Plate Nº 39 · recorded November 17, 2026
Health & Medicine ResearchReported finding
Vitamin K Refusals Rise, and Newborn Bleeding Cases Rise With Them
As vitamin K refusals rose from 2.3% to 5% of U.S. newborns, deficiency-related bleeding climbed 347% — a warning shot in the debate over routine newborn care.
By James Calloway4 min read779 words
In brief
- VKDB cases rose 347% (24.0 to 107.4 per 100,000 infants) from 2017 to 2025.
- Injectable vitamin K administration fell from 97.7% to 95.0% of newborns over the same period.
- Late VKDB carries a 30–60% risk of intracranial hemorrhage.
- Only 19.2% of infants diagnosed with VKDB had received the vitamin K injection.
- The studies used EPIC's Cosmos database spanning 2,300 hospitals and 50,000 outpatient settings.

Bleeding linked to vitamin K deficiency in infants jumped 347% between 2017 and 2025, while the share of newborns receiving the preventive injection fell. That is the headline finding from two studies, published in JAMA Pediatrics and JAMA, built on one of the largest U.S. health record databases ever assembled for this question.
The numbers move in opposite directions, and that is the point. Injectable vitamin K administration dropped from 97.7% of newborns in 2017 to 95.0% in 2025. Over the same period, vitamin K deficiency (VKD) rose 216%, from 11.8 to 37.3 cases per 100,000 infants. Vitamin K–deficient bleeding (VKDB) climbed 347%, from 24.0 to 107.4 cases per 100,000. In absolute terms, 2,151 infants were diagnosed with deficiency and 5,217 with deficiency-related bleeding.
The largest single-year surge in diagnoses came in 2025 — the same year vitamin K administration hit its lowest point. Correlation is not causation, a caveat the authors themselves flag. But they argue other explanations don't hold up, noting that behaviors and diseases raising VKD risk "remained constant or decreased during this period."
Steven Woolf, MD, director emeritus of the Virginia Commonwealth University Center for Society and Health, put the motivation plainly: "We knew from prior studies that vitamin K refusal had been increasing in the U.S., and we wanted to learn if there were any real-life consequences of this trend."
Why are newborns vulnerable in the first place?
Vitamin K powers the clotting factors that plug leaks in the circulatory system. Newborns arrive with very little of it. The vitamin crosses the placenta only sparingly, and a baby's gut and liver haven't yet ramped up production of the clotting factors that depend on it.
Breast milk, normally the gold standard for infant nutrition, is also low in vitamin K. That matters. According to the American Academy of Pediatrics, the most common real-world cause of VKD is no prophylaxis at birth combined with exclusive breastfeeding. Certain maternal medications — anticonvulsants, some antibiotics, and the older blood thinner warfarin — also raise risk.
VKDB comes in three forms:
- Early, within 24 hours of birth, driven mainly by maternal medications.
- "Classical", in the first week, often showing up as bruising or umbilical bleeding.
- Late, between one week and 6 months, mostly in breastfed babies or infants with intestinal problems that block absorption.
Late VKDB is the frightening one. It carries a 30% to 60% risk of intracranial hemorrhage — bleeding inside the skull — which can cause permanent brain damage, seizures, or death. In the new data, 6.9% of infants with VKD and 2.4% of those with VKDB had documented intracranial or intracerebral hemorrhage.
Notably, only 39.7% of infants diagnosed with VKD and 19.2% of those with VKDB had received the injection. Most bleeding happened to unprotected babies.
Are the shots actually safe?
This is where the story gets contested. Some older formulations contained phenol, which causes tumors in animals — but a systematic review found no increased cancer risk from the intramuscular injection. The FDA label carries a black box warning for allergic reactions and anaphylaxis, though those events stem primarily from intravenous use, and no evidence shows similar reactions in newborns getting the shot.
A 1950s-era version of the vitamin was linked to red blood cell damage and, indirectly, to kernicterus, a rare form of brain damage. Today's version, phytonadione, has not shown that effect. Sweden's experience adds another caution: switching from injections to oral prophylaxis in 1986 triggered a resurgence of late VKDB cases, partly because parents don't complete multi-dose oral courses and absorption varies.
What does this change for parents?
The practical takeaway is modest but real. One injection, given once, covers the vulnerable first months. It's cheap, well-tested, and can't be targeted to just the babies who need it — doctors can't reliably identify which infants would bleed without it.
Still, this remains a preliminary read of observational data, drawn from the Cosmos EPIC database covering 2,300 hospitals and 50,000 outpatient settings, and covering infants born at 35 weeks' gestation or later, aged 0 to 6 months. Database studies can miss context that chart reviewers would catch.
The harder question is ethical, not statistical. Unlike vaccination, there's no herd immunity argument here — the shot primarily protects the newborn, not the community. Parents exercise the autonomy; the baby bears the consequences. Physicians, the analysis argues, should recommend the shot without equivocation, keep explaining the evidence when parents refuse, and never pretend refusal is an equally safe choice. Where coercion becomes justified, though, remains unresolved.
via acsh.org (Source)
More from James Calloway
Show full bio
Staff writer covering regulation, policy and their impact on general science news at SciBeat.
419 articles
Nearby plates
- Vitamin D2 Supplements May Lower the Body's D3 Levels, Study Finds
- Daily Vitamin C Boosts Survival in Pre-Leukemia Trial
- Vitamin C Shows Survival Signal in Blood Disorder Trial
- Major Review Finds Calcium and Vitamin D Do Little to Prevent Fractures
- Nearly 17% of Children on GLP-1 Drugs Develop Nutritional Deficiencies