The Number That Stands Out
In an Israeli health system, adults with severe vitamin D deficiency had an adjusted all-cause mortality hazard ratio of 2.57 (95% CI 2.38-2.79) compared with adults who were vitamin D sufficient. A large US network produced the same direction of result, with a smaller effect: HR 1.59 (1.55-1.63).
The study, published in Aging Cell by Israel and colleagues, also asked a second question. When people bought vitamin D, did their outcomes look different? The answer was intriguing, but it comes with important caveats.
Source: Israel et al., "Vitamin D Deficiency, Supplementation, and Aging-Related Morbidity and Mortality Across Two Longitudinal Population Cohorts," Aging Cell (DOI: 10.1111/acel.70752; PubMed 42850752).
First, the Fine Print: This Is Observational EHR Data
This was not a randomized trial. Nobody was assigned to take vitamin D or a placebo. The researchers analyzed longitudinal electronic health records from two systems:
- Leumit Health Services (LHS): a nationwide Israeli health organization, with more than 1.67 million serum 25-hydroxyvitamin D [25(OH)D] measurements from over 468,500 adults (2009-2020).
- TriNetX: a federated US research network, used for external validation with 223,175 propensity score-matched pairs.
The matched cohorts were about 54 years old on average, 65%-67% female, and followed for up to 10 years. Observational designs can show associations and strengthen hypotheses, but they cannot fully rule out confounding. People with severe deficiency may differ from others in ways the models did not capture.
What Severe Deficiency Was Linked To
The researchers compared severe deficiency (<10 ng/mL)** with **sufficiency (>30 ng/mL). In both populations, severe deficiency was associated with higher risks across multiple aging-related outcomes. The abstract highlights all-cause mortality as the headline result.
The gap between the two cohorts (HR 2.57 vs 1.59) is worth noting. Both point the same way, but the size of the association differed between populations. The authors present TriNetX as external validation of the direction of the finding, not as an identical replication.
The authors frame severe deficiency as "an important and potentially modifiable determinant of systemic aging vulnerability and age-related multimorbidity." The word "potentially" matters there.
The Supplementation Twist
To test whether the risk could be changed, the team modeled supplementation over time using pharmacy-dispensed purchases. They calibrated these purchases to predict changes in serum 25(OH)D. This is a modeled estimate of vitamin D exposure, not a measured one for each person at each moment.
The findings:
- Supplementation was independently associated with dose-dependent reductions in mortality and multiple aging-related outcomes.
- For supplementation equivalent to more than 10 ng/mL of 25(OH)D increase, the mortality HR was 0.72 (0.68-0.76).
- There was no corresponding association for skin malignancy.
The skin cancer result works as a kind of internal check. If supplementation simply marked healthier, more health-conscious people, you might expect benefits to show up everywhere. A null result for one outcome suggests the associations were not uniformly inflated, though it does not prove causation.
Did They Try to Address Causality?
Yes, partly. The authors ran inverse-probability-of-treatment-weighted marginal structural Cox models for mortality. These methods are designed to handle exposures and confounders that change over time. The results were consistent with the conventional adjusted models.
This strengthens the analysis, but it does not turn an observational study into a trial. Possible remaining problems include:
- Reverse causation: serious illness can lower vitamin D levels.
- Healthy-user effects: people who buy and use supplements may differ in other behaviors.
- Exposure estimation: purchases are a proxy for actual intake.
Practical Takeaways, Only as Reported
- Severe deficiency: <10 ng/mL 25(OH)D
- Sufficiency (comparison group): >30 ng/mL
- The supplementation benchmark: an estimated rise of >10 ng/mL-equivalent
Dose and timing The abstract does not report a specific daily dose in IU or an optimal time of day. It reports supplementation as a modeled, calibrated effect on serum levels. Any dose recommendation would go beyond what the paper reports, so it is best settled with a clinician based on a measured 25(OH)D level.
Who this may be most relevant to The strongest signal concerns people at the severe-deficiency end (<10 ng/mL). If you have never had your level measured, this study is an argument for asking about testing, not for self-prescribing.
Who should be cautious or skip the takeaway
- Anyone looking for proof that supplements extend life. This study cannot provide it.
- People already in the sufficient range. The paper compares severe deficiency with sufficiency and does not show added benefit from going higher.
- Anyone hoping for skin cancer protection. The study found no association for skin malignancy.
- Anyone with medical conditions or medications that affect calcium or vitamin D handling. The abstract does not address safety, so that conversation belongs with a doctor.
The Bottom Line
This large, two-country analysis adds weight to the idea that severe vitamin D deficiency is a marker of, and possibly a contributor to, poorer aging outcomes. It also suggests that correcting the deficit may be associated with better survival. But the evidence is observational, the supplementation exposure is modeled, and the abstract gives no dosing protocol. Randomized trials in severely deficient populations would be the real test.
