Vitamin D3 for Life

Live Enlightened

Research libraryVitamin D3 For Life
Study design and methods

Primary relevance to vitamin D Vitamin D is a stated subject of this paper.

Vitamin D is a stated subject of this paper; the supporting source evidence is shown below.

Why this paper is included

Supporting source evidence:

In the Basic model, a significant effect of vitamin D supplementation was observed only for DISYPS Proxy (Table 2).

Source: Effect modification by baseline CRP levels. View original paper

Relevance classification does not establish study quality, causation or editorial approval.

Study design Secondary analysis.

Secondary analysis

Analyses existing study data; this paper is not automatically another independent trial.

Meta-analysis unclear · Double-blind reported · Human research reported

How this paper was classified

Automatic source-based classification; awaiting editorial verification. This is not a quality score.

All participants were deficient in vitamin D [25(OH)D ≤ 30 nmol/l] and at least mild depressive symptoms [Beck Depression Inventory-II (BDI-II) ≥ 14].

Source: Abstract.

This secondary analysis is based on a randomized, double-blind, placebo-controlled trial conducted at the Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, University Hospital Essen, Germany (LVR-Universitätsklinik Essen), between June 2016 and May 2018 [].

Source: Methods.

This secondary analysis is based on a randomized, double-blind, placebo-controlled trial conducted at the Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, University Hospital Essen, Germany (LVR-Universitätsklinik Essen), between June 2016 and May 2018 [].

Source: Methods.

View supporting evidence in the original paper

What the study types mean
Systematic review
Uses explicit methods to identify, assess and summarise research on a defined question.
Scoping review
Maps the extent and characteristics of research on a topic.
Narrative review
Discusses existing research; systematic-review methods have not been established by this classification.
Review / evidence synthesis
Reviews existing research; the exact review methods need further assessment.
Randomised controlled trial
Assigns participants or groups to comparison interventions by chance.
Non-randomised intervention study
Examines an intervention without random allocation.
Cohort study
Examines exposures and outcomes over time within a defined group.
Case-control study
Compares people with an outcome against people without it, examining exposures.
Cross-sectional study
Examines a population at a particular point in time.
Observational study
Observes exposures and outcomes without assigning the exposure; the subtype is not established here.
Case report
Describes an individual case.
Case series
Describes a group of individual cases.
Secondary analysis
Analyses existing study data; this paper is not automatically another independent trial.
Laboratory study
Investigates biological processes in a laboratory setting.
Qualitative study
Examines experiences or meanings using qualitative methods.
Genetic instrumental-variable study
Uses genetic variation as an instrument to investigate relationships; this is not a randomised treatment trial.
Study protocol
Describes planned methods and is not evidence that the planned study has reported results.
Multiple reported designs
More than one design is explicitly reported; examine the supporting methods.
Study design awaiting verification
The available source does not establish a sufficiently clear study design.

Meta-analysis combines numerical results from studies. Blinding describes who was unaware of treatment allocation. Neither is a separate quality grade. Study design, vitamin D relevance and document format are independent filters.

Full studies

Does inflammation moderate the effect of vitamin D supplementation on depressive symptoms? Results of an interventional study in children and adolescents

Secondary analysis · European Journal of Nutrition · 2026

Original publication title

Does inflammation moderate the effect of vitamin D supplementation on depressive symptoms? Results of an interventional study in children and adolescents

Laura Schlarbaum; Nicole Jankovic; Judith Bühlmeier; Eva Hohoff; Rhea Dankers; Harald Engler; Raphael Hirtz; Denise Zwanziger; Corinna Grasemann; Triinu Peters; Anke Hinney; Jochen Antel; Manuel Föcker; Lars Libuda

© The Author(s) 2026

Original publication · Publisher record · CC BY 4.0

Source wording retained. Formatting, section navigation and display-title punctuation adapted for this research record.

Authors’ conclusion

Overall, our findings do not support the hypothesis that the antidepressant effect of vitamin D supplementation in children and adolescents depends on CRP status: at least within a sample characterized by non-elevated inflammatory levels. Larger vitamin D RCTs with longer intervention periods are needed to investigate the role of more pronounced inflammation, ideally assessed using a broader range of inflammatory markers.

An extract from the source publication. Read it in context.

At a glance

Abstract: What objectives does the paper state?

Studies in adults suggest antidepressant effects of vitamin D, possibly via anti-inflammatory pathways, but evidence in youth is lacking. In the first randomized controlled trial (RCT) in depressed, vitamin D-deficient adolescents (DRKS00009758), symptom reduction emerged in only one of three outcomes. This secondary analysis investigates whether baseline inflammatory status modifies the effect of vitamin D.

Source: Purpose

Study methods

What methods does the paper describe?

Study design, participants and intervention

This secondary analysis is based on a randomized, double-blind, placebo-controlled trial conducted at the Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, University Hospital Essen, Germany (LVR-Universitätsklinik Essen), between June 2016 and May 2018 [25]. The study was approved by the local Ethics Committee (No. 15 to 6363-BO), registered at the German Clinical Trials Register (DRKS00009758), and conducted in accordance with the Declaration of Helsinki. Eligible participants were in: or daycare patients aged 11.0 to 18.9 years with written informed consent who were at risk for vitamin D deficiency (serum 25(OH)D < 12 ng/mL [< 30 nmol/L]) [26] and had at least mild depressive symptoms (Beck Depression Inventory-II (BDI-II) score ≥ 14) [27].

Exclusion criteria were severe somatic or renal disease, intellectual disability (IQ < 70), established hypocalcemia, and/or parathyroid hormone level > 130 ng/mL. Screening for eligibility occurred within the first two days after admission [25]. While data from 100 participants were analyzed in the original RCT [11], only those with available CRP data (n = 92) were included in the present analysis.

Details of the intervention and randomization process are described elsewhere [25]. In short, all participants received treatment as usual (TAU) according to current clinical guidelines for their respective mental disorders [25]. Using computer-generated lists, participants were randomized (1:1) to receive either oral vitamin D₃ supplementation (2640 IU/day, i.e., 66 µg/day; verum group [VG]) or placebo (placebo group [PG]) for 28 days, starting no later than the fifth day after admission. Vitamin D₃ and placebo were provided as identical pills (Dr. B.

Scheffler Nachf. GmbH & Co. KG, Bergisch Gladbach, Germany) in containers labeled with the randomization number. Participants were instructed to take three pills each morning after breakfast. Participants discharged from psychiatric treatment at the clinic at their own request before day 28 (n = 5) were instructed to continue supplementation. Furthermore, they received weekly telephone follow-ups and attended the end-of-study examination on day 28 post-randomization.

Outcome and covariate assessments

The severity of depressive symptoms was assessed at baseline (T0) and post-intervention (T1) using the BDI-II (n = 92) [28], and the Diagnostic System for Mental Disorders in Childhood and Adolescence (DISYPS) based on self-reports (DISYPS Self; n = 92), and external assessment (DISYPS Proxy; n = 91) [29]. The BDI-II is a 21-item self-report questionnaire based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) criteria, rated on a 0 to 3 Likert scale per item, with higher scores reflecting greater severity [28].

The DISYPS comprises 29 items (0 to 3 points each), referencing DSM-IV and the International Statistical Classification of Diseases and Related Health Problems Revision 10 (ICD-10) criteria [29]. It was conducted both as a self-assessment (DISYPS Self) and as an external assessment by guardians (DISYPS Proxy). Total scores were converted into stanine values, with scores ≥ 8 considered clinically relevant [29].

At T0 and T1, fasting blood samples were collected in the early morning from an antecubital vein into monovettes (Sarstedt, Nümbrecht, Germany), transferred to the Central Laboratory of the University Hospital Essen, and analyzed within one hour [25]. Serum 25(OH)D was measured using the Siemens ADVIA Centaur® Immunoassay-system (Siemens Healthineers, Erlangen, Germany). Serum CRP concentrations (mg/L) were determined at T0 using the Siemens Atellica® CH Analyzer (Siemens Healthineers, Erlangen, Germany). Clinical assessments at admission included assessments of body weight and height, pubertal status, and general health screening [25].

For participants under 18.25 years, the BMI standard deviation scores (SDS, kg/m2) were calculated using German national reference data according to the Lambda-Mu-Sigma (LMS) method [30].

Statistical analysis

Sample characteristics at T0 and T1 were summarized by median (P25, P75) values for continuous variables, and percentages for categorical variables (Table 1). To address potential sex-specific differences, interaction analyses (treatment × sex assigned at birth) were conducted for each outcome (BDI-II, DISYPS Self, DISYPS Proxy) separately, with CRP, treatment, and sex assigned at birth as covariates in the model. As these analyses did not provide evidence of sex interactions, data from both sexes were pooled for further analyses.

Intervention effects on depressive symptoms were analyzed separately for each outcome using analyses of covariance (ANCOVA) (Table 2). In the basic model, post-intervention values (T1) served as dependent variables, with treatment group (PG vs. VG) as the independent variable. Covariates were age, sex assigned at birth, baseline 25(OH)D concentration, and corresponding baseline values of the outcome, following the approach of Libuda et al. (2020) [11]. To examine the role of inflammation, baseline CRP concentration (T0) was added as an additional covariate (Model 1, Table 2).

In subsequent interaction analyses, CRP was tested as a potential effect modifier via treatment × CRP interaction terms with CRP and treatment as covariates in the model. These analyses aimed to determine whether treatment effects on depressive symptoms significantly differed depending on baseline CRP levels. CRP was examined using three thresholds: (1) the clinically established threshold of 5 mg/L [21], (2) the 3 mg/L threshold proposed in recent research on depression [15, 31], and (3) the sample-specific median (0.845 mg/L).

To account for multiple testing, p-values were corrected using the false discovery rate (FDR) method [32]. Afterwards, stratified analyses were conducted to assess treatment effects in CRP-defined subgroups using the thresholds defined above. All stratified models were adjusted for the covariates (Table 2). To assess the robustness of results, sensitivity analyses were conducted on four subsets: (1) participants not taking antidepressants [n = 78], due to potential effects on inflammation [33, 34], (2) participants without anorexia nervosa [n = 84], due to its impact on immunological processes [35, 36], (3) participants without potential current infection [n = 89], defined as CRP ≥ 10 mg/L [15], and (4) participants with clinically confirmed depression [n = 76], based on psychiatric assessment (data not shown).

Analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, USA). The alpha level was set to 0.05.

Table 1

Characteristics of N = 92 randomized participants in placebo (n = 45) or Vitamin D supplementation group (n = 47) at baseline (T0) and post-intervention (T1)

Placebo groupVerum group
nMedian or %nMedian or %
T0T0
Age [years]4516.1 (14.8; 17.4)4716.3 (15.2; 17.1)
Females [%]3782.23574.5
Body height [cm]45164.7 (160.5; 169.4)47165.7 (162.0; 175.0)
Body weight [kg]4556.4 (50.4; 65.2)4765.0 (53.9; 75.0)
BMI-SDS [kg/m2]a440.24 (−0.54; 1.49)450.21 (−0.16; 1.79)
Antidepressants (Yes)511.1919.1
CRP [mg/L]450.84 (0.32; 2.23)470.85 (0.24; 2.19)
T0T1T0T1
25(OH)D [ng/mL]b458.5 (6.8; 10.2)9.4 (7.3; 11.5)479.0 (7.7; 10.3)24.7 (20.5; 27.6)
BDI-II (Total scores)4528 (23; 34)23 (13; 36)4728 (22; 43)25 (14; 42)
DISYPS Self (stanine)459 (8; 9)9 (8; 9)479 (8; 9)8 (8; 9)
DISYPS Proxy (stanine)449 (8; 9)9 (8; 9)479 (8; 9)8 (7; 9)

25(OH)D, 25-hydroxy-cholecalciferol; BDI-II, beck depression inventory II; BMI, body mass index; CRP; C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

Values are presented as median (P25; P75) for interval-scaled variables, and percentages for categorical variables

aBMI-SDS calculated using German reference data (LMS method) for participants < 18.25 years [30]

b1 ng/mL serum 25(OH)D equates to 2.5 nmol/L

Table 2

ANCOVAa results for Vitamin D intervention (verum) effects on depressive symptoms: interaction and stratified analyses by multiple CRP thresholds

ModelPredictorBDI-IIDISYPS SelfDISYPS Proxy
nβ (CI)pnβ (CI)pnβ (CI)P
Basic modelaverum921.17 (−2.59; 4.93)0.5492−0.14 (−0.69; 0.40)0.5991−0.62 (−1.22; −0.03)0.04
Model 1bverum921.46 (−2.35; 5.26)0.4592−0.11 (−0.66; 0.45)0.7091−0.61 (−1.21; −0.01)0.047
CRP−0.29 (−0.89; 0.31)0.34−0.04 (−0.13; 0.04)0.32−0.02 (−0.11; 0.08)0.32
Basic model stratified by CRP ≥ 5.0 mg/Lc
CRP ≥ 5.0 mg/Laverum81.90 (−13.84; 17.64)0.738−1.28 (−4.24; 1.68)0.268−0.45 (−3.37; 2.48)0.66
CRP < 5.0 mg/Laverum841.24 (−2.79; 5.27)0.5484−0.07 (−0.66; 0.53)0.8283−0.64 (−1.29; 0.01)0.05
p value for interactione = 0.79p value for interactione = 0.37p value for interactione = 0.74
Basic model stratified by CRP ≥ 3.0 mg/Ld
CRP ≥ 3.0 mg/Laverum172.34 (−2.34; 7.02)0.3017−0.76 (−2.10; 0.58)0.2417−1.25 (−3.36; 0.85)0.22
CRP < 3.0 mg/Laverum751.73 (−2.57; 6.03)0.42750.10 (−0.49; 0.69)0.7374−0.45 (−1.07; 0.18)0.16
p value for interactione = 0.26p value for interactione = 0.047fp value for interactione = 0.13
Basic model stratified by CRP median
CRP ≥ 0.845 mg/Laverum46−0.01 (−5.29; 5.27)0.9946−0.29 (−1.15; 0.56)0.4946−0.78 (−1.56; 0.003)0.05
CRP < 0.845 mg/Laverum463.88 (−2.18; 9.95)0.20460.23 (−0.50; 0.95)0.5345−0.45 (−1.52; 0.61)0.40
p value for interactione = 0.34p value for interactione = 0.53p value for interactione = 0.70

BDI-II, beck depression inventory II; CRP, C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

aBasic model adjusted for age, sex assigned at birth, baseline 25(OH)D, and baseline value of the respective outcome

bModel 1: Basic model further adjusted for CRP

cThreshold of 5 mg/L, as clinically established [21]

dThreshold of 3 mg/L, as proposed in recent depression research [15, 31]

ep value for the interaction term (treatment × CRP) with CRP and treatment as covariates

fFalse discovery rate: adjusted p value = 0.423

Source: Materials and methods Source: Materials and methods Source: Materials and methods Source: Materials and methods Source: Materials and methods

Results: Sample characteristics

A total of N = 92 participants aged 12.7 to 18.9 years were included in the analysis. Baseline characteristics (T0) were largely comparable between groups, but the PG had a higher proportion of females (82.2%) than the VG (74.5%) (Table 1). 15.2% were taking antidepressants (PG: 11.1%; VG: 19.1%). At T0, depressive symptomatology was pronounced in both groups, with median BDI-II scores of 28 and stanine scores of 9 on both DISYPS Self and Proxy: the highest possible score. From T0 to T1, median BDI-II scores improved slightly in both groups (VG: 28 to 25; PG: 28 to 23).

While DISYPS Self and DISYPS Proxy scores remained stable in the PG, scores in the VG showed a slight improvement (Table 1). Baseline 25(OH)D levels were similarly low across groups, with median concentrations ≤ 9 ng/mL. After intervention (T1), the VG reached median levels in the “adequate” range (≥ 20 ng/mL). For CRP, baseline concentrations were comparable between groups (PG: 0.84 mg/L; VG: 0.85 mg/L; sample-specific median: 0.845 mg/L) (Table 1). Most participants had CRP levels below the clinical threshold of 5 mg/L (91.3%; n = 84) and below the 3 mg/L threshold for low-grade inflammation (81.5%; n = 75) (Table 2).

Table 1

Characteristics of N = 92 randomized participants in placebo (n = 45) or Vitamin D supplementation group (n = 47) at baseline (T0) and post-intervention (T1)

Placebo groupVerum group
nMedian or %nMedian or %
T0T0
Age [years]4516.1 (14.8; 17.4)4716.3 (15.2; 17.1)
Females [%]3782.23574.5
Body height [cm]45164.7 (160.5; 169.4)47165.7 (162.0; 175.0)
Body weight [kg]4556.4 (50.4; 65.2)4765.0 (53.9; 75.0)
BMI-SDS [kg/m2]a440.24 (−0.54; 1.49)450.21 (−0.16; 1.79)
Antidepressants (Yes)511.1919.1
CRP [mg/L]450.84 (0.32; 2.23)470.85 (0.24; 2.19)
T0T1T0T1
25(OH)D [ng/mL]b458.5 (6.8; 10.2)9.4 (7.3; 11.5)479.0 (7.7; 10.3)24.7 (20.5; 27.6)
BDI-II (Total scores)4528 (23; 34)23 (13; 36)4728 (22; 43)25 (14; 42)
DISYPS Self (stanine)459 (8; 9)9 (8; 9)479 (8; 9)8 (8; 9)
DISYPS Proxy (stanine)449 (8; 9)9 (8; 9)479 (8; 9)8 (7; 9)

25(OH)D, 25-hydroxy-cholecalciferol; BDI-II, beck depression inventory II; BMI, body mass index; CRP; C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

Values are presented as median (P25; P75) for interval-scaled variables, and percentages for categorical variables

aBMI-SDS calculated using German reference data (LMS method) for participants < 18.25 years [30]

b1 ng/mL serum 25(OH)D equates to 2.5 nmol/L

Table 2

ANCOVAa results for Vitamin D intervention (verum) effects on depressive symptoms: interaction and stratified analyses by multiple CRP thresholds

ModelPredictorBDI-IIDISYPS SelfDISYPS Proxy
nβ (CI)pnβ (CI)pnβ (CI)P
Basic modelaverum921.17 (−2.59; 4.93)0.5492−0.14 (−0.69; 0.40)0.5991−0.62 (−1.22; −0.03)0.04
Model 1bverum921.46 (−2.35; 5.26)0.4592−0.11 (−0.66; 0.45)0.7091−0.61 (−1.21; −0.01)0.047
CRP−0.29 (−0.89; 0.31)0.34−0.04 (−0.13; 0.04)0.32−0.02 (−0.11; 0.08)0.32
Basic model stratified by CRP ≥ 5.0 mg/Lc
CRP ≥ 5.0 mg/Laverum81.90 (−13.84; 17.64)0.738−1.28 (−4.24; 1.68)0.268−0.45 (−3.37; 2.48)0.66
CRP < 5.0 mg/Laverum841.24 (−2.79; 5.27)0.5484−0.07 (−0.66; 0.53)0.8283−0.64 (−1.29; 0.01)0.05
p value for interactione = 0.79p value for interactione = 0.37p value for interactione = 0.74
Basic model stratified by CRP ≥ 3.0 mg/Ld
CRP ≥ 3.0 mg/Laverum172.34 (−2.34; 7.02)0.3017−0.76 (−2.10; 0.58)0.2417−1.25 (−3.36; 0.85)0.22
CRP < 3.0 mg/Laverum751.73 (−2.57; 6.03)0.42750.10 (−0.49; 0.69)0.7374−0.45 (−1.07; 0.18)0.16
p value for interactione = 0.26p value for interactione = 0.047fp value for interactione = 0.13
Basic model stratified by CRP median
CRP ≥ 0.845 mg/Laverum46−0.01 (−5.29; 5.27)0.9946−0.29 (−1.15; 0.56)0.4946−0.78 (−1.56; 0.003)0.05
CRP < 0.845 mg/Laverum463.88 (−2.18; 9.95)0.20460.23 (−0.50; 0.95)0.5345−0.45 (−1.52; 0.61)0.40
p value for interactione = 0.34p value for interactione = 0.53p value for interactione = 0.70

BDI-II, beck depression inventory II; CRP, C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

aBasic model adjusted for age, sex assigned at birth, baseline 25(OH)D, and baseline value of the respective outcome

bModel 1: Basic model further adjusted for CRP

cThreshold of 5 mg/L, as clinically established [21]

dThreshold of 3 mg/L, as proposed in recent depression research [15, 31]

ep value for the interaction term (treatment × CRP) with CRP and treatment as covariates

fFalse discovery rate: adjusted p value = 0.423

Source: Sample characteristics Source: Sample characteristics

Results: Effect modification by baseline CRP levels

In the Basic model, a significant effect of vitamin D supplementation was observed only for DISYPS Proxy (Table 2). Adjusting for baseline CRP levels did not alter the results (Model 1, Table 2). No effect of baseline CRP was observed on either outcome. No significant interaction between treatment and CRP status at baseline was found when stratified by the threshold of 5 mg/L. In CRP-defined subgroups, no significant treatment effects were observed across all outcomes, including DISYPS Proxy.

At the 3 mg/L threshold, a statistically significant interaction was found for DISYPS Self (p = 0.047), however, this did not remain significant after FDR correction (p = 0.423). Stratified analyses by baseline CRP level indicated opposing trends between groups, with a trend toward symptom improvement in the CRP ≥ 3 mg/L subgroup receiving verum, though no effects in CRP-defined subgroups reached significance. Analyses using the sample-specific CRP median (0.845 mg/L) revealed no significant interactions for any outcome.

For DISYPS Self, the stratified results nevertheless followed a descriptively similar pattern to that observed at the 3 mg/L threshold. However, none of these subgroup effects reached significance (Table 2). These findings were consistent across sensitivity analyses (data not shown).

Table 2

ANCOVAa results for Vitamin D intervention (verum) effects on depressive symptoms: interaction and stratified analyses by multiple CRP thresholds

ModelPredictorBDI-IIDISYPS SelfDISYPS Proxy
nβ (CI)pnβ (CI)pnβ (CI)P
Basic modelaverum921.17 (−2.59; 4.93)0.5492−0.14 (−0.69; 0.40)0.5991−0.62 (−1.22; −0.03)0.04
Model 1bverum921.46 (−2.35; 5.26)0.4592−0.11 (−0.66; 0.45)0.7091−0.61 (−1.21; −0.01)0.047
CRP−0.29 (−0.89; 0.31)0.34−0.04 (−0.13; 0.04)0.32−0.02 (−0.11; 0.08)0.32
Basic model stratified by CRP ≥ 5.0 mg/Lc
CRP ≥ 5.0 mg/Laverum81.90 (−13.84; 17.64)0.738−1.28 (−4.24; 1.68)0.268−0.45 (−3.37; 2.48)0.66
CRP < 5.0 mg/Laverum841.24 (−2.79; 5.27)0.5484−0.07 (−0.66; 0.53)0.8283−0.64 (−1.29; 0.01)0.05
p value for interactione = 0.79p value for interactione = 0.37p value for interactione = 0.74
Basic model stratified by CRP ≥ 3.0 mg/Ld
CRP ≥ 3.0 mg/Laverum172.34 (−2.34; 7.02)0.3017−0.76 (−2.10; 0.58)0.2417−1.25 (−3.36; 0.85)0.22
CRP < 3.0 mg/Laverum751.73 (−2.57; 6.03)0.42750.10 (−0.49; 0.69)0.7374−0.45 (−1.07; 0.18)0.16
p value for interactione = 0.26p value for interactione = 0.047fp value for interactione = 0.13
Basic model stratified by CRP median
CRP ≥ 0.845 mg/Laverum46−0.01 (−5.29; 5.27)0.9946−0.29 (−1.15; 0.56)0.4946−0.78 (−1.56; 0.003)0.05
CRP < 0.845 mg/Laverum463.88 (−2.18; 9.95)0.20460.23 (−0.50; 0.95)0.5345−0.45 (−1.52; 0.61)0.40
p value for interactione = 0.34p value for interactione = 0.53p value for interactione = 0.70

BDI-II, beck depression inventory II; CRP, C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

aBasic model adjusted for age, sex assigned at birth, baseline 25(OH)D, and baseline value of the respective outcome

bModel 1: Basic model further adjusted for CRP

cThreshold of 5 mg/L, as clinically established [21]

dThreshold of 3 mg/L, as proposed in recent depression research [15, 31]

ep value for the interaction term (treatment × CRP) with CRP and treatment as covariates

fFalse discovery rate: adjusted p value = 0.423

Source: Effect modification by baseline CRP levels

Discussion: Strength and limitations

Recruitment from a clinical setting enhances internal validity, though generalizability is limited, as only about 20% of children with mental disorders receive psychiatric care [42Not reported44]. However, this clinical setting allows for a more homogenous and well-characterized sample. All participants met the criteria for at least mild depression [27, 29], and baseline vitamin D levels indicated deficiency risk [26]. Information on acute infections or inflammatory conditions at the time of assessment was not systematically collected.

However, following Osimo et al. (2019), CRP ≥ 10 mg/L was used as an indicator of potential current infection [15]. Only n = 3 participants exceeded this threshold, and a sensitivity analysis excluding these cases did not alter the study results (data not shown).

Depressive symptoms were assessed using three instruments, including self: and proxy-rated ratings [28, 29]. Discrepancies between self: and proxy-reports are common in child and adolescent psychiatry [45Not reported47], as parents often emphasize symptom severity while adolescents tend to under-report symptoms [48]. This may lead to inconsistencies and potential misclassification. In our study, only parent-reported (DISYPS Proxy) symptoms showed significant effects in the unstratified analyses, whereas a nominal effect emerged for self-reported DISYPS in the interaction analyses, but not for the BDI-II.

To account for such discrepancies, we followed best practices in youth mental health through a multi-informant approach [45, 46]. As the applied instruments assess symptom severity rather than diagnostic status [28, 29], depression was additionally verified by clinical evaluation, which classified 76 of 92 participants as depressed. A sensitivity analysis restricted to these 76 participants did not change the results (data not shown).

Focusing solely on CRP limits interpretation, as it reflects only part of the immune response. Additional humoral and cellular markers, such as cytokines (e.g., TNF-α, IFN-γ, IL-1β, IL-6, IL-8, IL-10) and immune cell subsets, as proposed in other studies exploring the etiology of depression [16, 18, 49Not reported53], may capture distinct pathways that CRP alone cannot represent. These markers could provide a more nuanced understanding of the immune mechanisms underlying adolescent depression and how inflammation influences depressive symptoms and treatment response.

While additional markers could enhance the analysis, CRP remains a valid and well-supported choice for investigating inflammation in depression, with associations to depression severity [16, 18], and well-established cut-offs [15, 31]. CRP was only analyzed at baseline, limiting insights into temporal dynamics. While this was sufficient for examining the potential modulating effect of inflammation on vitamin D supplementation, repeated assessments could provide a deeper understanding of how inflammation develops over time following vitamin D treatment.

Given evidence that depression and inflammation may influence each other during childhood and adolescence [18], future studies incorporating multiple time points could clarify how these dynamics impact the long-term effects of vitamin D supplementation on depression.

While sex-specific patterns in the inflammation: depression link have been suggested [54], other studies report no association between CRP levels and sex or gender [15]. In our study, no interaction between sex and treatment was observed. Yet, the overall sample size was too limited to allow for meaningful sex-stratified analyses. Such stratification would have further reduced statistical power and reliability. Nonetheless, we adjusted for sex in all ANCOVA models. The relatively small sample size, particularly in inflammation-defined subgroups (e.g., n = 8 for CRP ≥ 5 mg/L), limited statistical power and may have obscured subtle effects.

While this constraint must be acknowledged, it reflects the practical challenges of recruiting clinically well-characterized adolescents for biological studies. Nonetheless, the present findings provide a valuable foundation for future trials, highlighting key methodological aspects: such as adequate stratification by inflammatory status and longer intervention durations: that can guide more targeted study designs.

Source: Strength and limitations Source: Strength and limitations Source: Strength and limitations Source: Strength and limitations

Results

Sample characteristics

A total of N = 92 participants aged 12.7 to 18.9 years were included in the analysis. Baseline characteristics (T0) were largely comparable between groups, but the PG had a higher proportion of females (82.2%) than the VG (74.5%) (Table 1). 15.2% were taking antidepressants (PG: 11.1%; VG: 19.1%). At T0, depressive symptomatology was pronounced in both groups, with median BDI-II scores of 28 and stanine scores of 9 on both DISYPS Self and Proxy: the highest possible score. From T0 to T1, median BDI-II scores improved slightly in both groups (VG: 28 to 25; PG: 28 to 23).

While DISYPS Self and DISYPS Proxy scores remained stable in the PG, scores in the VG showed a slight improvement (Table 1). Baseline 25(OH)D levels were similarly low across groups, with median concentrations ≤ 9 ng/mL. After intervention (T1), the VG reached median levels in the “adequate” range (≥ 20 ng/mL). For CRP, baseline concentrations were comparable between groups (PG: 0.84 mg/L; VG: 0.85 mg/L; sample-specific median: 0.845 mg/L) (Table 1). Most participants had CRP levels below the clinical threshold of 5 mg/L (91.3%; n = 84) and below the 3 mg/L threshold for low-grade inflammation (81.5%; n = 75) (Table 2).

Table 1

Characteristics of N = 92 randomized participants in placebo (n = 45) or Vitamin D supplementation group (n = 47) at baseline (T0) and post-intervention (T1)

Placebo groupVerum group
nMedian or %nMedian or %
T0T0
Age [years]4516.1 (14.8; 17.4)4716.3 (15.2; 17.1)
Females [%]3782.23574.5
Body height [cm]45164.7 (160.5; 169.4)47165.7 (162.0; 175.0)
Body weight [kg]4556.4 (50.4; 65.2)4765.0 (53.9; 75.0)
BMI-SDS [kg/m2]a440.24 (−0.54; 1.49)450.21 (−0.16; 1.79)
Antidepressants (Yes)511.1919.1
CRP [mg/L]450.84 (0.32; 2.23)470.85 (0.24; 2.19)
T0T1T0T1
25(OH)D [ng/mL]b458.5 (6.8; 10.2)9.4 (7.3; 11.5)479.0 (7.7; 10.3)24.7 (20.5; 27.6)
BDI-II (Total scores)4528 (23; 34)23 (13; 36)4728 (22; 43)25 (14; 42)
DISYPS Self (stanine)459 (8; 9)9 (8; 9)479 (8; 9)8 (8; 9)
DISYPS Proxy (stanine)449 (8; 9)9 (8; 9)479 (8; 9)8 (7; 9)

25(OH)D, 25-hydroxy-cholecalciferol; BDI-II, beck depression inventory II; BMI, body mass index; CRP; C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

Values are presented as median (P25; P75) for interval-scaled variables, and percentages for categorical variables

aBMI-SDS calculated using German reference data (LMS method) for participants < 18.25 years [30]

b1 ng/mL serum 25(OH)D equates to 2.5 nmol/L

Effect modification by baseline CRP levels

In the Basic model, a significant effect of vitamin D supplementation was observed only for DISYPS Proxy (Table 2). Adjusting for baseline CRP levels did not alter the results (Model 1, Table 2). No effect of baseline CRP was observed on either outcome. No significant interaction between treatment and CRP status at baseline was found when stratified by the threshold of 5 mg/L. In CRP-defined subgroups, no significant treatment effects were observed across all outcomes, including DISYPS Proxy.

At the 3 mg/L threshold, a statistically significant interaction was found for DISYPS Self (p = 0.047), however, this did not remain significant after FDR correction (p = 0.423). Stratified analyses by baseline CRP level indicated opposing trends between groups, with a trend toward symptom improvement in the CRP ≥ 3 mg/L subgroup receiving verum, though no effects in CRP-defined subgroups reached significance. Analyses using the sample-specific CRP median (0.845 mg/L) revealed no significant interactions for any outcome.

For DISYPS Self, the stratified results nevertheless followed a descriptively similar pattern to that observed at the 3 mg/L threshold. However, none of these subgroup effects reached significance (Table 2). These findings were consistent across sensitivity analyses (data not shown).

Table 2

ANCOVAa results for Vitamin D intervention (verum) effects on depressive symptoms: interaction and stratified analyses by multiple CRP thresholds

ModelPredictorBDI-IIDISYPS SelfDISYPS Proxy
nβ (CI)pnβ (CI)pnβ (CI)P
Basic modelaverum921.17 (−2.59; 4.93)0.5492−0.14 (−0.69; 0.40)0.5991−0.62 (−1.22; −0.03)0.04
Model 1bverum921.46 (−2.35; 5.26)0.4592−0.11 (−0.66; 0.45)0.7091−0.61 (−1.21; −0.01)0.047
CRP−0.29 (−0.89; 0.31)0.34−0.04 (−0.13; 0.04)0.32−0.02 (−0.11; 0.08)0.32
Basic model stratified by CRP ≥ 5.0 mg/Lc
CRP ≥ 5.0 mg/Laverum81.90 (−13.84; 17.64)0.738−1.28 (−4.24; 1.68)0.268−0.45 (−3.37; 2.48)0.66
CRP < 5.0 mg/Laverum841.24 (−2.79; 5.27)0.5484−0.07 (−0.66; 0.53)0.8283−0.64 (−1.29; 0.01)0.05
p value for interactione = 0.79p value for interactione = 0.37p value for interactione = 0.74
Basic model stratified by CRP ≥ 3.0 mg/Ld
CRP ≥ 3.0 mg/Laverum172.34 (−2.34; 7.02)0.3017−0.76 (−2.10; 0.58)0.2417−1.25 (−3.36; 0.85)0.22
CRP < 3.0 mg/Laverum751.73 (−2.57; 6.03)0.42750.10 (−0.49; 0.69)0.7374−0.45 (−1.07; 0.18)0.16
p value for interactione = 0.26p value for interactione = 0.047fp value for interactione = 0.13
Basic model stratified by CRP median
CRP ≥ 0.845 mg/Laverum46−0.01 (−5.29; 5.27)0.9946−0.29 (−1.15; 0.56)0.4946−0.78 (−1.56; 0.003)0.05
CRP < 0.845 mg/Laverum463.88 (−2.18; 9.95)0.20460.23 (−0.50; 0.95)0.5345−0.45 (−1.52; 0.61)0.40
p value for interactione = 0.34p value for interactione = 0.53p value for interactione = 0.70

BDI-II, beck depression inventory II; CRP, C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

aBasic model adjusted for age, sex assigned at birth, baseline 25(OH)D, and baseline value of the respective outcome

bModel 1: Basic model further adjusted for CRP

cThreshold of 5 mg/L, as clinically established [21]

dThreshold of 3 mg/L, as proposed in recent depression research [15, 31]

ep value for the interaction term (treatment × CRP) with CRP and treatment as covariates

fFalse discovery rate: adjusted p value = 0.423

Source: Results Source: Results

Conclusions

Overall, our findings do not support the hypothesis that the antidepressant effect of vitamin D supplementation in children and adolescents depends on CRP status: at least within a sample characterized by non-elevated inflammatory levels. Larger vitamin D RCTs with longer intervention periods are needed to investigate the role of more pronounced inflammation, ideally assessed using a broader range of inflammatory markers.

Source: Conclusions
Authors’ abstract

Purpose

Studies in adults suggest antidepressant effects of vitamin D, possibly via anti-inflammatory pathways, but evidence in youth is lacking. In the first randomized controlled trial (RCT) in depressed, vitamin D-deficient adolescents (DRKS00009758), symptom reduction emerged in only one of three outcomes. This secondary analysis investigates whether baseline inflammatory status modifies the effect of vitamin D.

Methods

Data from 92 participants (78.3% ♀) of the RCT conducted at the Department of Child and Adolescent Psychiatry were analyzed. All participants were deficient in vitamin D [25(OH)D ≤ 30 nmol/l] and at least mild depressive symptoms [Beck Depression Inventory-II (BDI-II) ≥ 14]. Participants received 2640 IU/day of vitamin D or placebo for 28 days plus treatment as usual. Depressive symptoms were assessed via BDI-II and the Diagnostic System for Mental Disorders in Childhood and Adolescence (DISYPS), self: and proxy-rated.

C-reactive protein (CRP) was measured at baseline as an inflammatory marker. Its impact on treatment effects was examined via interaction analyses and stratification by CRP status.

Results

In 81.5% of the participants, CRP levels were below the threshold for low-grade inflammation (3 mg/L). No consistent interaction between treatment and baseline CRP was observed across outcomes. A nominal interaction for self-rated DISYPS at the 3 mg/L threshold lost significance after correction for multiple testing. Stratified analyses using different CRP thresholds revealed no significant subgroup effects.

Conclusions

Overall, our findings do not support the hypothesis that the antidepressant effects of vitamin D supplementation in children and adolescents depend on baseline CRP status: at least within a sample characterized by minimal baseline inflammation.

Read the full paper

Introduction

Depression is a prevalent mental health disorder with a complex etiology [1] and one of the leading causes of disability worldwide [2]. In youth, depression is associated with impairments in academic performance, social interactions [3, 4], and quality of life [5], as well as increased morbidity and mortality risks [6, 7]. Notably, over 50% of all initial depression diagnoses are made before age 14, and 75% occur before age 24 [8], highlighting adolescence as a critical window for intervention.

Vitamin D supplementation has shown promise as a treatment approach in meta-analyses of randomized controlled trials (RCTs) in adults [9, 10]. However, evidence in youth remains limited. The only RCT in children and adolescents conducted by our study group reported inconsistent results: only one of three depression measures showed significant symptom reduction [11]. While an inverse relationship has been observed in adults, the effects in youth were less clear, suggesting that vitamin D efficacy may depend on distinct biological subgroups and mechanisms.

There is increasing clinical and experimental evidence linking depression to inflammation [12, 13], with neuroinflammatory processes emerging as a key factor in its pathophysiology [14, 15]. Elevated levels of the inflammation marker C-reactive protein (CRP) have been reported in a subset of adults with depression [16], particularly in those with treatment-resistant symptoms [17]. Similar associations have been observed in children and adolescents [18]. Moreover, longitudinal studies associate elevated CRP levels in adolescence with the development of depressive episodes later in life, highlighting its potential as a biomarker for inflammatory pathways involved in the etiology of depression [19, 20].

While CRP ≥ 5 mg/L has been defined as the clinical threshold for acute inflammation in most medical conditions [21], low-grade inflammation in depression is defined as CRP ≥ 3 mg/L according to Centers for Disease Control and Prevention / American Heart Association (CDC/AHA) guidelines [22].

Considering the potential role of inflammation in the etiology of depression, anti-inflammatory agents, including vitamin D, could represent an option for the treatment of depression. Since previous studies indicated that vitamin D supplementation reduces CRP [23], vitamin D may exert antidepressant effects via immunomodulation. However, the proposed mechanisms of vitamin D in the context of depression are largely based on preclinical research [24]. While it has been emphasized that inflammatory markers should be evaluated alongside vitamin D in depression studies [9], longitudinal data on the interplay between vitamin D, inflammation, and depressive symptoms in children and adolescents are currently lacking.

It remains unclear whether the antidepressant effects of vitamin D supplementation are more pronounced in subgroups with increased inflammation. Therefore, we conducted a secondary analysis of the RCT in adolescents [11] to examine whether baseline CRP levels modify the effect of vitamin D supplementation on depressive symptoms in this age group.

Materials and methods

Study design, participants and intervention

This secondary analysis is based on a randomized, double-blind, placebo-controlled trial conducted at the Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, University Hospital Essen, Germany (LVR-Universitätsklinik Essen), between June 2016 and May 2018 [25]. The study was approved by the local Ethics Committee (No. 15 to 6363-BO), registered at the German Clinical Trials Register (DRKS00009758), and conducted in accordance with the Declaration of Helsinki. Eligible participants were in: or daycare patients aged 11.0 to 18.9 years with written informed consent who were at risk for vitamin D deficiency (serum 25(OH)D < 12 ng/mL [< 30 nmol/L]) [26] and had at least mild depressive symptoms (Beck Depression Inventory-II (BDI-II) score ≥ 14) [27].

Exclusion criteria were severe somatic or renal disease, intellectual disability (IQ < 70), established hypocalcemia, and/or parathyroid hormone level > 130 ng/mL. Screening for eligibility occurred within the first two days after admission [25]. While data from 100 participants were analyzed in the original RCT [11], only those with available CRP data (n = 92) were included in the present analysis.

Details of the intervention and randomization process are described elsewhere [25]. In short, all participants received treatment as usual (TAU) according to current clinical guidelines for their respective mental disorders [25]. Using computer-generated lists, participants were randomized (1:1) to receive either oral vitamin D₃ supplementation (2640 IU/day, i.e., 66 µg/day; verum group [VG]) or placebo (placebo group [PG]) for 28 days, starting no later than the fifth day after admission. Vitamin D₃ and placebo were provided as identical pills (Dr. B.

Scheffler Nachf. GmbH & Co. KG, Bergisch Gladbach, Germany) in containers labeled with the randomization number. Participants were instructed to take three pills each morning after breakfast. Participants discharged from psychiatric treatment at the clinic at their own request before day 28 (n = 5) were instructed to continue supplementation. Furthermore, they received weekly telephone follow-ups and attended the end-of-study examination on day 28 post-randomization.

Outcome and covariate assessments

The severity of depressive symptoms was assessed at baseline (T0) and post-intervention (T1) using the BDI-II (n = 92) [28], and the Diagnostic System for Mental Disorders in Childhood and Adolescence (DISYPS) based on self-reports (DISYPS Self; n = 92), and external assessment (DISYPS Proxy; n = 91) [29]. The BDI-II is a 21-item self-report questionnaire based on the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (DSM-IV) criteria, rated on a 0 to 3 Likert scale per item, with higher scores reflecting greater severity [28].

The DISYPS comprises 29 items (0 to 3 points each), referencing DSM-IV and the International Statistical Classification of Diseases and Related Health Problems Revision 10 (ICD-10) criteria [29]. It was conducted both as a self-assessment (DISYPS Self) and as an external assessment by guardians (DISYPS Proxy). Total scores were converted into stanine values, with scores ≥ 8 considered clinically relevant [29].

At T0 and T1, fasting blood samples were collected in the early morning from an antecubital vein into monovettes (Sarstedt, Nümbrecht, Germany), transferred to the Central Laboratory of the University Hospital Essen, and analyzed within one hour [25]. Serum 25(OH)D was measured using the Siemens ADVIA Centaur® Immunoassay-system (Siemens Healthineers, Erlangen, Germany). Serum CRP concentrations (mg/L) were determined at T0 using the Siemens Atellica® CH Analyzer (Siemens Healthineers, Erlangen, Germany). Clinical assessments at admission included assessments of body weight and height, pubertal status, and general health screening [25].

For participants under 18.25 years, the BMI standard deviation scores (SDS, kg/m2) were calculated using German national reference data according to the Lambda-Mu-Sigma (LMS) method [30].

Statistical analysis

Sample characteristics at T0 and T1 were summarized by median (P25, P75) values for continuous variables, and percentages for categorical variables (Table 1). To address potential sex-specific differences, interaction analyses (treatment × sex assigned at birth) were conducted for each outcome (BDI-II, DISYPS Self, DISYPS Proxy) separately, with CRP, treatment, and sex assigned at birth as covariates in the model. As these analyses did not provide evidence of sex interactions, data from both sexes were pooled for further analyses.

Intervention effects on depressive symptoms were analyzed separately for each outcome using analyses of covariance (ANCOVA) (Table 2). In the basic model, post-intervention values (T1) served as dependent variables, with treatment group (PG vs. VG) as the independent variable. Covariates were age, sex assigned at birth, baseline 25(OH)D concentration, and corresponding baseline values of the outcome, following the approach of Libuda et al. (2020) [11]. To examine the role of inflammation, baseline CRP concentration (T0) was added as an additional covariate (Model 1, Table 2).

In subsequent interaction analyses, CRP was tested as a potential effect modifier via treatment × CRP interaction terms with CRP and treatment as covariates in the model. These analyses aimed to determine whether treatment effects on depressive symptoms significantly differed depending on baseline CRP levels. CRP was examined using three thresholds: (1) the clinically established threshold of 5 mg/L [21], (2) the 3 mg/L threshold proposed in recent research on depression [15, 31], and (3) the sample-specific median (0.845 mg/L).

To account for multiple testing, p-values were corrected using the false discovery rate (FDR) method [32]. Afterwards, stratified analyses were conducted to assess treatment effects in CRP-defined subgroups using the thresholds defined above. All stratified models were adjusted for the covariates (Table 2). To assess the robustness of results, sensitivity analyses were conducted on four subsets: (1) participants not taking antidepressants [n = 78], due to potential effects on inflammation [33, 34], (2) participants without anorexia nervosa [n = 84], due to its impact on immunological processes [35, 36], (3) participants without potential current infection [n = 89], defined as CRP ≥ 10 mg/L [15], and (4) participants with clinically confirmed depression [n = 76], based on psychiatric assessment (data not shown).

Analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, USA). The alpha level was set to 0.05.

Results

Sample characteristics

A total of N = 92 participants aged 12.7 to 18.9 years were included in the analysis. Baseline characteristics (T0) were largely comparable between groups, but the PG had a higher proportion of females (82.2%) than the VG (74.5%) (Table 1). 15.2% were taking antidepressants (PG: 11.1%; VG: 19.1%). At T0, depressive symptomatology was pronounced in both groups, with median BDI-II scores of 28 and stanine scores of 9 on both DISYPS Self and Proxy: the highest possible score. From T0 to T1, median BDI-II scores improved slightly in both groups (VG: 28 to 25; PG: 28 to 23).

While DISYPS Self and DISYPS Proxy scores remained stable in the PG, scores in the VG showed a slight improvement (Table 1). Baseline 25(OH)D levels were similarly low across groups, with median concentrations ≤ 9 ng/mL. After intervention (T1), the VG reached median levels in the “adequate” range (≥ 20 ng/mL). For CRP, baseline concentrations were comparable between groups (PG: 0.84 mg/L; VG: 0.85 mg/L; sample-specific median: 0.845 mg/L) (Table 1). Most participants had CRP levels below the clinical threshold of 5 mg/L (91.3%; n = 84) and below the 3 mg/L threshold for low-grade inflammation (81.5%; n = 75) (Table 2).

Table 1

Characteristics of N = 92 randomized participants in placebo (n = 45) or Vitamin D supplementation group (n = 47) at baseline (T0) and post-intervention (T1)

Placebo groupVerum group
nMedian or %nMedian or %
T0T0
Age [years]4516.1 (14.8; 17.4)4716.3 (15.2; 17.1)
Females [%]3782.23574.5
Body height [cm]45164.7 (160.5; 169.4)47165.7 (162.0; 175.0)
Body weight [kg]4556.4 (50.4; 65.2)4765.0 (53.9; 75.0)
BMI-SDS [kg/m2]a440.24 (−0.54; 1.49)450.21 (−0.16; 1.79)
Antidepressants (Yes)511.1919.1
CRP [mg/L]450.84 (0.32; 2.23)470.85 (0.24; 2.19)
T0T1T0T1
25(OH)D [ng/mL]b458.5 (6.8; 10.2)9.4 (7.3; 11.5)479.0 (7.7; 10.3)24.7 (20.5; 27.6)
BDI-II (Total scores)4528 (23; 34)23 (13; 36)4728 (22; 43)25 (14; 42)
DISYPS Self (stanine)459 (8; 9)9 (8; 9)479 (8; 9)8 (8; 9)
DISYPS Proxy (stanine)449 (8; 9)9 (8; 9)479 (8; 9)8 (7; 9)

25(OH)D, 25-hydroxy-cholecalciferol; BDI-II, beck depression inventory II; BMI, body mass index; CRP; C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

Values are presented as median (P25; P75) for interval-scaled variables, and percentages for categorical variables

aBMI-SDS calculated using German reference data (LMS method) for participants < 18.25 years [30]

b1 ng/mL serum 25(OH)D equates to 2.5 nmol/L

Effect modification by baseline CRP levels

In the Basic model, a significant effect of vitamin D supplementation was observed only for DISYPS Proxy (Table 2). Adjusting for baseline CRP levels did not alter the results (Model 1, Table 2). No effect of baseline CRP was observed on either outcome. No significant interaction between treatment and CRP status at baseline was found when stratified by the threshold of 5 mg/L. In CRP-defined subgroups, no significant treatment effects were observed across all outcomes, including DISYPS Proxy.

At the 3 mg/L threshold, a statistically significant interaction was found for DISYPS Self (p = 0.047), however, this did not remain significant after FDR correction (p = 0.423). Stratified analyses by baseline CRP level indicated opposing trends between groups, with a trend toward symptom improvement in the CRP ≥ 3 mg/L subgroup receiving verum, though no effects in CRP-defined subgroups reached significance. Analyses using the sample-specific CRP median (0.845 mg/L) revealed no significant interactions for any outcome.

For DISYPS Self, the stratified results nevertheless followed a descriptively similar pattern to that observed at the 3 mg/L threshold. However, none of these subgroup effects reached significance (Table 2). These findings were consistent across sensitivity analyses (data not shown).

Table 2

ANCOVAa results for Vitamin D intervention (verum) effects on depressive symptoms: interaction and stratified analyses by multiple CRP thresholds

ModelPredictorBDI-IIDISYPS SelfDISYPS Proxy
nβ (CI)pnβ (CI)pnβ (CI)P
Basic modelaverum921.17 (−2.59; 4.93)0.5492−0.14 (−0.69; 0.40)0.5991−0.62 (−1.22; −0.03)0.04
Model 1bverum921.46 (−2.35; 5.26)0.4592−0.11 (−0.66; 0.45)0.7091−0.61 (−1.21; −0.01)0.047
CRP−0.29 (−0.89; 0.31)0.34−0.04 (−0.13; 0.04)0.32−0.02 (−0.11; 0.08)0.32
Basic model stratified by CRP ≥ 5.0 mg/Lc
CRP ≥ 5.0 mg/Laverum81.90 (−13.84; 17.64)0.738−1.28 (−4.24; 1.68)0.268−0.45 (−3.37; 2.48)0.66
CRP < 5.0 mg/Laverum841.24 (−2.79; 5.27)0.5484−0.07 (−0.66; 0.53)0.8283−0.64 (−1.29; 0.01)0.05
p value for interactione = 0.79p value for interactione = 0.37p value for interactione = 0.74
Basic model stratified by CRP ≥ 3.0 mg/Ld
CRP ≥ 3.0 mg/Laverum172.34 (−2.34; 7.02)0.3017−0.76 (−2.10; 0.58)0.2417−1.25 (−3.36; 0.85)0.22
CRP < 3.0 mg/Laverum751.73 (−2.57; 6.03)0.42750.10 (−0.49; 0.69)0.7374−0.45 (−1.07; 0.18)0.16
p value for interactione = 0.26p value for interactione = 0.047fp value for interactione = 0.13
Basic model stratified by CRP median
CRP ≥ 0.845 mg/Laverum46−0.01 (−5.29; 5.27)0.9946−0.29 (−1.15; 0.56)0.4946−0.78 (−1.56; 0.003)0.05
CRP < 0.845 mg/Laverum463.88 (−2.18; 9.95)0.20460.23 (−0.50; 0.95)0.5345−0.45 (−1.52; 0.61)0.40
p value for interactione = 0.34p value for interactione = 0.53p value for interactione = 0.70

BDI-II, beck depression inventory II; CRP, C-reactive protein; DISYPS Self, diagnostic system for mental disorders in childhood and adolescence based on self-assessment; DISYPS Proxy, DISYPS based on external assessment; n, number of participants

aBasic model adjusted for age, sex assigned at birth, baseline 25(OH)D, and baseline value of the respective outcome

bModel 1: Basic model further adjusted for CRP

cThreshold of 5 mg/L, as clinically established [21]

dThreshold of 3 mg/L, as proposed in recent depression research [15, 31]

ep value for the interaction term (treatment × CRP) with CRP and treatment as covariates

fFalse discovery rate: adjusted p value = 0.423

Discussion

To our knowledge, this is the first study to examine whether baseline inflammation modifies the effect of vitamin D supplementation on depressive symptoms in children and adolescents. Among 92 participants with vitamin D levels in the deficiency-risk range and at least mild depression, 81.5% had CRP levels below the threshold for low-grade inflammation (3 mg/L). No consistent interaction between CRP status and treatment effects emerged across outcomes and CRP cut-offs. An interaction for DISYPS Self at the 3 mg/L threshold was nominally significant but did not remain after FDR correction.

Depressive symptoms were assessed using three instruments, with only marginal reductions observed across all evaluation tools in both groups. These minimal changes may reflect the short intervention period, as previous studies suggest that four weeks may be insufficient to induce substantial neurobiological or antidepressant effects, with more improvements typically observed in longer trials [37]. However, adequate vitamin D levels (≥ 20 ng/mL) were reached in this short-time intervention in a population initially at risk for deficiency and DISYPS Proxy indicated a significant effect with adjunctive vitamin D treatment within this relatively short-term intervention.

Thus, the study offers potential for preliminary insights into the relationship between vitamin D and depression.

Considering results from a previous meta-analysis [14], we defined low-grade inflammation as CRP ≥ 3 mg/L. Using this threshold, Osimo et al. (2019) reported elevated odds of low-grade inflammation in depressed patients (OR = 1.46). Furthermore, this meta-analysis showed that mildly elevated CRP ≥ 1 mg/L could be a relevant threshold in the etiology of depression as well [15]. Studies have shown elevated CRP levels in a subset of individuals with depression [15, 16, 18], particularly in those with treatment-resistant depression [17].

In contrast, despite only marginal reductions in depressive symptoms following the intervention, which included treatment as usual in both groups and suggests treatment resistance, we observed notably low inflammation levels in our cohort, and no effect of CRP on either outcome was observed. Only 18.5% exceeded CRP ≥ 3 mg/L, and 45.7% exceeded ≥ 1 mg/L, compared with the 27% and 58% reported by Osimo et al. in adults [15]. These lower rates in our adolescent cohort underscore that inflammation is not a universal feature of depression, particularly in youth, and suggest that alternative mechanisms may be more relevant in this age group.

Further research is needed to clarify the role of inflammation across developmental stages and depressive subtypes [38]. The low overall CRP levels and limited variance in our sample may have reduced the ability to detect modifying effects. Furthermore, given that anti-inflammatory treatments appear effective mainly in patients with elevated inflammation [15], our sample with a predominantly subclinical inflammatory profile may have been inappropriate to elicit more obvious treatment effects of vitamin D. Future vitamin D interventions might focus on depressed individuals with CRP ≥ 3 mg/L, as it is already being done in current RCTs on pharmacotherapy [39].

Regarding the role of inflammation on the effect of vitamin D supplementation, adjusting for CRP did not alter the overall treatment effects, suggesting that inflammation, as measured by CRP, does not influence vitamin D efficacy in this cohort. Although different CRP cut-offs for inflammation have been proposed, including CRP > 5 mg/L [21], CRP ≥ 3 mg/L, and CRP ≥ 1 mg/L [15] (with the latter being close to our sample-specific median of 0.845 mg/L), none of these thresholds revealed interactions with treatment effects.

Additionally, an interaction test with CRP as a continuous variable also showed no significant results, reinforcing that baseline inflammation does not appear to modify the effect of vitamin D supplementation. Stratified analyses showed a slight reduction in depressive symptoms in participants with CRP ≥ 3 mg/L in the DISYPS Self (Δ = −0.76), while those of participants with lower CRP remained unchanged. A similar pattern was observed at the median CRP threshold. While these findings may support the hypothesis that individuals with higher CRP levels benefit more from vitamin D supplementation, it must be emphasized that none of the subgroup effects reached significance.

These findings are consistent with those of Kaviani et al. (2022), who observed BDI-II reductions after 8 weeks of high-dose vitamin D supplementation in adults despite lacking changes in inflammatory markers [40]. Similar to our cohort, baseline CRP levels in this study were low, indicating no elevated inflammatory status prior to the intervention. Although they did not formally test interaction effects, their results suggest that the antidepressant effects of vitamin D may occur independently of inflammation [40].

Likewise, Dogan-Sander et al. (2021) examined the relationship between inflammation and the effect of vitamin D on depressive symptomatology in a cohort of adults, with similarly low baseline CRP levels. Despite observing a correlation between low vitamin D levels and increased depressive symptoms, no moderating effect of inflammation was found [41], supporting our suggestion that vitamin D effects on depression may be independent of baseline inflammation.

Overall, our results suggest that CRP as a proxy for inflammation, plays a limited role in modulating vitamin D’s antidepressant effects in the presented cohort. This challenges the hypothesis that anti-inflammatory effects explain vitamin D effect on depression, at least in populations with low grade inflammation. While the question of defining the optimal CRP threshold remains unresolved, the absence of significant interactions suggests that other mechanisms of vitamin D may be more relevant in adolescents with non-elevated inflammation. These could include the antioxidative and neuromodulatory properties of vitamin D, as well as its role in serotonin synthesis and receptor modulation, as highlighted in previous mainly preclinical studies [24].

However, it remains unclear to what extent these mechanisms translate to clinical effects in children and adolescents, and whether mechanisms of action differ from those observed in adults [24].

Strength and limitations

Recruitment from a clinical setting enhances internal validity, though generalizability is limited, as only about 20% of children with mental disorders receive psychiatric care [42Not reported44]. However, this clinical setting allows for a more homogenous and well-characterized sample. All participants met the criteria for at least mild depression [27, 29], and baseline vitamin D levels indicated deficiency risk [26]. Information on acute infections or inflammatory conditions at the time of assessment was not systematically collected.

However, following Osimo et al. (2019), CRP ≥ 10 mg/L was used as an indicator of potential current infection [15]. Only n = 3 participants exceeded this threshold, and a sensitivity analysis excluding these cases did not alter the study results (data not shown).

Depressive symptoms were assessed using three instruments, including self: and proxy-rated ratings [28, 29]. Discrepancies between self: and proxy-reports are common in child and adolescent psychiatry [45Not reported47], as parents often emphasize symptom severity while adolescents tend to under-report symptoms [48]. This may lead to inconsistencies and potential misclassification. In our study, only parent-reported (DISYPS Proxy) symptoms showed significant effects in the unstratified analyses, whereas a nominal effect emerged for self-reported DISYPS in the interaction analyses, but not for the BDI-II.

To account for such discrepancies, we followed best practices in youth mental health through a multi-informant approach [45, 46]. As the applied instruments assess symptom severity rather than diagnostic status [28, 29], depression was additionally verified by clinical evaluation, which classified 76 of 92 participants as depressed. A sensitivity analysis restricted to these 76 participants did not change the results (data not shown).

Focusing solely on CRP limits interpretation, as it reflects only part of the immune response. Additional humoral and cellular markers, such as cytokines (e.g., TNF-α, IFN-γ, IL-1β, IL-6, IL-8, IL-10) and immune cell subsets, as proposed in other studies exploring the etiology of depression [16, 18, 49Not reported53], may capture distinct pathways that CRP alone cannot represent. These markers could provide a more nuanced understanding of the immune mechanisms underlying adolescent depression and how inflammation influences depressive symptoms and treatment response.

While additional markers could enhance the analysis, CRP remains a valid and well-supported choice for investigating inflammation in depression, with associations to depression severity [16, 18], and well-established cut-offs [15, 31]. CRP was only analyzed at baseline, limiting insights into temporal dynamics. While this was sufficient for examining the potential modulating effect of inflammation on vitamin D supplementation, repeated assessments could provide a deeper understanding of how inflammation develops over time following vitamin D treatment.

Given evidence that depression and inflammation may influence each other during childhood and adolescence [18], future studies incorporating multiple time points could clarify how these dynamics impact the long-term effects of vitamin D supplementation on depression.

While sex-specific patterns in the inflammation: depression link have been suggested [54], other studies report no association between CRP levels and sex or gender [15]. In our study, no interaction between sex and treatment was observed. Yet, the overall sample size was too limited to allow for meaningful sex-stratified analyses. Such stratification would have further reduced statistical power and reliability. Nonetheless, we adjusted for sex in all ANCOVA models. The relatively small sample size, particularly in inflammation-defined subgroups (e.g., n = 8 for CRP ≥ 5 mg/L), limited statistical power and may have obscured subtle effects.

While this constraint must be acknowledged, it reflects the practical challenges of recruiting clinically well-characterized adolescents for biological studies. Nonetheless, the present findings provide a valuable foundation for future trials, highlighting key methodological aspects: such as adequate stratification by inflammatory status and longer intervention durations: that can guide more targeted study designs.

Conclusions

Overall, our findings do not support the hypothesis that the antidepressant effect of vitamin D supplementation in children and adolescents depends on CRP status: at least within a sample characterized by non-elevated inflammatory levels. Larger vitamin D RCTs with longer intervention periods are needed to investigate the role of more pronounced inflammation, ideally assessed using a broader range of inflammatory markers.

Manuel Föcker and Lars Libuda have contributed equally to this work.

Acknowledgements

The authors thank the participants and their parents for their efforts and the clinic and laboratory personnel for their support in conducting the study.

Author contributions

LS: conceptualization, formal analysis, visualization, writing: original draft, writing: review and editing. NJ: writing: review and editing, supervision. JB, EH, RD, NT, HE, RH, DZ, DFS, CG, TP, AH, JA, and JH: critical input for manuscript. MF: conceptualization and critical input for manuscript. LL: conceptualization, writing: review and editing, supervision. All authors have read and agreed to the published version of the manuscript.

Funding

Open Access funding enabled and organized by Projekt DEAL.

Data Availability

No data was used for the research described in the article.

Declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest. Dr. B. Scheffler Nachf. GmbH & Co. KG provided vitamin D supplements as well as placebo free of charge. The company was neither involved in designing and conducting the study nor in data analysis, interpretation and manuscript preparation.

References

1. Malhi GS, Mann JJ (2018) Depression. Lancet 392:2299 to 2312. 10.1016/S0140-6736(18)31948-2 30396512 10.1016/S0140-6736(18)31948-2
2. Lim GY, Tam WW, Lu Y et al (2018) Prevalence of depression in the community from 30 countries between 1994 and 2014. Sci Rep 8:2861. 10.1038/s41598-018-21243-x 29434331 10.1038/s41598-018-21243-x PMC5809481
3. Birmaher B, Arbelaez C, Brent D (2002) Course and outcome of child and adolescent major depressive disorder. Child Adolesc Psychiatr Clin N Am 11:619 to 637. 10.1016/S1056-4993(02)00011-1 12222086 10.1016/s1056-4993(02)00011-1
4. Kovacs M (1996) Presentation and course of major depressive disorder during childhood and later years of the life span. J Am Acad Child Adolesc Psychiatry 35:705 to 715 8682751 10.1097/00004583-199606000-00010
5. Keenan-Miller D, Hammen CL, Brennan PA (2007) Health outcomes related to early adolescent depression. J Adolesc Health 41:256 to 262. 10.1016/j.jadohealth.2007.03.015 17707295 10.1016/j.jadohealth.2007.03.015 PMC2034364
6. Markkula N, Härkänen T, Perälä J et al (2012) Mortality in people with depressive, anxiety and alcohol use disorders in Finland. Br J Psychiatry 200:143 to 149. 10.1192/bjp.bp.111.094904 22157799 10.1192/bjp.bp.111.094904
7. Osby U, Brandt L, Correia N et al (2001) Excess mortality in bipolar and unipolar disorder in Sweden. Arch Gen Psychiatry 58:844 to 850. 10.1001/archpsyc.58.9.844 11545667 10.1001/archpsyc.58.9.844
8. Kessler RC, Berglund P, Demler O et al (2005) Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the national comorbidity survey replication. Arch Gen Psychiatry 62:593 to 602. 10.1001/archpsyc.62.6.593 15939837 10.1001/archpsyc.62.6.593
9. Vellekkatt F, Menon V (2019) Efficacy of vitamin D supplementation in major depression: a meta-analysis of randomized controlled trials. J Postgrad Med 65:74 to 80. 10.4103/jpgm.JPGM_571_17 29943744 10.4103/jpgm.JPGM_571_17 PMC6515787
10. Xie F, Huang T, Lou D et al (2022) Effect of vitamin D supplementation on the incidence and prognosis of depression: an updated meta-analysis based on randomized controlled trials. Front Public Health 10:903547. 10.3389/fpubh.2022.903547 35979473 10.3389/fpubh.2022.903547 PMC9376678
11. Libuda L, Timmesfeld N, Antel J et al (2020) Effect of vitamin D deficiency on depressive symptoms in child and adolescent psychiatric patients: results of a randomized controlled trial. Eur J Nutr 59:3415 to 3424. 10.1007/s00394-020-02176-6 32108263 10.1007/s00394-020-02176-6 PMC7669774
12. Goldsmith DR, Rapaport MH, Miller BJ (2016) A meta-analysis of blood cytokine network alterations in psychiatric patients: comparisons between schizophrenia, bipolar disorder and depression. Mol Psychiatry 21:1696 to 1709. 10.1038/mp.2016.3 26903267 10.1038/mp.2016.3 PMC6056174
13. Engler H, Brendt P, Wischermann J et al (2017) Selective increase of cerebrospinal fluid IL-6 during experimental systemic inflammation in humans: association with depressive symptoms. Mol Psychiatry 22:1448 to 1454. 10.1038/mp.2016.264 28138158 10.1038/mp.2016.264
14. Haan P, Klein HC, Hart BA (2017) Autoimmune aspects of neurodegenerative and psychiatric diseases: a template for innovative therapy. Front Psychiatry 8:46. 10.3389/fpsyt.2017.00046 28421005 10.3389/fpsyt.2017.00046 PMC5378775
15. Osimo EF, Baxter LJ, Lewis G et al (2019) Prevalence of low-grade inflammation in depression: a systematic review and meta-analysis of CRP levels. Psychol Med 49:1958 to 1970. 10.1017/S0033291719001454 31258105 10.1017/S0033291719001454 PMC6712955
16. Osimo EF, Pillinger T, Rodriguez IM et al (2020) Inflammatory markers in depression: a meta-analysis of mean differences and variability in 5,166 patients and 5,083 controls. Brain Behav Immun 87:901 to 909. 10.1016/j.bbi.2020.02.010 32113908 10.1016/j.bbi.2020.02.010 PMC7327519
17. Chamberlain SR, Cavanagh J, de Boer P et al (2019) Treatment-resistant depression and peripheral C-reactive protein. Br J Psychiatry 214:11 to 19. 10.1192/bjp.2018.66 29764522 10.1192/bjp.2018.66 PMC6124647
18. Colasanto M, Madigan S, Korczak DJ (2020) Depression and inflammation among children and adolescents: a meta-analysis. J Affect Disord 277:940 to 948. 10.1016/j.jad.2020.09.025 33065836 10.1016/j.jad.2020.09.025
19. Zalli A, Jovanova O, Hoogendijk WJG et al (2016) Low-grade inflammation predicts persistence of depressive symptoms. Psychopharmacology 233:1669 to 1678. 10.1007/s00213-015-3919-9 25877654 10.1007/s00213-015-3919-9 PMC4828485
20. Wium-Andersen MK, Ørsted DD, Nielsen SF et al (2013) Elevated C-reactive protein levels, psychological distress, and depression in 73, 131 individuals. JAMA Psychiat 70:176 to 184. 10.1001/2013.jamapsychiatry.102 10.1001/2013.jamapsychiatry.102 23266538
21. Oremek GM, Holzgreve F, Wanke EM et al (2024) C-reaktives protein (CRP): ein diagnostischer Entzündungsmarker am Beispiel ausgewählter Indikationen. Zbl Arbeitsmed 74:140 to 144. 10.1007/s40664-023-00523-y 10.1007/s40664-023-00523-y
22. Pearson TA, Mensah GA, Alexander RW et al (2003) Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the centers for disease control and prevention and the American heart association. Circulation 107:499 to 511. 10.1161/01.CIR.0000052939.59093.45 12551878 10.1161/01.cir.0000052939.59093.45
23. Jamilian H, Amirani E, Milajerdi A et al (2019) The effects of vitamin D supplementation on mental health, and biomarkers of inflammation and oxidative stress in patients with psychiatric disorders: a systematic review and meta-analysis of randomized controlled trials. Prog Neuropsychopharmacol Biol Psychiatry 94:109651. 10.1016/j.pnpbp.2019.109651 31095994 10.1016/j.pnpbp.2019.109651
24. Kouba BR, Camargo A, Gil-Mohapel J et al (2022) Molecular basis underlying the therapeutic potential of vitamin D for the treatment of depression and anxiety. Int J Mol Sci. 10.3390/ijms23137077 35806075 10.3390/ijms23137077 PMC9266859
25. Föcker M, Antel J, Grasemann C et al (2018) Effect of an vitamin D deficiency on depressive symptoms in child and adolescent psychiatric patients: a randomized controlled trial: study protocol. BMC Psychiatry 18:57. 10.1186/s12888-018-1637-7 29490621 10.1186/s12888-018-1637-7 PMC5831612
26. Ross AC, Taylor CL, Yaktine AL et al. (2011) Dietary reference intakes for calcium and vitamin D. 10.17226/13050 21796828
27. Kumar G, Steer RA, Teitelman KB et al (2002) Effectiveness of beck depression inventory-II subscales in screening for major depressive disorders in adolescent psychiatric inpatients. Assessment 9:164 to 170. 10.1177/10791102009002007 12066831 10.1177/10791102009002007
28. Hautzinger M, Keller F, Kühner C. (2006) Beck depressions-inventar (BDI-II). BDI II. Revision. Harcourt Test Services, Frankfurt am Main.
29. Döpfner M, Görtz-Dorten A, Lehmkuhl G (2008) DISYPS-II Diagnostik-system für psychische Störungen nach ICD-10 und DSM-IV für Kinder und Jugendliche: II. Hans Huber, Bern
30. Kromeyer-Hauschild K, Wabitsch M, Kunze D et al (2001) Perzentile für den body-mass-index für das Kindes: und Jugendalter unter Heranziehung verschiedener deutscher Stichproben. Monatsschr Kinderheilkd 149:807 to 818. 10.1007/s001120170107 10.1007/s001120170107
31. Orsolini L, Pompili S, Tempia Valenta S et al (2022) C-Reactive protein as a biomarker for major depressive disorder? Int J Mol Sci. 10.3390/ijms23031616 35163538 10.3390/ijms23031616 PMC8836046
32. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Ser B Stat Methodol 57:289 to 300. 10.1111/j.2517-6161.1995.tb02031.x 10.1111/j.2517-6161.1995.tb02031.x
33. García-García ML, Tovilla-Zárate CA, Villar-Soto M et al (2022) Fluoxetine modulates the pro-inflammatory process of IL-6, IL-1β and TNF-α levels in individuals with depression: a systematic review and meta-analysis. Psychiatry Res 307:114317. 10.1016/j.psychres.2021.114317 34864233 10.1016/j.psychres.2021.114317
34. Fernandes BM, Scotti-Muzzi E, Soeiro-de-Souza MG (2022) Effects of antidepressant drug therapy with or without physical exercise on inflammatory biomarkers in major depressive disorder: a systematic review and meta-analysis of randomized controlled trials. Eur J Clin Pharmacol 78:339 to 349. 10.1007/s00228-021-03240-8 34708271 10.1007/s00228-021-03240-8
35. Dalton B, Bartholdy S, Robinson L et al (2018) A meta-analysis of cytokine concentrations in eating disorders. J Psychiatr Res 103:252 to 264. 10.1016/j.jpsychires.2018.06.002 29906710 10.1016/j.jpsychires.2018.06.002
36. Xu C, Mutwalli H, Haslam R et al (2025) C-reactive protein (CRP) levels in people with eating disorders: a systematic review and meta-analysis. J Psychiatr Res 181:653 to 662. 10.1016/j.jpsychires.2024.12.039 39742796 10.1016/j.jpsychires.2024.12.039
37. Gowda U, Mutowo MP, Smith BJ et al (2015) Vitamin D supplementation to reduce depression in adults: meta-analysis of randomized controlled trials. Nutrition 31:421 to 429. 10.1016/j.nut.2014.06.017 25701329 10.1016/j.nut.2014.06.017
38. Khandaker GM, Dantzer R, Jones PB (2017) Immunopsychiatry: important facts. Psychol Med 47:2229 to 2237. 10.1017/S0033291717000745 28418288 10.1017/S0033291717000745 PMC5817424
39. Khandaker GM, Oltean BP, Kaser M et al (2018) Protocol for the insight study: a randomised controlled trial of single-dose tocilizumab in patients with depression and low-grade inflammation. BMJ Open 8:e025333. 10.1136/bmjopen-2018-025333 30244217 10.1136/bmjopen-2018-025333 PMC6157523
40. Kaviani M, Nikooyeh B, Etesam F et al (2022) Effects of vitamin D supplementation on depression and some selected pro-inflammatory biomarkers: a double-blind randomized clinical trial. BMC Psychiatry 22:694. 10.1186/s12888-022-04305-3 36368945 10.1186/s12888-022-04305-3 PMC9651884
41. Dogan-Sander E, Mergl R, Willenberg A et al (2021) Inflammation and the association of vitamin D and depressive symptomatology. Nutrients. 10.3390/nu13061972 34201276 10.3390/nu13061972 PMC8229745
42. Boyle MH, Duncan L, Georgiades K et al (2019) Tracking children’s mental health in the 21st century: lessons from the 2014 OCHS. Can J Psychiatry 64:232 to 236. 10.1177/0706743719830025 30978140 10.1177/0706743719830025 PMC6463360
43. Georgiades K, Duncan L, Wang L et al (2019) Six-month prevalence of mental disorders and service contacts among children and youth in Ontario: evidence from the 2014 Ontario Child Health Study. Can J Psychiatry 64:246 to 255. 10.1177/0706743719830024 30978138 10.1177/0706743719830024 PMC6463361
44. Merikangas KR, He J-P, Burstein M et al (2010) Lifetime prevalence of mental disorders in U.S. adolescents: results from the national comorbidity survey replication-adolescent supplement (NCS-A). J Am Acad Child Adolesc Psychiatry 49:980 to 989. 10.1016/j.jaac.2010.05.017 20855043 10.1016/j.jaac.2010.05.017 PMC2946114
45. Los Reyes A, Augenstein TM, Wang M et al (2015) The validity of the multi-informant approach to assessing child and adolescent mental health. Psychol Bull 141:858 to 900. 10.1037/a0038498 25915035 10.1037/a0038498 PMC4486608
46. Martel MM, Markon K, Smith GT (2017) Research review: multi-informant integration in child and adolescent psychopathology diagnosis. J Child Psychol Psychiatry 58:116 to 128. 10.1111/jcpp.12611 27492280 10.1111/jcpp.12611 PMC5247337
47. Meister R, Abbas M, Antel J et al (2020) Placebo response rates and potential modifiers in double-blind randomized controlled trials of second and newer generation antidepressants for major depressive disorder in children and adolescents: a systematic review and meta-regression analysis. Eur Child Adolesc Psychiatry 29:253 to 273. 10.1007/s00787-018-1244-7 30535589 10.1007/s00787-018-1244-7 PMC7056684
48. Salbach-Andrae H, Klinkowski N, Lenz K et al (2009) Agreement between youth-reported and parent-reported psychopathology in a referred sample. Eur Child Adolesc Psychiatry 18:136 to 143. 10.1007/s00787-008-0710-z 19129966 10.1007/s00787-008-0710-z
49. Harsanyi S, Kupcova I, Danisovic L et al (2022) Selected biomarkers of depression: what are the effects of cytokines and inflammation? Int J Mol Sci. 10.3390/ijms24010578 36614020 10.3390/ijms24010578 PMC9820159
50. Beurel E, Toups M, Nemeroff CB (2020) The bidirectional relationship of depression and inflammation: double trouble. Neuron 107:234 to 256. 10.1016/j.neuron.2020.06.002 32553197 10.1016/j.neuron.2020.06.002 PMC7381373
51. Kelly KM, Smith JA, Mezuk B (2021) Depression and interleukin-6 signaling: a mendelian randomization study. Brain Behav Immun 95:106 to 114. 10.1016/j.bbi.2021.02.019 33631287 10.1016/j.bbi.2021.02.019 PMC11081733
52. Perry BI, Upthegrove R, Kappelmann N et al (2021) Associations of immunological proteins/traits with schizophrenia, major depression and bipolar disorder: a bi-directional two-sample mendelian randomization study. Brain Behav Immun 97:176 to 185. 10.1016/j.bbi.2021.07.009 34280516 10.1016/j.bbi.2021.07.009 PMC7612947
53. Sørensen NV, Frandsen BH, Orlovska-Waast S et al (2023) Immune cell composition in unipolar depression: a comprehensive systematic review and meta-analysis. Mol Psychiatry 28:391 to 401. 10.1038/s41380-022-01905-z 36517638 10.1038/s41380-022-01905-z
54. Jarkas DA, Villeneuve AH, Daneshmend AZB et al (2024) Sex differences in the inflammation-depression link: a systematic review and meta-analysis. Brain Behav Immun 121:257 to 268. 10.1016/j.bbi.2024.07.037 39089535 10.1016/j.bbi.2024.07.037
Publication and source details

Original title: Does inflammation moderate the effect of vitamin D supplementation on depressive symptoms? Results of an interventional study in children and adolescents

Laura Schlarbaum; Nicole Jankovic; Judith Bühlmeier; Eva Hohoff; Rhea Dankers; Harald Engler; Raphael Hirtz; Denise Zwanziger; Corinna Grasemann; Triinu Peters; Anke Hinney; Jochen Antel; Manuel Föcker; Lars Libuda

European Journal of Nutrition · 2026

© The Author(s) 20262026https://creativecommons.org/licenses/by/4.0/Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.Universität Paderborn (3159)

Open Access funding enabled and organized by Projekt DEAL.

Repository JATS snapshot; version as supplied

Source retrieved: 2026-11-10T16:30:57.023952+00:00

DOI: 10.1007/s00394-026-03920-0 · PMC13038648