Vitamin D is one of the most purchased supplements in the world. It is also one of the most misused.

Roughly 25% of American adults fall below the National Academy of Medicine's clinical threshold for deficiency, defined as 20 ng/mL (50 nmol/L) on a serum test. When you use the Endocrine Society's higher threshold of 30 ng/mL (75 nmol/L), that number climbs substantially. Globally, a 2023 pooled analysis of 7.9 million participants confirmed deficiency is widespread across most regions. The groups at highest risk are people over 65, those with higher melanin levels (sun-driven synthesis is significantly reduced), people carrying excess body weight (vitamin D is fat-soluble and sequesters in adipose tissue), anyone who avoids sun exposure, anyone using high-SPF sunscreen as their primary or only UV exposure management, and people living above approximately 37 degrees north latitude during winter months, when UV-B angle is too low to drive meaningful skin synthesis regardless of time spent outdoors.

Here is the structural problem. The US Preventive Services Task Force concluded in 2021 that evidence is insufficient to recommend routine screening in asymptomatic adults. That gives most primary care physicians no clear mandate to order the test. The result: a large, measurable population is running chronically deficient, experiencing real downstream consequences across multiple body systems, and either not supplementing at all or supplementing blind at doses that may be too low, just right, or unnecessarily high.

One more piece of honest setup before we run it through the filter. The large randomized controlled trials on vitamin D, VITAL, ViDA, D-HEALTH, produced mostly null results on hard outcomes like cardiovascular events and all-cause mortality. Those results are real. They also have a specific design limitation that reframes the entire conversation, and we will address it directly.

Let's run it through the filter.

Yes, specifically for confirmed deficiency. The consequences of staying deficient are documented across bone, muscle, immune, and metabolic systems. The intervention required to find out where you stand costs almost nothing.

This is where the causal evidence is clearest and most replicated across multiple study designs.

Vitamin D is required for calcium absorption from the gut. Without adequate levels, the body compensates by pulling calcium from bone, progressively reducing bone mineral density and increasing fracture risk. In older adults, deficiency is also tied to sarcopenia, the progressive loss of muscle mass and strength that accelerates after 60. Vitamin D receptors are expressed directly in muscle tissue; the pathway from low levels to reduced muscle protein synthesis is well-established mechanistically. A 2025 study in adults over 50 found vitamin D deficiency associated with meaningfully higher odds of low muscle mass and sarcopenia, with muscle function, not just bone density, identified as the primary mediating mechanism for hip fracture risk.

This matters for the over-50 population specifically because sarcopenia and falls interact: weaker muscles lead to falls, falls lead to fractures, and hip fracture in older adults carries significant morbidity and mortality risk, sometimes called "the last fracture of life." Meta-analyses of RCTs confirm that supplementation in deficient individuals reduces fall risk, with the protective effect strongest in those with confirmed low baseline levels.

This connects directly to the resistance training and creatine work covered in Edition #006 and Edition #003, the muscle preservation picture has multiple legs, and vitamin D adequacy is one of them.

Vitamin D receptors are expressed on virtually every immune cell type. The mechanism is established: vitamin D induces production of antimicrobial peptides including cathelicidin and beta-defensin, and modulates both innate and adaptive immune responses, including dampening excessive inflammatory signaling.

The landmark 2017 BMJ meta-analysis by Martineau et al., covering 25 RCTs and over 11,000 participants, found that vitamin D supplementation reduced acute respiratory tract infections overall, with the protective effect dramatically stronger in participants with confirmed deficiency at baseline. An updated 2021 Lancet meta-analysis showed more heterogeneity across trials and more modest overall effects, with daily supplementation consistently outperforming high-dose bolus strategies. The pattern that emerges: the immune signal is real, but concentrated in people who are genuinely deficient.

Consistent and biologically plausible associations exist between low vitamin D and insulin resistance and type 2 diabetes. Vitamin D receptors are expressed on pancreatic beta cells, and the proposed mechanism involves impaired insulin secretion and reduced insulin sensitivity at low serum levels. Correcting deficiency in type 2 diabetics with confirmed low levels has shown improvements in glycemic markers in some trials.

Magnesium is also relevant here and worth flagging directly: it is a required cofactor for vitamin D hydroxylation and activation in both the liver and kidneys. A deficiency in both, which is common, compounds the problem, since unactivated vitamin D cannot exert its biological effects regardless of how much you supplement. Edition #005 covers magnesium in full.

Vitamin D receptors are present throughout the brain. Observational studies consistently associate deficiency with higher rates of depression and cognitive decline in older adults, and the receptor biology is mechanistically sound.

Here is the honest limitation, and it matters. Mendelian randomization studies, which use genetic variants to test causality and minimize confounding, have largely not confirmed that low vitamin D causes depression or cognitive decline. One large analysis using UK Biobank data (over 307,000 participants) found evidence that the causal arrow may actually run the other direction: depression leads to lower sun exposure and physical activity, which lowers vitamin D levels, not the other way around.

A 2024 meta-analysis of 31 RCTs found that vitamin D3 supplementation reduced depressive symptoms, with stronger effects in people who were symptomatic at baseline. But that does not resolve the causality question definitively. The bottom line: the association is consistent, the mechanism is plausible, the Mendelian randomization evidence does not confirm causation, and correcting deficiency carries no downside. It belongs in the picture with appropriate uncertainty, not as a settled claim.

Critics of vitamin D supplementation point to VITAL, ViDA, and D-HEALTH as proof it doesn't work. They are partially correct, for supplementation in people who are already sufficient.

VITAL's participants entered the trial with a mean baseline 25(OH)D level of approximately 73 nmol/L (about 29 ng/mL), already above the NAM deficiency threshold. Supplementing vitamin D-replete adults with 2,000 IU/day predictably produced little additional benefit, because the downstream proteins are already largely saturated. The null results in sufficient populations do not tell us what correction of genuine deficiency does. They tell us not to expect vitamin D supplementation to function as a pharmaceutical add-on above adequate levels. That is a reasonable and important finding. It is also precisely why this edition's central recommendation is: test first.

The one consistent signal across meta-analyses of RCTs: an approximately 13% reduction in cancer mortality over longer follow-up periods. Not cancer incidence, mortality specifically. Suggestive, not definitive, but consistent enough to mention.

High-dose supplementation without knowing your baseline is not a neutral act. Vitamin D is fat-soluble and accumulates. The National Academy of Medicine's established tolerable upper intake level is 4,000 IU/day. A 3-year RCT found hypercalcemia, elevated blood calcium and the primary toxicity marker, in 9% of participants taking 10,000 IU/day. The risks at that level are real, and include kidney damage and soft tissue calcification. The protection is straightforward: know your level first, dose to correct it, retest in 3 months to confirm. This is not complicated.

One more calibration: D3 (cholecalciferol) raises and sustains serum 25(OH)D levels significantly more effectively than D2 (ergocalciferol). If your doctor prescribes vitamin D2, it is worth asking about D3.

A serum 25(OH)D test is inexpensive and widely available. It typically runs $30 to $60 USD out of pocket if not covered by insurance, and is often included in comprehensive panels at no additional cost. This is not a meaningful barrier to finding out where you stand.

Vitamin D3 supplementation at corrective doses is among the least expensive interventions in this entire evidence base. Budget options run a few dollars per month; mid-range and third-party certified brands cost more but remain well under $30 USD for several months of supply. For those supplementing D3, combination D3/K2 products are widely available at comparable prices, see the sidebar on K2 below.

When you supplement vitamin D3, it increases calcium absorption from the gut. That calcium needs to go somewhere useful.

Vitamin K2, specifically the MK-7 form, activates two vitamin K-dependent proteins: osteocalcin, which drives calcium into bone matrix, and matrix Gla protein (MGP), which inhibits calcium deposition in arterial walls. The concern when supplementing D3 without adequate K2 is that absorbed calcium may end up deposited in soft tissue rather than routed to bone.

A 2023 randomized controlled trial published in JACC Advances (AVADEC study, 389 older men, 4 Danish hospitals) found that combined K2 MK-7 plus D3 supplementation may slow coronary artery calcium progression in subgroup analyses (the primary outcome across the full trial population was null, with the signal concentrated in participants with higher baseline calcium scores and those on statins). A separate 3-year RCT in postmenopausal women found MK-7 at 180 mcg/day attenuated age-related bone density loss versus placebo.

Honest framing: the K2 evidence base is smaller than D3's and not yet definitive for long-term cardiovascular outcomes. The mechanistic logic is sound, the safety profile is excellent, and the cost is minimal. MK-7 is the form with the best bioavailability and evidence for once-daily dosing. If supplementing D3 to correct deficiency, adding K2 MK-7 is a low-cost, low-risk step.

One firm exception: anyone taking warfarin or other vitamin K-affecting anticoagulants must speak to their physician before adding K2. Vitamin K affects coagulation pathways and can alter the drug's effectiveness.

This recommendation applies to otherwise healthy adults pursuing deficiency correction. People with kidney disease, hyperparathyroidism, granulomatous conditions (sarcoidosis, some lymphomas), or hypercalcemia have altered vitamin D metabolism and should not self-direct supplementation at any dose without physician guidance. Pregnancy: consult your physician for appropriate dosing. If you take medications that affect calcium balance, flag it with your prescriber before adding vitamin D.

The vitamin D story has an unusual structure. The observational data showing associations with lower disease risk across a dozen conditions is enormous and spans decades. The large RCTs have been mostly null, but those trials largely enrolled people who were not deficient. The Mendelian randomization evidence is mixed: strong support for bone and muscle effects, less consistent for metabolic and cognitive outcomes, actively contested for mood.

What holds up across all of it: deficiency produces measurable, multi-system consequences that have no equivalent benefit to supplementing on top of adequate levels. The minimum effective action here is not "take vitamin D." It is "find out where you actually stand."

Request a serum 25-hydroxyvitamin D [25(OH)D] test at your next physical. You can ask for it explicitly. It is the correct test, not the active 1,25-dihydroxyvitamin D form. Target range: 30 to 50 ng/mL (75 to 125 nmol/L). This is the sufficiency range used by the Endocrine Society and consistent with most functional medicine targets. The NAM's 20 ng/mL floor is a bone-health minimum, not an optimal target.

Subscribe free — one edition in your inbox every Wednesday. No spam. No conflicts of interest. Any future sponsorships will be clearly disclosed and personally vetted.

After subscribing, check your junk/spam folder for the confirmation email and mark it as ‘Not Spam’ to ensure delivery.