This retrospective, cross-sectional study evaluated the prevalence and variation of serum 25-hydroxyvitamin D [25(OH)D] concentrations in a heterogeneous patient population from Changsha, China. Data were obtained from the Central Clinical Laboratory of Liuyang Maternal and Child Health Care Hospital, a tertiary academic institution serving both urban and suburban communities. All biochemical test results performed between March 2024 and September 2025 were extracted from the hospital’s electronic Laboratory Information System (LIS). This 18-month interval was selected to capture samples across all four meteorological seasons, thereby enabling the assessment of seasonal variation in vitamin D status. The study’s analytical framework was adapted from methods used in similar regional assessments, such as those conducted in Western Romania (19) and Northern China (20).

A total of 23,029 records were initially retrieved from the LIS. After rigorous data cleaning, unique cases were retained, of which 22,484 valid 25(OH)D determinations were available for statistical analysis. Inclusion criteria were: (1) complete demographic and biochemical data, including test date, age, gender, ordering department, and total 25(OH)D concentration; and (2) laboratory records within the defined sampling window (March 2024–September 2025). Exclusion criteria included missing demographic information (~545 entries), duplicate test identifiers (none detected), or physiologically implausible values (Figure 1).

Participants were stratified into seven age-related groups: infants/neonates (< 1 year), young children (1–5 years), school-age children (6–12 years), adolescents (13–17 years), adults aged 18–39 years in pregnancy-focused departments, adults aged 18–39 years in non-pregnancy departments, and older adults (> 40 years). These categories were adapted from prior pediatric and adult vitamin D studies in Asia and Europe (19). Hospital departments were aggregated into five categories; Neonatology, Children’s Health, Pre-Pregnancy, Adult Comprehensive, and Other to facilitate subgroup analyses by clinical setting.

All blood specimens were processed by the hospital’s accredited clinical laboratory under standardized operating procedures. Serum 25(OH)D and Vitamin A concentrations were measured using a two-dimensional liquid chromatography (2D-LC) (Fully Automatic 2D Liquid Chromatography Analysis System, Hunan Aidaimai Biotechnology Co., Ltd., QG3000). The analyzer simultaneously quantified 25(OH)D₂ and 25(OH)D₃, and their sum was recorded as the total 25(OH)D concentration. Assay performance was verified through internal quality control procedures, yielding intra-assay and inter-assay coefficients of variation < 5% and < 10%, respectively. All analyses adhered to the manufacturer’s specifications and the laboratory’s internal quality standards. Vitamin D status was classified in accordance with Endocrine Society and Chinese Nutrition Society criteria: Deficient: ≤ 20 ng/mL; Low: 21–29 ng/mL; Optimal: ≥ 30 ng/mL. These thresholds are consistent with internationally accepted reference intervals and comparable with previous reports (20, 21).

The following variables were extracted for each record: test date, gender, patient age, ordering department, 25(OH)D₂, 25(OH)D₃, total 25(OH)D, and vitamin A concentration (where available). Seasonal classification was based on the test date: Spring (March–May), Summer (June–August), Fall (September–November), and Winter (December–February), following established frameworks for vitamin D epidemiology in subtropical climates. Yearly comparisons (2023 vs. 2025) were evaluated to identify temporal consistency, although the two-year interval was not sufficient for formal longitudinal trend analysis.

All statistical analyses were performed using R version 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria). Continuous variables were expressed as mean ± standard deviation (SD), and categorical variables as counts and percentages. Normality was assessed using the Shapiro–Wilk test. Between-group comparisons were conducted using Student’s t-test or one-way ANOVA, as appropriate. Associations between categorical variables (vitamin D status by age, gender, department, and season) were analyzed using chi-square (χ²) tests. Significant group differences were confirmed as follows: χ² (6 df) = 2383.2, p < 0.001 for age group; χ² (8 df) = 1398.5, p < 0.001 for department group; χ² (2 df) = 536.0, p < 0.001 for gender; χ² (6 df) = 1253.9, p < 0.001 for season. Spearman’s rank correlation analysis evaluated relationships between continuous variables: 25(OH)D vs. age: p = −0.351, p < 0.001; 25(OH)D vs. vitamin A: p = −0.036, p = 0.011. A two-tailed p < 0.05 was considered statistically significant.

This study complied with the ethical principles of the Declaration of Helsinki and has been approved by the Ethics Committee of Liuyang Maternal and Child Health Care Hospital (HL20230115).

A total of 22,484 participants from Liuyang Maternal and Child Health Care Hospital, located in Changsha, China, were included in the final analysis following rigorous quality control and exclusion of incomplete entries. Participants spanned a wide age range, from newborns to 95 years, reflecting the hospital’s mixed pediatric, obstetric, and adult service profile. However, the low mean age (18.4 ± 15.7 years) and median age (12 years) underscore that the hospital-based cohort is predominantly composed of children and young adults, which is consistent with the hospital’s specialized focus on maternal and child health. Females accounted for the majority of participants (n = 13,187; 58.6%), consistent with the hospital’s focus on maternal and child health services. The age-group distribution was as follows: infants/neonates (< 1 year): 13.5% (n = 3039); young children (1-5 years): 24.0% (n = 5394); school-age children (6-12 years): 14.2% (n = 3196); adolescents (13-17 years): 2.3% (n = 518); pregnancy-focused adults (18-39 years): 18.4% (n = 4 131); non-pregnant adults (18-39 years): 12.9% (n = 2911); and older adults (> 40 years): 14.7% (n = 3295) (Figure 2).

The majority of 25-hydroxyvitamin D [25(OH)D] measurements were ordered from Children’s Health (n = 9,268; 41.2%) and Adult Comprehensive (n = 5,407; 24.0%) departments, followed by Pre-Pregnancy (n = 6,035; 26.9%), Other (n = 1,662; 7.4%), and Neonatology (n = 112; 0.5%) units. The total sample size for departmental analysis was N = 22,484 (Figure 2B). These distributions mirror the hospital’s integrated maternal and pediatric care model. Sample collection occurred continuously across all four seasons, with a modest predominance of summer and fall samples, consistent with the study’s 18-month sampling period (March 2024 - September 2025).

Serum 25-hydroxyvitamin D [25(OH)D] concentrations displayed wide interindividual variation, reflecting diverse vitamin D status among individuals tested at Liuyang Maternal and Child Health Care Hospital between March 2024 and September 2025. Overall, 62.0% of participants exhibited optimal 25(OH)D levels (≥ 30 ng/mL), whereas 27.7% were low (21–29 ng/mL) and 10.3% were deficient (≤ 20 ng/mL) (Figure 3A). These findings indicate that most individuals achieved adequate vitamin D status; however, a substantial minority remained below recommended thresholds.

Age-stratified analysis revealed pronounced developmental differences (Figure 3B). Infants and neonates (<1 year) exhibited the highest prevalence of optimal 25(OH)D levels (82.1%; 2,495/3,039), followed by young children aged 1–5 years (80.4%; 4,337/5,394). Prevalence declined with increasing age, falling to 55.8% (1,783/3,196) in school-age children (6–12 years) and reaching the lowest observed level of 37.7% (195/518) in adolescents (13–17 years). Among adults, pregnancy-focused women (18–39 years) showed 46.1% optimal status (1,904/4,131), while non-pregnant adults aged 18–39 years showed a comparable prevalence (46.3%; 1,347/2,911). Older adults (>40 years) demonstrated a higher proportion of optimal levels (57.3%; 1,888/3,295) (Figure 3B). Sex-stratified comparisons indicated higher optimal prevalence in males than females in infancy (82.8% vs 81.5%), young children (82.5% vs 78.5%), and school-age children (62.0% vs 50.0%); in adolescents, females showed slightly higher optimal prevalence than males (40.8% vs 32.9%). Across pediatric strata, deficiency was consistently more common in females than males (e.g., school-age children: 15.0% vs 10.9%), and this pattern persisted in older adults (9.8% vs 7.5%). Pregnancy-focused women also exhibited a notable burden of deficiency (15.3%). Overall, these results highlight age- and sex-related disparities in vitamin D status (Table 1).

Departmental comparisons further substantiated these trends (Figure 3C). The Children’s Health showed the most favorable distribution, where 75.1% of participants achieved an optimal vitamin D level (≥30 ng/mL). In contrast, other departments had a significantly lower prevalence of optimal status, ranging from 50.8% in Adult Comprehensive to 57.1% in Neonatology. Concurrently, the prevalence of deficiency (≤20 ng/mL) was lowest in the Children’s Health group (5.7%). The highest deficiency rates were observed in the Pre-Pregnancy (14.6%) and Adult Comprehensive (13.2%) groups. The Neonatology department had an intermediate deficiency rate of 10.7%. The prevalence of insufficiency (21–29 ng/mL) was also most favorable in the Children’s Health department (19.2%), while it exceeded 30% in the Pre-Pregnancy (33.2%) and Adult Comprehensive (36.0%) groups. These findings indicate that vitamin D sufficiency is significantly higher in the pediatric population served by the Children’s Health department, while adult-focused and pre-pregnancy care groups exhibit a greater burden of vitamin D deficiency and insufficiency.

Serum 25(OH)D levels exhibited significant seasonal fluctuations among the study population (ANOVA p < 0.001), indicating that environmental sunlight exposure and related behavioral factors strongly influenced circulating vitamin D concentrations (Figure 4A). The mean ± SD 25(OH)D concentrations increased progressively from Spring (30.5 ± 13.8 ng/mL) to Fall (38.6 ± 12.8 ng/mL) before declining slightly in Winter (33.9 ± 12.9 ng/mL). This pattern reflects the cumulative effect of greater ultraviolet B (UVB) exposure during summer and autumn months, consistent with trends reported in East Asian and European populations.

The distribution of vitamin D status categories also varied markedly by season (Figure 4B). The proportion of vitamin D deficiency (≤ 20 ng/mL) peaked in Spring (19.5%), declined sharply during Summer (6.0%) and Fall (4.5%), and rose moderately in Winter (11.3%). Conversely, the prevalence of optimal vitamin D levels (≥ 30 ng/mL) was highest during Fall (76.3%) and lowest in Spring (48.8%). These findings confirm a pronounced seasonal rhythm in vitamin D sufficiency, with maximum adequacy following summer sunlight exposure and transient declines during colder months. When stratified by gender (Figure 4C), men consistently demonstrated higher 25(OH)D concentrations than women across all four seasons (p < 0.001).

Table 2 summarizes seasonal variation in vitamin D status across all participants. Optimal 25(OH)D levels peaked during the fall (76.3%) and summer (64.7%), while spring had the highest proportion of deficiency (19.5%). Winter values were intermediate, consistent with reduced sunlight exposure but possible supplementation carryover. These results align with the patterns illustrated in Figure 3 and confirm significant seasonal effects (χ² = 1,253.9, p < 0.001). Supplementary Table S1 further stratifies vitamin D deficiency rates by both season and age group. The greatest seasonal fluctuation occurred in school-age children and pregnancy-focused adults, in whom deficiency rose sharply during spring (25.3 and 24.8%, respectively) and declined markedly in fall (6.5 and 8.7%). In contrast, infants and young children maintained stable sufficiency year-round. These data underscore differential seasonal vulnerability among older pediatric and adult subgroups. This gender disparity likely reflects combined effects of occupational sunlight exposure, outdoor activity levels, and physiological differences in vitamin D metabolism. Together, these seasonal and gender-related trends underscore the influence of environmental and demographic factors on vitamin D homeostasis in the population.

Analysis of serum 25(OH)D concentrations across different age categories revealed a pronounced age-dependent gradient in vitamin D status (p < 0.001; Figure 3B). The highest mean 25(OH)D levels were observed among infants and neonates (< 1 year; 82.1% optimal) and young children aged 1-5 years (80.4% optimal), reflecting widespread supplementation practices and regular monitoring in early life. In contrast, vitamin D sufficiency progressively declined with increasing age: school-age children (6–12 years) demonstrated 55.8% optimal, while adolescents (13–17 years) exhibited the lowest sufficiency (37.7%). Among adults, older adults (>40 years) showed a higher proportion of optimal levels (57.3%) than pregnancy-focused adults (18–39 years; 46.1%).

When analyzed by clinical department (Figure 3C), the distribution of vitamin D status differed significantly across patient categories (χ² p < 0.001). The Children’s Health department demonstrated the highest prevalence of optimal vitamin D levels (75.1%), followed by Neonatology (57.1%), reflecting strong adherence to preventive pediatric supplementation and nutritional guidelines. In contrast, the Pre-Pregnancy and Adult Comprehensive departments exhibited higher rates of deficiency (14.6 and 13.2%, respectively), consistent with limited vitamin D awareness and reduced routine screening among adults. Patients categorized under other departments (13.0%) displayed heterogeneous results due to diverse clinical indications for testing.

Collectively, these findings underscore the strong influence of developmental stage and clinical specialization on vitamin D status within the population. Pediatric and maternal groups benefit from structured supplementation programs and regular testing, while adolescents and non-pregnant adults remain vulnerable to sub-optimal vitamin D levels due to lower healthcare engagement and lifestyle factors such as reduced outdoor exposure. These demographic disparities justify targeted public health interventions aimed at improving vitamin D awareness and screening among adult and adolescent populations.

To explore potential biochemical interplay between fat-soluble vitamins, serum vitamin A concentrations were analyzed in relation to circulating 25(OH)D levels. As shown in Figure 5A, correlation analysis revealed a weak but statistically significant negative association between vitamin A and 25(OH)D (Pearson r = −0.05, p = 0.00), suggesting minimal inverse coupling between these nutrients. Comparison across vitamin D status groups (Figure 5B) showed no meaningful trend in median vitamin A levels among deficient, low, or optimal categories (p > 0.05 by ANOVA). Stratification by vitamin A tertiles (Figure 5C) further confirmed the absence of major shifts in vitamin D sufficiency prevalence across increasing vitamin A concentrations. Collectively, these findings indicate that within this population, vitamin A status exerts little measurable influence on serum 25(OH)D concentrations.

To extend the biochemical insights observed in our Changsha hospital-based cohort (Section 3.4), we validated these findings using data from the U. S. National Health and Nutrition Examination Survey (NHANES 2009–2012). This nationally representative dataset includes standardized assays for 25-hydroxyvitamin D [25(OH)D], calcium, phosphorus, alkaline phosphatase (ALP), and C-reactive protein (CRP)—parameters that reflect bone-mineral metabolism and systemic inflammation. By analyzing these variables across two consecutive survey cycles, we sought to determine whether the associations identified in the Chinese population were consistent at the population level. As shown in Figure 6A, serum 25(OH)D levels were positively correlated with calcium (r = 0.10, p < 0.001) and phosphorus (r = 0.04, p < 0.001), and inversely correlated with ALP (r = −0.05, p < 0.001) and CRP (r = −0.06, p < 0.001). Although modest in magnitude, these correlations were highly significant, suggesting a consistent biochemical link between vitamin D sufficiency, mineral balance, and reduced inflammatory activity (Supplementary Table S2).

Group-wise comparisons further demonstrated that participants with optimal vitamin D status (≥30 ng/mL) exhibited higher mean calcium and phosphorus levels (ANOVA p < 0.001 for both) and lower CRP concentrations (ANOVA p < 0.001) compared with those who were deficient or low (Figure 6B). ALP activity did not differ significantly among categories (ANOVA p = 0.15). This dose–response pattern indicates that physiological benefits of vitamin D become more apparent as serum concentrations approach sufficiency thresholds. In multivariable models adjusted for survey cycle, each one-standard-deviation increase in serum 25(OH)D was associated with higher calcium (β = +0.10, 95% CI 0.09–0.12, p < 0.001) and phosphorus (β = +0.04, 0.03–0.06, p < 0.001), and lower ALP (β = −0.05, −0.07 to −0.03, p < 0.001) and CRP (β = −0.06, −0.08 to −0.05, p < 0.001) (Figure 6C). These associations persisted after adjusting for inter-cycle variation, supporting their reproducibility across different sampling periods.

Taken together, the NHANES analyses corroborate the biochemical patterns identified in our Changsha hospital-based cohort. Across geographically and ethnically distinct populations, optimal vitamin D status consistently aligns with favorable calcium-phosphate regulation and reduced inflammatory burden. This convergence underscores the biological universality of vitamin D’s role in maintaining mineral and immune homeostasis.