Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Jun 15.
Published in final edited form as: Am J Physiol Regul Integr Comp Physiol. 2025 Jan 28;328(3):R262–R273. doi: 10.1152/ajpregu.00168.2024

Don’t Sleep on Vitamin D: Vitamin D is Associated with Sleep Variability in Apparently Healthy Adults

Meral N Culver a, Braxton A Linder c, Delaney E Lyons a, Zach J Hutchison a, Catherine L Garrett a, Jessica N McNeil b, Austin T Robinson a,c
PMCID: PMC12167621  NIHMSID: NIHMS2085204  PMID: 39873709

Abstract

Background:

Vitamin D is associated with sleep quality and duration, but it’s unclear whether vitamin D status influences sleep variability. Therefore, we sought to determine whether vitamin D status was associated with sleep variability in healthy adults.

Methods:

We assessed objective sleep, including timing and duration standard deviation (SD) using the Philips Actiwatch Spectrum and subjective sleep quality using the Pittsburgh Sleep Quality Index (PSQI) in 130 adults. We measured plasma 25(OH)D concentration to assess vitamin D. We used one-way ANOVAs and Kruskal-Wallis tests to compare sleep in participants characterized as vitamin D deficient (<20ng/mL), insufficient (21–29ng/mL), and sufficient (>30ng/mL). We used covariate-adjusted linear regression to assess associations between vitamin D status and sleep metrics. We compared differences in ‘low’ and ‘high’ sleep variability based on vitamin D status using Chi-Squared test.

Results:

There was an effect of vitamin D status on sleep timing SD (Kruskal-Wallis, p=0.021) and sleep duration SD (Kruskal-Wallis, p<0.001). There was an inverse association between vitamin D status with sleep duration SD (after covariate adjustment R2=0.267, p<0.001, deficient vs. sufficient p=0.050, insufficient vs. sufficient p=0.022). There was no effect of vitamin D status on objective sleep duration, efficiency, or PSQI scores (ps>0.05). We did not observe differences in ‘low’ and ‘high’ sleep timing SD based on vitamin D status (χ2=5.43, p=0.066), but we did for sleep duration SD (χ2=22.4, p<0.001).

Conclusion:

Our data indicate that individuals with poor vitamin D status exhibit greater objective sleep variability.

Keywords: Sleep Variability, Accelerometry, Sleep Duration, Sleep Timing, Vitamin D Status

Graphical Abstract

graphic file with name nihms-2085204-f0006.jpg

Introduction

Vitamin D is a fat-soluble vitamin that is primarily synthesized by the skin in response to sunlight exposure. To a lesser extent, vitamin D can be obtained from certain foods and dietary supplements (e.g., fish and liver oils, mushrooms, egg yolk, oysters, and dark chocolate)[1]. When in systemic circulation, vitamin D undergoes hydroxylation in the liver to 25(OH)D and then in the kidney to its active form 1,25(OH) D that binds to vitamin D receptors throughout the body [2]. Vitamin D receptors and the regulating enzymes controlling their activation are widely distributed throughout many tissues, including in the brain regions involved in sleep regulation, such as the supraoptic and paraventricular nuclei of the hypothalamus and within the substantia nigra[3]. Specifically, vitamin D may play a role in regulating sleep through its effects on the production of melatonin [3]. Melatonin is a hormone that helps regulate sleep-wake cycles[4]. Higher concentrations are produced at night to promote sleep while less melatonin is produced during the day to promote wakefulness [5, 6]. Vitamin D receptors have been identified in brain regions that directly and indirectly affect sleep in both humans [7] and rodents [8–10]. Individuals with deficient or insufficient circulating vitamin D concentrations (i.e., <30 ng/mL) are more likely to take longer to fall asleep (i.e., longer sleep latency and later sleep timing) and/or exhibit shorter sleep duration [11–16]. Additionally, low vitamin D is associated with sleep disorders such as insomnia, excessive daytime sleepiness, and narcolepsy [17–19].

In addition to mean sleep duration and sleep efficiency (i.e., the percentage of time in bed spent sleeping), sleep variability has recently emerged as an important dimension of sleep health [20]. Sleep variability represents night to night variation in sleep patterns including time of sleep onset and duration. Sleep duration variability, expressed as the intraindividual fluctuations in the amount of sleep time an individual gets each night over a period of time, is associated with increased risk of chronic diseases, such as cardiovascular disease [21, 22] and metabolic syndrome [23, 24]. Additionally, higher sleep variability is adversely associated with measures of macro- and microvascular structure and function [25–27] and early manifestations of blood pressure dysregulation even in younger adults [28].

However, to our knowledge, there is no available data on vitamin D status and sleep variability (sleep timing or duration). Therefore, the present investigation sought to determine whether vitamin D status was related to objectively measured sleep timing variability and sleep duration variability, in addition to objectively measured mean sleep duration and efficiency, and self-reported sleep quality. We hypothesized that individuals with deficient or insufficient concentrations of circulating vitamin D would exhibit greater sleep variability, shorter sleep duration, and poorer sleep quality than those with sufficient plasma vitamin D concentration.

Methods

We included data from 3 current and recently completed studies registered on clinicaltrials.gov (NCT04334135, NCT04244604, and NCT04576338). All studies were conducted in the same laboratory using the same questionnaires and sleep actigraphy devices to achieve a larger pooled sample size. Each of these studies had their separate purposes and hypotheses that were independent of the current investigation’s key objective to assess associations between vitamin D status with sleep variables. All participants provided their verbal and written consent and completed a health history form from which we obtained information regarding dietary supplement usage. The procedures and protocols were approved by the Institutional Review Board at Auburn University and are in accordance with the Declaration of Helsinki.

Participants

During an initial screening visit, participants self-reported their medical history, including the use of vitamin D or vitamin D containing supplements (i.e., daily multi-vitamin). During the same visit, we obtained height and body mass for the calculation of body mass index (BMI; kg/m2) and resting brachial blood pressure in triplicate following at least ten minutes of supine rest using a SpygmoCor XCEL (AtCor Medical, Naperville, IL, USA) [29]. Individuals were excluded for the following reasons: not being between the ages of 19-75; history of any overt or uncontrolled chronic diseases including cardio/cerebrovascular, renal, metabolic, autoimmune, or cancerous conditions; a recent history of COVID-19 infection (< six weeks); currently pregnant or breast feeding; diagnosis of a sleeping disorder (e.g., insomnia, obstructive sleep apnea, restless leg syndrome); a BMI ≥ 35 kg/m2; or greater than moderate hypertension (>150/90 mmHg).

Objective Sleep

Participants were instructed to wear a Phillips ActiWatch Spectrum PLUS for objective sleep monitoring on the dorsal side of their preferred wrist prior to coming to the laboratory for the experimental visit. The observation period ranged from 6-15 consecutive days depending on the original study the individual participated in, with a minimum of six nights to be included in the present analyses. These devices are validated against the gold standard assessment of sleep, polysomnography (sensitivity >0.90, accuracy >0.80) [30, 31]. Participants’ median wear time was 8, 6 (median, interquartile range) nights. All objective sleep data was downloaded and analyzed in 60-second epochs. Rest intervals were coded through visual inspection of the accelerations in the daily actograms within the Actiware software along with the information from the event markers and the nightly sleep logs provided by the participants. If the monitor was removed, participants were asked to record the date and time of removal on their sleep log. Upon visual inspection, sleep data were then manually adjusted with the consideration of the subjective information by an investigator (MNC or DEL). As previously described [28], actigraphy-derived sleep duration was defined as total nightly sleep time, not including wake after sleep onset, as that would reference total time in bed. Sleep duration was quantified for each night the device was worn and averaged to determine mean values. Sleep efficiency was operationalized as the percent of time in bed the participant was sleeping (sleep duration divided by total time spent in bed) and sleep midpoint was defined as the clock time between when an individual fell asleep and woke up. As described previously [25, 28], sleep variability was defined as the standard deviation (SD) of nightly sleep onset timing (sleep timing SD), sleep duration (sleep duration SD) and sleep midpoint SD. Two participants were excluded from variability measures due to technical issues during a software license update, which prevented recovery of their raw data files. While we have their average sleep duration and sleep efficiency, daily files needed for variability calculations were unavailable.

Self-reported Sleep

All participants completed the 20 item Pittsburgh Sleep Quality Index (PSQI) to characterize their sleep quality and disturbances over the previous month. The PSQI is scored 0 [better sleep quality] to 21 [worse sleep quality] [32]. There are seven components in the PSQI: sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. These components are summed to generate the global PSQI score of sleep quality.

Experimental Visits

All participants arrived at the laboratory for the experimental visits having abstained from caffeine, alcohol, and exercise for ≥ 12 hours and fasted from food or caloric beverages (i.e., only water) for ≥ 4 hours. After ≥ 10-minute supine rest, we performed a venous blood draw. We collected blood into vacutainers treated with dipotassium ethylenediaminetetraacetic acid. We measured blood glucose in the whole unspun blood immediately following collection using an Alere Cholestech LDX analyzer per the manufacturer’s instructions (Abbott Laboratories, Abbott Park, IL, USA) [28]. Blood samples were then centrifuged at 1500 x g for 10 minutes at 4° C and subsequently aliquoted into cryogenic tubes and stored at −80°.

Circulating Vitamin D

We used an enzyme-linked immunosorbent assay kit (Catalog # 80987, Crystal Chem, Elk Grove Village, IL USA) to assess previously frozen plasma concentration of 25(OH)D, the primary circulating metabolite of vitamin D, per the manufacturer’s instructions. We used the plasma samples from the experimental visit and the samples were assayed in triplicate. The average intra-sample coefficient of variation was 4.8 ± 3.0%. We also confirmed that our quality control samples provided by the manufacturer were within the expected range when performing each assay. Participants were classified as deficient (< 20 ng/mL), insufficient (21 – 29 ng/mL), or sufficient (> 30 ng/mL) based on reference ranges recommended by the Endocrine Society. The evidence for these vitamin D ranges has been reviewed at length elsewhere [33, 34].

Assessment of Skin Pigmentation

We quantified skin pigmentation primarily because it influences vitamin D production, but we also acknowledge it influences social experiences (i.e., colorism) [35, 36]. As previously described [37], we measured skin pigmentation by reflectance spectrophotometry (DSM III DermaSpectrometer; Cortex Technology, Hadsund, Denmark) [37, 38]. Specifically, we determined the melanin index (M-index) of the skin on the inner aspect of the participant’s upper arm. We performed measurements in triplicate and reported the average of the three readings. The M-index was measured in this region because of its ease of access and because it represents constitutive skin pigmentation due to its relatively low sun exposure [39]. Darker skin pigmentation can lead to reduced ultraviolet radiation mediated synthesis of vitamin D [40, 41].

Physical Activity

Participants were instructed to wear a validated tri-axial accelerometer [42] (ActiGraph wGT3X-BT, Pensacola, FL, USA) on their right hip for seven to eight consecutive days, with a minimum of 1000 minutes of wear time per day. Data were analyzed using our previously published parameters and variables of interest in this present investigation included average weekly steps and moderate and vigorous physical activity (MVPA) [43].

Caffeine Consumption

Participants were instructed to complete a 3-6 day food and fluid record, depending on the study they were initially enrolled in, including at least 2 weekdays and 1 weekend day. Diet records were analyzed, and nutrient intakes were derived using the Nutrition Data System for Research (version 2020, University of Minnesota, Minneapolis, MN, USA) by a registered dietitian.

Statistical Analyses

We inspected all variables for normality using the Shapiro-Wilk test and QQ plots to compare the shapes of distributions. Descriptive characteristics are presented as means and standard deviations for normally distributed data or as medians and interquartile ranges for non-normally distributed data. We compared sleep measures in participants based on vitamin D status (i.e., deficient, insufficient, or sufficient). Our primary analysis entailed performing One-way ANOVAs for normally distributed data and Kruskal-Wallis tests for non-normally distributed data. We report eta squared (η2) as a measure of effect size for ANOVAs and epsilon squared (ε2) for Kruskal-Wallis tests. When appropriate, we used the Tukey post hoc test or the Dwass-Steel-Critchlow-Fligner test for pairwise comparisons between vitamin D groups. Additionally, we utilized unadjusted linear regression models (model 1) and linear regression models adjusted for age, sex, race, BMI, all other objective sleep variables when not the predictor of interest, and PSQI (model 2). The vitamin D sufficient group was classified as the reference group. We report beta coefficients (β) and 95% confidence intervals [95% CI] for the regressions. We also performed the Chi-Square test to analyze potential differences in the proportion of low and high sleep variability based on vitamin D status. To our knowledge, there are no published reference ranges for sleep timing SD, so we performed a median split to define “low” and “high” sleep timing SD. Consistent with prior studies [25] we operationalized a sleep duration SD value of < 1.2 hours as “low” and ≥ 1.2 hours as “high”. Coincidentally, a sleep duration SD value of 1.2 hours was also the median value in our sample. Lastly, on an exploratory basis, we assessed associations between continuous measures of sleep with vitamin D concentrations (as opposed to vitamin D status) using simple bivariate Spearman’s rho (ρ) correlations. A subset of analyses was performed using the exact statistical procedures described above with individuals who were supplementing with vitamin D removed. Statistical significance for all analyses was defined a priori as p ≤ 0.05. Statistical analyses were performed using jamovi 2.3.28 and GraphPad Prism 10.0.3 (GraphPad, San Diego, CA).

Results

Participants

We included 130 participants with complete plasma vitamin D measures and sleep actigraphy in these analyses. Of the 130, 17 individuals did not have satisfactory sleep actigraphy (see methods for criteria) data, and one did not completely fill out their PSQI form (see Figure 1). A full description of study participants disaggregated by vitamin D status can be found in Table 1. Resting brachial blood pressure, physical activity and caffeine consumption are reported in Table 1 as descriptors of our cohort. Notably, there were age, race, M-index, physical activity, and caffeine consumption differences between vitamin D status groups such that the sufficient vitamin D group tended to be older, have a higher proportion of White participants, and have a lower M-index. Importantly, the number of days the sleep watches were worn was not different between vitamin D status groups, and neither was BMI. Additionally, there were no sex differences between any of the sleep variables of interest except sleep efficiency (p = 0.038), and no differences were detected for vitamin D status or concentration (ps ≥ 0.081). In addition to reporting no difference is days monitored via actigraphy (Table 1), the ratio of weekday days to weekend days for the sleep observation period split by vitamin D status were as follows (median, interquartile range): for the vitamin D deficient group: 1.33, 0.417 vitamin D insufficient group: 1.33, 0.333, vitamin D sufficient group: 1.33, 0.333. There were no differences in the weekday days to weekend days ratio detected across groups using the Kruskall Wallis test (p = 0.873).

Figure 1:

Figure 1:

Open in a new tab

Schematic of study inclusion and final sample sizes. Two participants were excluded from variability measures (sleep timing and duration SD) due to technical issues during a software license update, which prevented recovery of their raw data files.

Table 1:

Participant characteristics are split based on vitamin D status and described as mean ± standard deviation for normally distributed data and median, interquartile range (IQR) for non-normally distributed data. Racial breakdown by vitamin D status is included B = Black, W = White, BR = Biracial, LX = Latinx, N = Native). For biological sex, F = Female and M = Male. For race & ethnicity and sex comparisons, we performed the Chi-Square test for the p values and Cramer’s V for effects. For all other variables, we performed One-way ANOVAs for normally distributed data and Kruskal-Wallis tests for non-normally distributed data. We report eta squared (η2) as a measure of effect size for ANOVAs and epsilon squared (ε2) for Kruskal-Wallis tests. For the physical activity variable, including average weekly steps and moderate to vigorous physical activity (MVPA), the sample sizes were: deficient (n = 20), insufficient (n = 35), sufficient (n = 54). For caffeine consumption: deficient (n = 21), insufficient (n = 40), sufficient (n = 63). Regarding pairwise comparisons for significant Kruskal Wallis tests, the vitamin D sufficient group was older (p = 0.041), exhibited lower M-index (p < 0.001), exhibited earlier sleep midpoint (p = 0.023), exhibited greater daily MVPA (p = 0.014) and steps (p = 0.006), and reported higher caffeine consumption (p < 0.001) compared to the vitamin D deficient group. The vitamin D sufficient group also reported higher caffeine consumption (p = 0.009) compared to the vitamin D insufficient group. All other pairwise comparisons were null (ps ≥ 0.061).

Participant Characteristics Deficient
n = 22
n = 1		 Insufficient
n = 42
n = 3		 Sufficient
n = 66
n = 17		 p Effect Size
 Supplementing
Race/ethnicity (14 B, 8 W) (18 B, 19 W, 4 BR,1 LX) (12 B, 49 W, 3 BR,1 LX, 1 N) 0.006 0.288
Biological sex (female/male) (15F/7M) (18/F24M) (32F/34M) 0.148 0.172
Age, yrs 21.0, 1.8 22.0, 5.8 22.5, 14.8 0.040 0.049
Height, cm 164, 16 170, 15 173, 15 0.126 0.032
Mass, kg 77 ± 16 76 ± 14 75.3 ± 14 0.848 0.003
BMI, kg/m2 27, 9 26, 6 25, 4 0.292 0.019
M-Index 49, 24 34, 26 29, 8 <0.001 0.141
Brachial Systolic BP, mmHg 119 ± 11 118 ± 9 117 ± 10 0.777 0.005
Brachial Diastolic BP, mmHg 71 ± 8 71 ± 8 70 ± 7 0.971 0.001
Glucose, mg/dL 86, 13 88, 12 83, 10 0.604 0.008
Sleep Onset, clock hour 00:54 ± 1:33 11:54 ± 1:30 11:54 ± 1:17 0.159 0.982
Days monitored, sleep actigraphy 7, 1 8, 6 8, 7 0.205 0.203
Sleep Midpoint SD 0.926, 1.02 0.864, 0.532 0.989, 0.591 0.655 0.008
Sleep Midpoint, clock hour 28.50, 2.50 27.90, 1.90 27.50, 1.68 0.023 0.068
Average daily steps 3713,2741 5243,2839 6187,4318 0.005 0.097
Average daily MVPA 28.9,25.3 34.3,29.9 43.3,23.6 0.015 0.078
Days monitored, PA 8,1 8,1 8,1 0.768 0.005
Caffeine consumption, mg 16.2,43 32.2,107 92,181 <0.001 0.138
Days monitored, caffeine 3,3 6,3 6,3 0.201 0.026
Open in a new tab

In Figure 2, we present comparisons of sleep timing SD and sleep duration SD between vitamin D status groups. There was a difference in sleep timing SD across vitamin D status groups with a significant post hoc difference between sufficient and deficient groups (Figure 2A). There was also a difference in sleep duration SD across vitamin D status with significant post hoc differences between sufficient and insufficient groups and between sufficient and deficient groups (Figure 2B). In Figure 3, we present comparisons between objectively measured sleep duration and efficiency and the self-reported PSQI global score across vitamin D status. There were no differences between vitamin D groups and any of these sleep metrics (Figure 3 A–C).

Figure 2:

Figure 2:

Open in a new tab

Comparisons of objective sleep timing SD (deficient n = 15, insufficient n = 36, sufficient n = 58) (A) and objective sleep duration SD (deficient n = 17, insufficient n = 37, sufficient n = 59) (B) between vitamin D status groups. SD = standard deviation. Data are presented as individual data points are super imposed on median and interquartile range.

Figure 3:

Figure 3:

Open in a new tab

Comparisons of the sleep metrics between objectively measured sleep duration (deficient n = 17, insufficient n = 37, sufficient n = 59) (A) and efficiency (deficient n = 17, insufficient n = 37, sufficient n = 59) (B) and the self-reported PSQI global score (deficient n = 21, insufficient n = 42, sufficient n = 66) (C) across vitamin D status groups. Data are presented as individual data points are super imposed on median and interquartile range.

In Table 2 we present results from regression models to assess associations between vitamin D status with sleep timing SD and sleep duration SD. Being classified as vitamin D deficient was associated with higher sleep timing SD (model 1), but not after covariate adjustment (model 2). Regarding the covariates, none of them were independently associated with sleep timing SD. Being classified as vitamin D deficient or insufficient was independently associated with higher sleep duration SD. Regarding the covariates, age and sleep efficiency were negatively associated with sleep duration SD while sleep timing SD was positively associated. No other covariate was independently associated with sleep duration SD. In Table 3 we present regression models to assess associations between vitamin D status with sleep duration, efficiency, and PSQI global score. There were no independent associations between vitamin D status with sleep duration, efficiency, or PSQI global score. Additionally, the overall models were not predictive of sleep duration or PSQI global score. The overall model for sleep efficiency was significant, with independent negative associations between PSQI global score and sleep duration SD with sleep efficiency. Vitamin D concentration (i.e., continuous variable) was not associated with any of the sleep metrics in regression models (ps ≥ 0.05).

Table 2:

Represents unadjusted (model 1) and regression adjusted for age, sex, race, BMI, and sleep variables (model 2) to assess independent associations between sleep timing SD and sleep duration SD with vitamin D status.

Sleep Timing SD Sleep Duration SD
Model 1 Model 2 Model 1 Model 2
B SEM p B SEM p B SEM p B SEM p
Vitamin D
 Deficient-Sufficient 0.343 0.149 0.023 −0.089 0.169 0.597 −0.159 0.159 0.027 0.280 0.135 0.040
 Insufficient-Sufficient 0.175 0.109 0.110 −0.077 0.122 0.530 −0.503 0.195 0.011 0.190 0.095 0.049
Age −0.003 0.004 0.483 −0.009 0.003 0.009
Sex 0.070 0.088 0.426 0.067 0.089 0.457
Race 0.025 0.048 0.609 −0.059 0.049 0.230
BMI 0.002 0.088 0.838 0.016 0.011 0.148
Sleep Duration SD 0.506 0.088 <0.001 - - -
Sleep Timing SD - - - 0.513 0.089 <0.001
Duration −0.056 0.047 0.225 −0.015 0.048 0.759
Efficiency −0.000 0.008 0.946 −0.017 0.008 0.033
PSQI 0.037 0.023 0.118 −0.030 0.024 0.208
Overall Model Fit R2 p R2 p R2 p R2 p
0.238 0.046 0.436 <0.001 0.131 <0.001 0.524 <0.001
Open in a new tab

Table 3:

Represents unadjusted (model 1) and regression adjusted for age, sex, race, BMI, and sleep variables (model 2) to assess independent associations between sleep duration, sleep efficiency, and PSQI with vitamin D status.

Sleep Duration Sleep Efficiency PSQI
Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
B SEM p B SEM p B SEM p B SEM p B SEM p B SEM p
Vitamin D
 Deficient-Sufficient 0.264 0.267 0.327 −0.228 0.291 0.434 0.956 1.72 0.584 0.735 1.755 0.676 0.167 0.475 0.726 −0.441 0.583 0.451
 Insufficient - sufficient 0.378 0.250 0.134 0.087 0.206 0.675 1.19 1.61 0.464 0.903 1.238 0.468 0.113 0.445 0.801 0.008 0.413 0.985
Age 0.002 0.007 0.813 0.040 0.045 0.374 −0.006 0.015 0.706
Sex 0.286 0.188 0.132 2.293 1.119 0.043 0.520 0.377 0.171
Race −0.212 0.103 0.042 −0.068 0.631 0.915 −0.095 0.210 0.651
BMI −0.027 0.023 0.270 −0.008 0.145 0.958 0.054 0.048 0.265
Duration - - - 0.757 0.608 0.216 −0.027 0.204 0.897
Efficiency 0.021 0.017 0.216 - - - −0.092 0.033 0.006
PSQI −0.006 0.051 0.897 −0.826 0.2949 0.006 - - -
Sleep Time SD −0.277 0.218 0.207 0.062 1.323 0.963 0.681 0.435 0.121
Sleep Duration SD −0.067 0.217 0.759 −2.752 1.273 0.033 −0.545 0.430 0.208
Overall Model Fit R2 p R2 p R2 p R2 p R2 p R2 p
0.021 0.319 0.102 0.076 0.070 0.764 0.235 0.003 9.78e−4 0.940 0.139 0.135
Open in a new tab

As presented in Figure 4, we did not observe differences in the proportion of high and low sleep timing SD participants between vitamin D status groups (Figure 4A) as indicated by the Chi-Square test (χ2 = 5.43, p = 0.066). However, we did observe differences in the proportion of high and low sleep duration SD participants between vitamin D status groups (Figure 4B) as indicated by the Chi-Square test (χ2 = 22.4, p <0.001). In Figure 5, we present exploratory bivariate correlations using Spearman’s ρ to assess associations between vitamin D concentration with sleep timing SD and sleep duration SD. Plasma vitamin D concentration was inversely associated with both sleep timing SD (Figure 5A) and sleep duration SD (Figure 5B). In contrast, plasma vitamin D concentration was not associated with sleep duration (ρ = 0.150, p = 0.115), efficiency (ρ = 0.061, p = 0.524), or PSQI global score (ρ = 0.001, p = 0.997).

Figure 4:

Figure 4:

Open in a new tab

Proportion of high and low objectively measured sleep timing SD (deficient n = 15, insufficient n = 36, sufficient n = 58) (A) and objectively measured sleep duration SD (deficient n = 17, insufficient n = 37, sufficient n = 59) (B) across vitamin D status groups. SD = standard deviation

Figure 5:

Figure 5:

Open in a new tab

Bivariate correlations using Spearman’s ρ to assess associations between sleep timing SD (deficient n = 15, insufficient n = 36, sufficient n = 58) (A) and sleep duration SD (deficient n = 17, insufficient n = 37, sufficient n = 59) (B) with vitamin D concentration. SD = standard deviation All data are presented as individual data points with unique symbols and colors based on vitamin D status.

Removal of Participants Supplementing with Vitamin D

Among our cohort of 130 participants, 16% (n = 21) reported supplementing with either a multi-vitamin containing vitamin D (n = 13) or with vitamin D alone (n = 8). To ensure that our findings were not driven by these individuals, we excluded them from our sample in a series of separate confirmatory analyses. Removing these participants altered some of our findings regarding vitamin status and sleep timing SD. Specifically, the results of the Kruskall-Wallis test (Figure 2A) were no longer significant (p = 0.245) upon removal of supplement users. Removal of these participants did not alter the association between vitamin D status and sleep timing SD in the fully adjusted regression model as there was a lack of an association in both the entire cohort and after removing supplement users. The results of Chi-Square test for sleep timing SD, which was approaching significance (Figure 4A) in the whole sample, were convincingly null after removal of supplement users (χ2 = 1.86, p = 0.395). Regarding the exploratory bivariate correlation analysis, there was no longer a correlation (Figure 5A) between vitamin D concentration and sleep timing SD after removal of supplement users. (Spearman’s ρ = −0.175, p = 0.095)

In contrast, our overall findings for vitamin D status and sleep duration SD remained consistent after removal of supplement users. The results of the Kruskall-Wallis test for sleep duration SD (Figure 2B) remained significant (p = 0.005, ε2 = 0.110), with post hoc analyses indicating pairwise differences between deficient vs. sufficient groups (p = 0.018) and insufficient vs. sufficient groups (p = 0.031). Similarly, in regression models, the independent association between vitamin D status and sleep duration SD remained (overall model, R2 = 0.301, p <0.001, sleep duration SD: deficient – sufficient β = 0.276, SE = 0.157, p = 0.082; insufficient – sufficient β = 0.253, SE = 0.119, p = 0.037). The Chi-Square test for sleep duration SD remained significant (Figure 4B) after removal of supplement users (χ2 = 7.52, p = 0.023). Furthermore, for continuous vitamin D analysis, the correlation between vitamin D concentration and sleep duration SD (Figure 5B) remained after removal of supplement users (Spearman’s ρ = −0.294, p = 0.004). Lastly, all statistical tests (Kruskall-Wallis, ANOVA, regression models, Chi-Square tests, and bivariate correlations) yielded similar, non-significant findings between vitamin D status or concentration with sleep duration, sleep efficiency, or the PSQI global score (ps > 0.05).

Discussion

This study examined associations between vitamin D status and sleep variability, inclusive of sleep timing SD and sleep duration SD, along with other objective sleep metrics such as mean sleep duration, efficiency, and self-reported sleep quality. Our key findings include that vitamin D was associated with sleep duration SD when assessed using several unique analyses whether we included individuals taking supplements containing vitamin D or not. There was some evidence that vitamin D status was associated with sleep timing SD. However, after adjusting for covariates, this association no longer remained. When we removed individuals who took supplements containing vitamin D, there was no longer evidence to support an association between vitamin D status and sleep timing SD. While sleep midpoint SD was not an a priori focus of our paper, it was discordant from our sleep timing SD and sleep duration SD, in that there was no effect of vitamin D status on sleep midpoint SD (Table 1). Sleep midpoint measures the variability in the midpoint of sleep (the halfway point between bedtime and awake time) across nights. The sleep midpoint SD between groups was not substantial, suggesting that despite differences in timing or duration variability across groups, the average variability in midpoint across nights remained more consistent across groups. Our primary focus was to examine potential differences among participants classified as deficient or insufficient compared with those classified as sufficient, based on well-established reference ranges [31, 32] recommended by the Endocrine Society. Additionally, our exploratory analyses indicated that plasma vitamin D concentration was inversely correlated with sleep variability. However, this correlation only remained for sleep duration SD after removal of participants who reported consuming supplements containing vitamin D. For instance, individuals with deficient or insufficient plasma vitamin D concentrations exhibited higher sleep variability.

Specifically, vitamin D may play a role in regulating sleep through its effects on the production of melatonin [44]. Vitamin D receptors have been identified in brain regions that directly and indirectly affect sleep in both humans [7] and rodents [8–10]. For example, in human brain samples, there is high expression of the vitamin D receptor in the supraoptic and paraventricular nuclei of the hypothalamus and within the substantia nigra [7]. Interestingly, vitamin D expression reported in the human study was similar to that of prior rodent studies [9, 10]. Regarding the expression of vitamin D receptors in the substantia nigra in the human brain, recent rodent data indicate that GABA release in the lateral substantia nigra pars reticulata is negatively correlated with slow wave sleep and positively correlated with awake time [45]. Sleep control systems in the hypothalamus interact with the circadian pacemaker in the suprachiasmatic nuclei (SCN)[46]. The SCN is the principal circadian pacemaker in mammals, responsible for generating circadian rhythms [47]. Light receptors in the eyes (i.e., photosensitive retinal ganglion cells) enable the SCN to coordinate cellular clocks throughout the body that influence homeostasis [3, 48]. These light receptors are activated by sunlight, which is the major factor influencing vitamin D status, albeit through separate mechanisms. Additionally, in vitro data indicate vitamin D synchronizes circadian clock gene expression in adipose-derived stem cells [49], further suggesting a potentially important role for vitamin D in regulating circadian clocks throughout the body. Although there is a need for more data in humans, and studies on the mechanisms underpinning the relations between vitamin D with sleep and circadian regulation, these findings indicate that vitamin D receptors in the brain could influence sleep and circadian rhythms.

Contrary to our hypothesis, there was no association between vitamin D status with sleep duration, efficiency, or PSQI global score. Consistent with our vitamin D and sleep duration findings, data from nearly 3000 adults in the National Health and Nutrition Examination Survey demonstrated that vitamin D concentration was not associated with self-reported sleeping hours [16]. There are a few prior studies using objective sleep assessments that reported inverse associations between 25(OH)D concentration and sleep duration and efficiency [12, 14]. However, both studies examined participants who were generally older than our cohort. The participants in the Multi-Ethnic Study of Atherosclerosis sleep study were 68 years old on average. While there was an association between vitamin D and sleep duration, the effect sizes were modest. Specifically, after adjusting for demographics, obesity, and health habits, individuals who were vitamin D deficient slept an average of 13 minutes shorter than sufficient individuals [14]. In a study of ~650 adults (52% female participants) that also utilized the gold standard polysomnography, male participants with short sleep (< 6 hours) had 4-fold increased odds of being vitamin D deficient (25(OH)D < 20 ng/mL). In contrast, there was no association between short sleep duration with vitamin D deficiency in female participants. The average age of the cohort was 50 years old, which is also older than our cohort. Another study using polysomnography in children demonstrated that those with vitamin D deficiency or insufficiency exhibited less total sleep time and poorer sleep efficiency compared with children with sufficient vitamin D [50]. It is worth noting that this investigation took place in a children’s hospital. Thus, the reasons for the divergent findings regarding vitamin D and sleep duration between the present study and prior investigations are not entirely clear, but they may be due to methodological differences or differences in the general age and/or health of the cohorts across studies.

The implications regarding vitamin D and sleep health are not entirely clear and require future investigation. One interpretation could be that strategies to optimize vitamin D may improve sleep. Vitamin D receptors being present in brain regions that directly and indirectly affect sleep in both humans [7] and rodents [8–10] support this biological plausibility. However, studies on vitamin D supplementation and sleep outcomes have reported mixed findings. A placebo-controlled study that administered high dose vitamin D (50,000 IU) every two weeks for an eight-week intervention reported improvements in self-reported sleep quality assessed via the PSQI [51]. In contrast, a placebo-controlled study that administered high dose vitamin D (20,000 IU) every week for 16 weeks reported no improvement in any metric of self-reported sleep [52]. Another important consideration is that many studies examining sleep and vitamin D, including our own, have been cross-sectional in nature. Therefore, causality between vitamin D status and sleep outcomes should not be inferred. It could be that short sleep, or irregular sleep may contribute to vitamin D status by influencing other health behaviors such as outdoor leisure time, physical activity, and diet, which directly impact vitamin D status. Thus, the connection between sleep and vitamin D, as well as the underlying mechanisms, remain to be completely understood.

Strengths, Limitations, and Future Directions

This present study has several strengths. One key strength is the use of a diverse (racial, sex, and age) cohort. This diversity contributes to the generalizability of our findings across healthy adults. Our participants were also well characterized including physical activity and caffeine consumption. Regarding some evidence of a healthy user bias towards the vitamin D sufficient group, they were more physically active. We would have speculated that the vitamin D sufficient group consumed similar or less caffeine than the deficient and insufficient group, but they consumed more caffeine. Moreover, this study benefited from a relatively long observation period for objectively assessing nightly sleep averaging eight nights across the multiple studies This observation period enables a more comprehensive and presumably more valid assessment of sleep variability, duration, and efficiency [53–56].

However, it is important to acknowledge the limitations of the present study. Our findings may not be generalizable to those with chronic health conditions or populations in lower socioeconomic positions. To our knowledge, our participants were free from sleeping disorders (e.g., obstructive sleep apnea, insomnia, etc.). However, we only have verbal confirmation, and we did not have any participants complete any additional screening tools (i.e., STOP-BANG or Insomnia Severity Index), which is a limitation. We also have reported on racial differences in social determinants of health that may influence sleep variability [36–38, 43, 57], such as discrimination and neighborhood socioeconomic disadvantage. Unfortunately, these instruments were not harmonized across the three protocols used in this analysis, and we could not assess these factors in relation to sleep variability. Also, in the individuals who were supplementing vitamin D there was irregularity in dosing vitamin D based on type of supplement and brand. Additionally, the time of year when sleep was assessed, and blood samples were obtained for vitamin D were not controlled for. Time of year can impact an individual’s vitamin D status due to differences in sunlight exposure, which is the primary method for vitamin D synthesis in the body [2]. There are also seasonal variations in sleep patterns [58, 59]. Thus, an ideal design for a future study may be to utilize a repeated measures design. Participants would report to the lab during each of the four seasons for testing to account for some of the confounding factors that may influence the relation between vitamin D status and the sleep parameters. Although one could also posit that the current investigation has higher ecological validity given that data were collected throughout the year.

Future studies can also overcome the limitations of our cross-sectional study design, which restricts our ability to determine directionality of the observed associations between vitamin D status and sleep variability. For example, short- to long-term vitamin D supplementation interventions could determine whether increasing vitamin D improves sleep variability or other dimensions of sleep health. Investigations into whether potential seasonal differences in sleep are mediated by vitamin D could also provide additional insight. Alternatively, interventions targeted at improving sleep variability while monitoring potential changes in circulating vitamin D could help provide insight on whether sleep variability may casually influence vitamin D. Pursuing these future directions has the potential to advance our understanding of the interplay between vitamin D and sleep and inform potential strategies to optimize aspects of overall health and well-being.

Conclusions

In our cohort of apparently heathy adults, those who had insufficient or deficient vitamin D displayed higher sleep variability, most notably sleep duration variability, compared with those who were vitamin D sufficient. This observation further supports evidence of an association between vitamin D and sleep health. Further evidence is needed to assess these associations within the context of cardiometabolic disease risk factors given the strong associations between sleep patterns and cardiometabolic diseases.

Table 4:

Participants who did not use multi-vitamins or vitamin D supplements’ characteristics are split based on vitamin D status and described as mean ± standard deviation for normally distributed data and median, interquartile range (IQR) for non-normally distributed data. Racial breakdown by vitamin D status is included B = Black, W = White, BR = Biracial, LX = Latinx, N = Native). For biological sex, F = Female and M = Male. For race & ethnicity and sex comparisons, we performed Chi-Square test for the p values and Cramer’s V for effects. For all other variables, we performed One-way ANOVAs for normally distributed data and Kruskal-Wallis tests for non-normally distributed data. We report eta squared (η2) as a measure of effect size for ANOVAs and epsilon squared (ε2) for Kruskal-Wallis tests.

Participant Characteristics Deficient
n = 21		 Insufficient
n = 39		 Sufficient
n = 49		 p Effect Size
Race/ethnicity (13 B, 8 W) (17 B, 18 W, 3 BR,1 LX) (9 B, 36 W, 3 BR, 1 N) 0.023 0.286
Biological sex (female/male) (15F/6M) (17F/22M) (25F/24M) 0.117 0.199
Age, yrs 21.0, 2.0 22.0, 5.5 22.0, 4.0 0.166 0.033
Height, cm 164, 16 164, 16 172, 16 0.202 0.030
Weight, kg 76 ± 22 76 ± 17 73 ± 15 0.732 0.006
BMI, kg/m2 27 ± 8 26 ± 5 24, 5 0.269 0.024
M-Index 48 ± 23 34, 26 29, 8 <0.001 0.140
Brachial Systolic BP mmHg 120 ± 12 118 ± 9 116 ± 14 0.355 0.020
Brachial Diastolic BP, mmHg 71 ± 8 70 ± 8 69 ± 7 0.474 0.009
Glucose, mg/dL 86, 12 88, 14 83, 8 0.197 0.032
Sleep Onset, clock hour 00:54 ± 1:55 00:24 ± 1:41 11:54 ± 1:26 0.077 0.056
Days monitored, sleep actigraphy 7, 1 8, 6 8, 7 0.293 0.026
Midpoint SD, hour 0.926, 1.02 0.863, 0.521 0.812, 0.725 0.910 0.002
Sleep Midpoint, clock hour 28.50, 2.50 27.90, 1.93 27.50, 1.60 0.061 0.061
Average daily steps 3658,3048 5243,2937 6070,3845 0.013 0.094
Average daily MVPA 28.8,24.0 34.3,30.3 43.6,26.4 0.042 0.069
Days monitored, PA 8,1 8,1 8,1 0.614 0.011
Caffeine Intake, mg 15.6,35.3 28.7,83.4 96.6,184 <0.001 0.150
Days monitored, caffeine 3,3 6,3 6,3 0.151 0.037
Open in a new tab

New and Noteworthy.

Our findings reveal that individuals with lower circulating vitamin D concentrations experience greater sleep variability compared to those with higher circulating concentrations This supports the growing body of evidence suggesting an important link between vitamin D status and sleep health.

Acknowledgements

We would like to thank Alex Barnett, Clayton Ivie, Isaac Fields, and Soolim Jeong for assistance with Vitamin D assays. We would also like to thank participants for their time and efforts in facilitating data collection. No conflicts of interest, financial or otherwise, are declared by the other authors.

Funding

This work was supported by the National Institutes of Health (NIH) Grant K01HL147998 and (to A.T.R.), UL1TR003096 (Pilot funding to A.T.R. and TL-1 Fellowship to B.A.L.), and an Auburn University Presidential Graduate Research Fellowship (to B.A.L.).

Abbreviations and Acronyms:

BMI

body mass index

BP

blood pressure

PSQI

Pittsburgh Sleep Quality Index

SD

standard deviation

Sleep Duration SD

sleep duration standard deviation

Sleep Time SD

sleep onset time standard deviation

Midpoint SD

sleep midpoint standard deviation

Footnotes

Clinical Trial Registry: URL: https://www.clinicaltrials.gov/; Unique identifiers NCT04334135, NCT04244604, and NCT04576338

Data Availability Statement

The authors are willing to share the data upon request, contingent upon the requester completing a Data & Material Transfer Agreement.

References:

  • 1.Benedik E, Sources of vitamin D for humans. International Journal for Vitamin and Nutrition Research, 2021. [DOI] [PubMed] [Google Scholar]
  • 2.Christakos S, et al. , Vitamin D: Metabolism, Molecular Mechanism of Action, and Pleiotropic Effects. Physiol Rev, 2016. 96(1): p. 365–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Muscogiuri G, et al. , The lullaby of the sun: the role of vitamin D in sleep disturbance. Sleep Med, 2019. 54: p. 262–265. [DOI] [PubMed] [Google Scholar]
  • 4.Czeisler CA, et al. , Stability, precision, and near-24-hour period of the human circadian pacemaker. Science, 1999. 284(5423): p. 2177–81. [DOI] [PubMed] [Google Scholar]
  • 5.Ghareghani M, Zibara K, and Rivest S, Melatonin and vitamin D, two sides of the same coin, better to land on its edge to improve multiple sclerosis. Proceedings of the National Academy of Sciences, 2023. 120(14): p. e2219334120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Claustrat B and Leston J, Melatonin: Physiological effects in humans. Neurochirurgie, 2015. 61(2-3): p. 77–84. [DOI] [PubMed] [Google Scholar]
  • 7.Eyles DW, et al. , Distribution of the vitamin D receptor and 1 alpha-hydroxylase in human brain. J Chem Neuroanat, 2005. 29(1): p. 21–30. [DOI] [PubMed] [Google Scholar]
  • 8.Musiol IM, et al. , Vitamin D nuclear binding to neurons of the septal, substriatal and amygdaloid area in the Siberian hamster (Phodopus sungorus) brain. Neuroscience, 1992. 48(4): p. 841–8. [DOI] [PubMed] [Google Scholar]
  • 9.Stumpf WE and O’Brien LP, 1,25 (OH)2 vitamin D3 sites of action in the brain. An autoradiographic study. Histochemistry, 1987. 87(5): p. 393–406. [DOI] [PubMed] [Google Scholar]
  • 10.Prüfer K and Jirikowski GF, 1.25-Dihydroxyvitamin D3 receptor is partly colocalized with oxytocin immunoreactivity in neurons of the male rat hypothalamus. Cell Mol Biol (Noisy-le-grand), 1997. 43(4): p. 543–8. [PubMed] [Google Scholar]
  • 11.Jung YS, et al. , The relationship between serum vitamin D levels and sleep quality in fixed day indoor field workers in the electronics manufacturing industry in Korea. Ann Occup Environ Med, 2017. 29: p. 25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Massa J, et al. , Vitamin D and actigraphic sleep outcomes in older community-dwelling men: the MrOS sleep study. Sleep, 2015. 38(2): p. 251–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.de Oliveira DL, et al. , Sleep duration as an independent factor associated with vitamin D levels in the EPISONO cohort. J Clin Sleep Med, 2021. 17(12): p. 2439–2449. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Bertisch SM, et al. , 25-Hydroxyvitamin D Concentration and Sleep Duration and Continuity: Multi-Ethnic Study of Atherosclerosis. Sleep, 2015. 38(8): p. 1305–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Liu X, et al. , Sleep duration is associated with vitamin D deficiency in older women living in Macao, China: A pilot cross-sectional study. PLoS One, 2020. 15(3): p. e0229642. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Shiue I, Low vitamin D levels in adults with longer time to fall asleep: US NHANES, 2005–2006. Int J Cardiol, 2013. 168(5): p. 5074–5. [DOI] [PubMed] [Google Scholar]
  • 17.Gao Q, et al. , The Association between Vitamin D Deficiency and Sleep Disorders: A Systematic Review and Meta-Analysis. Nutrients, 2018. 10(10). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Dauvilliers Y, et al. , Vitamin D deficiency in type 1 narcolepsy: a reappraisal. Sleep Med, 2017. 29: p. 1–6. [DOI] [PubMed] [Google Scholar]
  • 19.Carlander B, et al. , Low vitamin D in narcolepsy with cataplexy. PLoS One, 2011. 6(5): p. e20433. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Makarem N, et al. , Variability in Sleep Patterns: an Emerging Risk Factor for Hypertension. Current Hypertension Reports, 2020. 22(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Huang T, Mariani S, and Redline S, Sleep Irregularity and Risk of Cardiovascular Events: The Multi-Ethnic Study of Atherosclerosis. J Am Coll Cardiol, 2020. 75(9): p. 991–999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Morris CJ, et al. , Circadian misalignment increases cardiovascular disease risk factors in humans. Proc Natl Acad Sci U S A, 2016. 113(10): p. E1402–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Huang T and Redline S, Cross-sectional and Prospective Associations of Actigraphy-Assessed Sleep Regularity With Metabolic Abnormalities: The Multi-Ethnic Study of Atherosclerosis. Diabetes Care, 2019. 42(8): p. 1422–1429. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Zuraikat FM, et al. , Sleep Regularity and Cardiometabolic Heath: Is Variability in Sleep Patterns a Risk Factor for Excess Adiposity and Glycemic Dysregulation? Current Diabetes Reports, 2020. 20(8). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Hoopes EK, et al. , Sleep duration regularity, but not sleep duration, is associated with microvascular function in college students. Sleep, 2021. 44(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Hoopes EK, et al. , Sleep Variability, Eating Timing Variability, and Carotid Intima-Media Thickness in Early Adulthood. J Am Heart Assoc, 2023. 12(19): p. e029662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Nakashima H, et al. , Association of sleep-wake rhythm and sleep quality with endothelial function in young adults. Sleep Sci, 2022. 15(3): p. 267–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Culver MN, et al. , Sleep duration irregularity is associated with elevated blood pressure in young adults. Chronobiology International, 2022: p. 1–9. [DOI] [PubMed] [Google Scholar]
  • 29.Schultz MG, et al. , Validation Study to Determine the Accuracy of Central Blood Pressure Measurement Using the Sphygmocor Xcel Cuff Device. Hypertension, 2020. 76(1): p. 244–250. [DOI] [PubMed] [Google Scholar]
  • 30.Quante M, et al. , Actigraphy-based sleep estimation in adolescents and adults: a comparison with polysomnography using two scoring algorithms. Nat Sci Sleep, 2018. 10: p. 13–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Kahawage P, et al. , Validity, potential clinical utility, and comparison of consumer and research-grade activity trackers in Insomnia Disorder I: In-lab validation against polysomnography. J Sleep Res, 2020. 29(1): p. e12931. [DOI] [PubMed] [Google Scholar]
  • 32.Buysse DJ, et al. , The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res, 1989. 28(2): p. 193–213. [DOI] [PubMed] [Google Scholar]
  • 33.Pludowski P, et al. , Vitamin D effects on musculoskeletal health, immunity, autoimmunity, cardiovascular disease, cancer, fertility, pregnancy, dementia and mortality-a review of recent evidence. Autoimmun Rev, 2013. 12(10): p. 976–89. [DOI] [PubMed] [Google Scholar]
  • 34.Bischoff-Ferrari HA, et al. , Estimation of optimal serum concentrations of 25-hydroxyvitamin D for multiple health outcomes. Am J Clin Nutr, 2006. 84(1): p. 18–28. [DOI] [PubMed] [Google Scholar]
  • 35.Nguyen AW, et al. , Discrimination and social isolation among African Americans: The moderating role of skin tone. Cultur Divers Ethnic Minor Psychol, 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Grosicki GJ, et al. , Racial and ethnic disparities in cardiometabolic disease and COVID-19 outcomes in White, Black/African American, and Latinx populations: Social determinants of health. Prog Cardiovasc Dis, 2022. 71: p. 4–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Robinson AT, et al. , Cross-sectional analysis of racial differences in hydration and neighborhood deprivation in young adults. Am J Clin Nutr, 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Grosicki GJ, et al. , Acute beetroot juice reduces blood pressure in young Black and White males but not females. Redox Biol, 2023. 63: p. 102718. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Wolf ST, et al. , Four weeks of vitamin D supplementation improves nitric oxide-mediated microvascular function in college-aged African Americans. Am J Physiol Heart Circ Physiol, 2020. 319(4): p. H906–H914. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Wolf ST and Kenney WL, Skin pigmentation and vitamin D-folate interactions in vascular function: an update. Curr Opin Clin Nutr Metab Care, 2021. 24(6): p. 528–535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Libon F, Cavalier E, and Nikkels AF, Skin color is relevant to vitamin D synthesis. Dermatology, 2013. 227(3): p. 250–4. [DOI] [PubMed] [Google Scholar]
  • 42.Karaca A, et al. , Validation of the ActiGraph wGT3X-BT Accelerometer for Step Counts at Five Different Body Locations in Laboratory Settings. Measurement in Physical Education and Exercise Science, 2022. 26(1): p. 63–72. [Google Scholar]
  • 43.Jeong S, et al. , Interplay of race and neighborhood deprivation on resting and ambulatory blood pressure in young adults. Am J Physiol Heart Circ Physiol, 2024. 327(3): p. H601–h613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Muscogiuri G, et al. , The lullaby of the sun: the role of vitamin D in sleep disturbance. Sleep medicine, 2019. 54: p. 262–265. [DOI] [PubMed] [Google Scholar]
  • 45.Lai YY, et al. , Substantia nigra pars reticulata-mediated sleep and motor activity regulation. Sleep, 2021. 44(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Mignot E, Taheri S, and Nishino S, Sleeping with the hypothalamus: emerging therapeutic targets for sleep disorders. Nat Neurosci, 2002. 5 Suppl: p. 1071–5. [DOI] [PubMed] [Google Scholar]
  • 47.Hastings MH, Maywood ES, and Brancaccio M, Generation of circadian rhythms in the suprachiasmatic nucleus. Nat Rev Neurosci, 2018. 19(8): p. 453–469. [DOI] [PubMed] [Google Scholar]
  • 48.Hu J, et al. , Melanopsin retinal ganglion cells mediate light-promoted brain development. Cell, 2022. 185(17): p. 3124–3137.e15. [DOI] [PubMed] [Google Scholar]
  • 49.Gutierrez-Monreal MA, et al. , A role for 1α,25-dihydroxyvitamin d3 in the expression of circadian genes, in J Biol Rhythms. 2014, © 2014 The Author(s). United States. p. 384–8. [DOI] [PubMed] [Google Scholar]
  • 50.Al-Shawwa B, Ehsan Z, and Ingram DG, Vitamin D and sleep in children. J Clin Sleep Med, 2020. 16(7): p. 1119–1123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Majid MS, et al. , The effect of vitamin D supplement on the score and quality of sleep in 20–50 year-old people with sleep disorders compared with control group. Nutr Neurosci, 2018. 21(7): p. 511–519. [DOI] [PubMed] [Google Scholar]
  • 52.Larsen AU, et al. , No improvement of sleep from vitamin D supplementation: insights from a randomized controlled trial. Sleep Med X, 2021. 3: p. 100040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Aili K, et al. , Reliability of Actigraphy and Subjective Sleep Measurements in Adults: The Design of Sleep Assessments. J Clin Sleep Med, 2017. 13(1): p. 39–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Buysse DJ, et al. , Night-to-night sleep variability in older adults with and without chronic insomnia. Sleep Med, 2010. 11(1): p. 56–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Anaya G, et al. , Optimal Instruments for Measurement of Dietary Intake, Physical Activity, and Sleep Among Adults in Population-Based Studies: Report of a National Heart, Lung, and Blood Institute Workshop. J Am Heart Assoc, 2024. 13(21): p. e035818. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Knutson KL, et al. , CARDIA sleep ancillary study: study design and methods. Sleep Adv, 2024. 5(1): p. zpae071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Fuller-Rowell TE, et al. , Racial disparities in sleep health between Black and White young adults: The role of neighborhood safety in childhood. Sleep medicine, 2021. 81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Yetish G, et al. , Natural sleep and its seasonal variations in three pre-industrial societies. Curr Biol, 2015. 25(21): p. 2862–2868. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 59.Maxwell J, Seasonal variation in vitamin D. Proceedings of the Nutrition Society, 1994. 53(3): p. 533–543. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The authors are willing to share the data upon request, contingent upon the requester completing a Data & Material Transfer Agreement.

RESOURCES