Vitamin D3 for Life

Live Enlightened

Article

Free Versus Total Vitamin D Measurement and Interpretation: Full Study

Previous related pageFree and Total Vitamin D: Binding, Measurement and Interpretation: Evidence SynthesisNext related pageFree Versus Total Vitamin D Measurement and Interpretation: Study Summary
Study at a glance

Does measuring free vitamin D give more useful information than the total?

This page asks whether measuring free vitamin D gives more useful information than measuring the total amount. The studies measured 25-hydroxyvitamin D. Most of this form is carried by proteins in blood. Free means unattached to these proteins; total includes both attached and free amounts.

Who was studied to compare free and total vitamin D tests?

Several human studies included healthy adults, pregnant women, people with liver disease and nursing-home residents. A study of 155 people compared free amounts measured directly with amounts estimated using calculations. Other studies examined inherited carrier differences, supplement responses and rare cases of missing carrier protein.

When did free and total vitamin D results differ?

Free and total measurements often changed together, but their relationship differed in some health conditions. People with severe liver disease could have low total levels without low free levels. Calculated free amounts did not closely match direct measurements in the 155-person study. In another study, the laboratory method created large apparent differences between population groups that other methods did not reproduce.

Should free vitamin D testing replace total testing for everyone?

Total measurements remain useful, but carrier proteins and laboratory methods affect how results should be understood. Free measurements may add information when carrier proteins change greatly. The evidence does not show that free testing should replace total testing for everyone or that it leads to better health. Neither measurement directly shows how vitamin D acts in every tissue.

Read the full article below ↓

The clinical question is not simply free versus total

Total 25(OH)D is widely used because it captures the main circulating reservoir and is supported by established laboratory practice. Free 25(OH)D asks a narrower physical question: how much 25(OH)D is unbound at the time of measurement? Because most 25(OH)D is protein-bound, the two measures can respond differently when DBP or albumin changes. See Study Summary.

That difference creates a tempting hierarchy in which free must be more biologically meaningful. The human evidence does not support that blanket conclusion. The relevant question is conditional: when does carrier biology or assay behaviour make total concentration less representative of the free pool, and does measuring free improve interpretation enough to matter clinically?

Direct measurement and calculation are different analytical routes

Free 25(OH)D can be measured directly or estimated mathematically from total 25(OH)D, DBP, albumin and assumed binding constants. Those routes should not be treated as equivalent. A calculation inherits every error in the measured inputs and every assumption in the affinity model. If the DBP assay is biased by genotype, the calculated free value can be biased even when the mathematics is executed perfectly.

Schwartz and colleagues compared directly measured and calculated free 25(OH)D in 155 people, including 24 with cirrhosis and 20 pregnant women. Calculated free 25(OH)D correlated with direct measurement, but it explained only about 13% of the variability and tended to be higher. This is a major warning against treating calculated free 25(OH)D as a transparent substitute for direct measurement.

The wider distinction between analytical variability and biology is developed in Variability in Vitamin D Measurements and Limitations of Vitamin D Blood Tests.

DBP assay choice can create an apparent biological difference

Nielson and colleagues examined 1,057 men from US, UK and Gambian populations using several ways of assessing DBP, including monoclonal and polyclonal immunoassays and proteomic measurement. The monoclonal assay produced large apparent ancestry-associated differences in DBP that were not reproduced by the other methods. This mattered because DBP values had been used to calculate free or “bioavailable” vitamin D.

Directly measured free 25OHD correlated strongly with total 25OHD, at roughly r = 0.84, and the dramatic racial-genotypic pattern inferred from the problematic DBP assay was not supported across the alternative methods. The lesson is not that genotype never matters. It is that an assay can manufacture or magnify a biological-looking difference when its binding to DBP variants is unequal.

This finding materially changed how earlier claims about ancestry, DBP and calculated bioavailable vitamin D should be read. It is a good example of why measurement architecture belongs inside scientific interpretation rather than in a footnote.

Ordinary DBP phenotypes do not create two separate monitoring systems

Sollid and colleagues used samples from a randomised supplementation study involving 472 adults. They measured total 25(OH)D by LC-MS/MS, directly measured free 25(OH)D and DBP phenotype. Common phenotypes were associated with differences in DBP and total 25(OH)D, while differences were smaller for free 25(OH)D.

Vitamin D supplementation increased the measured vitamin D variables, but the response did not materially diverge according to DBP phenotype, sex or age. This is important negative evidence. It argues against assuming that ordinary genetic variation in DBP requires a completely different clinical monitoring strategy.

The study also shows why the free-total question cannot be answered by rare genetics alone. Common variation is a much more relevant test of routine physiology, and here the two measures remained substantially coupled.

Cirrhosis is a clear example of altered binding context

Cirrhosis can reduce albumin and DBP, making it a strong test of what happens when carrier capacity changes. In the 2014 Schwartz study, total 25(OH)D, DBP and albumin were lowest in the cirrhosis group, yet directly measured free 25(OH)D was highest. Earlier liver-disease work likewise found lower total concentrations with a higher percentage free and broadly preserved absolute free 25(OH)D.

These observations show that total 25(OH)D can become harder to interpret in a major altered-protein state. They do not establish a universal cirrhosis-specific free 25(OH)D threshold, nor do they prove that free measurement predicts clinical outcomes better. They establish divergence in the measured compartments.

The measurement lesson is that a low total result can partly reflect reduced carrier capacity rather than a proportionate fall in every circulating fraction. This does not make the total result meaningless; it changes the physiological context in which the number should be interpreted.

Pregnancy changes DBP without producing a simple free-hormone prediction

Pregnancy provides the opposite carrier perturbation because DBP rises. A simple calculation might predict that more binding protein should necessarily drive free 25(OH)D downward. In the 2014 direct-measurement study, DBP was highest in pregnant participants, but measured free 25(OH)D did not differ significantly from the comparison group.

That result is useful because it resists an overly mechanical interpretation of binding equations. Physiology can alter total concentration, carrier abundance and equilibrium together. Direct measurement can therefore reveal a relationship that is less intuitive than a calculation based on fixed assumptions.

The conclusion here remains restricted to how pregnancy demonstrates that binding-state changes can complicate free-total interpretation. The study does not establish a pregnancy-specific universal free 25(OH)D target or a general rule that total 25(OH)D becomes uninformative during pregnancy.

A larger multicentre dataset maps where the relationship changes

The 2018 Schwartz analysis combined data from seven academic sites and 1,661 adults. The population included healthy adults, people with prediabetes, general outpatients, people with cirrhosis, pregnant women and nursing-home residents. Total 25(OH)D was measured by LC-MS/MS and free 25(OH)D by a direct immunoassay, with DBP, albumin, creatinine, PTH and DBP haplotype also available.

Across this diverse population, free and total 25(OH)D remained related, but the relationship was not identical across health states. Cirrhosis and nursing-home groups showed notable differences, and BMI contributed additional variation. The study therefore supports a model in which total is usually informative while particular physiological or clinical states alter the conversion between total and free.

Crucially, the investigators did not claim that this established clinical superiority of free 25(OH)D. They concluded that the clinical importance of free measurement still needed outcome-based investigation. That boundary remains central.

Rare DBP deficiency proves divergence but not routine superiority

The Henderson GC-deletion case is an extreme stress test of total 25(OH)D interpretation. With essentially no circulating DBP, total 25(OH)D was extraordinarily low while free 25(OH)D was far less extreme and mineral homeostasis was comparatively preserved. Banerjee’s later unrelated GC loss-of-function case reinforced the point that severe reductions in total vitamin D metabolites caused by absent DBP do not necessarily produce the phenotype predicted from total concentration alone.

These cases are compelling because they remove the carrier rather than merely correlating carrier concentration with vitamin D measurements. They are also clinically exceptional. A rare homozygous loss-of-function genotype cannot be used to conclude that a person with normal DBP and an ordinary low total 25(OH)D has adequate free vitamin D or adequate tissue action.

The physiological basis for that distinction is developed in Vitamin D Binding Protein Explained and Free, Albumin-Bound and Total Vitamin D in the Blood: Full Study.

Does free 25(OH)D predict physiology better?

The evidence is mixed and does not justify a universal yes. Some studies in altered binding states find free measures more closely aligned with the physiological situation than total alone. Within the selected human evidence, PTH relationships do not establish a consistent advantage for free measurement, and supplementation data show that total and free often move together.

This is exactly what the binding model predicts. When DBP and albumin behave within ordinary ranges, total concentration carries substantial information about the free pool. When carrier concentration, affinity or assay measurement is abnormal, the mapping can loosen. A test can therefore be more informative in a subgroup without being a universally superior replacement.

This distinction is consistent with What Vitamin D Levels Represent: a circulating concentration is a marker within a physiological system, not a direct measurement of every downstream biological effect.

Why directly measured free 25(OH)D still has limitations

Direct measurement avoids some assumptions required by calculation, but it does not eliminate analytical uncertainty. Free 25(OH)D exists at very low concentrations, making measurement technically demanding. Direct immunoassays were widely used in the human studies discussed here, and the studies themselves show that method choice and calculation assumptions materially affect comparison across datasets.

Analytical method matters at both ends of the comparison. In these studies, total 25(OH)D was commonly measured by LC-MS/MS, while DBP measurement was strongly method-dependent and direct free immunoassay results could not simply be replaced by calculated values. A result labelled “free vitamin D” therefore needs its method understood before it is compared across studies or converted into a clinical threshold.

For readers trying to understand the difference between the metabolite being measured and the assay used to measure it, Vitamin D Testing: 25-OH-D vs 1,25-OH₂-D provides the metabolite-level testing context.

What can and cannot be concluded clinically

The evidence supports continued use of total 25(OH)D as a broadly useful conventional marker. It also supports recognising that total concentration is carrier-dependent and that free measurement can add interpretive information when DBP or albumin is markedly altered. It supports scepticism toward calculated free values when DBP assay performance or binding assumptions are uncertain.

The evidence does not establish a universal free 25(OH)D target, does not show that direct free measurement should replace total 25(OH)D in routine screening, and does not show that free concentration directly measures tissue vitamin D action. It also does not justify interpreting a rare DBP-null patient as evidence that ordinary low total 25(OH)D is clinically irrelevant.

The most defensible interpretation is therefore conditional. Ask what was measured, how it was measured, whether binding proteins are likely to be unusual, and what clinical question the number is being asked to answer.

Primary studies

Sollid ST et al. European Journal of Endocrinology, 2016. Effects of vitamin D binding protein phenotypes and vitamin D supplementation on serum total 25(OH)D and directly measured free 25(OH)D.

Nielson CM et al. Journal of Clinical Endocrinology & Metabolism, 2016. Free 25-Hydroxyvitamin D: Impact of Vitamin D Binding Protein Assays on Racial-Genotypic Associations.

Schwartz JB et al. Journal of Clinical Endocrinology & Metabolism, 2014. A Comparison of Measured and Calculated Free 25(OH) Vitamin D Levels in Clinical Populations.

Schwartz JB et al. Journal of Clinical Endocrinology & Metabolism, 2018. Determination of Free 25(OH)D Concentrations and Their Relationships to Total 25(OH)D in Multiple Clinical Populations.

Henderson CM et al. New England Journal of Medicine, 2019. Vitamin D-Binding Protein Deficiency and Homozygous Deletion of the GC Gene.

Banerjee RR et al. Journal of the Endocrine Society, 2021. Very Low Vitamin D in a Patient With a Novel Pathogenic Variant in the GC Gene That Encodes Vitamin D-Binding Protein.

Bikle DD et al. Journal of Clinical Investigation, 1986. Free 25-hydroxyvitamin D levels are normal in subjects with liver disease and reduced total 25-hydroxyvitamin D levels.

Related reading

Library

Diverse Effects of Combinations of Maternal-Neonatal VDR Polymorphisms and 25-Hydroxyvitamin D Concentrations on Neonatal Birth Anthropometry: Functional Phenocopy Variability Dependence, Highlights the Need for Targeted Maternal 25-Hydroxyvitamin D Cut-Offs during Pregnancy

The aim of this study was to evaluate the combined effect of maternal and neonatal VDR polymorphisms (ApaI, TaqI, BsmI, FokI, Tru9I) and different maternal and neonatal 25(OH)D cut-offs on neonatal birth anthropometry.

Original paper ↗Read more →

Original sources