Conceived and designed the experiments: AJD MAB DWE DK JJM. Performed the experiments: DK TH WMJP. Analyzed the data: AJD TH WMJP JJM. Contributed reagents/materials/analysis tools: MAB DWE DK JJM. Wrote the paper: AJD MAB DWE DK TH WMJP JJM.
The authors have declared that no competing interests exist.
Epidemiological research links vitamin D status to various brain-related outcomes. However, few trials examine whether supplementation can improve such outcomes and none have examined effects on cognition. This study examined whether Vitamin D supplementation led to improvements in diverse measures of cognitive and emotional functioning, and hypothesised that supplementation would lead to improvements in these outcomes compared to placebo.
Healthy young adults were recruited to a parallel-arm, double-blind trial conducted at The University of Queensland. Participants were randomly allocated to receive Vitamin D (one capsule daily, containing 5000 IU cholecalciferol) or identical placebo capsule for six weeks. All participants and outcome assessors were blinded to group assignment. Primary outcome measures assessed at baseline and 6 weeks were working memory, response inhibition and cognitive flexibility. Secondary outcomes were: hallucination-proneness, psychotic-like experiences, and ratings of depression, anxiety and anger. 128 participants were recruited, randomised and included in primary analyses (vitamin D n = 63; placebo n = 65). Despite significant increases in vitamin D status in the active group, no significant changes were observed in working memory (F = 1.09;
Our findings indicate that vitamin D supplementation does not influence cognitive or emotional functioning in healthy young adults. Future controlled trials in targeted populations of interest are required to determine whether supplementation can improve functioning in these domains.
Australian and New Zealand Clinical Trials Registry;
Vitamin D is a nuclear steroid hormone with diverse physiological roles. In addition to its well-established role in calcium homeostasis, vitamin D is being reconsidered as a neuroactive steroid
Epidemiological studies link vitamin D status to a range of brain-related outcomes. Low concentrations of vitamin D have been associated with impairments in cognitive functions such as memory and orientation
In response to such findings, The Institute of Medicine
Participants were healthy volunteers recruited through advertising at the University of Queensland. Inclusion criteria were at least 18 years of age, with sufficient English language skills required to complete the study protocol. Individuals were excluded if they met any of the following criteria: current use of vitamin D or calcium supplements; history of adverse reactions to vitamin supplements; current or past diagnosis of a mood or psychotic disorders; history of neurologic illnesses including cerebrovascular accident, CNS tumours, head trauma, multiple sclerosis, epilepsy, movement disorders or migraine treatment; current or recent (1 year) history of dependence on alcohol or illicit substances; intellectual disability; pregnancy or current breast feeding, or potential to become pregnant during the trial; history of severe renal impairment. After telephone screening, potentially eligible participants were invited to attend the research clinic for assessment. After they had been provided with verbal and written information about the study (including the patient information sheet), eligible participants provided written informed consent. No changes to eligibility criteria or other methods were made after trial commencement. The trial was approved by the University of Queensland Medical Research Ethics Committee, and pre-registered with the Australian and New Zealand Clinical Trials Registry (ACTRN12610000318088). The study was funded by Queensland Health. The protocol for this trial and supporting CONSORT checklist are available as supporting information; see
This study was a randomised controlled, parallel-arms trial, comparing vitamin D supplementation to placebo (1∶1 allocation ratio).
Participants were randomly assigned to receive either Vitamin D (5000IU cholecalciferol) or placebo. These were provided as identical microcellulose capsules, prepared by an external clinical trials service. Placebo capsules contained lactose. Participants were provided with 6 weeks of study medication, and instructed to take one capsule daily. To optimise adherence, participants were sent weekly reminders via email or text message.
Randomisation sequence was generated by the external clinical trials site. To ensure that each treatment group was uniformly represented over the time course of the study, a varying-block randomisation protocol was used (randomly determined block sizes of 4 or 6). After provision of written consent, each participant was assigned to the next consecutive participant number by two researchers not involved in the generating the randomisation sequence. All investigators, outcome assessors and participants were blinded to treatment allocation procedures and treatment group throughout the study. After completion of the final assessment, participants were asked to guess whether they had received vitamin D or placebo based on their overall subjective impression of change during the study.
The following assessments were conducted at baseline and after 6 weeks of vitamin D supplementation. The cognitive measures were selected based on their limited capacity to generate ceiling effects in healthy populations.
The N-Back task is a widely used computer-based test of visuospatial working memory
Response inhibition is a specific executive function which involves suppression of behavioural impulses and is measured using tasks such as the Stop Signal Task
This is a computer-based measure that requires alternating between responses to different categories of stimuli
The Peters Delusion Inventory-21 (PDI-21) is a 21-item self-report measure of delusional-like experiences
The white noise task involves presenting participants with one of three types of auditory stimuli via headphones: (a) white noise only; (b) white noise and audible speech of neutral content; and (c) white noise and barely audible speech of neutral content. Twenty-five fragments of each group were presented in random order. Participants were asked to indicate whether they had heard a voice or not. The primary outcome for analysis was the number of times participants indicated hearing a voice in the white noise only condition. This measure of speech illusion is thought to indicate proneness to psychosis, and higher scores have been associated with schizotypy compared to controls
The Beck Depression Inventory (BDI) is a 21-item self-report questionnaire with item scores ranging from 0 to 3 and a total score ranging from 0 to 63. It has been verified as a reliable and valid screening instrument to detect intensity of depression in a variety of populations and has also been employed to measure treatment response when used in a pre- and post-test design
The State-Trait Anxiety Inventory (STAI) is a 40-item self-report questionnaire used to measure both current anxiety (state) (20 items) and anxiety as a more enduring stable personality characteristic (trait) (20 items). In this study we utilised the State Anxiety subscale
The State-Trait Anger Expression Inventory (STAXI-2) is a 57-item self report questionnaire which measures the experience, expression and control of anger. In this study, the key outcome was the State Anger subscale
The Treatment Emergent Symptom Scale is a 31-item self-report questionnaire widely used in psychopharmacology trials
Blood samples were collected as finger prick capillary whole blood spots. These were collected at baseline and at the 6-week follow-up using a single-use, automated lancet device. Vitamin D status is internationally quantified as 25-hydroxyvitamin D3 (25OHD3). 25OHD was extracted from the dried blood spots, and assayed using a highly sensitive tandem mass spectroscopy assay optimised for these samples
For the primary two-group comparisons, a sample size of approximately 128 will have a power of 80% to detect a small-medium effect size (equivalent to
Treatment outcomes were analysed using mixed-effects modelling. A mixed effects model was created for each outcome variable. Primary analysis examined the effect of treatment allocation group (vitamin D versus placebo) on treatment outcome, with time and treatment group entered as fixed effects. The adequacy of each model was assessed by examining residuals for heterogeneity and normality. Primary analysis included all randomised participants (intention to treat analysis). Two secondary analyses were conducted:
The effect of change in vitamin D concentrations on treatment outcome. Two groups were determined based on the mean change in concentrations of 25OHD3; change in vitamin D group and time were entered as fixed factors in mixed models.
The effect of change in vitamin D concentrations on treatment outcome, in the subset of individuals who had lower serum concentrations of 25OHD3 at baseline. Vitamin D deficiency is often defined by serum concentrations of 25OHD3 less than 50 nmol/L. However, results indicated that too few participants met this criterion. As such, for this analysis, participants were classified
Trial recruitment began in February 2010 and was follow-up was completed in September 2010 after successful recruitment of the target sample size. One hundred and thirty nine individuals contacted the recruitment team and were screened for eligibility. Eight individuals declined to participate (not available for the follow-up session), and three were excluded (one had an acquired brain injury and two were taking psychotropic medication). This left 128 eligible participants who consented to participate. Sixty-three were randomly allocated to vitamin D capsules, and 65 were allocated to placebo capsules. Only one participant was lost to follow-up. All participants were included in analysis of primary outcomes. Participant screening and flow is described in
The mean age of participants was 21.8 years (s.d. = 2.9; Range 18–30), and more than half were female (57%; 73/128).
Vitamin D | Placebo | Statistic | |
(n = 63) | (n = 65) | ||
Age (years) | 21.45 (2.96) | 22.06 (2.74) | t = 1.21 |
Gender (% female) | 61.9% (39/63) | 52.3% (34/65) | χ2 = 1.20 |
Ethnicity | European 41.9% (26/62) | European 33.8% (22/65) | χ2 = 1.33 |
Asian 45.2% (28/62) | Asian 55.4% (36/65) | ||
Other 12.9% (8/62) | Other 10.8% (7/65) | ||
Outdoor time (hours/week) | 16.95 (4.19) | 16.46 (5.44) | t = −0.57 |
25OHD3 (nmol/L) | 76.25 (19.63) | 77.23 (20.95) | t = 0.26 |
Working memory (Correct hits) | 0.57 (0.18) | 0.55 (0.20) | t = −0.61 |
Response inhibition (SSRT) | 210.42 (40.77) | 207.15 (39.22) | t = −0.45 |
Cognitive flexibility (ms) | 246.34 (166.29) | 256.93 (185.25) | t = 0.34 |
PDI total score | 6.11 (3.05) | 7.45 (4.33) | t = 2.02 |
White noise task (count) | 0.29 (0.73) | 0.31 (0.81) | t = 0.20 |
BDI | 7.24 (7.82) | 5.72 (5.56) | t = −1.27 |
State anxiety | 36.29 (10.10) | 34.15 (8.31) | t = −1.31 |
State anger | 16.41 (3.95) | 16.26 (3.76) | t = −.22 |
*p<0.05.
In all participants, mean concentrations of 25OHD3 at baseline were 76.6 nmol/L (SD 19.9; Range 41.1–149.3 nmol/L; median 75.0 nmol/L). Only ten participants had baseline concentrations lower than 50 nmol/L, the cut-off typically used for insufficiency
There was a significant difference between treatment groups with regard to changes in vitamin D status over time (F = 21.44; p<0.001;
Vitamin D | Placebo | |||||||
Mean (SE) | 95% CI (mean) | Mean (SE) | 95% CI (mean) | F |
|
|
||
25OHD3 (nmol/L) | Baseline | 76.2 (2.6) | 71.0–81.4 | 77.2 (2.6) | 72.0–82.4 | 21.44 | <0.001 | 0.83 |
Follow-up | 98.0 (3.3) | 91.4–104.6 | 75.4 (3.3) | 68.9–81.9 | ||||
Working memory | Baseline | 0.57 (0.02) | 0.52–0.62 | 0.55 (0.02) | 0.50–0.59 | 1.09 | 0.30 | 0.19 |
(N Back - Correct hits) | Follow-up | 0.62 (0.02) | 0.57–0.66 | 0.62 (0.02) | 0.58–0.67 | |||
Response inhibition | Baseline | 211.13 (5.18) | 200.87–221.38 | 208.83 (5.08) | 198.77–218.88 | 0.82 | 0.37 | 0.16 |
(Stop Signal SSRT, ms) | Follow-up | 195.71 (4.67) | 186.45–204.96 | 198.85 (4.59) | 189.78–207.93 | |||
Cognitive flexibility | Baseline | 246.34 (22.18) | 202.44–290.24 | 255.59 (21.96) | 212.12–299.05 | 1.37 | 0.24 | 0.21 |
(Switch cost RT, ms) | Follow-up | 143.91 (17.15) | 109.98–177.84 | 185.75 (16.98) | 152.16–219.34 | |||
White noise task | Baseline | 0.29 (0.10) | 0.09–0.48 | 0.31 (0.10) | 0.12–0.50 | 0.02 | 0.88 | 0.03 |
(Hallucination count) | Follow-up | 0.37 (0.16) | 0.05–0.68 | 0.35 (0.16) | 0.04–0.67 | |||
PDI (total) | Baseline | 6.11 (0.48) | 5.16–7.06 | 7.45 (0.47) | 6.51–8.38 | 1.01 | 0.32 | 0.18 |
Follow-up | 5.58 (0.48) | 4.62–6.53 | 6.49 (0.48) | 5.56–7.43 | ||||
BDI (total) | Baseline | 7.24 (0.84) | 5.58–8.90 | 5.72 (0.83) | 4.09–7.36 | 0.44 | 0.51 | 0.12 |
Follow-up | 6.40 (0.85) | 4.73–8.07 | 5.38 (0.83) | 3.74–7.02 | ||||
State anxiety | Baseline | 36.29 (1.20) | 33.91–38.66 | 34.15 (1.19) | 31.81–36.50 | 1.44 | 0.23 | 0.21 |
Follow-up | 36.68 (1.21) | 34.28–39.08 | 36.08 (1.19) | 33.73–38.42 | ||||
State anger | Baseline | 16.41 (0.49) | 15.45–17.37 | 16.26 (0.48) | 15.32–17.21 | 1.30 | 0.26 | 0.20 |
Follow-up | 15.81 (0.33) | 15.15–16.47 | 16.43 (0.32) | 15.79–17.07 |
The following cut-offs were used for secondary analysis: (i) the mean change in 25OHD3 concentrations was 10.04 nmol/L and outcomes were compared between those scoring above this and those scoring below; (ii) the median baseline concentration of 25OHD3 was 75.0 nmol/L and the effects of supplementation was examined in the sub-group of participants scoring below this value (n = 59).
Primary analysis demonstrated that vitamin D supplementation was associated with no change on any of the outcome measures. For example, working memory (as measured by number of correct hits on the N-back task) exhibited an improvement over time in all participants (F = 16.31; p<0.001) but no significant difference between groups over time (F = 1.09; p = 0.30). Similar results were observed with all other measures, with no significant differences over time between participants receiving vitamin D and those receiving placebo for response inhibition (F = 0.82; p = 0.37), cognitive flexibility (F = 1.37; p = 0.24), hallucination proneness (F = 0.02; p = 0.88), delusions (F = 1.01; p = 0.32), depressive symptoms (F = 0.44; p = 0.51), state anxiety (F = 1.44; p = 0.23), and state anger (F = 1.30; p = 0.26). Results for all treatment × time analyses are provided in
Secondary analysis examining the effect of change in vitamin D concentrations on treatment outcome in (a) all participants, and (b) participants with low serum concentrations of 25OHD3 at baseline (<75.00 nmol/L) reported similar findings. No differences between participants exhibiting higher than average change in 25OHD3 concentrations over time and remaining participants were observed on any outcome measure (data not shown).
Vitamin D was well tolerated. There was one report of transient rectal bleeding from one participant receiving placebo. No other adverse effects were reported during the trial. There were no differences between vitamin D and placebo groups on total score on the Treatment Emergent Symptom Scale (t = −1.13; p = 0.26). There was no difference in participant perceptions of their treatment allocation, with the majority believing they received placebo: 73.8% (45/61) of participants receiving vitamin D and 76.6% (49/64) of those receiving placebo believed they received placebo (χ2 = 0.13; p = 0.84).
This trial was the first study, to our knowledge, specifically designed to assess the effects of vitamin D supplementation on key measures of cognitive and emotional functioning. Our findings indicate that vitamin D supplementation or increasing serum concentrations of 25OHD3 had no beneficial effects on (i) core executive functions of working memory, response inhibition or cognitive flexibility, (ii) psychotic-like experiences and hallucination proneness, or (iii) ratings of depression, anxiety or anger. This trial utilised an adequate dose of vitamin D, was placebo-controlled, was adequately powered for primary analyses and displayed high retention rates. As such, our findings appear to represent a true non-effect of vitamin D supplementation on cognitive and emotional functioning in healthy young adults.
These findings should be viewed in the context of an increasing number of studies associating low vitamin D status with impairments in mood and cognitive function and subsequent recommendations of widespread supplementation
Our cognition findings support other studies which have examined the relationship between vitamin D status and cognitive functioning in young adults. For example, the NHANES study
Our study has a number of limitations. Our sample consisted of healthy young adults, who were free of major psychiatric illness and cognitive impairment. As such, our findings may not generalise to clinical populations exhibiting cognitive impairment or emotional disorders. Further controlled trials need to be conducted in key populations of interest, including those with established deficiency. Existing studies have not identified a putative ‘effective dose’ or ‘necessary threshold’ for vitamin D to improve brain function. The dose used in this study was higher than that used in many other studies, and thus, our negative findings are unlikely to have resulted from inadequate doses. The putative mechanism of action of vitamin D on the adult brain is not established. It is unclear whether vitamin D is active only in individuals who are deficient, or whether it also exerts specific pharmacological effects in those with adequate concentrations of vitamin D. Whilst our sample was adequately powered for primary analyses and the first stage of secondary analyses, it is possible that too few participants exhibited low baseline concentrations of 25OHD3 to conduct well-powered analysis in this potentially important subgroup. Analyses were not adjusted for multiple outcome analyses; however, this strengthens the likelihood that our negative findings represent a true lack of effect in this group. It is also feasible that low vitamin D status operates over many years, and that brain-related outcomes are ‘long latency’ disorders
In conclusion, our findings indicate that vitamin D supplementation does not influence cognitive or emotional functioning in healthy young adults. Despite promising clues from observational studies, there are currently no clinical data that supports the use of vitamin D supplementation as a treatment for cognitive or emotional impairments. Although detection and treatment of vitamin D insufficiency remains important for a range of health outcomes (e.g. bone health), future controlled trials in targeted populations of interest are required to elucidate the causal contribution of vitamin D status to brain-related outcomes and determine whether supplementation can improve functioning in these domains.
(DOC)
(PDF)