It’s become almost reflexive: a healthy adult comes in for a physical, and a 25-hydroxyvitamin D level goes on the panel alongside the CBC and metabolic panel “might as well check it.” It’s one of the highest-volume lab tests in medicine. And in healthy adults, it’s a test that generates a problem, triggers a treatment cascade, and improves no outcome we can point to.
The practice
Ordering serum 25(OH)D levels routinely in healthy, asymptomatic adults as part of general check-ups or “wellness” panels including reflexively testing groups assumed to be at risk, such as adults with obesity or dark skin outside of a specific clinical indication.
Why we think we should
“Low vitamin D is associated with everything.” Decades of observational data tie low 25(OH)D to cardiovascular disease, cancer, diabetes, infections, mortality so checking it feels like prudent prevention.
It seems harmless and cheap, and correcting a “deficiency” feels like an easy win.
Patients ask for it, and it’s easier to order than to explain why not.
The intuition is that finding and fixing low vitamin D must help. Randomized trials say it mostly doesn’t and the testing is where the low-value cascade begins.
Why there’s no good reason
Two things dismantle the routine-screening habit.
First, supplementation doesn’t deliver the promised outcomes. The large VITAL trial (25,871 adults, 2,000 IU/day, 5.3 years) found no significant benefit on cancer, cardiovascular events, or fractures. When the interventions that screening leads to don’t improve outcomes, screening to find candidates for them can’t either.
Second and this is the deprescribing insight testing itself drives harm through the cascade it forces. As the 2024 Endocrine Society guideline’s author put it, when clinicians measure vitamin D, they’re forced to decide what to do about it. A level comes back “low” against a threshold not tied to any outcome; now you supplement, re-test to confirm “correction,” re-dose, re-test again a chain of visits, labs, and patient anxiety about a “deficiency,” with no demonstrated downstream benefit. You manufactured a problem (a number below a line) and a treatment for it, and helped no one.
There’s a deeper reason the observational promise misled us: low 25(OH)D is largely a marker of poor health, not a cause of it the ill, obese, inactive, and housebound have low levels because of those states. Supplementing the number treats the readout, not the cause, which is exactly why the randomized trials came back null.
Accordingly, the 2024 Endocrine Society guideline recommends against routine 25(OH)D testing in healthy adults, explicitly including those with obesity or dark complexion, because outcome-specific benefit thresholds have never been established.
What to do instead
Test for a reason, not by reflex. Reserve 25(OH)D for genuine indications: malabsorption, established metabolic bone disease (osteoporosis, osteomalacia), hyperparathyroidism workups, certain medications, or specific clinical suspicion of deficiency.
For the empiric-supplementation groups the guideline does endorse (pregnancy, adults >75, high-risk prediabetes, children/adolescents), supplement without testing the recommendation is empiric precisely to avoid the test-and-cascade loop.
Skip it on wellness panels for healthy adults, and be ready with the one-line explanation: correcting the number in healthy people hasn’t been shown to improve outcomes, so we don’t routinely measure it.
The takeaway
Routine vitamin D screening in healthy adults is the textbook case of a test that shouldn’t be ordered: the treatment it prompts doesn’t improve outcomes, and the measurement itself launches a low-value cascade of supplementation and re-testing. In 2024 a major society said so plainly. The generalizable lesson is worth carrying everywhere: before you measure a number, ask whether an abnormal result would trigger an action that actually helps the patient. If “low” only leads to an unproven treatment, the evidence-based move is not to order the test.