In 2017, 2.92% of American newborns left the delivery room without the vitamin K shot. By 2024 that figure was 5.18%, a 77% increase across the eight years, drawn from the records of more than five million births.
Those numbers come from a study by researchers at the Children’s Hospital of Philadelphia, published in the Journal of the American Medical Association on 8 December 2025. They are the first nationwide measurement of something pediatricians had been noticing anecdotally for years: a routine injection that has been standard practice since 1961 is quietly losing ground, one delivery room at a time. And unlike the childhood vaccine debate it has been swept into, this one is not about vaccines at all.
What the shot actually does
Vitamin K is the nutrient the body uses to build blood clots. Adults get plenty of it from leafy greens, meat, dairy and eggs, and the bacteria living in the gut manufacture more. Newborns have neither advantage. Vitamin K crosses the placenta poorly, a newborn’s intestines have barely been colonized by bacteria, and breast milk carries only trace amounts.
The result is a window of vulnerability that closes only when a baby starts solid food, usually somewhere between four and six months. During that window, a baby low on vitamin K can bleed and not stop: sometimes visibly, from the umbilical stump or the nose, and sometimes invisibly, into the gut or the skull. The American Academy of Pediatrics describes brain bleeding in about half of the babies who develop the condition, known as vitamin K deficiency bleeding, or VKDB.
A single intramuscular dose in the thigh, given within six hours of birth, closes the window. The liver stores it and releases it slowly over months.
The arithmetic of that dose is unusually stark. Without the shot, the risk of late-onset VKDB (the form that strikes between one week and six months, when parents are home and no longer watching for anything) runs somewhere between 1 in 14,000 and 1 in 25,000 births. With the shot, it falls below 1 in 100,000. Put differently: babies who don’t get the injection are 81 times more likely to develop late VKDB than babies who do.
Why refusals are climbing
The CHOP team pulled de-identified records from Epic Cosmos, a database of hospital and outpatient charts, covering births between January 2017 and December 2024 at gestational ages of 35 to 43 weeks. Of 5 million-plus newborns, 3.92% did not receive the injection. The trajectory, not the average, is what alarmed the authors.
“There may be a growing perception among parents that vitamin K is unnecessary. Unfortunately, opting out of Vitamin K for a newborn is akin to gambling with a child’s health, forgoing a straightforward and safe measure that effectively prevents severe complications.”
Dr. Kristan Scott, neonatologist and lead author, in a Children’s Hospital of Philadelphia statement on the study
Part of the explanation is a paradox familiar to anyone who works in public health. The intervention worked so well that the disease it prevents effectively vanished, and a disease nobody has seen starts to look like a disease that never existed.
Dr. Ivan Hand, director of neonatology at Kings County Hospital Center in New York and co-author of the AAP’s 2022 policy statement on vitamin K, put it plainly to ProPublica, in an investigation of VKDB deaths: We’re a victim of our own success. Since we’ve been treating babies with vitamin K, we haven’t seen much deficiency bleeding, so people think it doesn’t exist.
The other part is informational. Refusals track closely with the broader post-pandemic slide in childhood immunization, the same current visible in falling school vaccination rates across Texas districts and in the tangle of motives behind why parents hesitate in the first place. Vitamin K got pulled in despite not being a vaccine, largely through social media claims the AAP has had to answer point by point: the injection contains no mercury, does not cause cancer, and is not overdosed for a newborn’s size.
Politics has not helped. At a House subcommittee hearing this spring, Rep. Kim Schrier, a Washington Democrat and a pediatrician, asked Health and Human Services Secretary Robert F. Kennedy Jr. to publicly reassure parents that the shot is safe. He declined, saying he had never said, literally never said, anything about it.
Schrier’s response: That’s exactly the point. You don’t say anything about it, but the doubt you’ve created about all of medicine and science is causing parents to make dangerous decisions.
An HHS spokesperson later told ProPublica that vitamin K at birth remains the standard of care.
Is the vitamin K shot a vaccine?
No. A vaccine trains the immune system to recognize a pathogen. The vitamin K injection supplies a nutrient the infant is temporarily short of, and nothing more. It is grouped with the hepatitis B vaccine and the antibiotic eye ointment simply because all three are given in the same hours after birth. The conflation is a category error, and it is doing real damage.
Can a baby take oral vitamin K instead?
Parents can ask, and some hospitals will accommodate. The pediatrics academy’s position is that infants absorb the oral form poorly, so it does not reliably prevent VKDB. The injection remains, in the AAP’s framing, the safest and best option.
What about the 1990s cancer study?
A single study in the early 1990s suggested a possible link between the injection and childhood cancer. It worried doctors as much as parents, and it prompted a wave of follow-up research. None of that research reproduced the finding. Three decades later the claim survives mainly in Facebook comment threads, which is a fair description of how a great deal of medical misinformation now propagates.
What does vitamin K deficiency bleeding look like?
Frequently, it looks like nothing at all until it is an emergency. When signs do appear, the AAP lists bleeding from the umbilical stump or nose, unusual bruising around the head and face, pale skin or gums, bloody or black tarry stool, vomiting blood, jaundice appearing three weeks or more after birth, and seizures or unusual lethargy. The Centers for Disease Control and Prevention estimates that 1 in every 5 babies who develop VKDB will die.
The number nobody has
Here is the uncomfortable part: no federal or state agency tracks vitamin K refusal, and none tracks VKDB. Deaths directly attributed to the condition on death certificates number fewer than a dozen a year, though that count has been rising. But death certificates rarely name a nutrient deficiency; they name the bleed. More than 700 newborns died of spontaneous brain hemorrhage in 2024, a figure that also captures prematurity and liver disease. Six medical specialists and a CDC official told ProPublica that a meaningful share of those deaths was likely caused by vitamin K deficiency.
The gap is not neutral. Dr. Jaspreet Loyal, a pediatric hospitalist at Yale Medicine, described it to the same reporters as a kind of false comfort: The lack of data is almost acting like a reassurance for families that this risk is worth taking.
Which is the trap. A parent researching the injection at 2 a.m. finds passionate testimony from families whose babies skipped the shot and came through fine, because the overwhelming majority of them do, and finds almost no counting of the ones who didn’t. The evidence for the shot is a century old and settled. The evidence of what happens without it is scattered across autopsy reports that nobody aggregates.
Heather Burris, a CHOP neonatologist and one of the study’s authors, reaches for the most ordinary comparison she can find. We use car seats, we make sure they wear seatbelts, and we hold our kids’ hands when we cross the street,
she said. Another step we take to keep our children safe is to give them newborn Vitamin K.
Nobody asks for a control group before buckling a car seat. The vitamin K shot has six decades of evidence behind it, and it is losing ground anyway.
