WhatsAppinfo@ri-hub.it
Ri-Hub Junior

Infant Colic: What the Studies Really Say

Team Ri-Hub14 min lettura

Revisione clinica: Daniele Carretto, Fisioterapista (Albo 2019) · Pubblicato il 11 agosto 2026

Bambino carino sdraiato su un tappeto da gioco a Poznań, con un'espressione facciale serena.
Condividi

It's three in the morning. You have been holding them for forty minutes, they have fed, they are dry, they have no fever, and they keep crying with their legs drawn up to their belly. You have already read fifteen pages on your phone, and every one of them offered you something to buy or book.

Many parents don't search for the word "colic" but for what they see: inconsolable crying in a newborn. It's the same thing, and the definition used in studies starts exactly there: prolonged crying with no apparent cause, in a baby who is otherwise healthy.

We'll tell you two uncomfortable things up front, which are more useful than another list of remedies.

The first: colic is not our field. We are physiotherapists, and colic is a paediatric matter: the diagnosis is made by the paediatrician.

The second: on one thing we can genuinely help, which is reading the studies behind what is being offered to you: osteopathy, manipulation, probiotics, fascia, and telling you what they really hold up. Most of these fall within our professional area, and you deserve to know the state of the literature before you pay.

What "colic" means (and why the definition has changed)

The definition still used almost everywhere is the "rule of three": crying for at least 3 hours a day, at least 3 days a week, for at least 3 weeks. It is also the definition the Cochrane review on dietary modifications uses.

The point is that many specialists have largely abandoned it. In 2016 the international reference criteria (the Rome IV criteria) were revised precisely for colic, and the reason given in a dedicated review is as simple as it is freeing: the 3-hour threshold was arbitrary, because there is no evidence that a baby crying for 2 hours and 50 minutes is any different from one crying for 3 hours. The word "paroxysmal" was also dropped, because colic crying does not appear to sound any different from normal crying. The threshold remains in research; in real life the question "is it colic or not?" matters much less than it seems.

Something else matters a great deal: colic is a diagnosis of exclusion. A clinical review in American Family Physician says so explicitly: you only get there after a careful clinical history and examination have ruled out worrying causes. That is why the first step is the paediatrician, not the first professional who answers you on WhatsApp.

How much a normal newborn cries

This is probably the most useful part of the article, and nobody tells you it.

A systematic review with meta-analysis published in the Journal of Pediatrics pooled 28 diary studies, 33 samples, 8,690 babies, to answer a simple question: how much does a newborn actually cry?

Age Crying and fussing (average, 24 hours) How many fall into the "colic" definition
1-6 weeks 117-133 minutes 17-25%
8-9 weeks not reported 11%
10-12 weeks 68 minutes 0.6%

Read it calmly. Two hours a day of crying and fussing in the first month and a half is the average: not the difficult case, the average. The same analysis also dismantled the famous peak at 6 weeks, for which it found no statistical evidence: crying stays stable for six weeks and then falls.

Then look at the last column: 0.6% at three months. Colic ends, practically always, on its own. Keep this in mind, because it is the key to everything that follows: any treatment started in the eighth week will look like it works.

Una madre sorridente culla il suo neonato in un accogliente soggiorno, mostrando amore e connessione familiare.

Manual therapies and osteopathy: what the reviews really say

Now to what you are being offered.

The Cochrane review, read to the end

The Cochrane review on manipulative therapies for colic (chiropractic, osteopathy, cranial manipulation) included 6 randomised trials covering 325 babies. Pooling the five studies that measured hours of crying, the result looks excellent: 1 hour and 12 minutes less crying a day (mean difference -1.20; 95% CI: -1.89 to -0.51).

If the article stopped there, we would tell you to book. But the authors did the right thing: they recalculated, keeping only the studies in which the parents did not know whether the baby had received the real treatment. And there the result is no longer statistically significant: -0.57 hours (95% CI: -2.24 to 1.09), an interval that includes both a large improvement and a worsening.

This is not a technical detail; it is everything. In these studies the parent keeping the crying diary is the parent: the same person who brought the baby, who hoped, who paid. Only 2 of the 6 studies were at low risk of this bias. And for "complete recovery", the three available studies give an odds ratio of 11.12 with a confidence interval from 0.46 to 267.52: an interval this wide means we know nothing.

The authors' conclusion is worded plainly: small studies prone to bias, which makes it impossible to reach a definitive conclusion.

Later reviews say the same

A systematic review with meta-analysis in PLoS One assessed spinal manual therapy in infants, children and adolescents: 26 studies, GRADE method. Verdict: very low quality of evidence, to the point that the authors state they cannot conclude whether gentle mobilisations reduce symptoms in infants with colic or torticollis.

On safety it gives two pieces of information to hold together: low-velocity gentle mobilisations appear safe in infants, but there are 5 reports of serious harm after high-velocity, low-amplitude manipulation (4 infants and one child), probably linked, the authors observe, to an underlying condition that had not been recognised. The real risk is not the hand: it is skipping the visit that would have found something else.

There is also the most favourable review of all, in BMJ Open: 32 reviews and guidelines, written by researchers at an osteopathy college. That is precisely what makes it interesting, because even they rate the evidence for manual therapies as moderate to low (33-76 minutes less crying in 24 hours), and they give the highest rating to probiotics in breastfed babies. The three national guidelines examined all unanimously recommended only one thing: information, reassurance for parents, and clinical assessment of mother and baby.

And the randomised trial from 2025?

A multicentre randomised trial of 103 infants with colic according to the Rome IV criteria was published recently: 52 received three osteopathic treatments a week apart, and 51 waited three weeks and were treated afterwards. The results are striking: on parental psychological stress, the main outcome, the difference between groups reaches 3.5 points on a 0 to 10 scale (95% CI: 2.6 to 4.4), and crying improves in a similar way.

But look at the design: the control group received nothing, there was no sham treatment, the parents knew which group they were in, and the main measure was how stressed they felt. It is the same flaw the Cochrane review had asked to be corrected twelve years earlier. That study says something true about how much it helps to be taken seriously for three weeks; it says little about what the manipulation itself does.

The things that have some numbers behind them

  • Probiotics. The only item with good-quality evidence, and only for some babies. A meta-analysis of individual data from 4 double-blind studies (345 babies) found that Lactobacillus reuteri DSM 17938 gave 25.4 fewer minutes of crying a day at 21 days (95% CI: -47.3 to -3.5) and almost double the chance of success, but only in breastfed babies, with a non-significant effect in those fed formula. To be honest, several authors declare financial ties with milk and probiotic companies. It is a question for the paediatrician, not something to decide from a blog.
  • Parent training. Here is the largest number in the article, and almost nobody cites it. The Cochrane review of programmes aimed at parents (soothing techniques, advice on sleep and feeding, stress reduction) found, across 3 studies and 157 babies, 113 fewer minutes of crying a day (95% CI: -144 to -83). Same caveat: low certainty, small studies, same blinding problem. But it is more than one and a half times the 72 minutes the Cochrane review estimated for manual therapies.
  • Diet. The Cochrane review on dietary modifications (15 randomised trials, 1,121 babies) rated the evidence as very low quality and did not feel able to recommend any intervention. Put simply: do not cut out dairy on your own initiative if you are breastfeeding, and do not change formula on your own.
  • Lactase. A 2024 review of 6 randomised trials and 394 babies: three find a benefit, three do not. It is a picture of data that does not exist.
  • Simethicone and acid suppressants. For simethicone, the BMJ Open review reports moderate-to-low evidence of no benefit or a negative effect, and no support for proton pump inhibitors. The American clinical review is more direct: ineffective.

Position and containment: the practical part

Here we enter the territory we know best, with the same rule: distinguish what is proven from what is merely reasonable.

Carry them more. A historic randomised trial from 1986, of 99 mother-baby pairs: increasing the time spent being held throughout the day (not only in response to crying) reduced crying and fussing by 43% at six weeks (1.23 against 2.16 hours a day) and by 51% in the evening hours. Lovely. Then the same group tested it in babies with colic, in 1991: 66 mothers, a real increase in holding time of 2.2 hours a day, resulting in 3 fewer minutes of crying a day (95% CI: from 37 minutes less to 32 more). In other words, nothing.

The honest reading is this: carrying reduces normal crying, not colic. It is still worth doing, because it cuts the other two hours and lets you get through the evening on your feet instead of watching the clock. If you use a wrap or carrier, wear it well, for the baby's hips and for your back.

Swaddling. A systematic review in Pediatrics found that swaddled babies woke less and slept more, and that those with excessive crying cried less when swaddled than when massaged. The same review lists the risks, and this is what you need to know before trying it:

  1. the legs must not be held straight and tight: that position is associated with a higher risk of hip dysplasia, and they must be able to stay bent and apart;
  2. stop as soon as the baby tries to roll, because being swaddled face down increases the risk of death in the cot: always sleep the baby on their back;
  3. do not overwrap: if swaddled badly the baby can overheat, and there is some evidence that the tighter the swaddle, the higher the risk of respiratory infections.

And not everyone agrees: the 2015 clinical review places swaddling among the things not supported by the evidence for colic.

White noise. A small randomised trial of 40 one-month-old babies: white noise reduced crying and lengthened sleep more than rocking. Forty babies is few: it is a clue, not proof. But it is free and does no harm.

Positions to try tonight, without promises: upright against your chest, their head level with your neck; lying along your forearm, face down, head near the elbow (while held, never for sleep); curled on one side, arms brought in towards the middle and knees bent. The common thread is containment, not the manoeuvre: arms gathered in, legs flexed, broad and steady contact. And one thing at a time, for a few minutes: if in ten minutes you try six, you will never know what calmed them.

When it is not colic

Colic is a diagnosis of exclusion: it means someone has to rule out everything else first. Call the paediatrician, without waiting until tomorrow, if the crying comes with fever, repeated vomiting or green vomit, blood in the stool or vomit, refusing feeds, weight that is not going up, a baby who is limp or hard to wake, a swelling in the groin or scrotum, or if the crying changes character and becomes high-pitched or whimpering.

The same goes if the intense crying starts after 4-5 months or has not stopped beyond 4 months: outside the typical window, the "colic" explanation holds up much less well.

And this brings back the earlier point: the few serious harms described after manipulation were probably linked to an unrecognised condition. That is the strongest argument for the paediatric check before any manual treatment, not after.

The part that is rarely discussed: you

Across all this literature one thing is constant: guidelines unanimously recommend information and reassurance for parents, and the biggest number of all (the 113 minutes) comes from programmes aimed at parents, not from therapies applied to the baby.

Then something to say directly, even if it is hard to read at three in the morning: a baby must never be shaken. This is not a judgement on you: inconsolable crying is the typical trigger, and it happens to ordinary, exhausted parents who love their child. If you feel you are about to lose control, put the baby down in the cot, on their back, and leave the room for five minutes: crying alone for five minutes does them no harm.

Here, too, the honest data: a meta-analysis of educational programmes to prevent abusive head trauma found no convincing evidence that they reduce cases, and only weak evidence on stepping away from a crying baby. Knowing is not enough: you need a concrete plan and someone who can take over for you. Ask for that before you reach your limit, not after.

What a physiotherapist can do (and what they cannot)

Let's say it so there is no confusion: physiotherapy does not cure colic, and no study allows us to claim otherwise. We do not take bookings for colic.

There is, however, a nearby situation where we can be useful: when something motor and asymmetric appears alongside the crying. The head always turned to the same side or tilted towards one shoulder; latching that only works on one side; an area of the skull that is flattening. That is not the colic chapter, it is the chapter on congenital torticollis in newborns, where timing matters and the evidence is far stronger than what you have read above.

If that is your situation, with Ri-Hub Junior you start with a free 20-minute introductory consultation by video call with a physiotherapist on the Italian professional register who works with children and teenagers: you show us the baby live, we watch how they move, and we tell you frankly whether there is anything to follow up. For children up to 10 years old, sessions last 50 minutes and cost 59,90 €. If it is only colic, we will say so and refer you to the paediatrician: we are not interested in selling sessions that are not needed.

And if there is a complex neurological condition behind it, the right place is not us: those pathways require in-person assessment and specialist centres.

Massage is one of the first things recommended: what the evidence really says is in Massage for newborns.

If the crying does not follow the typical colic pattern and you are wondering what else might be behind it, we have gathered that in Newborn who cries all the time.

If the baby arches backwards during feeds, that deserves separate discussion: Newborn arching backwards.

And if reflux is the suspicion, be careful with what is said online about sleeping positions: Infant reflux.

Sources

Last source check: September 2026.

Domande frequenti

Does osteopathy work for infant colic?

The evidence is weak, and it is worth knowing why. The Cochrane review on manipulative therapies for colic pooled 6 studies and 325 babies. In the five that measured hours of crying, the reduction was 1 hour and 12 minutes a day (mean difference -1.20; 95% CI: -1.89 to -0.51). But when the authors kept only the studies in which the parents did not know whether the baby had been treated, the result was no longer statistically significant (-0.57; 95% CI: -2.24 to 1.09). The authors' conclusion is worded plainly: it is impossible to reach a definitive conclusion. That does not mean the person offering it to you is deceiving you. It means the benefit measured disappears when the people measuring do not know who was treated.

How much does a normal newborn cry?

A lot more than you might expect. A meta-analysis of 28 diary studies with 8,690 babies measured an average of 117-133 minutes a day of crying and fussing in the first 6 weeks, falling to an average of 68 minutes at 10-12 weeks. Two hours a day of crying in the first month and a half is the average: not an exception. And the same analysis found no statistical evidence of the famous 'peak at 6 weeks' that supposedly applies to every baby.

When do colics end?

Almost always on their own, usually by 3-4 months. In the diary meta-analysis, the share of babies who met the definition of colic was 17-25% in the first 6 weeks, 11% at 8-9 weeks and 0.6% at 10-12 weeks. A clinical review from the United States places resolution between 3 and 6 months. It is the only thing in this whole article that can be said with some confidence, and it is also the reason any treatment started at week eight seems to work.

Do probiotics really help?

They have the best evidence, but only for some babies, and the decision belongs to the paediatrician. A meta-analysis that re-analysed the individual data of 4 double-blind studies with 345 babies found, with Lactobacillus reuteri DSM 17938, 25.4 fewer minutes of crying a day at 21 days (95% CI: -47.3 to -3.5) and roughly double the chance of success. The effect was clear in breastfed babies (NNT 2.6) and not significant in those fed formula. To be honest, note that several authors declare financial ties with companies in the sector.

Does swaddling help with colic?

It can calm, but it has to be done properly, and not all experts agree. A systematic review in Pediatrics found that swaddled babies woke less and slept more, and that babies with excessive crying cried less when swaddled than when massaged. The same review also lists the risks: hip dysplasia if the legs are swaddled straight and tight, overheating if it is overdone, and above all an increased risk of death in the cot if a swaddled baby ends up face down, so swaddling stops as soon as the baby tries to roll. A 2015 clinical review, by contrast, does not consider swaddling supported by the evidence for colic.

Do tummy massage and bicycle legs work?

They are harmless and give you something to do, but do not expect them to fix it. A 2015 clinical review lists infant massage among the things the evidence does not support for colic, alongside chiropractic and osteopathic manipulation, acupuncture and herbal supplements. If your baby settles while you do it, carry on: you are sharing skin-to-skin contact and breath against breath, and that counts. But it is not a therapy, and there is no gas to 'release'.

Can a physiotherapist treat my baby's colic?

No, and anyone who tells you otherwise is going beyond what the studies allow us to say. Colic is a paediatric matter: the diagnosis is made by the paediatrician, who rules out other causes before calling it colic. Physiotherapy in childhood makes sense for something else: if alongside the crying you notice that the baby always keeps the head turned to the same side, latches well on one breast only, or that an area of the skull is flattening, then there is something motor to look at, and that is where we can help.

Dal singolo esercizio al percorso

Questi esercizi sono generici. Il tuo no.

Quelli che hai appena letto valgono per tutti, e per questo non valgono per nessuno in particolare. Un fisioterapista iscritto all'Albo guarda il tuo caso nel primo colloquio — gratuito, venti minuti — e da lì costruisce il programma su di te: da casa quando basta, in uno studio vicino a te quando servono le mani. Il dolore che torna non toglie solo il movimento: toglie il sonno, la pazienza, la voglia di uscire di casa. Rimetterti in movimento è il nostro mestiere.

Trova il tuo fisioterapista →

Hai solo una domanda? Scrivila su WhatsApp →

Fino a 10 anni 50 minuti a 59,90€ · dagli 11 ai 18 sedute da 29,90€ · nessun abbonamento

Ri-Hub ti mette in contatto con un fisioterapista online: primo colloquio gratuito in videochiamata con un professionista iscritto all'Albo, che ti segue per tutto il percorso. E se al tuo caso servono le mani, è chi ti segue a dirtelo e a mandarti da uno dei nostri, vicino a te. Se preferisci partire dalla tua zona, guarda il fisioterapista vicino a te, città per città. Se ti stai chiedendo quanto costa una seduta di fisioterapia, abbiamo confrontato studio privato, domicilio, ticket SSN e online.

#neonati#coliche#pianto#età evolutiva#osteopatia

Articoli correlati