Clinical Trials Characterizing the Serum 25-Hydroxyvitamin D Response to Vitamin D Supplementation in Children: A Systematic Review and Meta-Analysis

Angelina Lamberti ^1,2, Sinan Haddad ^1,3,*, Stefan Wagenpfeil ⁴, Thomas Vogt ¹, Jörg Reichrath ¹

1. Department of Dermatology, Saarland University Medical Center, 66421 Homburg, Germany

2. German Red Cross Hospital Saarlouis, 66740 Saarlouis, Germany

3. Cantonal Hospital Thurgau, Spital Thurgau Hospital Group, Thurgau, Switzerland

4. Institute for Medical Biometry, Epidemiology and Medical Informatics, Saarland University Medical Center, 66421 Homburg, Germany

* Correspondence: Sinan Haddad

Academic Editor: Mauro Fisberg

Special Issue: Vitamin D and Human Health

Received: September 13, 2024 | Accepted: October 22, 2025 | Published: October 29, 2025

Recent Progress in Nutrition 2025, Volume 5, Issue 4, doi:10.21926/rpn.2504024

Recommended citation: Lamberti A, Haddad S, Wagenpfeil S, Vogt T, Reichrath J. Clinical Trials Characterizing the Serum 25-Hydroxyvitamin D Response to Vitamin D Supplementation in Children: A Systematic Review and Meta-Analysis. Recent Progress in Nutrition 2025; 5(4): 024; doi:10.21926/rpn.2504024.

© 2025 by the authors. This is an open access article distributed under the conditions of the Creative Commons by Attribution License, which permits unrestricted use, distribution, and reproduction in any medium or format, provided the original work is correctly cited.

1. Introduction

Vitamin D deficiency affects over one billion children and adults worldwide [1]. Because of its high prevalence and its association with many acute and chronic diseases (including autoimmune-, infectious-, cardiovascular- diseases, cancer, diabetes, and neurological disorders), vitamin D deficiency is now increasingly recognized as a severe health problem [2,3,4]. The human body can fulfill its requirements in vitamin D by uptake of vitamin D₂ (e.g., derived from irradiated fungal sources) or vitamin D₃. Under most living conditions of populations in temperate parts of the world, 70-80% of the requirements in vitamin D need to be fulfilled by the ultraviolet B (UV-B)-induced synthesis in human skin, while the remaining 20-30% can be obtained by the diet or by supplements [2,3,4]. It is generally accepted that the best laboratory investigation to assess a person's vitamin D status is measuring their serum 25-hydroxyvitamin D (25(OH)D) concentration. The definition of vitamin D deficiency varies from country to country. Still, mostly values below 50 nmol are defined as vitamin D insufficiency and 25(OH)D concentrations below 30 nmol/L refer to a (severe) deficiency [5]. Beyond the cutaneous UV-induced production and intake of vitamin D, which are obviously primary inputs to vitamin D nutritional status, several other factors influence the serum 25(OH)D concentration of children, including several diseases, e.g. Asthma, Type-1-Diabetes, Celiac Disease, Cystic Fibrosis; the level and functional integrity of distinct serum proteins (Vitamin D binding protein, Albumin); age; BMI; skin type and pigmentation; cultural habits (e.g. minimizing exposure to sunshine); geographic location [6,7]. Recommendations for vitamin D supplementation are crucial for the prevention and management of vitamin D deficiency-associated diseases. However, for children, clear, evidence-based guidance on dosage and administration remains lacking, underscoring the need for robust data to inform health authority recommendations. This meta-analysis aims to provide clarity regarding the impact of vitamin D supplementation on increasing serum 25-hydroxyvitamin D levels in children.

2. Materials and Methods

We conducted a systematic literature search in MEDLINE (via PubMed) on vitamin D supplementation in children using the following search terms: ((vitamin D) AND (supplementation OR intake OR treatment OR injection OR dietary)) AND (serum 25-hydroxyvitamin D). In addition, cross-references from previously published reviews were screened.

Eligible studies were randomized controlled trials published up to June 6, 2016, in English, German, or French. The number of participants had to be reported, with at least 50 individuals in each intervention and control group. The study population was required to have a mean age of 18 years or younger, with a maximum age of 20 years. The vitamin D dose, supplementation schedule, and administration form (e.g., pill, fortified juice) had to be specified, and the mean difference in serum 25-hydroxyvitamin D concentrations had to be reported or calculable. Control groups were required to receive either a placebo or no intervention. We excluded all non-randomized study designs, such as cohort or case-control studies, as well as studies in which 25-hydroxyvitamin D values were not reported or were only reported as medians. Studies with undefined dose or supplementation duration were excluded, as were studies involving infants, to avoid confounding by maternal vitamin D supplementation during pregnancy or lactation. For each included study, we documented whether the study population consisted of healthy children or whether vitamin D supplementation was investigated in the context of a specific disease. The season in which the studies were carried out was explored. For this purpose, the months of the beginning and end of the studies were recorded. Studies in the Northern Hemisphere, that were conducted between September and April only, were classified to the "conducted during the winter" category, as opposed to studies undertaken all year round or which did not report the season. If several measurement times were specified during the study, the measurement time that was set at the end or during the winter was considered. The latitude was noted. If there were multiple study locations, the mean latitude was calculated. We ensured that the studies were not conducted across various countries. If a study had multiple study arms, the study arm in which the highest total dose of vitamin D was administered, or which was conducted in winter (to minimize the influence of sunlight), was included in the overall analysis. Descriptive statistics of the study population, study design, and 25-hydroxyvitamin D baseline values were created. The most essential details and results of the studies were recorded to provide an overview of each study's content and to show any positive or negative effects of vitamin D supplementation. If side effects of vitamin D supplementation were reported in the studies, they were mentioned in the study summaries. To evaluate the safety of supplementation, we searched for excessive 25-hydroxyvitamin D levels in the supplementation groups as well as hypercalcemia or hypercalciuria. We conducted a meta-analysis to demonstrate the effect of vitamin D supplementation on serum 25-hydroxyvitamin D levels and to identify possible factors that are associated with a greater increase. Baseline serum 25-hydroxyvitamin D level, study duration, vitamin D dosage, vitamin D medium, season, latitude, and sex were considered possible confounding factors. BMI, skin type, ethnicity, and UV exposure habits were not examined, but they can certainly influence serum vitamin D levels. Regarding BMI, there appears to be a complex interplay between obesity and vitamin D metabolism: (1) Volumetric dilution - since vitamin D is stored in adipose tissue, serum concentrations are lower in obese individuals. On the other hand, the larger vitamin D reservoir in adipose tissue may act as a continuous supply that helps maintain bone turnover. Nevertheless, multiple studies suggest adjusting vitamin D supplementation by 30-40% in obese individuals. (2) Vitamin D metabolism is impaired in obesity, with reduced hepatic 25-hydroxylation activity and altered gene expression of vitamin D-metabolizing enzymes [8].

Using a random-effects model, meta-estimates and 95% confidence intervals (CIs) were derived for the increase in serum 25-hydroxyvitamin D levels in the intervention groups compared to the control groups. The significance level was set at p < 0.05. Forest plots were created to provide an initial indication of heterogeneity. To detect qualitative and quantitative heterogeneity, Cochran's Q and I² were determined as heterogeneity measures [9,10]. Meta-regression and subgroup analyses were used to identify sources of heterogeneity across the studies. The increase in 25-hydroxyvitamin D levels at different time points was analyzed using subgroups as well as other types of administration (fortified food versus pills). The Egger's test was intended to rule out possible publication bias. [11] All meta-analyses were carried out using the "metafor-package" in R (version 3.2.4). The meta-analysis was performed according to the PRISMA-P recommendations [12]. Based on the formulas of Rupprecht et al. and Mo et al. [13,14], the vitamin D dose was calculated that would raise (or maintain) 97.5% of the study population into the 25-hydroxyvitamin D target range of over 50 nmol/L. The serum increase in nmol/L per 2.5 μg/day vitamin D(α) was calculated using the mean achieved 25-hydroxyvitamin D serum concentration of the intervention groups(b), the initial serum concentration of the intervention groups(c), and the vitamin D dose in μg/day(d). Rupprecht created a formula based on a formula from Mo et al. [13,14]: α = [(b-c)/d] * 2.5. To raise or to maintain 97.5% of the population at the target serum value through the dose recommendation, the 5th percentile(β) of the weighted mean 25-hydroxyvitamin D initial value of the intervention groups(e) and standard deviation(f) was calculated using the standard normal, which is calculated with the following formula: β = [e - (1.96 * f)]. 1.96 is the quantile value of the standard normal distribution (Ulm University [15]). The subsequent dose recommendation, which is necessary for 97.5% of the population to achieve at least 50 nmol/L 25(OH)D serum concentration(γ), was then determined with the target serum value of 50 nmol/L, the 5th percentile of the weighted mean 25-hydroxyvitamin D value of the studies (β) and the weighted mean serum increase per 2.5 μg/day vitamin D(α) using the following formula from Rupprecht et al. [13]: γ = [(50 nmol/L - β)/(α/2.5)]. Study quality was assessed using the Jadad scale for randomized controlled trials [16]. Finally, the quality of evidence of the key theses was assessed with the GRADE method [17]. The results were checked for risk of bias, inconsistency, indirectness, lack of precision and publication bias and then evaluated according to the following gradation: High ++++ (We are very confident that the actual effect is close to the effect estimate); Moderate +++ (We have moderate confidence in the effect estimate: the exact effect is probably close to the effect estimate, but there is a possibility that it is relevantly different); Low ++ (Our confidence in the effect estimate is limited: the actual effect may be relevantly different from the effect estimate); Very low + (We have very little confidence in the effect estimate: the exact effect is probably relevantly different from the effect estimate) [18].

3. Results

3.1 Literature Research and Descriptive Statistics

A total of 4435 articles were found during the literature search, with the search in MEDLINE (via PubMed) yielding 4428 hits. Seven further studies were found in the literature references of leading reviews on vitamin D, of which two studies were ultimately included in the analysis. Most studies were excluded due to the age of participants (older than 18 years or younger than one year) or because of the study design (no RCTs). Another reason for exclusion was missing data on 25-hydroxyvitamin D at the beginning or end of the studies, or low-dose supplementation of the control groups. In total, seventeen studies were included in the analysis. Except for one study by Duhamel et al., all the studies were placebo-controlled [19]. The intervention groups included 25 study arms with a total of 2198 children who received oral vitamin D. 17 of the 25 study arms were included in the overall analysis (if a study contained multiple arms, the study arm with the highest vitamin D dose was considered). The intervention groups included 1397 children, and the control arms included 17 groups with 1262 children, of which 1234 children received a placebo. Most studies were conducted in healthy children, focusing on bone turnover parameters or only 25-hydroxyvitamin D levels. Vitamin D supplementation regularly led to parathyroid hormone suppression; other parameters, like calcium level or bone density, showed mixed effects under supplementation [19,20,21,22,23,24,25,26,27,28,29]. Four studies examined vitamin D intake among children with various medical conditions. In the study by Arpadi et al. the influence of vitamin D supplementation on HIV disease in children was examined. [20] Ganmaa et al. reviewed the influence of vitamin D supplementation on the conversion rate in the tuberculin skin test in Mongolia, where children still often come into contact with Mycobacterium tuberculosis. They reported a significantly lower conversion rate in the intervention group indicating a lower rate of new tuberculosis infections [30]. The effect of vitamin D substitution on the course of acute otitis media was reported by Marchisio et al. They found that children with lower vitamin D levels below 30 ng/ml (equivalent to 75 nmol/L) had a higher incidence of otitis media. Vitamin D supplementation was associated with a significantly reduced risk of uncomplicated otitis media. Complicated otitis media (with spontaneous otorrhea after tympanic membrane perforation) occurred at similar rates in the placebo and intervention groups [31]. Thacher et al. examined the influence of vitamin D supplementation on calcium deficiency rickets. They treated children in Nigeria with proven calcium deficiency rickets with vitamin D₂ and calcium or with calcium substitution alone. This study was the only one to administer vitamin D₂, i.e. ergocalciferol. It was shown that vitamin D supplementation promoted faster healing of rickets [28]. The studies differed in terms of administration schedule and dose of vitamin D, with vitamin D administered daily in most studies. The smallest dose administered was 133 IU of vitamin D, the most significant single dose was 2000 IU per day. Five studies examined high-dose administration: two of these studies administered 50,000 IU of vitamin D monthly, two of the studies administered 100,000 IU monthly, and one study administered 100,000 IU every three months. In some studies, vitamin D was supplemented using vitamin D-fortified foods such as juice or milk; In the majority of studies, it was administered as a vitamin D preparation (in the form of a chewable tablet, soft capsule, etc.). In one study, vitamin D₂ was administered instead of vitamin D₃ [28]. The shortest study duration was seven weeks (Rich-Edwards et al.); the longest study by Du et al. was carried out over two years [22,32]. All other studies were conducted all year round, or no information was given about the time of year. The latitude at which the studies were conducted varied between 10° N and 60° N, with the average being 39.52° N +/- 10.46° (SD).

3.2 25(OH)D Baseline Values

The baseline serum 25-hydroxyvitamin D concentration was 42.48 +/- 17.82 nmol/L (mean +/- standard deviation) in the intervention groups and 41.42 +/- 16.74 nmol/L (mean +/- standard deviation) in the control groups. In only five of the seventeen studies, the baseline values were higher than the recommended minimum value of 50 nmol/L. In 41% of the studies, there was a vitamin D deficiency at the beginning, and in 30% of the studies, the average 25-hydroxyvitamin D value was in the insufficient range. The influence of latitude on the baseline serum 25-hydroxyvitamin D level was explored graphically, and no significant influence of latitude was shown. In a total of seven studies, the results were reported gender-specific: the average 25-hydroxyvitamin D concentration was 31.50 nmol/L (CI: 20.75, 43.15 nmol/L) for girls. The 25-hydroxyvitamin D concentration of the boys was on average 42.61 nmol/L (CI: 34.23, 50.99 nmol/L).

3.3 25(OH)D Increase

There was a significant increase in serum 25-hydroxyvitamin D concentration in the children in the intervention groups compared to the children in the control groups. This effect was also observed in all individual studies except for Abrams et al. and Economos et al. [20,32]. In Figure 1, the results are shown as a forest plot. The serum 25-hydroxyvitamin D level increased by an average of 25.57 nmol/L (CI: 19.59, 31.56 nmol/L, p < 0.001) in the intervention groups compared to the control groups.

Figure 1 Increase of the serum 25(OH)D concentration (intervention vs. control groups) [18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34].

The serum 25-hydroxyvitamin D concentration increased significantly in the studies with shorter study durations (<16 weeks) (pooled estimate: 26.23 nmol/L, p < 0.001). Vitamin D administration over a more extended period did not show a greater increase in 25-hydroxyvitamin D concentrations (Table 1).

Table 1 Meta-analysis regarding subgroups with different study durations.

There was significant heterogeneity both overall and in subgroups (I² > 90%). Using meta-regression, total dose, vitamin D medium, latitude, and 25-hydroxyvitamin D baseline values were tested as possible influencing factors. None of the factors were significantly associated with the increase in 25-hydroxyvitamin D. A dose-effect relationship could not be calculated due to the heterogeneous results. After supplementation, there was no longer any vitamin D deficiency in any of the intervention groups, but in 18% of the studies, the vitamin D level was still in the insufficient range below 50 nmol/L.

The analysis of the eight studies conducted only in winter (between September and April) resulted in a slightly larger increase in 25-hydroxyvitamin D (pooled estimate: 27.69 nmol/L (CI: 19.29, 36.09 nmol/L)) [18,23,24,26,27,30,31,32]. A meta-regression on studies conducted only in winter could not demonstrate any influence of total dose, baseline value, or study duration on the increase in 25-hydroxyvitamin D. The 25-hydroxyvitamin D levels of the control groups were documented to draw conclusions about the 25-hydroxyvitamin D half-life. Still, there was only limited data available, so we decided not to conduct an analysis.

The group analysis of the studies that administered a higher daily dose of over 800 IU of vitamin D also showed a slightly greater increase in serum levels (Pooled estimate 29.56 nmol/L (CI: 21.13, 37.99 nmol/L)) [20,21,23,24,27,28,30,31,34,35].

Supplementation of vitamin D-fortified foods led to a significant increase of an average of 30.36 nmol/L (CI: 11.82, 48.90 nmol/L, p = 0.0013). The vitamin D supplements increased the 25-hydroxyvitamin D level by an average of 25.67 nmol/L (CI: 19.40, 31.93 nmol/L, p < 0.001). In comparison, the fortified foods increased the 25-hydroxyvitamin D level even more. In most food studies, milk was fortified with vitamin D, except for the study by Economos et al., where orange juice supplemented with vitamin D and vitamins A and E was administered. However, in the study arm of Economos et al., which was included in our analysis, no significant increase in 25-hydroxyvitamin D was detected for the fortified juice [33]. To test the influence of gender on the rise, a subgroup analysis was carried out for boys and girls. The increase in boys (pooled estimate: 40.50 nmol/L (CI: 36.35, 44.65 nmol/L)) was significantly more substantial compared to the girls, whose pooled estimate was 25.87 nmol/L (CI: 16.22, 35.52 nmol/L). Since only a few studies were available, this statement must be interpreted with caution (see Figure 2: Evidence level according to the GRADE method: very low +) because this effect was not evident in the individual studies by Ghazi, Khadgawat, and Maalouf et al., which directly compared girls and boys [23,34,35].

Figure 2 Quality of evidence. Interpretation/Grading: Not severe/no severe limitations 0, not likely 0, no (significant) evidence 0, serious -1, likely -1, large evidence +1, very serious -2, very likely -2, substantial evidence +2. a) Many more studies could have been included to analyze the prevalence of vitamin D deficiency in children (no RCTs necessary). b) Only a few studies are available. c) considerable heterogeneity. d) no significantly larger increase. e) very different dosage regimens.

3.4 Egger's Test and Jadad Score

The Egger's test was used to exclude any possible publication bias (Figure 3). The quality of the randomized controlled studies was assessed using the Jadad score. Only one study (Thacher et al. [28]) was rated as a study of poor quality according to the Jadad scale.

Figure 3 Egger's test.

3.5 Safety of Vitamin D Supplementation and Daily Dose Estimation

To evaluate the safety of vitamin D administration, Figure 4 was created. We searched for excessive 25-hydroxyvitamin D values as well as potential side effects such as hypercalcemia (total calcium >2.65 mmol/l or ionized calcium >1.35 mmol/l) and hypercalciuria (5-7.5 mmol (200-250 mg) calcium/24 h urine, calcium/creatinine ratio < 0.57) after vitamin D administration. Only one study by Maalouf et al., who supplemented 14,000 IU of vitamin D weekly, showed excessive 25-hydroxyvitamin D levels, and one case of hypercalcemia was reported [35]. Hypercalcemia also occurred in two further studies (Du et al. and Thacher et al. [22,28]), but they substituted calcium as well. In four further studies, serum calcium was not assessed, and in two studies, serum calcium was measured, but whether hypercalcemia occurred was not reported. Calcium excretion was measured in only seven studies. Hypercalciuria did not happen in these studies. Several studies administered doses of 1000 IU of vitamin D per day, and no side effects were reported.

Figure 4 Safety of vitamin D supplementation [19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35].

The serum increase per 2.5 µg/d of vitamin D was calculated with the formulas listed in the materials and methods section using the average initial values of the intervention groups, the serum 25-hydroxyvitamin D concentrations achieved after supplementation, and the calculated vitamin D dose per day. The 5th percentile of the average initial values and standard deviation was then determined using the standard normal distribution, and finally, the dose recommendation was calculated to ensure that 97.5% of the study population achieves a 25-hydroxyvitamin D concentration of at least 50 nmol/L. For all studies, a vitamin D dose of 1145 IU per day was calculated. However, the dosage recommendations vary significantly across the individual studies, in line with the different increases in 25-hydroxyvitamin D levels.

3.6 Quality of Evidence

In Figure 2, the quality of evidence of the results was assessed using the GRADE method. The results were checked for the risk of bias, inconsistency, indirectness, lack of precision, and publication bias. High-quality evidence was available on the effectiveness of vitamin D supplementation and the safety of a daily dose up to 1000 IU [36].

4. Discussion

Because of its high prevalence and its association with many independent diseases, vitamin D deficiency represents a major obstacle to human health worldwide. To generate or re-evaluate recommendations for the oral uptake of vitamin D, health authorities in Europe and other countries at present have an urgent need to obtain reliable data on the effect of vitamin D supplementation on vitamin D status, especially in children. However, because most intake recommendations for children are based on individual studies, reliable data remain lacking in this age group [37,38]. Therefore, this work aimed to obtain reliable data to enable health authorities to re-evaluate recommendations for oral vitamin D supplementation in children. We conducted a systematic literature search in MEDLINE/PubMed on vitamin D supplementation. In addition, we searched for cross-references in previously published reviews. A total of seventeen RCTs met the inclusion criteria. On average, the baseline 25-hydroxyvitamin D concentration was 42.48 +/- 17.82 nmol/L in the intervention groups and 41.42 +/- 16.74 nmol/L in the control groups (mean value +/- standard deviation). In 41% of the studies, vitamin D deficiency at the start of the study period; in 30%, the average 25-hydroxyvitamin D value was in the insufficient range at this time. The baseline level of girls was 31.50 nmol/L (CI: 20.75, 43.15 nmol/L) and lower than the baseline level of boys (42.61 nmol/L (CI: 34.23, 50.99 nmol/L)). The average serum 25-hydroxyvitamin D level increased in the intervention groups by an average of 25.57 nmol/L (CI: 19.59, 31.56 nmol/L, p < 0.001). Interestingly, this effect was already significant after a few weeks (studies with a duration shorter than 16 weeks showed a mean increase of 26.23 nmol/L), and vitamin D administration over a longer period of time did not result in a greater increase in 25-hydroxyvitamin D concentrations (average increase of 25.45 nmol/L after 52 weeks). There appears to be an initial steep increase followed by a later flattening of serum 25(OH)D. An explanation could be that when vitamin D supplementation is initiated, the liver quickly converts it to 25(OH)D (by CYP2R1), so serum levels rise steeply. As 25-hydroxylase becomes saturated and breakdown pathways (CYP24A1) are activated, the increase decelerates. At the same time, vitamin D is stored in adipose tissue and recycled by binding proteins in the kidney, so the blood level gradually stabilizes, usually reaching a plateau within 6-12 weeks [36]. Notably, male gender was associated with a significantly higher increase in serum 25-hydroxyvitamin D. In boys, serum 25-hydroxyvitamin D concentration increased on average by 40.50 nmol/L (CI: 36.35, 44.65 nmol/L); in girls by 25.87 nmol/L (CI: 16.22, 35.52 nmol/L). However, these findings should be interpreted with caution due to the inconsistent data available. RCTs applying foods fortified with vitamin D achieved a slightly greater increase in serum 25-hydroxyvitamin D (30.36 nmol/L (CI: 11.82, 48.90 nmol/L)) as compared with studies in which a vitamin D preparation was administered (25.67 nmol/L (CI: 19.40, 31.93 nmol/L)). This observation can be explained by the fact that fortified foods may show a greater increase in serum 25(OH)D than supplements because the food matrix, especially its lipid content, enhances micellar solubilization and intestinal absorption. Vitamin D incorporated into fat-containing foods is more bioaccessible and stable than when delivered in supplement formulations [39]. In addition, consuming fortified foods with meals improves compliance and coincides with bile secretion. Interestingly, vitamin D medium, latitude, and baseline were tested in the meta-regression as factors that may influence serum 25-hydroxyvitamin D levels; none were significantly associated with the increase in serum 25-hydroxyvitamin D. After supplementation, there was no vitamin D deficiency in any of the intervention groups, but in 18% of the studies, the 25-hydroxyvitamin D level was still in the insufficiency range (<50 nmol/L). Another important result of this study was the finding that no signs of severe vitamin D intoxication were observed, highlighting the safety of vitamin D supplementation, even in children. The topic of toxicity should generally be considered in the area of vitamin D supplementation [14]. A vitamin D dose should be targeted that improves vitamin D status while avoiding or minimizing the risks of potential toxicity associated with overdose [13]. Vitamin D intoxication is associated with hypercalcemia, hyperphosphatemia, and suppressed PTH levels, which typically occurs with the excessive consumption of vitamin D [14]. Here, we show that doses of up to 1000 IU of vitamin D per day or 100,000 IU of vitamin D per month did not lead to excessive serum 25-hydroxyvitamin D levels in any RCT, although hypercalcemia was found in 3 studies. However, in these trials, calcium was substituted in parallel, or a relatively high dose of vitamin D (2000 IU per day) was substituted. Hypercalciuria was not observed, but it should be noted that calcium excretion was examined in only 7 studies.

A significant weakness of the study is that only relatively few studies could be included, and only studies up to 2016 were included. Further limitations were the considerable heterogeneity, for which reasons could not be found, and we were unable to estimate both a dose-effect relationship and the serum 25-hydroxyvitamin D half-life.

In conclusion, Vitamin D deficiency in childhood is not a marginal finding but a pervasive, clinically relevant problem with consequences for skeletal development, (auto-)immune, metabolic, and neurocognitive health. This meta-analysis provides the most comprehensive synthesis of randomized controlled trials to date. It demonstrates with high-level evidence that oral vitamin D supplementation is both safe and reliably normalizes serum 25-hydroxyvitamin D concentrations in the majority of children within weeks. Importantly, the analysis quantifies the intake required to achieve sufficiency in nearly all children and shows that current recommendations of 800 IU/day are safe but may be insufficient for complete population coverage, with approximately 1,100 IU/day needed to reach sufficiency in 97.5% of children. These findings supply a robust data-driven basis for revising pediatric vitamin D guidelines and call for harmonized, evidence-based dosing recommendations worldwide. Future large-scale, well-designed trials should define the optimal regimen for preventing deficiency-related disease beyond bone health. Interestingly, supplementation with fortified foods and vitamin D preparations was both effective. In addition, subgroup analysis showed that girls had, on average, lower 25-hydroxyvitamin D levels that increased less than those of boys after supplementation, indicating that female gender is also a risk factor for vitamin D deficiency in children. There was significant heterogeneity (I² > 90% according to Higgins/Thompson), so the results must be interpreted with caution. This meta-analysis was not yet able to provide answers to some questions, such as the serum half-time of 25(OH)D, the dose-response relationship, or other factors influencing the increase described for adults, such as latitude, baseline value, or ethnicity. Since only one study supplemented vitamin D₂, no comparison could be drawn between vitamin D₂ and D₃. However, vitamin D₃ supplementation had previously been recommended, with a presumably shorter half-life and weaker potency in increasing serum 25-hydroxyvitamin D levels compared to vitamin D₂. Future RCTs should aim to answer these still unresolved questions. Our work contributes to the increasing body of evidence that provides health authorities with reliable data to re-evaluate recommendations for oral vitamin D supplementation in children.

Abbreviations

Author Contributions

A. Lamberti: Conceptualization, writing - original draft, formal analysis, writing - review and editing. S. Haddad: Methodology, writing - review and editing. S. Wagenpfeil: Software, writing - review and editing. J. Reichrath: Methodology, writing - review and editing. T. Vogt: Review. All authors have read and approved the published version of the manuscript.

Competing Interests

The authors have declared that no competing interests exist.