Vitamin D3 for Life

Live Enlightened

Physiology

Interpreting Vitamin D Status in Older Adults: Full Evidence Review

Previous related pageInterpreting Vitamin D Status in Older Adults: Study SummaryNext related pageAge-Related Changes in Vitamin D Metabolism, Activation and Clearance: Full Evidence Review
Study at a glance

What question did the studies ask?

What do vitamin D blood tests tell us about older adults? Researchers examined total 25-hydroxyvitamin D, a form that cells can change into active hormone. They also examined the portion of this compound not bound to blood proteins, hormone results and comparisons between related vitamin D compounds. Most studies observed people without assigning treatment.

What did they find?

Some measurements were associated with fractures or bone loss. Free vitamin D had mixed associations with health outcomes and was not shown to guide treatment better than total vitamin D. Normal calcium or related hormone results did not reliably exclude a low vitamin D result in one selected group over 70.

What does this mean?

A blood marker associated with a health problem does not necessarily identify a treatment that prevents it. Proposed new cutoffs and seasonal blood levels have not been established as treatment targets. The findings do not justify routinely replacing the usual test.

What are the limits?

Several papers reused participants. Most findings relied on one blood sample. Kidney function, health differences and test methods affected interpretation. Some studies included middle-aged people or relatively healthy older adults, limiting what they show about severe frailty or advanced kidney disease.

Read the full article below ↓

Interpreting vitamin D status in older adults requires more than comparing a blood result with a single threshold. This review asks what total vitamin D measurements, hormone responses, free concentrations and metabolite ratios can establish about function and future health. Studies in older populations show that these measures can give different answers. They do not yet establish a replacement test that reliably identifies who will benefit from treatment. Read the shorter account, Interpreting Vitamin D Status in Older Adults: Study Summary.

What does a vitamin D result measure?

The usual status measurement is total 25-hydroxyvitamin D, abbreviated 25(OH)D. It describes a circulating precursor that cells can convert into the active hormone, calcitriol. Total concentration combines vitamin D2-derived and D3-derived forms when the assay measures both adequately. It is not a direct reading of vitamin D receptor activity in every tissue. Nor does a single concentration describe all the effects of kidney function, binding proteins, recent intake and timing on vitamin D metabolism.

Several proposed alternatives measure different aspects of this system. Free 25(OH)D is the small fraction not bound to blood proteins. It may be measured directly or estimated from total concentration and binding-related assumptions. A vitamin D metabolite ratio compares a breakdown product with its precursor. Parathyroid hormone, which helps regulate blood calcium, provides another biological response measure. These are related quantities, but they are not interchangeable tests of a single hidden state.

Five core observational study populations inform this review: the Cardiovascular Health Study, Health ABC, ESTHER, a Salford clinical sample and SarcoPhAge. Several publications reuse the same cohorts. Their findings therefore provide complementary analyses rather than independent confirmation from every paper. A separate short trial supplies additional context. Their participants also differ substantially: relatively healthy community residents, people selected for clinical testing and adults ranging from late middle age to older adulthood cannot simply be pooled into one representative ageing population. The broader functional-status concept is introduced in Functional Vitamin D Status.

A normal hormone result does not exclude a low vitamin D concentration

Heald and colleagues examined laboratory requests for 241 people over 70 in Salford, with complete results available for 236. The study excluded people with an estimated glomerular filtration rate below 60 mL/min/1.73 m², a measure of kidney filtration adjusted for body surface area and some specialist referrals. Vitamin D2 and D3 metabolites were measured using liquid chromatography with mass spectrometry, a method that separates and identifies compounds. Calcium, phosphate, alkaline phosphatase and parathyroid hormone were measured in the same samples.

Many people with low 25(OH)D had normal routine bone-profile results or normal parathyroid hormone. This supports a narrow but useful interpretation: normal results for those other tests did not reliably rule out a low vitamin D concentration in this selected sample. It does not establish that everyone over 70 should be tested, that a low result explains a person’s symptoms, or that screening improves health. The study did not compare testing strategies or treatment outcomes.

Clinical referrals also differ from a population survey. Diagnoses, medicines and supplement use were unavailable, and people with reduced kidney function were excluded. Some patients overlapped with an earlier general-adult Salford report. Consequently, neither the frequency of low values nor the biochemical relationships should be presented as a universal description of older people, especially those with advanced kidney disease. Analytical limitations are examined in Circulating Vitamin D Measurements: Physiological and Analytical Limitations: Full Study.

Bone outcomes do not all follow the same marker

Barbour and colleagues studied fractures in Health ABC participants originally recruited at ages 70 to 79 with relatively good physical function. The principal analysis included 2,614 people. Vitamin D and parathyroid hormone were measured once, and subsequent fractures were followed prospectively. This design can examine whether a baseline marker is associated with later events, but it cannot determine what would happen if treatment changed that marker.

The lowest vitamin D category did not show a clear, precise increase in hip-fracture risk across the analysis. The fully adjusted estimate comparing the lowest and highest quartiles was 1.73, with a 95% confidence interval from 0.80 to 3.75. That interval was compatible with markedly different effects, and there was no clear trend across the categories. Nonspine fractures and parathyroid hormone also did not establish a consistent relationship. No significant fracture threshold was identified.

This does not prove that vitamin D is irrelevant to bone. The number of hip fractures was limited, and the cohort was healthier than many frail or institutionalised populations. It does show why a hormonal response, a bone-density change, a fall and a fracture cannot be treated as equivalent outcomes. A proposed biochemical threshold needs evidence linking it to the particular clinical decision being considered. The skeletal physiology background is introduced in Vitamin D and Bone.

What does the metabolite ratio add?

Ginsberg and colleagues examined the ratio of 24,25-dihydroxyvitamin D3 to 25-hydroxyvitamin D3 in the Cardiovascular Health Study. This ratio uses D3-specific compounds, not a denominator combining all D2 and D3 forms. The breakdown product reflects metabolism, but the ratio is not a direct measurement of receptor signalling inside cells. Kidney physiology and other influences remain relevant to its interpretation.

In the fracture analysis, a higher ratio was associated with lower fracture risk after adjustment: the hazard ratio per standard-deviation increase was 0.74, with a 95% confidence interval of 0.61 to 0.88. Total 25(OH)D did not show a clear association in the corresponding model. However, parathyroid hormone associations were similar across the markers, and the ratio did not show a bone-density association. Bone density in the smaller subset had been measured about two years before the metabolite sample, further limiting a direct comparison.

A later Health ABC analysis provided a separate cohort comparison, although it reused the Health ABC population discussed above. Among 786 participants, a 50% lower metabolite ratio was associated with approximately 0.3 percentage points more annual hip bone-density loss and with higher fracture risk. The adjusted fracture hazard ratio was 1.49, with a 95% confidence interval of 1.06 to 2.08. The corresponding 25-hydroxyvitamin D3 association was uncertain.

These results make the ratio scientifically interesting. They do not establish that it is a clinically superior test. Both studies used observational measurements; the Health ABC sample was selected partly around kidney-function decline and analysed with weighting. A stronger association with an outcome does not by itself show better prediction for an individual, a useful treatment threshold, or improved health when clinicians act on the result. Other proposed biological response measures are assessed in Functional Vitamin D Status: Biological Response Beyond Circulating Levels: Full Study.

A proposed functional-deficiency threshold remains exploratory

Ladang and colleagues studied the metabolite ratio in SarcoPhAge, a Belgian community cohort of adults aged 65 or older. The analysis used 204 participants with available samples, substantially fewer than the original recruited cohort. The ratio could be assessed in 199, and only 14 fell into the proposed low-ratio category. The cutoff of 4% was a proposed classification rather than a threshold validated by a treatment trial.

Lower ratios were associated with higher subsequent mortality. However, the low-ratio group contained only six deaths, estimates were imprecise, and the reported mortality models adjusted for age, sex and body mass index rather than the full set of relevant kidney, dietary and exposure factors. Comparing the size of that association with an association for low total vitamin D does not establish superior prediction: the categories themselves differ.

The biochemical findings also qualify a simple functional-deficiency interpretation. The ratio correlated with parathyroid hormone, but the group below the proposed cutoff did not have significantly higher parathyroid hormone. Bone-turnover markers did not establish a clear relationship either. A correlation across a continuous range and a useful diagnostic cutoff are different claims. The study supports further investigation, not a validated label that can be applied to every older patient.

Free vitamin D: measured and calculated results differ

Zhu and colleagues analysed 5,899 ESTHER participants aged 50 to 75. They calculated free and bioavailable 25(OH)D from total vitamin D, binding protein, albumin and assumptions about binding affinity. These fractions were not directly measured. The strong correlations between total and calculated fractions were partly expected because total concentration was an input to the calculation. They cannot serve as independent validation of a better test.

The ESTHER analysis described distributions and determinants rather than proving improved prediction of fractures or response to treatment. It also largely represented late middle age and younger older adulthood in a relatively homogeneous population. The paper’s label for binding-protein-bound vitamin D should not be taken to mean that this fraction can never become available to tissues. A calculation’s terminology does not establish a universal biological boundary.

Cheng and colleagues instead directly measured free 25(OH)D in a Health ABC sample of 786 people, with a mean age of 75. The free assay was less precise than the assay for total concentration. Higher free values were associated with lower heart-failure risk but higher risk of kidney-function decline; associations with overall cardiovascular disease and fractures were uncertain. Total vitamin D also showed an inverse heart-failure association.

These mixed findings do not establish that free vitamin D protects the heart or damages the kidney. They are observational associations based on a single measurement, with residual confounding and several outcomes examined. Nor did the analysis demonstrate that measuring free vitamin D improves decisions beyond total concentration. It is another Health ABC analysis, not an additional independent population confirming the earlier metabolite findings. The wider evidence for free and total measurements is reviewed in Free Versus Total Vitamin D Measurement and Interpretation: Full Study.

Season and thresholds need careful interpretation

De Boer and colleagues studied 1,621 White older adults in the Cardiovascular Health Study who were free of selected major diseases at baseline. They used the data to derive season-specific vitamin D thresholds for a combined outcome including hip fracture, myocardial infarction, cancer or death. Values below the selected thresholds were associated with a higher combined event rate.

Those thresholds were observational risk categories, not established treatment targets. The same dataset helped choose and evaluate them, and the individual outcomes had different biological meanings. The estimated improvement in risk reclassification was uncertain: 2.4%, with a 95% confidence interval from −0.6% to 5.3%. The analysis therefore did not demonstrate a statistically clear improvement in reclassification. It also shared the wider CHS population with the metabolite-ratio report.

Season can help explain the context of a blood sample without automatically requiring a different clinical target for every season. Similarly, an association between a low measurement and later illness may reflect differences in health, activity or exposure that statistical adjustment cannot fully remove. It cannot show that raising the measurement will reverse the associated risk. Evidence for natural changes over time is reviewed in Physiological Variability in Vitamin D Measurements: Seasonal, Diurnal and Metabolic Effects: Full Study.

From measurement to demonstrated usefulness

A functional interpretation should distinguish five questions: what an assay measures; how that measurement relates to biological processes; whether it is associated with future events; whether it improves prediction; and whether using it improves clinical outcomes. The evidence is much stronger for some of the first three questions than for the last two. Calling a marker functional does not complete those remaining steps.

A short post hoc analysis from the Styrian trial illustrates the distinction. Among hypertensive adults classified using proposed functional-deficiency criteria, supplementation changed vitamin D metabolites and parathyroid hormone without establishing clear changes in the main bone-turnover outcomes. This was a small, short study in adults mostly younger than the geriatric cohorts, not validation of a testing strategy for older people. It neither proves the proposed ratio useless nor establishes clinical superiority.

The reviewed studies also use different assays, stored samples, age ranges and selection criteria. Few findings can be assumed to apply unchanged to the oldest-old, severe frailty or advanced kidney disease. Single measurements and incomplete adjustment limit causal conclusions. Measurement precision, biological interpretation and evidence of treatment benefit must therefore remain separate. These studies provide a more informed way to understand uncertainty, but they do not establish a universal blood target, a personal supplement dose or a reason to replace total 25(OH)D testing routinely. The wider ageing subject is introduced in Vitamin D and Age-Related Decline.

Sources

Related reading

Original sources

Source

Serum 25-hydroxyvitamin D concentration and risk for major clinical disease events in a community-based population of older adults: a cohort study

Published in de Boer IH et al (2012).

Publication details

Serum 25-hydroxyvitamin D concentration and risk for major clinical disease events in a community-based population of older adults: a cohort study. de Boer IH et al. 2012. Annals of internal medicine. Volume 156(9); pages/article 627-634. DOI: 10.7326/0003-4819-156-9-201205010-00004.

Read more →
Source

Serum free 25(OH)D concentrations and cardiovascular disease, heart failure, kidney function decline, and fracture: the health, aging, and body composition study

We evaluated the relationship of free vitamin D with incident cardiovascular disease (CVD), heart failure (HF), kidney function decline (KFD) and fracture, among 786 participants in the Health Aging and body composition study.

Publication details

Serum free 25(OH)D concentrations and cardiovascular disease, heart failure, kidney function decline, and fracture: the health, aging, and body composition study. Cheng JH et al. 2025. JBMR Plus. Volume 9(3); pages/article ziaf001. DOI: 10.1093/jbmrpl/ziaf001.

Read more →
Source

Classification of Vitamin D Status Based on Vitamin D Metabolism: A Randomized Controlled Trial in Hypertensive Patients

In this work, we aim to evaluate the hypothesis that the definition of functional vitamin D deficiency is able to identify individuals who significantly benefit from vitamin D supplementation.

Publication details

Classification of Vitamin D Status Based on Vitamin D Metabolism: A Randomized Controlled Trial in Hypertensive Patients. Zelzer S et al. 2024. Nutrients. Volume 16(6); pages/article 839. DOI: 10.3390/nu16060839.

Read more →