Vitamin D dosing by blood level, not by bottle

Take the same vitamin D capsule as the person next to you and you will not end up at the same blood level. Of everything one review could measure, dose relative to body weight explained the most: 34.5% of the gap between groups of people taking supplements.¹

Why the same capsule lands in different places

A vitamin D test measures 25-hydroxyvitamin D, the form your liver makes from what you swallow or make in sunlight. The studies here report it in nanomoles per liter. A US lab prints nanograms per milliliter, which runs on a smaller scale, so those numbers will not line up with these.

Two things shape that gap. The lower your level to begin with, the more each capsule moves it, and the rise flattens as you climb. And a heavier body distributes the same dose through more tissue, so it lands lower.

The vitamin D data behave that way. One review gathered 94 studies to ask why identical doses produce such different results. Dose per kilogram of body weight explained 34.5% of the variation.¹ Which form people took explained 9.8%, age 3.7%, calcium taken alongside 2.4%, and the starting level 1.9%.¹ And the line bent rather than ran straight. Each extra unit moved the blood level less than the one before.¹

What explained the gap between groups on the same dose% of variation

Dose per kg of body weight34.5%

Form taken (D2 vs D3)9.8%

Age3.7%

Weight matters because a capsule holds a fixed amount, not an amount matched to you. Take 800 IU a day and that is about 13 IU per kilogram for someone who weighs 60 kg, and 8 for someone who weighs 100 kg.^1,2 That is much of the reason two people sharing a bottle end up in different places.

What a dose does to the number

In trials of vitamin D3, people who started at an average of 50.4 nmol/L gained about 37.² Four things predicted the size of the rise: the dose, how long it was taken, the starting level and age.²

The starting level worked backwards, and that is the whole argument for dosing by blood level rather than by label. The lower you begin, the further the same capsule moves you. People who began under 50 nmol/L gained 39.6.²

Two other patterns are worth having. Rises were bigger after six to twelve months than after short courses, at 41.7 nmol/L, so a level checked after a few weeks may understate where a dose will settle.² And adults over 80 still gained 40.5 nmol/L.² That cuts against the usual assumption that older bodies get less out of a given dose.

What the blood number does not tell you

A blood level is not the point of anything on its own. It is worth reaching only if something further down the line depends on it.

The clearest look at that comes from people with type 2 diabetes or prediabetes. Set against people at 50 nmol/L or above, those sitting between 25 and 50 were 36% more likely to die during follow-up, and those below 25 were 58% more likely.³ Risk was lowest at around 60 nmol/L.³

Then the trials in that same population gave some of them vitamin D, and deaths did not fall.³ So a low level travels with dying. Raising it with a capsule has not been shown to change that. Dosing by blood level is an efficient way to hit a target. It is not evidence the target is worth hitting.

D3, D2 and how often you take it

Two forms sit on the shelf: D3, mostly from animal sources, and D2, from yeast and mushrooms. How much the choice matters depends on the schedule. An early analysis found D3 clearly beat D2 when both were given as an occasional large dose, and the advantage disappeared when either was taken daily.⁴

A newer look at daily dosing found a gap after all. On the same daily amount, D2 raised total 25-hydroxyvitamin D by 10.39 nmol/L less than D3, a shortfall of about 40%.⁵ Body size looked like the reason the two analyses disagreed: the difference between the forms faded in people with a body mass index above 25.⁵

Daily D2 against daily D3: change in total 25(OH)D, nmol/L0 = no effect

D2 group minus D3 group-10.39 (-14.62 to -6.16)

The form question keeps turning back into the body-size question.

The ceiling, and when a test earns its place

The ceiling sits higher than most bottles, which is a fact about safety margins and not a dose to aim at. One risk assessment worked through the human trials and found no toxicity in healthy adults at 250 micrograms a day, which is 10,000 IU of D3.⁶ It argued that the ceiling standing at the time, 2,000 IU, was too restrictive.⁶ That is one group's reading of the trials, not a dose to take, and ordinary dosing decisions happen far below it, in the stretch where the response is still steep.

Testing is where the guidance is narrower than its reputation. The Endocrine Society suggested measuring 25-hydroxyvitamin D as the first test in patients at risk of deficiency, and treating deficiency with either form.⁷ It also said there was not enough evidence to screen people who are not at risk, or to prescribe vitamin D for heart protection.⁷ The guideline most often cited to justify a test declines to recommend one for people with no reason to suspect a problem.

What this leaves out

Most of the difference between people is still unexplained. The five things that review could measure account for a minority of the variation between groups, which leaves the rest to genetics, sunlight, gut absorption and things nobody has isolated.¹

The comparison of the two forms rests on thin numbers, and the older analysis could not check whether age, sex or ethnicity changed the answer.^4,5

The bigger gap is at the far end. The link between a low level and dying comes from watching people rather than assigning them a dose, and from a group selected for diabetes.³ Nobody has taken two groups, dosed one from its blood levels and the other from a fixed label, and followed both for years. Until that exists, dosing by blood level is a claim about arithmetic, not about health.