Can a flat head still improve after four months?

There is a misconception worth clearing up first, because it explains everything else. Most people assume a baby's skull is soft like clay, and that softness is what makes it responsive. Gary Rogers, the craniofacial surgeon whose work underpins most of this article, rejects that outright. If softness were the cause, every baby placed on their back would flatten, and in reality fewer than a quarter do.¹ Lay a newborn on a firm surface and the head does not deform.

What actually happens is growth. The skull does not push itself outward. It grows passively, in response to the brain expanding underneath it, and where that growth meets a resting surface it is resisted, so the extra volume is displaced elsewhere. Rogers uses a pumpkin to explain it: a pumpkin cannot grow into the ground, so it grows along it and flattens underneath.¹

That has a direct consequence for timing. How much improvement is possible depends entirely on how much rapid brain growth is still to come. Skull growth is fastest in the first four to six months of life and tapers sharply after the first year. The three to six month period matters because most of the growing is still ahead of you. Growth is what any approach has to work with, and cranial growth is largely complete by early childhood. It is the last stretch where there is plenty of it left.

This is the mechanism many parents are trying to work with at home: taking pressure off the resting surface while that growth window is still open. The Perfect Noggin is a contoured head shaping surface that redistributes pressure away from flat spots, used alongside your physiotherapist's work rather than instead of it.

Four things shift across this window, and they pull in different directions.

Your baby stops adding to the problem. Rogers notes most term babies develop enough strength to hold their head against gravity by around three months, and that further flattening after that age is unusual.¹ So by three months you are generally no longer watching it get worse. You are watching what has already accumulated.

The flattening looks its worst at four months. In the Hutchison cohort, which followed 200 New Zealand babies from birth, prevalence peaked at 19.7 per cent at four months.² That is the point of maximum visibility, which is why so many parents come looking for answers at exactly this age.

Rolling starts, and it takes the pressure off. Raising Children Network puts rolling both ways at around four to five months.⁹ A baby who rolls in their sleep stops resting on one spot, and the measured rates in the Hutchison cohort fall from 19.7 per cent at four months to 9.2 per cent at eight.² That is a count of how many babies sat above a measurement threshold, though, so it tells you the pressure is easing rather than that the flat spots have gone. Rolling is also the point at which the cot needs to be completely clear, with nothing extra in the sleep space once your baby can turn themselves over.

The neck finally shows itself. This is the one most parents are not warned about. Before three to four months the head is supported by a surface most of the time, so a neck imbalance can be invisible. Once your baby holds their own head up, it shows. Fenton and Gaetani report congenital muscular torticollis in 70 to 95 per cent of infants with deformational plagiocephaly¹¹, so if a preference or a tilt appears in this window, that is the thing to get assessed.

This is the reassurance most parents get, and it deserves a straight answer rather than a comfortable one.

There is a real effect behind it, sometimes called the dilutional effect of growth. As the head gets bigger, a flat area becomes a smaller proportion of the whole and looks less pronounced. But the head has not changed shape. Once your baby has independent head mobility at around four to five months, the pressure that caused the flattening lifts, and both sides of the back of the skull then grow at a similar rate. Because neither side grows faster, the physical indentation stays. The head gets bigger and the flat spot travels with it.

Measured rates do fall as babies grow. The Hutchison cohort dropped to 3.3 per cent by two years², and the Royal Children's Hospital notes that plagiocephaly and brachycephaly are present in around 1 per cent of teenagers.⁵ But those figures count how many babies sat above a measurement threshold, not how many flat spots disappeared, and the studies followed babies getting ordinary care with no untreated group to compare against. Nobody can tell you in advance which group your baby is in.

But measurable asymmetry persists more often than visible concern does. Steinbok and colleagues followed children with positional plagiocephaly to beyond five years of age.³ Residual asymmetry was noticed by 58 per cent of parents, though only 21 per cent were concerned about their child's appearance. Those two numbers belong together. Something is usually still measurable.

In that same study, parents of children who had used a cranial orthotic did not report better long-term cosmetic outcomes than parents of children managed with counter-positioning alone.³ The authors were careful about how far that goes: only 65 of 278 families responded, the numbers were too small to rule out a real difference being missed, and the children given orthoses had more pronounced asymmetry to begin with. It is a reminder that an orthotic is one option among several rather than the only route to a good outcome, and that repositioning and physiotherapy carry real weight. Your paediatric physiotherapist or paediatrician is the right person to weigh those options for your baby.

So the honest position is this. Waiting and hoping is a poor strategy, because if the shape is not improving on its own you will usually only know that once the useful window has passed. Early, structured, low-tech action is the sensible place to start: repositioning, tummy time, and getting the neck assessed. The evidence behind it is strongest for stopping flattening getting worse rather than for undoing flattening that has already formed, which is why the neck assessment is the part that matters most. What that will achieve for your baby is not something anyone can promise in advance, which is the argument for starting now rather than later.

Not every unusual head shape is positional flattening. Some are caused by craniosynostosis, where the joints between the skull bones fuse too early, and that needs medical assessment rather than repositioning. This article cannot tell you which one you are looking at. A GP or paediatrician can, and it is a quick thing to rule out.

When to seek medical assessment for your baby's head shape goes through the warning signs in more detail.

No. Four months sits comfortably inside a useful window and is worth acting on promptly.

What has changed by four months is that time is no longer working as hard in your favour. Each week from here, brain growth slows slightly and that process weakens a little. That is an argument for booking an assessment this month rather than next, not an argument for panic.

The most useful thing a parent at four months can do is get a paediatric physiotherapy assessment, get the severity measured properly rather than estimated, and take structured action instead of waiting to see what develops.

Book a paediatric physiotherapy assessment this week.

Depending on your baby's rolling status and how pronounced the flattening is, there are usually still options. A physiotherapist can tell you where things actually stand, what is realistic at your baby's age, and what the sensible next step is. Starting at five months is later than ideal, but it is not the same as starting at nine. There is still brain growth to work with, and repositioning, tummy time and treating any neck tightness are all still worth doing.

For the small number of babies where a helmet is appropriate, this is also the age range where that conversation happens. The Royal Children's Hospital notes a specialist may suggest one if a baby's head is severely and persistently misshapen at six months of age, and that helmet therapy is most effective if started between six and eight months of age.⁶ Pregnancy, Birth and Baby notes fewer than one in ten babies with plagiocephaly will need one.⁸

What is not useful is inaction. The slope is still there. You are further along it, which is why moving now matters more, not less.

There is no single day when head shape locks in. It sets gradually, as the growth described above slows down.

• Birth to 3 months. The fastest head growth of your baby's life. The WHO growth standards show more head growth from birth to three months than in the nine months that follow.¹³
• After that. Growth slows but does not stop. It carries on until the soft spot on top of the head, the anterior fontanelle, closes. Pregnancy, Birth and Baby notes that can happen any time between 4 and 26 months.¹²

Less growth means smaller, slower change. That is why a flat spot that is still there at around two years tends to stay, and why the measured rates covered under Will my baby's flat head round out on its own? fall more slowly with age.

Get the assessment. Book with your child and family health nurse or GP and ask specifically for a referral to a paediatric physiotherapist. Ask for it by name. A physio measures the asymmetry objectively, tests neck rotation properly, and works out whether you are dealing with tightness, weakness or both.

Photograph the head from directly above every two weeks, in the same spot and the same light.

Keep the daily habits going. Supervised tummy time every day while your baby is awake, alternate which end of the cot you use, alternate which arm you feed from, and limit stretches in car seats, capsules and bouncers where the head rests in one spot. The NSW Health guideline for allied health professionals, GL2020_013, lists infant handling, counter positioning, stimulation of motor development and parental education as having strong evidence behind them.⁷

Always place your baby on their back to sleep. Red Nose Australia advises against pillows in the cot, and against products that restrict a baby's movement or hold the head in position.¹⁰ The Royal Children's Hospital similarly advises against using a pillow during sleep to prevent or manage flattening, because of the risk of sudden infant death syndrome.⁵

Deal with the neck, not just the head. The Royal Children's Hospital notes head shape typically improves once torticollis is treated.⁵ If your baby has a side preference, that is the thing to raise.

If you want the full stage-by-stage picture from birth to twelve months, baby flat head progression month by month covers each age band. If you are not sure what you are looking at, the four-point check walks through it in two minutes.