Skipping newborn vitamin K tied to 47% higher infant bleeding risk

As more US parents decline newborn intramuscular (IM) vitamin K prophylaxis, a large national cohort study now quantifies the cost: infants who did not receive the injection had a 47% higher risk of bleeding during their first 6 months of life and more than double the risk of intracranial bleeding.¹

Published in JAMA, the retrospective study of more than 5.2 million infants also found that vitamin K nonreceipt more than doubled, from 3.1% in 2017 to 6.8% in 2025.¹ In an accompanying editorial, Laura R. Kair, MD, MAS, of University of California Davis School of Medicine, and colleagues wrote that, with these data, "communicating the risk of bleeding with vitamin K declination just got a little easier."²

Epic Cosmos cohort design for newborn vitamin K and infant bleeding

Investigators led by Kristan Scott, MD, MSHP, of the Children's Hospital of Philadelphia, analyzed Epic Cosmos electronic health record data for infants born at 35 to 43 weeks' gestation from January 2017 through December 2025.¹ After quality filters, the cohort included 5,240,780 infants; 234,825 (4.5%) did not receive IM vitamin K.¹

The primary outcome was any bleeding from birth through 6 months, a composite of ICD-10 codes for intraventricular hemorrhage, intracranial hemorrhage, newborn hemorrhage, and vitamin K deficiency bleeding (VKDB).¹ Models adjusted for gestational age, sex, small for gestational age status, race and ethnicity, insurance, maternal age and parity, social vulnerability, rurality, and year.¹

Rising vitamin K refusal and infant bleeding trends, 2017-2025

Bleeding increased in both groups over the study period, but faster among infants who did not receive vitamin K.¹ Among nonrecipients, bleeding by 6 months rose from 225.3 to 394.0 per 10,000 births, compared with 136.6 to 221.8 per 10,000 among recipients.¹ The annual increase in bleeding odds was 8.7% vs 5.6%, respectively (P < .001 for interaction).¹ The authors noted the overall rise could reflect greater ascertainment or sicker maternal-infant dyads.¹

Vitamin K nonreceipt, intracranial hemorrhage, and mortality risk

Overall, 288.0 vs 185.9 per 10,000 nonrecipients and recipients, respectively, had a bleeding event by 6 months (adjusted risk ratio [ARR], 1.47; 95% CI, 1.43-1.51).¹ The adjusted number needed to harm was 114, meaning 1 additional infant would bleed for every 114 newborns who did not receive vitamin K.¹

Nonreceipt was also associated with higher risk of intracranial bleeding (33.0 vs 13.2 per 10,000; ARR, 2.43; 95% CI, 2.26-2.62) and death with concurrent bleeding during the birth hospitalization (2.9 vs 1.1 per 10,000; ARR, 2.69; 95% CI, 2.09-3.47).¹ Bleeding after discharge through 6 months, the window for late VKDB, roughly doubled (10.3 vs 4.8 per 10,000; ARR, 2.11; 95% CI, 1.85-2.41).¹ Findings held in sensitivity analyses that relaxed hospital inclusion criteria.¹

Swedish cohort and VKDB trend data support the findings

The relative risks closely mirrored those from a Swedish cohort of more than 2 million infants, which reported adjusted odds ratios of 1.52 for any bleeding and 2.91 for intracranial bleeding.¹⁴ A companion JAMA Pediatrics research letter using Cosmos data from more than 11 million infants found VKDB diagnoses rose from 24.0 to 107.4 per 100,000 between 2017 and 2025, as injectable vitamin K receipt fell from 97.7% to 95.0%.³ Only 19.2% of infants with VKDB in that analysis had received the injection.³

Study limitations and implications for vitamin K counseling

The cohort excluded infants born before 35 weeks and home births, and birth centers were likely underrepresented, potentially underestimating nonreceipt.¹ Investigators could not determine whether infants received oral vitamin K or later IM doses, and data lacked bleeding timing, severity, and etiology.¹ Clinicians may also code bleeding more readily when vitamin K was withheld, which could inflate estimates.¹

Kair and colleagues said the number needed to harm offers a concrete figure that may resonate with parents more than percentages.² They called on obstetricians, nurses, and public health agencies to reinforce IM vitamin K recommendations, urged dedicated surveillance of administration and outcomes, and called for an FDA-regulated oral formulation for families who firmly decline the injection.²

References

1. Scott K, Handley SC, Lo JY, et al. Nonreceipt of newborn vitamin K and risk of bleeding. JAMA. Published online November 5, 2026. doi:10.1001/jama.2026.21764

2. Kair LR, Goyal N, Phillipi CA. Communicating risk of intramuscular vitamin K declination in the US. JAMA. Published online November 5, 2026. doi:10.1001/jama.2026.23146

3. Wolf ER, French E, Shokouhi Niaki D, Woolf SH. Vitamin K deficiency and associated bleeding in infants. JAMA Pediatr. Published online November 5, 2026. doi:10.1001/jamapediatrics.2026.5794

4. Simatou E, Tsamantioti E, Hallström A, et al. Vitamin K prophylaxis in newborns and bleeding in infancy. JAMA Pediatr. 2026;180(9):995-1004. doi:10.1001/jamapediatrics.2026.2606