Three giant vitamin D trials found almost nothing. The author still takes it

The trials that deflated the hype, told honestly

The post — which the author said took about three weeks to write — is a careful walk through what randomized trials actually show. The old observational excitement traces back to Apperly (1941) and the Garland brothers (1980), who noticed sunnier US latitudes had lower overall cancer mortality; early correlations hinted at ~30% reductions in all-cause death.

Then the megatrials arrived and flattened the story. The author lays out three:

None is statistically significant. His title — "the worthlessness is mildly exaggerated" — is the thesis: the hazard ratios lean just below 1.0, so the honest reading isn't "nothing," it's "a weak signal we can't confirm." He estimates that detecting his hypothesized effect with 80% power would need roughly 570,000 participants, about 5× every trial run so far. If the true HR is 0.96, a lifetime of daily pills buys ~0.48 years — he frames it as about 8.6 minutes of life per pill. His bet: supplement if your levels are low, because the downside is small and humans likely evolved at far higher baselines (traditional East African populations average ~115 nmol/L vs. a "normal" 50–125).

The doctor who now sees overdoses, not deficiency

The thread's most-upvoted voice reframed the whole debate around a clinical observation that flips the usual fear. The strongest evidence, this commenter noted, is in the severely deficient — but the influencer pivot to "everyone is deficient" created a new problem:

"she's now actually seeing more people who have excess Vitamin D than too little… patients have listened to a podcast about Vitamin D and started taking it, unaware that they're pushing their levels into the range where it can start doing more harm than good." — a commenter relaying a physician who runs vitamin D labs in annual physicals

A cardiology-trained founder added the mechanistic frame the article underplays:

"It's pretty well-established science now that vitamin D is a hormone, not a true vitamin. Vitamin D binds a nuclear receptor that regulates roughly 1,000 to 2,000 genes" — co-founder of Empirical Health, ex-UCSF Cardiology and Google

That same founder flagged a newer in-progress trial, TARGET-D, that doses dynamically to keep blood levels in a 40–80 ng/mL band and is reporting a 52% reduction in heart-attack risk — his read being that vitamin D is harmful both too low and too high, so fixed mega-doses wash out in the averages.

The confounder the RCTs can't rule out

A second camp argued the trials may be measuring the wrong thing entirely. If blood vitamin D mostly marks how much someone exercises outdoors, then the exercise, not the pill, is causal — and swallowing D3 wouldn't reproduce the benefit. One commenter sharpened it into a general rule:

"this is one of those proxies for healthy lifestyle. Kinda like how grip strength is correlated with longevity but banging out tons of hand gripper sets isn't going to do much for you health." — a commenter on vitamin D as a lifestyle proxy

Others complicated even that: people supplementing 2,000 IU/day reported still testing under 30 ng/mL until they switched to 5,000 IU with K2 in oil, and several insisted the latitude problem is real — above ~42°N the winter sun is too low to make any. The practical consensus that emerged isn't "take it" or "skip it" — it's measure first. The recurring advice across the credentialed voices was to get a blood test rather than dose blind, precisely because this particular molecule lingers for months and the dose that fixes a deficiency can, doubled, start to hurt.