Study at a glance
What did long-term studies examine about vitamin D, inflammation and ageing?
This review examines three different studies, called VITAL, DO-HEALTH and TILDA. VITAL tested supplements and blood signs of inflammation. DO-HEALTH tested supplements and exercise using estimates of how the body was ageing, based on chemical patterns on DNA. TILDA followed people over time to compare starting vitamin D levels with later categories of blood-sugar levels. A finding for one of these outcomes cannot stand for the others.
Did supplements change inflammation or estimates of ageing?
In VITAL, one blood marker of inflammation was lower with vitamin D at two years, but not at four years. Other inflammatory substances showed no consistent benefit. In the DO-HEALTH analysis of 777 older adults, vitamin D alone did not clearly change the four ageing estimates studied. Some results involved omega-3 or combined treatments, so they cannot be credited to vitamin D alone.
Were vitamin D levels linked with later blood-sugar problems?
Lower starting vitamin D levels were linked with later prediabetes, meaning raised blood sugar below the diabetes range. The relationship with new diabetes cases was uncertain. TILDA did not randomly assign supplements, so it cannot show that low vitamin D caused the blood-sugar changes. A comparison with people who had higher vitamin D levels does not establish a treatment target.
Do the studies show that vitamin D slows ageing or prevents disease?
The studies do not show that vitamin D reduces inflammation and thereby slows ageing of the immune system or prevents metabolic disease, which affects how the body processes nutrients. They did not test that whole sequence. DNA-based ageing estimates do not directly measure how well the immune system works or how long someone lives. VITAL had fewer blood samples at later assessments, DO-HEALTH analysed a selected part of its study group, and other differences between people could affect the TILDA findings. These results do not establish an anti-ageing treatment.
Read the full article below ↓Long-term vitamin D research does not provide one continuous demonstration that low status causes chronic inflammation, accelerates immune ageing and then produces metabolic disease. The studies reviewed here measure different parts of that proposed sequence. VITAL examined inflammatory blood markers, DO-HEALTH examined DNA-methylation measures of biological ageing, and TILDA examined later glycaemic status. Their results must remain distinct before they can be interpreted together. Read the companion Vitamin D, Long-Term Inflammation, Biological Ageing and Metabolic Risk: Study Summary.
Three questions, three kinds of evidence
The VITAL biomarker analysis by Dong and colleagues tested randomised vitamin D assignment against changes in selected inflammatory markers over several years. The DO-HEALTH analysis by Bischoff-Ferrari and colleagues evaluated biological-age estimates in a subset of a factorial trial. McCarthy and colleagues’ TILDA analysis related baseline vitamin D status to later prediabetes and diabetes in an observational cohort.
These are independent participant programmes, but independence does not make their outcomes equivalent. A lower inflammatory protein concentration is not the same as a slower epigenetic clock, and neither is equivalent to a lower incidence of diabetes. A coherent physiological hypothesis can connect these topics without making each study evidence for every step.
The starting framework is Vitamin D and Chronic Inflammation. This review asks what the long-term evidence actually supports about inflammatory burden and related ageing or metabolic measurements. It also identifies the crucial missing link: these papers do not directly demonstrate that vitamin D preserves immune function over time through reduced chronic inflammation.
VITAL: inflammatory markers at two and four years
The VITAL analysis included 1,054 participants in its biomarker sample, with 520 assigned to vitamin D and 534 to placebo. Mean age was approximately 64.9 years, about half were women, and most participants were white. Mean baseline 25-hydroxyvitamin D was about 28.1 ng/mL; roughly 17% were below 20 ng/mL. This was not a trial restricted to severe deficiency.
Participants were assigned 2,000 IU of vitamin D daily within a factorial trial that also tested omega-3 supplementation. This dose describes the research intervention and is not a recommendation. Blood sampling at two and four years allowed the investigators to assess whether differences in inflammatory markers appeared and whether they persisted.
The numbers contributing samples declined over time: approximately 1,051 at baseline, 977 at two years and 685 at four years. That change affects precision and the composition of the available sample. It should not be ignored when comparing the two follow-up points or interpreting the absence of a later association.
The measured markers included high-sensitivity C-reactive protein, IL-6, IL-10 and TNF-α. These markers represent selected aspects of inflammation. They do not amount to a complete survey of immune activation in every tissue, and they do not directly quantify the duration or severity of an individual’s inflammatory symptoms.
The hs-CRP signal was not sustained
At two years, vitamin D assignment was associated with lower hs-CRP in the reported model. The coefficient on the natural-log hs-CRP scale was approximately −0.17, with a standard error of 0.06; the nominal P value was 0.007 and the multiple-comparison-adjusted value was 0.028. The repeated-measures model adjusted for omega-3 assignment, age, sex, race, BMI, smoking and medication use for hypertension, diabetes or high cholesterol. The finding applies to this marker and time point.
At four years, the comparable hs-CRP finding was absent. The other measured cytokines did not show a consistent beneficial pattern across follow-up. A result at one time point therefore cannot be presented as sustained suppression of systemic inflammation throughout the trial.
The paper’s percentage description and a simple exponentiation of the reported coefficient do not yield an identical percentage. To avoid implying a precision the reporting does not consistently support, this review retains the model coefficient and the direction of the finding rather than treating one percentage as an exact universal effect.
Other analysis choices also limit stronger claims. A compliance-related nominal finding did not remain persuasive after the reported adjustment for multiple comparisons. Subgroup interactions did not establish a reliable category of participants with a clearly different response. Selecting the most favourable analysis would overstate the consistency of the evidence.
The findings inform Vitamin D and Inflammatory Signalling, but they do not identify which intracellular pathway produced the hs-CRP difference. Neither a mechanistic laboratory result nor a plausible inflammatory cascade can substitute for a measured mediation analysis in the trial. The result is a marker difference at a specified time point.
Measurement limits in the inflammatory analysis
Approximately 29% of IL-6 measurements were outside the assay’s quantifiable range. This restricts the information available from that analyte. A null statistical result in such a setting should not be described as a precise demonstration that all participants had unchanged IL-6 biology.
The population’s baseline vitamin D distribution also matters. Most participants were not below the reported deficiency threshold. The study therefore supplies limited direct evidence about a long-term intervention restricted to markedly deficient individuals. At the same time, this limitation does not establish that a larger benefit would occur in such individuals; it leaves that question incompletely answered.
VITAL does not prove that supplementation prevents excessive immune activation, the broader topic discussed in Vitamin D and Immune Overactivation. It also does not demonstrate that a temporary hs-CRP difference caused fewer metabolic events or slower immune ageing. Those extensions require outcomes and analyses beyond the biomarker comparison reviewed here.
DO-HEALTH: biological-age estimates are a different endpoint
The DO-HEALTH analysis examined 777 Swiss participants from a larger trial of 2,157 older adults. The parent Swiss cohort included 1,006 participants. The analysed group had a mean age around 75 years, approximately 60% were women and most were physically active. Mean baseline vitamin D was about 23.6 ng/mL, with roughly one third below 20 ng/mL.
The trial tested vitamin D, omega-3 supplementation and a home exercise programme in a factorial design. Vitamin D was assigned at 2,000 IU daily for three years. The omega-3 and exercise assignments were separate interventions, which matters when interpreting any combination effect. The regimens describe research conditions and do not constitute a combined anti-ageing prescription.
The investigators used DNA-methylation-based measures to estimate aspects of biological ageing. Such measures draw on patterns in methylation data; they are not direct observations of lifespan, infection resistance or immune-cell competence. Different clocks can also capture different aspects of ageing-related variation.
Vitamin D alone did not produce a clear change in the four clock measures considered in this analysis. Omega-3 showed effects on several measures, and combined interventions showed an additive pattern for PhenoAge. The presence of vitamin D in a combination does not establish that vitamin D independently caused the combined result.
Why an ageing clock is not immune rejuvenation
The distinction is especially important for Vitamin D and Immune Ageing. A DNA-methylation estimate can be relevant to ageing research without directly measuring how effectively immune cells respond to infection, vaccination or tissue damage. The DO-HEALTH analysis did not demonstrate that vitamin D restored youthful immune function.
Some protein-related outcomes in the analysis were inferred from DNA-methylation patterns. They were not equivalent to measuring the corresponding proteins directly in circulation. Describing such an estimate as a measured blood-protein reduction would change the nature of the evidence.
The analysis was also conducted in a post hoc subset and was not a separately powered trial for every epigenetic endpoint. Selection into the available sample, multiplicity and the absence of technical replicates constrain interpretation. These limitations do not erase the reported findings, but they argue against treating them as a definitive clinical anti-ageing result.
The broader context in Vitamin D, Ageing, and Real-World Context helps distinguish research measures from everyday ageing outcomes. Here, the specific contribution is a qualification: the reviewed vitamin D intervention did not show an independent benefit on the selected clocks, and the combination findings cannot be assigned to vitamin D alone.
TILDA: baseline status and later glycaemic change
The TILDA analysis began with a cross-sectional sample of 5,272 participants and a four-year follow-up sample of 3,828. In the complete-covariate model restricted to those initially normoglycaemic, 3,373 participants contributed to the prospective analysis. There were 435 cases of prediabetes and 59 cases of diabetes in that model.
Unlike the two trials, TILDA did not randomly assign vitamin D supplementation. It measured baseline 25-hydroxyvitamin D and examined later glycaemic categories. The design can establish temporal ordering between the baseline measurement and the follow-up classification, but it cannot by itself establish that the vitamin D concentration caused the later outcome.
Relative to participants with baseline concentrations of at least 75 nmol/L, those below 30 nmol/L had a higher reported relative risk ratio for incident prediabetes: approximately 1.62, with a 95% confidence interval of 1.12 to 2.35. The 30 to 49.9 nmol/L category also showed an association, with a relative risk ratio around 1.56. The 50 to 74.9 nmol/L category did not show the same clear association. These estimates were adjusted for age, sex, education, BMI, smoking, physical activity, statin use and the season of vitamin D sampling.
The reference category matters. Replacing it with a different threshold would alter the claim. These results describe category comparisons in the reported model; they do not establish 75 nmol/L as a treatment target. A higher-status reference group in an observational analysis is not a randomised target-selection experiment.
Prediabetes and diabetes did not give the same result
The corresponding incident-diabetes association was not statistically clear. For the lowest vitamin D category, the reported relative risk ratio was approximately 1.36, with a wide confidence interval of about 0.55 to 3.36. The smaller number of diabetes cases contributes to uncertainty. The result should not be converted into either a confirmed increase or proof of no possible association.
Glycaemic classification used HbA1c together with relevant history and medication information. It did not directly measure insulin resistance using an oral glucose-tolerance test or a HOMA-based endpoint. The study is related to Vitamin D and Insulin Signalling, but its measurements should not be relabelled as direct demonstrations of altered insulin signalling.
A single baseline vitamin D measurement also cannot establish sustained exposure over the entire follow-up. Participants’ health, activity, diet, adiposity and other characteristics may change. Statistical adjustment reduces some measured confounding but cannot guarantee removal of all relevant differences between vitamin D categories.
Follow-up participants were also a selected subset, with a healthier profile than those not retained. That can affect how broadly the findings apply. The prospective association is therefore informative but remains observational, with uncertainty about causal direction and the mechanisms that might connect the measurements.
Does inflammation explain the metabolic association?
The reviewed TILDA analysis did not demonstrate that inflammation mediated the association with prediabetes. It did not show a sequence in which low vitamin D caused a measured inflammatory increase that then caused glycaemic deterioration. Joining its result to VITAL’s hs-CRP finding would create a causal chain that neither study tested.
The same caution applies to Vitamin D and Oxidative Stress. Oxidative stress may be relevant to metabolic and ageing research, but these studies do not establish it as the intermediary connecting their reported outcomes. A mechanistic explanation should remain a hypothesis unless the relevant process and its causal role were actually examined.
Similarly, biological-age estimates, inflammatory markers and glycaemic categories cannot be counted as repeated measurements of one common “ageing burden”. They answer neighbouring questions at different levels. Their combined value lies partly in revealing where the evidence does not connect, rather than in producing a single summary score or treatment narrative.
Conclusion
The long-term evidence is mixed and endpoint-specific. VITAL found a two-year hs-CRP difference that was not sustained at four years, without a consistent beneficial cytokine pattern. DO-HEALTH did not demonstrate an independent vitamin D effect on the selected epigenetic clocks. TILDA associated lower baseline vitamin D categories with later prediabetes, while the incident-diabetes comparison remained uncertain.
These findings do not establish that vitamin D supplementation reduces chronic inflammatory burden sufficiently to slow immune ageing or prevent metabolic disease. They support careful separation of randomised marker effects, surrogate ageing measures and observational metabolic associations, with the missing causal and functional links stated explicitly.
This page is an original Vitamin D3 For Life explanation and critical appraisal of the research cited below.
Primary source and publication details
Related reading
A Machine Learning Approach to Understanding the Genetic Role in COVID-19 Prognosis: The Influence of Gene Polymorphisms Related to Inflammation, Vitamin D, and ACE2
This study aimed to evaluate the incidence of polymorphisms in genes linked to the RAAS system, cytokine production, and vitamin D on COVID-19 severity, with the goal of gaining a deeper understanding of the genetic etiology related to COVID-19.
Original paper ↗Read more →Biological determinants of blood-based biomarker levels in Alzheimer’s disease: role of nutrition, inflammation, and metabolic factors
While the individual contributions of neuroinflammation, brain insulin resistance, and micronutrient deficiencies to AD pathology are well-established, a significant knowledge gap exists in understanding their intricate, synergistic interactions.
Original paper ↗Read more →Development and validation of a predictive model for sarcopenia in patients with type 2 diabetes based on the synergistic mechanism of insulin resistance and inflammation
The aim of this study was to develop and internally validate a clinically feasible, mechanism-based machine learning model for predicting sarcopenia risk in patients with T2DM.
Original paper ↗Read more →Effects of Probiotics and Vitamin D3 Supplementation on Sports Performance Markers in Male Mixed Martial Arts Athletes: A Randomized Trial
Thus, this study aimed to assess and compare the level of selected markers of sports performance of mixed martial arts (MMA) athletes supplemented with vitamin D3 or probiotics combined with vitamin D3.
Original paper ↗Read more →Epidemiological characteristics, SAP-associated sepsis, and prognostic analysis of persistent inflammation: immunosuppression-catabolism syndrome in severe acute pancreatitis patients after percutaneous drainage: a retrospective study
Multivariate logistic regression with sepsis as a covariate identified independent PICS risk factors; Kaplan: Meier and Cox regression evaluated prognostic impact.
Original paper ↗Read more →Integrative analysis of vitamin D, ferritin, and eosinophilic inflammation in predicting acute exacerbations of childhood asthma
This study investigated the associations of serum vitamin D, ferritin, and eosinophilic inflammation with acute asthma exacerbations in children and explored their potential interaction.
Original paper ↗Read more →Osteoporosis and sarcopenia are common and insufficiently diagnosed among chronic pancreatitis patients
We aim to study the prevalence of osteoporosis, sarcopenia and vitamin deficiency among CP patients.
Original paper ↗Read more →Procalcitonin metabolomics in the critically ill reveal relationships between inflammation intensity and energy utilization pathways
Our aim was to determine if patients with elevated procalcitonin have differences in the metabolomic response to early critical illness.
Original paper ↗Read more →Serum Vitamin D Levels Were Not Associated With the Risk of Aneurysmal Subarachnoid Hemorrhage: A Large Cohort Study With Propensity Score Matching and Mendelian Randomization Analysis
Furthermore, the causal relationship between VitD and aSAH was investigated using a two‐sample Mendelian randomization (MR) analysis.
Original paper ↗Read more →Effects of Vitamin D3 and Marine Omega-3 Fatty Acids Supplementation on Biomarkers of Systemic Inflammation: 4-Year Findings from the VITAL Randomized Trial
Background: The VITAL study was a nationwide, randomized, double-blind, placebo-controlled, 2 × 2 factorial trial of vitamin D3 (2000 IU/day) and marine n-3 FAs (1 g/day) supplements.
Original paper ↗Read more →Vitamin D, Long-Term Inflammation, Biological Ageing and Metabolic Risk: Study Summary
A concise summary of vitamin D findings on long-term inflammation, ageing markers and metabolic risk, separating trials from observational associations.
Read more →Persistent inflammation during anti-tuberculosis treatment with diabetes comorbidity
The analysis compared 17 markers of inflammation in tuberculosis patients with and without diabetes at diagnosis, and at two time periods during the 6-month course of treatment.
Original paper ↗Read more →The Role of Vitamin D and Its Molecular Bases in Insulin Resistance, Diabetes, Metabolic Syndrome, and Cardiovascular Disease: State of the Art
In the last decade, an increasing awareness was directed to the role of Vitamin D in non-skeletal and preventive roles for chronic diseases.
Original paper ↗Read more →Vitamin D and Chronic Inflammation
Explains how vitamin D relates to chronic inflammation, including cytokine balance, immune regulation, metabolic links, and long-term inflammatory signalling, without making treatment claims.
Read more →What Immune-Ageing Measurements Can and Cannot Tell Us
A guide to immune-cell counts, senescence markers, inflammation tests and recall responses, and the different questions each measurement answers.
Read more →Vitamin D and Immune Ageing
Immune ageing involves gradual changes in immune function across the lifespan.
Read more →Contextual Factors That Distort Vitamin D Blood Test Interpretation: Full Study
Review how illness, season, binding proteins, obesity, timing and biological variation can distort interpretation of vitamin D blood test results.
Read more →Vitamin D and Inflammatory Signalling
Learn how vitamin D relates to inflammatory signalling, including initiation, regulation, resolution, cytokine communication, metabolic links, and tissue specific responses.
Read more →Vitamin D and Epigenetic Regulation
How vitamin D influences epigenetic regulation through DNA methylation, histone modification, non-coding RNAs, and changes in gene accessibility across life stages.
Read more →Vitamin D, Ageing, and Real-World Context
How age, daily life, health status, and long-term context shape real-world vitamin D experience beyond research settings.
Read more →Characterization of vitamin D supplementation and clinical outcomes in a large cohort of early Parkinson’s disease
We analyzed data from the National Institutes of Health Exploratory Trials in Parkinson’s Disease (NET-PD) Long-term study (LS-1), a longitudinal study of 1741 participants.
Original paper ↗Read more →Original sources
Effects of Vitamin D3 and Marine Omega-3 Fatty Acids Supplementation on Biomarkers of Systemic Inflammation: 4-Year Findings from the VITAL Randomized Trial
Background: The VITAL study was a nationwide, randomized, double-blind, placebo-controlled, 2 × 2 factorial trial of vitamin D3 (2000 IU/day) and marine n-3 FAs (1 g/day) supplements.
Publication details
Effects of Vitamin D3 and Marine Omega-3 Fatty Acids Supplementation on Biomarkers of Systemic Inflammation: 4-Year Findings from the VITAL Randomized Trial. Yanbin Dong, Haidong Zhu, Li Chen, Ying Huang, William Christen, Nancy R Cook, Trisha Copeland, Samia Mora, Julie E Buring, I-Min Lee, Karen H Costenbader, JoAnn E Manson. Nutrients; 2022; 14; 24; 5307. DOI: 10.3390/nu14245307. CC BY. This page provides an original explanatory appraisal; no source figures are reproduced.
Individual and additive effects of vitamin D, omega-3 and exercise on DNA methylation clocks of biological aging in older adults from the DO-HEALTH trial
Individual and additive effects of vitamin D, omega-3 and exercise on DNA methylation clocks of biological aging in older adults from the DO-HEALTH trial.
Publication details
Individual and additive effects of vitamin D, omega-3 and exercise on DNA methylation clocks of biological aging in older adults from the DO-HEALTH trial. Heike A. Bischoff-Ferrari, Stephanie Gängler, Maud Wieczorek, Daniel W. Belsky, Joanne Ryan, Reto W. Kressig, Hannes B. Stähelin, Robert Theiler, Bess Dawson-Hughes, René Rizzoli, Bruno Vellas, Laure Rouch, Sophie Guyonnet, Andreas Egli, E. John Orav, Walter Willett, Steve Horvath. Nature Aging; 2025; 5; 3; 376 to 385. DOI: 10.1038/s43587-024-00793-y. CC BY. This page provides an original explanatory appraisal; no source figures are reproduced.
Association between vitamin D deficiency and the risk of prevalent type 2 diabetes and incident prediabetes: A prospective cohort study using data from The Irish Longitudinal Study on Ageing (TILDA)
Association between vitamin D deficiency and the risk of prevalent type 2 diabetes and incident prediabetes: A prospective cohort study using data from The Irish Longitudinal Study on Ageing (TILDA).
Publication details
Association between vitamin D deficiency and the risk of prevalent type 2 diabetes and incident prediabetes: A prospective cohort study using data from The Irish Longitudinal Study on Ageing (TILDA). Kevin McCarthy, Eamon Laird, Aisling M O’Halloran, Cathal Walsh, Martin Healy, Annette L Fitzpatrick, James B Walsh, Belinda Hernández, Padraic Fallon, Anne M Molloy, Rose Anne Kenny. eClinicalMedicine; 2022; 53; 101654. DOI: 10.1016/j.eclinm.2022.101654. CC BY. This page provides an original explanatory appraisal; no source figures are reproduced.