Birth injury, in whole numbers
In a Finnish population study, 132 arm-nerve injuries in 41,980 births, of which 27 had not healed by the first birthday.
Birth injury is usually discussed, if at all, as a legal category. That leaves parents with no sense of scale at all. Here is the scale.
The nerve injury to the arm, which is the one people mean. A population study in Helsinki, Finland followed every baby born between 2003 and 2006, with repeated clinical examinations and ultrasound scans at 1, 3, 6 and 12 months.
So in that population, about 3 babies in every 1,000 had the injury and fewer than 1 in 1,000 still had it at their first birthday. Two cautions travel with those figures. They are Finnish, and the current US rate of the injury is about a third of the Finnish one, so the proportion that heals may not transfer. And you will see a widely quoted recovery range attached to this injury in other sources; its original source could not be retrieved and read for this guide, so it is not repeated here. The two Finnish numbers above are what can be stood behind.
The US rate of that injury, and its direction. Across US hospital data covering 23,385,597 births, the rate fell from 1.7 per 1,000 live births in 1997 to 0.9 per 1,000 in 2012, and has stayed between 0.9 and 1.1 per 1,000 from 2006 through 2019. About 1 in 1,000, roughly half what it was in the late 1990s. Shoulder dystocia raises the risk enormously, but 55.0% of cases had no identifiable risk factor at all, which is why it is not something you can screen your way out of in advance.
The overall trend, which runs two ways at once. In a US national sample of hospital births from 2006 to 2014, all coded birth trauma rose from 25.3 to 31.1 per 1,000. Scalp injuries, the mildest end of the category and 80% of the total, rose from 19.87 to 26.46 per 1,000. Major birth trauma fell, from 5.44 to 4.67 per 1,000. More is being recorded; less of what matters is happening.
Injury at cesarean, which most parents assume is zero. A prospective study counted every injury to the baby in all 37,110 cesarean deliveries at 13 university centres over two years.
| In 37,110 US cesarean deliveries | Number | Rate |
|---|---|---|
| Any identified injury to the baby | 418 | 1.1%, about 1 in 90 |
| Skin laceration | 272 | 0.7% |
| Cephalohematoma | 88 | |
| Facial nerve palsy | 11 | |
| Collarbone fracture | 11 | |
| Arm-nerve injury | 9 | |
| Skull fracture | 6 |
Two thirds of those injuries were a skin cut. But the rate depended enormously on why the operation was done: 6.9% after a first cesarean that followed a failed forceps or vacuum attempt, 1.7% for a repeat cesarean after a failed instrumental attempt, and 0.5% for a repeat cesarean planned before labour, the lowest of any group. The order of those three numbers is the honest answer to whether a cesarean is safer for the baby: it depends entirely on which cesarean.
Collarbone fracture. No US population figure could be verified. Finnish national data covering 629,457 vaginal births at 37 weeks or later found the rate falling from 17.6 per 1,000 vaginal live births in 2004 to 2010, to 6.2 per 1,000 in 2011 to 2017. Note the denominator: vaginal births, not all births.
The one scalp bleed that is genuinely dangerous, and the number that does not exist. Subgaleal haemorrhage is bleeding into the space under the scalp, and unlike the common lumps it can be serious. What can be verified is in words, not numbers. One review describes it as a rare condition of the neonate often associated with instrumental delivery
and a potentially fatal condition that is often underreported and underdiagnosed
. A 2025 review states: Although infrequent, the incidence of SH has not decreased over the past several decades, despite improvements in obstetrical care.
No incidence figure with a stated denominator could be traced to a retrievable source for this guide, including a pair of figures that is repeated constantly in textbooks and online. Those numbers are therefore not printed here. Rare, potentially serious, associated with vacuum delivery, and not falling. That is the whole verified statement.
"How many of these have you done, and what is your own success rate with them?"
"What are you watching my baby's head for afterwards, and for how long?"
"Will any mark or lump be written into the notes before we go home?"