- Design
- nationwide register-based cohort study with multivariable adjustment
- Population
- 2,020,302 live births at 35 or more weeks' gestation in Sweden, 2003 to 2021, of whom 24,089 had no record of intramuscular vitamin K
- Primary outcome
- bleeding diagnosis within the first 6 months of life
- Effect
- adjusted odds ratio 1.52 (95% CI 1.27 to 1.81) for any bleeding and 2.91 (2.13 to 3.96) for intracranial bleeding
A Swedish national cohort covered 2,020,302 live births at 35 weeks or more between 2003 and 2021, followed to six months. Of these, 24,089 infants had no record of intramuscular vitamin K at birth.
Those infants had adjusted odds of a bleeding diagnosis of 1.52 (95% CI 1.27 to 1.81) and of intracranial bleeding of 2.91 (2.13 to 3.96). The trend in non-receipt is the second finding: it fell from 1.32% in 2003 to 0.66% in 2006, then more than doubled to 1.50% by 2021.
Vitamin K deficiency bleeding is rare, and rarity is exactly what makes refusal feel reasonable to a parent who has never seen it. The number worth carrying into that conversation is the intracranial one - roughly three times the odds - because late vitamin K deficiency bleeding presents as intracranial haemorrhage in a previously well baby, often without warning and often with permanent damage. A doubling of refusal over fifteen years in a high-coverage system is a signal for every service, not only Sweden's.
- Record vitamin K administration explicitly, and record refusal explicitly - a missing entry and a refusal look identical in an audit
- Have the conversation antenatally rather than in the hour after birth, where refusal is hardest to revisit
- Where parents decline the injection, discuss the oral schedule and its requirement for repeat doses rather than accepting nothing
- Consider vitamin K deficiency bleeding in any infant under six months with unexplained bleeding or an intracranial haemorrhage, and ask what was given at birth
The statistics, in plain English
These are adjusted odds ratios from a national register, so the comparison is between infants whose parents declined or who otherwise missed the injection and those who received it - groups that differ in more than vitamin K. Home birth, less contact with services and other declined interventions all travel with refusal and could contribute. The intervals are narrow (1.27 to 1.81, and 2.13 to 3.96) because the denominator is two million births, which gives precision but does not address confounding. Note that the outcome is any bleeding diagnosis, not confirmed vitamin K deficiency bleeding, which will include unrelated events and dilute rather than inflate the association. Absolute risk stays very low in both groups.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for paediatrics, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free