Prenatal vitamin D trial: one significant result out of eleven, after correction

A Positive Headline from a Mostly Null Study

The JAMA Network Open study is a secondary analysis of COPSAC2010, a Danish randomized, blinded, placebo-controlled trial. The original trial was designed and powered to test whether high-dose prenatal vitamin D3 reduces childhood asthma and wheeze — it did not. Cognitive outcomes at age 10 were assessed as a post hoc secondary endpoint in 498 children, 247 in the high-dose arm (2,800 IU/day plus the standard 400 IU, from week 24 of pregnancy through one week postpartum) and 251 on standard dose alone.

What the Numbers Actually Say

Researchers tested eleven cognitive domains: estimated intelligence, processing speed, reaction time, sustained attention, motor speed, verbal memory, visual memory, verbal working memory, spatial working memory, planning, and flexibility/set-shift.

Of eleven measures:

• Visual memory: β = 0.24 SD (95% CI, 0.06–0.42), p = .01 — survived correction
• Verbal memory: β = 0.17 SD (95% CI, 0.03–0.32), p = .02 — survived correction
• Flexibility/set-shift: β = 0.19 SD, p = .04 — did not survive correction
• Estimated intelligence: no effect; p = .90
• All other eight measures: non-significant

The correction method was Benjamini-Hochberg false discovery rate, applied within cognitive domains rather than across all eleven functions. The two surviving hits — verbal and visual memory — happened to fall in the same two-item "memory" domain, so BH correction within that domain barely adjusted anything.

The Thread's Surgical Critique

In the thread, a high-karma contributor with extensive HN history put the methodological issue most plainly:

"They tested 11 different measures. Only 3 of those tests showed significance, but after they corrected for multiple testing the significance of 2 of those disappeared. Only 1 test remained below the significance threshold… This study isn't very convincing. It's a classic p-hacking trick to include many different smaller tests so if one of them pops up as an outlier you can claim that something was significant." — HN commenter with 52k karma

A commenter who went deeper into the statistics identified exactly why the correction was toothless:

"The two 'winning' functions, verbal memory (p = .02) and visual memory (p = .01), both sit inside the memory domain — which contains exactly those two functions. BH within a 2-member family barely adjusts anything: the most lenient threshold is just 0.05, and both p-values are already under it, so the q-values come back at .02 and .02. The correction was toothless by construction… A reviewer should have made them show the whole-family result side by side." — HN commenter on statistics

The co-founder of Apache Cassandra and DataStax made it even shorter:

"this is right there in the abstract, isn't that the entire game?" — jbellis, founder of Brokk AI

What This Study Can and Cannot Support

The Danish cohort is vitamin-D-sufficient by northern-European standards. Generalizing to populations with genuine deficiency — women of color, who require higher doses to achieve equivalent serum levels, or populations in low-sunlight environments with different baseline levels — would require separate trials. The study's authors acknowledge the post hoc status, report modest effect sizes, and apply some correction. That makes the rhetorical sleight of hand harder to spot: the careful-looking scaffolding surrounds a null primary outcome and a survival-by-construction secondary positive. The defensible conclusion from this data is that prenatal high-dose D3 in a D-sufficient cohort showed no meaningful whole-battery cognitive benefit. Further targeted replication on memory outcomes specifically is what the data warrants — not a supplementation recommendation.