Vitamin D3 for Life

Live Enlightened

Physiology

Vitamin D, Night Work and Time-of-Day Metabolic Measurements: Full Study

Previous related pageVitamin D, Night Work and Time-of-Day Metabolic Measurements: Study SummaryNext related pageFrom Vitamin D Gene Responses to Circadian Timing
Study at a glance

What did studies examine about vitamin D, night work and daily timing?

This page examines two studies about vitamin D and timing. One compared vitamin D blood levels in 425 Danish workers with different work schedules. The other gave 29 women aged 60 to 80 vitamin D3 or water without vitamin D on separate study days, in random order. Researchers measured 20 substances in their blood over 24 hours.

Did night-shift workers have lower vitamin D levels?

Workers on permanent night shifts had about 25% lower vitamin D levels than indoor workers after allowing for measured differences between groups. Workers with rotating night shifts did not show a clear difference. More recorded outdoor time on workdays was linked with higher levels in a smaller seasonal analysis. These links do not prove that night work or a disrupted body clock caused low vitamin D.

Did vitamin D change blood measurements over the day?

After allowing for starting levels on each study day, vitamin D produced a small average increase in acetone, one measured substance. It did not clearly change how any of the 20 substances varied across sampling times. The acetone result does not establish improved fat burning. Meals and fasting could help explain changes during the day.

Do these studies identify the best time to take vitamin D?

Neither study compared taking vitamin D in the morning with taking it in the evening. They cannot identify a best time to take a supplement or show that changing supplement time repairs a disrupted body clock. The permanent-night group was small, and groups were not studied equally across seasons. The treatment study was small and short. Its blood measurements do not establish better sleep or other health benefits.

Read the full article below ↓

Night work, daylight exposure, blood-sampling time and supplement timing describe different scientific questions. An occupational study and an acute crossover trial help separate them: a work-schedule association does not establish clock disruption as its cause, and a response measured after a morning dose does not establish morning as the best dosing time. For the shorter account, see Vitamin D, Night Work and Time-of-Day Metabolic Measurements: Study Summary.

This VD3FL analysis uses both complete principal texts and their tables, the published main figures and relevant supplementary comparisons. It distinguishes baseline-adjusted findings from raw concentration plots.

Why several meanings of timing need separating

In vitamin D research, timing may refer to the opportunity for sunlight exposure, the clock time of a blood sample, the interval since a dose, or the phase of an internal biological rhythm. Treating these as the same exposure can turn a limited result into an unsupported practical recommendation.

This analysis examines two complementary primary studies through that distinction. Daugaard and colleagues compared occupational groups and measured status. Espersen and colleagues compared vitamin D and placebo study days while repeatedly measuring blood metabolites. Their designs illuminate different parts of the question, rather than providing interchangeable tests of Vitamin D and Circadian Biology.

The reader’s practical question may be whether changing supplement time could compensate for an irregular schedule. Answering it requires a direct comparison of relevant timing strategies and meaningful outcomes. Neither simply observing night workers nor collecting samples after one scheduled dose provides that comparison.

What the occupational study measured

Daugaard’s cross-sectional Danish study included 425 workers: 162 indoor, 112 outdoor, 118 rotating-night and 33 permanent-night workers. Indoor and outdoor categories were based on reported outdoor working hours in summer, while night work required more than three working hours between midnight and 05:00. These classifications describe work patterns, not directly measured circadian phase.

Researchers collected a questionnaire and blood sample, then recorded light for seven days. Serum vitamin D status was the sum of 25-hydroxyvitamin D2 and D3 measured by liquid chromatography: tandem mass spectrometry. A recorder worn outside clothing on the upper arm measured illuminance; readings above 1,000 lux provided a proxy for outdoor time. Invalid recording periods and days with insufficient usable data were excluded.

The fully adjusted model in Table 3 estimated 25.3% lower vitamin D concentrations in permanent-night than indoor workers (95% confidence interval: 36.6% to 11.9% lower). Rotating-night workers were estimated 5.6% lower, but their interval ranged from 14.8% lower to 4.4% higher. Outdoor workers were estimated 2.5% higher, also with an interval spanning no difference. The permanent-night result therefore should not be extended to every night-work schedule.

Adjustment included season, age, sex, socioeconomic status, smoking, body mass index, supplement use and other measured factors. Permanent-night workers also had 14.5% higher parathyroid hormone concentrations than indoor workers (95% confidence interval: 0.1% to 31.1%). This additional biochemical association does not identify a clock mechanism or measure a sleep benefit.

Using the paper’s definition of vitamin D insufficiency, below 50 nmol/L, the adjusted odds ratio for permanent-night versus indoor workers was 4.55 (95% confidence interval: 1.39 to 14.94). This wide interval reflects considerable uncertainty. An odds ratio is not a risk ratio or an individual prediction, and the research cut-off should not be converted into a personal treatment target.

Outdoor-time results and limits of the comparison

Table 5 examined 186 workers with usable outdoor-time data who participated from April through September. In the model containing both job group and outdoor hours, each additional recorded outdoor hour per workday was associated with 5.2% higher vitamin D concentrations (95% confidence interval: 1.1% to 9.5%). The corresponding days-off estimate was 1.0% (−2.1% to 4.1%), which did not clearly differ from zero. These are adjusted associations, not predicted benefits from prescribing another hour outdoors.

The permanent-night group was small, and none of its participants were examined from July through September. Seasonal adjustment cannot fully replace overlapping observations across groups. The April: September analysis contained only 15 permanent-night workers, and its job-group confidence intervals were wide. This limits how precisely work schedule can be separated from season and correlated circumstances.

A work schedule may accompany differences in sleep, meals, activity, working conditions and socioeconomic circumstances. The paper measured status at one study visit; it did not assign night work, repeatedly track an internal clock shift or reverse that shift with vitamin D. Its design therefore cannot establish circadian disruption as the cause of the observed status difference.

The wider subject of Indoor Lifestyles and Vitamin D helps explain exposure opportunity. It should not imply that occupational groups differ only in daylight. Nor can a percentage difference in blood concentration be translated into an equally sized sleep or cardiovascular effect.

Visible light, UVB exposure and clock information

The arm-worn recorder provides environmental information, but its visible-light threshold is not a direct measure of the UVB dose reaching uncovered skin. Clothing, shade, sensor placement and the spectral composition of light can separate the recorded signal from the exposure relevant to vitamin D synthesis.

That distinction connects Vitamin D and Sun with Artificial Light and Vitamin D Rhythm Disruption. Light relevant to circadian timing and ultraviolet exposure relevant to vitamin D synthesis are different quantities. High illuminance does not establish that a particular UVB dose reached the skin.

The same schedule can also coincide with different environmental exposures across the year. Vitamin D and Season supplies that context. The observational outdoor-hours coefficient does not establish a safe or effective exposure duration for an individual.

How the acute crossover trial was conducted

Espersen’s analysis included 29 postmenopausal women aged 60 to 80, with a mean age of 69. It compared two arms of a larger five-arm crossover experiment: 500 mL of water with or without 200 micrograms (8,000 IU) of vitamin D3. Each woman received both conditions in randomised sequence, with at least ten days between study days.

A fasting baseline blood sample was obtained at 08:00, followed by the intervention and breakfast. Further samples were collected after 2, 4, 6, 8, 10, 12 and 24 hours; lunch followed the four-hour sample and supper the ten-hour sample. The final sample was fasting. Thus meals, fasting state, elapsed time and clock time all contributed to the conditions under which the profiles were recorded.

Participants had screened below 50 nmol/L for 25-hydroxyvitamin D, with a screening mean of 39 nmol/L. Baseline means were approximately 45 and 50 nmol/L on the first and second study days. Some women had already received vitamin D in other parent-trial arms, so screening status should not be described as an unchanged state throughout the experiment.

Nuclear magnetic resonance spectroscopy quantified 20 blood metabolites for analysis. This was a defined measurement panel, not every circulating metabolite. The principal treatment comparison was vitamin D versus placebo at the scheduled morning hour; it was not morning versus evening administration and cannot identify an optimal dosing time.

What the crossover and statistical models address

The crossover structure compares each participant under both conditions, reducing some between-person variation. The mixed model included treatment, time, their interaction, order, carryover and study day, with participant-related random effects. Measurements were adjusted by subtracting their study-day baseline. Formal period and carryover tests did not suggest such effects, although small samples and different preceding parent-trial exposures still limit precision.

A treatment-by-time interaction asks whether the profiles differ between conditions across sampling times. None of the 20 metabolites had a significant interaction; all interaction P values exceeded 0.15. This does not demonstrate that vitamin D reshaped the daily profile, and it also does not prove that every possible temporal effect is exactly zero.

This matters for Vitamin D Response. If every dose occurs at one hour, elapsed time advances alongside clock time. A placebo day helps interpret shared changes, but a direct dosing-time claim requires a comparison between alternative administration times and relevant outcomes.

Daily sampling does not isolate an internal rhythm

Many metabolites differed across the sampled conditions, with morning and fasting samples tending to separate from afternoon samples. The scheduled meals offer one reason for temporal variation. Without a design separating meals, activity, sleep and internal timing, this pattern describes the study day rather than proving an endogenous circadian mechanism.

There was also a long interval between the 12-hour evening and 24-hour next-morning samples. A line drawn across that gap does not establish the timing or size of an unmeasured overnight peak. These measurements cannot fully describe the overnight profile or establish a shift in clock phase.

The appropriate connection with Vitamin D and Metabolism concerns blood concentrations. Concentration reflects the balance of production, use, transport and clearance; it cannot identify a particular pathway’s rate without additional measurements.

Acetone, exploratory comparisons and multiplicity

Acetone was the only metabolite with a significant baseline-adjusted mean treatment difference: approximately 0.003 mmol/L higher with vitamin D (95% confidence interval: 0.002 to 0.005 mmol/L; P < 0.001). This survived the stated Bonferroni threshold of 0.0025 for 20 comparisons. An average difference and a treatment-by-time interaction test different questions; broadly parallel profiles can have different adjusted means.

The raw Figure 1 acetone concentrations were lower on vitamin D study days, including at baseline. The positive model estimate describes differences after baseline adjustment, not a higher raw concentration throughout the day. Keeping those two quantities distinct prevents the graph and model from being made to tell incompatible stories.

The exploratory supplementary comparisons also identified lower valine at 12 hours with vitamin D. The overall baseline-adjusted valine mean and treatment-by-time interaction were not significant. That isolated time-point comparison should not replace the overall result or be presented as an established reshaping of valine metabolism.

Testing many metabolites and time points creates multiple opportunities for apparently positive findings. The analysis used different correction procedures for exploratory and mixed-model comparisons. The relevant interpretation must retain the comparison being tested and its correction, rather than selecting one favourable point from the wider set.

Even the corrected acetone finding does not establish enhanced fat burning or improved metabolic flexibility. Those claims add unmeasured functional outcomes to a concentration difference. Replication and direct pathway or clinical measurements would be needed to support them.

How this relates to molecular clock research

Maissan and Carlberg’s separate blood-cell analysis found enriched overlap between vitamin D-responsive genes and a circadian gene list. That finding offers molecular context, but it did not include a randomised dosing-time comparison or a full circadian time series in the vitamin D participants.

The distinction within Vitamin D and Gene Expression is parallel to the metabolite distinction: a changed transcript at a sampled time is not equivalent to a changed oscillation. Nor does overlap with a circadian list establish that an occupational status association was mediated by that gene response.

The connection with Vitamin D and Sleep: Wake Regulation remains a question for direct investigation. Sleep timing, clock phase and nutritional status may be related, but measurements from different populations cannot be assembled into a demonstrated causal pathway merely because each concerns timing.

Conclusion

The occupational study identifies a status association in permanent-night workers, while the acute trial separates an isolated mean metabolite difference from unsupported claims of a changed daily profile. The combined lesson is to specify the exposure, clock and outcome being measured. Neither study establishes that changing supplement time repairs circadian disruption or identifies a universally best dosing hour.

Scientific sources

Daugaard S, Garde AH, Hansen ÅM, Vistisen HT, Rejnmark L, Kolstad HA. Indoor, outdoor, and night work and blood concentrations of vitamin D and parathyroid hormone. Scandinavian Journal of Work, Environment & Health. 2018;44(6):647-657. DOI 10.5271/sjweh.3745; PMID 29909424.

Espersen R, Correia BSB, Rejnmark L, Bertram HC. The Diurnal Blood Metabolome and Effects of Vitamin D Supplementation: A Randomised Crossover Trial in Postmenopausal Women. International Journal of Molecular Sciences. 2022;23(17):9748. DOI 10.3390/ijms23179748; PMID 36077145.

Maissan P, Carlberg C. Circadian Regulation of Vitamin D Target Genes Reveals a Network Shaped by Individual Responsiveness. Nutrients. 2025;17(7):1204. DOI 10.3390/nu17071204.

Related reading

Physiology

Vitamin D and Circadian Biology

Explains how vitamin D relates to circadian biology, including links with light exposure, hormone timing, metabolism, immune rhythms, and sleep: wake regulation, without making treatment claims.

Read more →

Original sources

Source

The Diurnal Blood Metabolome and Effects of Vitamin D Supplementation: A Randomised Crossover Trial in Postmenopausal Women

Published in International Journal of Molecular Sciences (2022).

Publication details

Espersen R, Correia BSB, Rejnmark L, Bertram HC. The Diurnal Blood Metabolome and Effects of Vitamin D Supplementation: A Randomised Crossover Trial in Postmenopausal Women. International Journal of Molecular Sciences. 2022;23(17):9748. DOI 10.3390/ijms23179748; PMID 36077145.

Read more →