The 8 month sleep regression: crawling, clinginess and the nap that won't happen

It’s twenty past four. She skipped the third nap again, or rather she lay in the cot for forty minutes practising sitting up, then cried until you got her. Now she’s pulling herself up on the sofa for the fifth time, rubbing her eyes, and you’re doing the maths: try the nap and accept bedtime sliding to half past eight, or push through and hope.

In short: around 7 to 10 months several things arrive together, and any of them can disturb sleep: big motor skills that get practised in the cot, the first real separation anxiety, teething for some, and a drop from three naps to two that often lands in the same weeks. We think a lot of what gets called the 8 month sleep regression is that last one: a schedule that no longer fits. The useful moves, we think, are daytime practice for the new skill, a settled way of saying goodnight, and holding bedtime steady when the third nap collapses. Many babies barely notice any of it.

This article is general information about a common phase, not medical advice. If your baby has a fever, seems in pain, is refusing feeds, or anything about them worries you, talk to your doctor or health nurse.

Why 7 to 10 months is a crowded few months for sleep

This is the stretch when many babies get moving: sitting, some form of crawling, pulling up on whatever’s nearest. They practise relentlessly, and the cot is a fine place to practise. A baby who has just worked out how to stand may stand, including at 3am, and then discover she doesn’t yet know how to sit back down.

At the same time, something changes in how she understands you. The NHS describes separation anxiety as common from about 6 months, a sign the baby now realises they are a separate person who depends on a small group of carers.¹ The American Academy of Pediatrics puts the sharper phase at around 9 months, with some babies showing it as early as 4 or 5.² For a baby in this phase, being put down and left is now a different experience from a month ago. She knows you’ve gone, and she minds. Add teething for some babies, usually mild and short (more on that below). The quieter shift underneath is the schedule: for a baby still on three naps, the third is often on its way out, and we think that alone can break the night.

None of this means night waking is new. A large review of infant sleep studies found babies of 7 to 11 months waking on average about once a night, with the normal range running from none to about three; the pooled average for that band was actually slightly higher than for 3 to 6 months, though the authors are clear that night waking was the most variable thing they measured.³ An 8 month old waking at night is ordinary. What parents notice as a regression is a baby who had settled starting to wake again, or a bedtime that used to take ten minutes taking an hour. The same cluster arriving a few weeks later gets called the 9 month sleep regression; we don’t think of them as two events.

Is it the regression, or a nap schedule that no longer fits?

In how we model it, a baby of 7 to 9 months can comfortably stay awake for about two and a half to three and a half hours between sleeps, and fits two naps into the day rather than three; from 10 months the windows lengthen again. The observational data points the same way: across that review, the average number of daytime naps fell from about three in the first six months to about two in the second half of the first year.³ The review can’t say when in that half-year the third nap goes, or for which babies. But its averages and our own tables agree that by 7 months two naps is the ordinary shape of the day, and our model makes the switch at about six months. We think that’s why so many “regression” days look like a schedule problem.

The third nap often doesn’t announce it’s leaving; it just gets harder. The next sleep isn’t timed from the last wake alone: how much she has slept today and last night counts too, and in our model a short night shortens the next day’s windows. Here’s one made-up baby’s day, worked through with that arithmetic; your times will differ, but the shape is easy to fall into:

1. Say she wakes at 6:30. Her naps at 9:00 and 12:30 go roughly as usual.
2. Her third nap, usually at 15:30, gets refused, then finally happens at 16:30 for forty minutes.
3. Now she’s awake at 17:10, and with her window of around three hours she’s not tired until after 20:00. Bedtime lands at 20:30.
4. The night is short and broken. She’s up at 5:30, overtired, and after a short night her windows shrink, so she goes down for the first nap at 8:00.
5. The early first nap drags everything forward, and the third nap “fits” again. Repeat.

At this point these parents are reading about the 8 month sleep regression, and what they have is a day with one nap too many. The night waking, the early rising and the nap fights can all be downstream of a schedule built for a younger baby.

The alternative day for the same baby: wake 6:30, nap around 9:15, a second nap around 13:30 that ends by about 15:00, then a long final window to a bedtime between about 18:30 and 19:00. Our model holds bedtime at this age to 20:00 at the latest by default, even when the afternoon has gone sideways (those numbers are our own age tables, not a clinical source), on the view that an early bedtime after a rough day costs less than a late one. The first week of two naps is often wobbly, and an 8 month old fighting naps in that week isn’t proof the switch was wrong. What you’re looking for is whether bedtime gets easier and the night gets longer.

If your baby is already on two naps and sleeping badly, this isn’t the explanation; the rest of the article is for you.

8 month old separation anxiety: sleep is where you notice it

Neither the NHS nor the AAP page on separation anxiety says anything about sleep, and we won’t put words in their mouths.¹² But the logic of what they describe points straight at bedtime. If the new thing your baby has learned is that you still exist when you’re out of sight, then being put down and left is the purest version of the thing she’s just started to mind.

So a baby who went down happily at seven months may now cry when you leave, or wake at 1am and not settle until she’s checked you’re still there. It’s the same story as the crawling: a new understanding of the world, worked through partly at night.

For daytime goodbyes the AAP’s advice is: quick, the same every time, no lingering.² We think that translates almost directly to the cot. The NHS suggests letting your baby hear your voice from the next room, and playing peek-a-boo in the day as a way of showing that you always come back.¹ We think a short goodnight she can predict does more here than extra soothing at the moment of leaving.

Teething gets blamed for a lot

Every 3am at this age gets attributed to a tooth by someone. The research is more modest.

The most careful study we found followed 125 babies with daily parent diaries from 4 to 12 months, covering 475 teeth. The authors looked for symptoms in an eight-day window around each tooth, four days before to three days after. Within it, wakefulness, irritability, drooling, gum-rubbing, ear-rubbing and a slight rise in temperature were genuinely more common. But “sleep disturbance” as parents recorded it was not, nor was a fever over 39°C; no single symptom occurred in more than about a third of teething babies, and no pattern of symptoms predicted a tooth.⁴

So: teething is real, mild, brief, and does seem to make some babies a bit more wakeful for a few days. What it doesn’t do is explain three weeks of broken nights. If the bad sleep outlasts the tooth, look at the schedule and the new skills first. The authors’ own caution is worth repeating: a proper fever or real distress should be checked for other causes rather than filed under teeth.⁴

What we think helps: daytime practice, a predictable goodbye, an earlier bedtime

Nothing here is a method; it’s what we think follows from the sources above and our own tables.

Give the new skill daytime hours. A baby who spends the afternoon crawling and pulling up on the sofa has less unfinished business at 3am. If she stands in the cot and can’t get down, practise that specific move, sitting from standing, during the day. (Cot setup as your baby gets mobile is a safety question for your doctor or health nurse.)

Say goodnight the same way every night. Short, warm, predictable, and then leave. Bath, pyjamas, one song, a sentence you always say, out. That’s the AAP’s daytime-goodbye advice² applied to bedtime.

Hold bedtime when the third nap goes. When the third nap is refused or collapses to twenty minutes late in the afternoon, the temptation is to let bedtime slide late. Do the opposite. Bring bedtime earlier for a few days (for the baby in the example, 18:30 to 19:00), take the long final window on the chin, and let the night absorb the tiredness.

Be patient with the clinginess. It passes on its own. Extra cuddles in the day are not a setback, we think, and the NHS is direct that leaving your baby with another carer will not damage them.¹ Writing the days down for a week, on paper or wherever, can show whether it’s the three-nap cascade or something else.

Why the extra nap and the 3am fix usually backfire

Reintroducing a third nap to fix a bad night. After a rough night the obvious move is to add daytime sleep back in. For a baby who has just moved to two naps, that can recreate the cascade above. An early bedtime does the same job without pushing bedtime late the way the example does.

A new way of going to sleep, invented at 3am. If you offer a feed, a rock or a lap of the flat at 3am, assume you may still be offering it next month. At 3am itself, do the boring thing you’d be happy doing again tomorrow, and fix the day in the morning.

Changing everything at once. If the bedtime, the routine and the nap count all change in the same week, you can’t tell which one helped. Schedule first, then the rest.

When the regression isn’t the explanation

The limits of all this:

• Many babies sail through. Plenty of 8 month olds learn to crawl, get clingy, cut two teeth and sleep exactly as before. If your nights are fine, you may not need this at all.
• If nights were already broken, you may not see a step change. If your baby has woken two or three times a night since birth, this window may just look like more of the same. The schedule advice still applies.
• If your baby is unwell, that comes first. Fever, pulling at an ear with real distress, refusing feeds, a rash that concerns you, or a baby who is simply not herself: these are a doctor’s conversation, not a sleep-phase one, and not a teething one either.⁴
• If separation anxiety is severe or isn’t easing. The NHS suggests talking to your health nurse if the distress is severe, prolonged, or goes on for more than a few weeks.¹
• If the three-nap cascade isn’t your pattern. Some babies still need three short naps at 8 months, and forcing two on them can produce an overtired mess by late afternoon. If three naps fit comfortably and bedtime is on time, keep them.

See the nap that broke the night, before bedtime pays for it

Most of this comes down to seeing the shape of the day: when the last nap ended, how much she’s slept today and last night, how long the window before bed was, whether tonight’s late bedtime is a one-off or the fourth this week. That’s what Tiny Sleeps is for. One tap when she goes down and one when she wakes, and a schedule for her age lays out the next nap and tonight’s bedtime from when she last woke and how much she’s slept today and last night, moving as the day goes and holding bedtime to a sensible latest time rather than chasing a collapsed third nap. Both parents see the same day when they open the app on their own phones, so whoever does bedtime can check what the afternoon looked like.

You can do all of this with a notebook. The app just makes it something you glance at rather than reconstruct.

Download Tiny Sleeps on the App Store for iPhone.

Related

• The 4 month sleep regression, explained
• Baby sleep tracker for two parents: how to hand over without the group chat
• Setting up a baby tracker for grandparents and nannies

Sources