by myostatin-follistatin imbalance and elevated inflammatory markers. The questionnaire-based assessment demonstrated potential as an initial screening tool to identify individuals more likely to have VD deficiency, supporting a feasible stepwise approach in community care.