A flat spot on the back or side of a baby's head is one of the most common things parents notice in the first six months, and one of the least dangerous.
- Positional flattening affects an estimated 20% to 50% of babies at 6 months
- About 16% of six-week-olds show flattening; by age two it is down to 3.3%
- An estimated 85% of babies with torticollis also have a positional skull deformity
- Tummy time builds up to about an hour a day, spread over several short sessions
You notice it in a photo, usually. The angle from above while he is in the car seat, or the bath shot where the hair is wet and the shape is suddenly obvious. Then you cannot unsee it, and you start looking at every other baby's head at the playgroup. That part is normal too.
Quick Reference
| Age | What is happening | What to do |
|---|---|---|
| 0 to 6 weeks | Skull is at its most mouldable, flattening starts | On the back for every sleep, alternating which way the head faces. Tummy time from day one, seconds at a time |
| 6 weeks to 4 months | The window where repositioning does the most work | Alternate sides, limit time in car seats and bouncers, build tummy time up |
| 4 to 6 months | Flattening usually peaks, then starts to correct | Keep going. This is when your doctor reassesses if nothing has shifted |
| 6 to 12 months | Sitting and crawling take pressure off the back of the head | Shape improves on its own as he spends less time lying down |
| 12 to 24 months | Hair grows, skull grows, most of it disappears | Only 3.3% still show flattening at two |
Plagiocephaly or brachycephaly: what you are looking at
The two words describe two shapes, and knowing which one you have makes the advice easier to follow.
Plagiocephaly is when the head is "asymmetrically (unevenly) flat in the back on one side", and the forehead on that same side may push forward, so from above the head reads "like a parallelogram" rather than an oval. Brachycephaly is when the head is "symmetrically (evenly) flat in the back and wide from side to side" (HealthyChildren, AAP).
The view that tells you most is from directly above, looking down, with the hair damp. Not the mirror, not a side-on photo. Take one every few weeks from the same angle so you are comparing pictures instead of comparing today against a memory of March.
Why it happens, and why back-sleeping is still not up for debate
A baby's skull is soft and designed to change shape. The NHS puts it plainly: "a baby's skull is soft and can easily change shape", and flattening follows wherever pressure is steady (NHS).
Flat spots became more common after back-sleeping became the standard advice. That fact gets repeated in a way that makes parents wonder whether the trade is worth it, so let us be direct about the arithmetic. Back-sleeping prevents infant deaths. Flattening is a cosmetic problem that mostly resolves. German public health guidance holds the line for the whole first year: "place your baby on their back to sleep throughout the first year", and in the same breath recommends tummy time while awake because it "prevents possible one-sided deformation of the skull" (kindergesundheit-info.de, BIÖG).
Put the baby on his back to sleep. Then work on everything else. If the back is still new territory and nights are the bigger worry right now, the newborn sleep picture is the better place to start than the head shape.
The pressure that shapes a head is not only sleep, and this is the part that gets missed. Car seat, bouncer, swing, pram, play gym, the carrier that gets skipped because it is easier to click the car seat into the pram base. Each one holds the same point of the skull against the same surface. On a busy day those hours add up faster than the night does.
The torticollis link most parents miss
Here is the thing to check before you buy a single product.
An estimated 85% of babies with torticollis, a tightness on one side of the neck, also have a positional skull deformity. The mechanism is not mysterious: if turning one way is uncomfortable, the baby lies on the same spot every time, and the flattening follows the preference rather than causing it.
So the real question is not "how do I fix this flat spot". It is "why does he always look that way". Watch him on his back for a minute. Does his head rest turned to one side by default? When you offer a toy on the other side, does he track it with his eyes but not turn his neck? The NHS lists exactly this as a reason to see a GP: "your baby's head is always tilted to one side, or they have difficulty turning their head".
Treated torticollis is a physiotherapy referral and some stretches, and it usually responds well. Untreated, you can reposition all you like and the head keeps going back to the comfortable side. The stiff neck comes first, and it is easy to see once you know to look for it. Motor development runs on the same track, which is why the crawling timeline is worth a look if head preference is part of a wider pattern.
What actually works
Repositioning and tummy time. That is the whole treatment for the large majority of babies, and it is unglamorous enough that parents keep looking for something else.
Put him on his back for every sleep, then alternate which way the head faces. The sleep position never changes: always on the back. What you vary is the side his head is turned toward while he lies there, and the direction the interesting things come from. Babies turn toward the room, so turning the cot end to end every week or so changes which side is the interesting one without you having to fight him.
Move the interesting things. The NHS advice is to "change the position of your baby's toys and mobiles when they're lying on their back". A baby will look at the door, the window and you. Change where those are relative to him.
Switch sides when you feed. Alternate arms at every bottle, the way breastfeeding naturally alternates.
Cut container time. Car seats, bouncers and swings hold the head in one fixed position. Necessary in the car. Not necessary in the living room for an hour after you get home.
Build tummy time up properly. Start with 3 to 5 minutes, 2 to 3 times a day, and "gradually work up to an hour per day in several short sessions". An hour sounds enormous when your baby screams after ninety seconds. It is not one hour, it is eight bad minutes repeated across a day, on your chest, over your lap, on a blanket on the floor. Always while your baby is awake and you are right there (AAP). The floor is not the only venue, but it is the safest one; a changing table is a place to change nappies, not to practise.
The stubborn one is evenings. Tummy time at 6pm with a tired baby is a fight nobody wins, and skipping it then is not failure, it is scheduling. Morning babies are better at this.
One more thing about container time, since it sits right next to a problem plenty of families already have. A baby who only settles when held spends a lot of hours against a chest, and a chest is the one surface that puts no pressure on the back of the head at all. That habit is hard on you and quietly good for his skull, as long as you are awake while he is in your arms. When he falls asleep, he still goes on his back on his own flat, firm surface.
What the helmet evidence really says
This is the part worth knowing before anyone quotes you a price.
Helmet therapy has an intuitive logic and a large market. It also has a randomised controlled trial, and the trial is not kind to it. Dutch researchers randomised 84 infants aged 5 to 6 months with moderate to severe skull deformation to either six months of helmet therapy or the natural course of the condition, and measured skull shape blind at 24 months. Full recovery was achieved in 26% of the helmet group and 23% of the untreated group, with no statistically significant difference between them (p=0.74). "All parents reported one or more side effects." The authors concluded that they "discourage the use of a helmet as a standard treatment for healthy infants with moderate to severe skull deformation" (van Wijk et al., BMJ, 2014).
The NHS reached the same place: "helmets and headbands are not available on the NHS to treat flat head syndrome because there's not enough evidence to show they work".
This is not the whole picture, and it would be dishonest to present it as one. The AAP still lists a helmet as something to consider for "moderate or severe head flattening that does not respond to treatment by 5 or 6 months of age", and German paediatricians take a similar line for severe cases, noting the child has to wear it almost continuously. The trial only included babies born after 36 weeks, aged 5 to 6 months, without torticollis or craniosynostosis, and all of them had physiotherapy before they joined, so it does not answer every case. A similar outcome in that group does not prove a helmet never makes sense. The US neurosurgeons' guideline recommends a helmet for persistent moderate to severe plagiocephaly after repositioning or physiotherapy, or when a baby is seen late, while rating its own certainty as uncertain (CNS).
What the evidence supports is a sequence, not a ban. Reposition properly first. Treat the neck if the neck is the problem. If a specialist recommends a helmet for a severe case that has not moved by six months, that is a legitimate conversation. Buying one at three months because a clinic's before-and-after photos are persuasive is not.
Reposition, refer, or wait
- Mild flattening, turns head both ways easily, under 4 months → repositioning and supervised tummy time, compared with photos from the same angle after 6 to 8 weeks. If anything worries you before then, ask; the home plan is not a waiting list.
- Head preference or a tilt, any age → ask about torticollis and a physiotherapy referral. This is the one that changes the outcome.
- Unusual head shape present at birth, or a palpable ridge → get it looked at rather than repositioned. Craniosynostosis produces "changes in the infant's head shape that are present at birth" and needs a specialist (AAP clinical report).
- Moderate or severe, no change by 5 to 6 months despite doing the work → this is the point where helmet therapy is a real discussion, with a specialist, with the trial evidence in the room.
- Improving slowly, baby is 7 months and starting to sit → keep going and stop measuring weekly. Sitting up is doing the work now.
Common mistakes
Buying a shaping pillow. Positioning pillows and head-shaping cushions add soft bedding to a sleep surface, which runs against safe sleep advice. This is the one place where the flat-head fix can create a real risk.
Repositioning the head while ignoring the neck. If turning one way is uncomfortable, every night's careful positioning is undone by morning.
Judging progress weekly. Skull shape changes over months. Weekly checks mostly measure your own anxiety.
Treating tummy time as floor time. Chest, lap, forearm and a blanket on the floor all count, while he is awake and watched. The requirement is time off the back of the head, not a specific mat.
Assuming it is too late at five months. Improvement continues through the first and second year as the head grows and he spends less time lying down. The window narrows, it does not slam.
Reading a head circumference percentile as a shape measurement. Circumference tells you about size, not shape, and a perfectly average number sits happily on a flattened head. Worth knowing what growth percentiles actually describe before using one to reassure yourself either way.
When to see your doctor
See your doctor if the head is always tilted, if he cannot turn it fully to both sides, if the shape was unusual at birth, if you can feel a firm ridge along the skull, or if things are getting worse rather than better after two solid months of repositioning.
Book the routine conversation if you simply do not know whether it has improved since summer. That is most parents, and it is a fair reason to ask. Bringing dated photos from the same angle, plus when the head preference started, turns a vague worry into something a doctor can actually assess, which is the whole argument in how to prepare for a pediatric visit using your child's data.
Frequently asked questions
Will my baby's flat head correct itself?
Usually, yes. The NHS says it "often gets better in around 1 to 2 years as your baby's head grows and they spend less time lying on their back". The numbers back that up: flattening shows in about 16% of six-week-olds and only 3.3% of two-year-olds. Correction is slow and mostly invisible week to week, which is why photos beat impressions.
Does a flat head affect brain development?
No. Positional skull deformities "do not affect brain growth", which is the key difference from craniosynostosis, where the skull bones fuse too early and the brain does need room. Positional flattening is a shape problem, not a pressure problem, and it needs no surgery.
How much tummy time does my baby actually need?
Start with 3 to 5 minutes, two or three times a day from the first weeks, and build gradually toward about an hour a day split into several short sessions. Consistency matters more than duration. Two minutes six times beats one miserable ten-minute block, and a baby who has just eaten will hate all of it.
Can a special pillow fix a flat head?
No, and it introduces a risk that the flat spot does not carry. Head-shaping pillows mean extra soft bedding in the sleep space, which is not compatible with safe sleep guidance. The approaches with evidence behind them are alternating head position, reducing time in car seats and bouncers, tummy time, and treating a tight neck if there is one.
Is my baby's head shape my fault?
No. It follows from a soft skull, a strong head-turning preference and a sleep position that prevents infant deaths. Many babies with a marked flat spot had attentive parents doing tummy time daily. Some babies simply have a stronger preference for one side, and some spend their first weeks in a NICU where position is not negotiable.
At what age is it too late to do anything?
There is no hard cutoff, but the skull is most mouldable in the first four to six months, so repositioning does most of its work early. After six months, sitting and crawling take over and the shape improves through natural growth. If a helmet is going to be discussed at all, that conversation happens around 5 to 6 months, not at 12.
How KidyGrow helps you
You will be asked when the head preference started, and whether it has improved. Six months later that is genuinely hard to answer. KidyGrow remembers the short notes you leave at the time, so the answer is a date rather than a guess. Milestones you tick off, like rolling both ways, keep their dates too. Before a check-up, appointment prep turns your recent notes, milestones and concerns into a summary with questions worth asking.
It does not measure head shape. Nothing in your pocket does. That is what the dated photo from above is for.
Sources
- American Academy of Pediatrics, HealthyChildren.org: When a Baby's Head is Misshapen: Positional Skull Deformities
- American Academy of Pediatrics: AAP Releases Clinical Report: Differentiating Infant Head Shape Abnormalities
- van Wijk RM, van Vlimmeren LA, Groothuis-Oudshoorn CG, et al. Helmet therapy in infants with positional skull deformation: randomised controlled trial. BMJ. 2014;348:g2741
- NHS: Plagiocephaly and brachycephaly (flat head syndrome)
- BIÖG, kindergesundheit-info.de: Die Schlafumgebung für das Baby
- BVKJ, Kinderärzte im Netz: Kopfverformung beim Baby: Meist hilft Lagerungstherapie
- Congress of Neurological Surgeons: Guideline on the Role of Cranial Molding Orthosis (Helmet) Therapy for Patients With Positional Plagiocephaly. Neurosurgery. 2016






