Children who broke a bone in a minor fall had lower vitamin D than children who fell and didn't. The doctors who found it say the blood test still isn't worth ordering after an ordinary fracture

A child falls off a chair or lands badly on a trampoline, and an hour later a parent is looking at an X-ray of a cracked wrist. The question that follows is almost always the same: are the bones weak, and should we be giving vitamin D? A small study from South Tyrol, the mountain province in Italy's far north, tried to answer it with a blood sample, and the answer it gives is more careful than the usual one.

Between July 2024 and November 2025, orthopaedic doctors at the hospitals of Merano and Bressanone invited children aged 3 to 15 who arrived after what is called low-energy trauma, meaning ordinary falls and knocks and not road accidents or falls from a height. The children whose X-rays showed a fracture became the cases. The children whose X-rays were clear became the comparison group. Within a week, each gave blood for vitamin D (measured as 25-hydroxyvitamin D), parathyroid hormone, calcium, phosphate and alkaline phosphatase, and parents filled in a questionnaire about calcium in the diet. The paper appeared in the European Journal of Pediatrics on 9 October 2026.

The final group was 100 children, 50 with a fracture and 50 without, average age about ten. Three in four fractures were in the arm, most often the wrist or forearm. The vitamin D gap was clear enough: a median of 53.9 nmol/L in the fracture group against 70.4 in the others. Taking the usual line for 'enough' as 50 nmol/L (20 ng/mL), 40% of the children with fractures were below it against 16% of those without. After allowing for age, sex and body size, being under that line went with about three times the odds of being in the fracture group.

That is the part that will be quoted. The rest of the paper pulls the other way, and the authors say so themselves. Real deficiency, below 30 nmol/L, was rare: by our arithmetic from the paper's percentages, two children in the fracture group and one in the other. Parathyroid hormone, calcium and phosphate were normal and the same in both groups, which is not what you would see if soft, under-mineralised bone were behind the breaks. Alkaline phosphatase was a little higher in the children with fractures, but it rises when a bone starts to heal, and the blood was drawn after the injury.

The team also asked how well a vitamin D result could tell the two groups apart. Not very well. The score they report, 0.653, sits on a scale where 0.5 is a coin toss and 1 is perfect. The best cut-off the data could find was 67.6 nmol/L, well above the line normally used for sufficiency, and even there it missed a quarter of the children with fractures and wrongly flagged four in ten of the children without. A test like that cannot say much about any one child.

Their more interesting claim is about weight. The link between low vitamin D and fracture was stronger in heavier children and close to absent in leaner ones. It fits what is already known: body fat soaks up vitamin D, and a heavier child lands harder. But here we would be cautious for a reason the paper does not raise. Heights and weights were not measured at the hospital. They came from the parents, reporting the child's last routine check-up. And the table of results shows that the fracture group's body-size scores spread twice as wide as the other group's and run down to minus 9.4, a figure far outside anything expected for a healthy child and more likely to be an old or mistaken measurement. In a study of 100 children, a handful of odd values can shape a finding of exactly this kind. That is our reading of the table, not the authors'.

The same goes for calcium. The paper says low vitamin D mattered more in children who ate less calcium, because the gap in fractures was statistically significant in the low-calcium half and not in the high-calcium half. Worked back from the percentages, each of those comparisons rests on 14 children with low vitamin D: 11 of the 14 had a fracture in one half and 9 of the 14 in the other. The two gaps are about the same size, and the paper does not test whether they differ. One passing a significance line and the other just missing it is not evidence that calcium changes anything.

There are plainer limits too, most of which the authors list. Thirty-nine children recruited for the comparison group never gave blood or returned the questionnaire and were dropped, so the comparison group is made of the families who followed through, and no one can check how they differed. Boys were 68% of the fracture group and 46% of the other. Nobody recorded puberty, sport, time outdoors or family income, all of which move both vitamin D and the chance of a fall. And because the study picked 50 children with fractures and 50 without, it cannot tell anyone the actual risk of a fracture at a given vitamin D level.

So what should a parent take from it? The authors' own conclusion is the useful one: in an otherwise healthy child, a broken bone after a fall is mostly about the fall, the child's age and how the body met the ground, and routine vitamin D testing after an uncomplicated fracture has limited value. That does not make vitamin D and calcium unimportant. Growing bones need both, and a child who is short of them should have that put right for its own sake. If your child keeps breaking bones, breaks one with very little force, or has a restricted diet or a long-term illness, that is a reason to ask a doctor for a proper look. One cast after one fall usually is not.