To read this article in Spanish, click here.

Lately there’s been a fair amount of talk surrounding Crying-It-Out (CIO).  After a piece was published on one woman’s admittance to doing CIO with her kids, others came forward to talk about how they too did it out of desperation.  A lack of sleep, no social support, and a screaming baby are usually enough to send most moms (or dads) over the edge.  We’ve all been taught that it’s okay to put a baby down and walk away when there’s a risk of harming your child (something I firmly believe in), but with this path of rationalization, I start to see people question the whole stance against CIO.  After all, if we’re not suggesting a mom who does it when stressed out and frazzled will have to reap the negative repercussions of it, why are we able to say it for the mom who does it intentionally?  It’s a good question and very tricky topic, one I hope I can navigate here without sticking my foot in mouth, but I make no guarantees, so please, just bear with me.

I want to first come out and say that CIO can have effects for your child, regardless of why you do it.  There’s no reason to hide this fact and it’s not stated out of a desire to make moms who are stressed feel worse, but because I do believe there is an alternative we need to approach in our society (which I will get to further down) and to get people on board requires accepting that our behaviours have consequences, even if they’re the lesser of two evils.   I’ve talked about the research in bits and pieces all over this site and I’ll attempt to give a summary here, along with some new research that’s in press as I type, ending with my own suggestion as to how to try and fix this.

1.  The Sleep Reviews (for a more detailed response, see Ask EP: Crying-It-Out)

Before I get into the evidence against CIO, I have to address the elephant in the room, which is that two reviews have been cited by sleep stores and trainers galore demonstrating no ill effects of sleep training.  The first, in Sleep Medicine Reviews, simply states “no evidence of deleterious effects was found in these studies”[1] (p.284).  So how many studies were cited as showing no deleterious effects?  Three.  And one of them is a case study of FOUR children – not really a study at all.  So two studies and one case study.  But what if two studies found no deleterious effects – that counts right?  You bet, if they actually used valid measures of child attachment or social-emotional health, but of course they didn’t.  Both studies[2][3] utilized parent-report measures of “attachment”, in quotes because one of them ([2]) utilized a scale that doesn’t even claim to measure attachment, but rather parental stress while other ([3]) modified a measure that is supposed to be administered by a trained interviewer (in order to be objective and delve deeper when necessary) into a self-report measure, in essence invalidating the measure.  Furthermore, neither study included long-term assessments of behaviour, only pre- and post-sleep training.  Finally, the youngest child in these studies was 16 months of age, a far cry from the 4 months that many parents are encouraged to start sleep training.

The second review, and probably more commonly cited, was in the journal Sleep[4] and included 52 studies while claiming to find no secondary effects for sleep training.  The first problem here is that not all 52 studies actually examined side effects so it’s not like 52 studies didn’t find deleterious effects, only the ones that did study it found no deleterious effects.  What are these potentially deleterious secondary effects examined?  They are: attachment status, predictability, irritability, and crying/fussiness.  Personally, predictability make no sense to me as secondary effect on the child’s well-being, but so be it.  However, the examination of attachment status is highly interesting, if it has a chance at being different from the previous review.  Again, though, only three articles were found that assessed attachment, not 52.  Unfortunately, they all suffered the same flaws as mentioned above (and one of them was the same study, [3], as cited above), including the use of the exact same self-report measure for attachment (that’s supposed to be administered by a trained professional).  Personally, I see none of this as providing much evidence of anything, except that sleep training helps parents.  However, that’s me, but even if you accept the results of the reviews, no research that is based on children in the 16+ month range (the age range for this research) is at all applicable to children who are younger, which is when the vast majority of CIO techniques are used.

2.  Neurological Evidence

I first covered the neurological stress response in My Baby Cries Too so you can read more there if you like.  The stress response is high and immediate to all stressors at birth (even minor ones), but slowly starts to attenuate over the first year[5].  It does not, however, completely diminish, and this attenuation seems to be related to strong social regulation or parental buffering of stress (in the form of responsiveness to distress), suggesting that if during this period parents do not provide a buffer against the activation of the HPA axis, the attenuation will be less[5][6].  Furthermore, reviews of the effects of stress on the developing brain of human infants has been consistent in demonstrating that maternal separation continues to cause stress responses in the form of cortisol increases even in toddlerhood[5], and as you know, the basis of CIO is the separation of mother (or father) and baby.

The long term effects of heightened activity in the HPA axis (or situations known to result in heightened activity of the HPA axis) include depression as early as adolescence, reduced empathy, and behavioural problems[7][8].  What causes this heightened activity?  It is not, as many sleep trainers would have you believe, due to “extreme” neglect and abuse.  Sadly, many of these studies were done with normal families where the mother suffered post-partum depression and thus children fell into a category of slight neglect.  In some cases, the stressor was simply when children had to separate from their parents at daycare.  (Notably, this raises the very good question of what is the cost of sleep deprivation on PPD and a mother’s functionality versus something like CIO which I will get to further on.)

But the fact remains that no matter how logical it is to place the results of these other studies on a practice like CIO, none of them actually examined the neurological reactions of infants to CIO routines.  So what is the specific neurological effect of CIO?  The aforementioned reviews said there were no deleterious effects and while I hope I’ve convinced you it’s a bit of a disingenuous statement given what the research actually was, it also turns out that it might not be the case.  In all of these previous sleep studies, no one had bothered to check the physiological stress responses of the infants, they simply made the assumption that if the infant stopped crying, the distress was gone.  Now, if you read Educating the Experts – Lesson One: Crying, you know that there are various reasons why an infant ceases to cry, many of which are unrelated to the infant actually being “fine”.  So I was thrilled to find out (from Eileen Joy over at Live With Purpose) that one researcher has actually started to do just this…

Wendy Middlemiss from the University of North Texas recently completed a study on infants’ stress responsivity to CIO or “extinction” programs[9].  She and her colleagues examined infants aged 4 months to 10 months who were in an in-patient sleep training program in New Zealand.  In the program, mothers and infants are separated at night and infants are left to CIO while nurses check on them every 10-15 minutes to ensure they are safe and swaddled.  At the start of the program, mother and infant cortisol levels were recorded and found to be highly synchronous.  What does this mean?  It means mom and baby were “in tune” with each other as their cortisol levels affects each other—as one’s went up or down, so did the other’s.  In fact, the correlation between mother and infant cortisol levels was a whopping r=.776 during the day and r=.748 at night (note that the maximum value of a correlation is 1).  This type of synchrony has been found to be associated with an infant’s ability to learn self-regulation and to develop emotionally[10], as well as being the foundation to secure attachment[11].  So despite mom being tired and run down, she was in a state of being able to form positive attachments with her child.  Notably, during the day in this program, moms and babies spent their time together, ensuring there were lots of positive interactions (sound familiar, sleep experts?).  However, by day three, the level of synchrony between mom and baby had decreased during the daytime to a correlation of r=.582 and at night to r=.422 (which was not significant, meaning it was not statistically different from 0).  This lowered synchrony may result in worse attachment, particularly from the infant’s point-of-view as the manner in which infants affect mom’s cortisol is by signaling (or crying) and if they have stopped because they feel mom won’t respond, it should come as no surprise that attachment should be affected.

Even more telling, though, was that while the infants had ceased crying while going to sleep, presumably having learned to “self-settle”, their cortisol levels continued to spike.  In fact, their cortisol levels were exactly the same as they were on day one when they were separated from their mother and everyone acknowledged the infant was in distress.  So while they stopped showing any outward display of distress, internally they were highly stressed.  This puts a wrench in the popular notion that the stress response for a baby undergoing CIO only lasts as long as the crying does.  As the authors’ state:

“Although infants exhibited no behavioral cue that they were experiencing distress at the transition to sleep, the infants continued to experience high levels of physiological distress, as reflected in their cortisol scores.”

It will be interesting to see more of Dr. Middlemiss’ research and hopefully more longitudinal studies going forward.  But at the very least, we should take note of her current findings.  After all, most sleep sites and trainers try to use cortisol spikes associated with sleep deprivation to justify CIO methods (even though infants don’t develop regular night-wake cortisol cycles for 6-9 months, sometimes later[12]) and now it seems that spike might happen during CIO training.

3.  Moms’ Mental Health versus Baby’s Brain Development

So now we get to the meaty issue – what about moms who have to leave their baby to CIO while they try to gain some sanity through sleep?  The moms suffering post-partum depression who are at risk of so much more if left with a child who doesn’t sleep.  First I want to address the issue of CIO for training purposes and for mental health purposes.  While I stated above that there are effects of our actions, I also want to add that there still is a distinction between these two forms of CIO.  Women who are trying to be as responsive as possible and thus losing sleep to wake with a baby every hour or two (or more) and utilize CIO every once in a while to get some sanity back are most likely more responsive in general to their children’s distress than parents who feel they need to train their children to behave as they would like.  The reason for CIO becomes a confound, and a notable one.  Yes, your child will show physiological distress to the CIO session, but a child whose parents feel he or she needs to be trained will most likely (but not always) show less responsiveness during non-sleep hours, compounding the distress and HPA-axis activation for the infant.  And after all, it is through repeated behaviours that the most lasting damage is done.

But why should we pit it as mom’s mental health versus a baby’s brain development?  Making it an either-or argument means someone has to lose.  Doesn’t that just suck?  And it doesn’t have to be that way (in the long run).  This is where I’m going to go on again about community and the role us moms can play in helping other moms out.  We need to find ways to be there for new moms so that mothers don’t have to make this awful choice.  Personally I find it ridiculous that the only people you meet when going through pregnancy are other new moms.  Yes, you can have a little support group which is nice, but really you can’t do much besides vent together and offer social support.  And don’t get me wrong – social support is essential, but not as essential as instrumental support[13].  Mothers need someone there to help with the little things and perhaps even give mom a chance a sleep for a couple hours uninterrupted (because even if babe cries, an infant being held during a distressing time does not show the level of HPA-axis activation, if they show any activation at all [5][6]), and who better to do this than other moms?  We know what it’s like – we’ve been there and have the empathy necessary to help new moms handle this transition.

Yes, there are post-partum and night doulas for new moms, and they are wonderful in myriad ways, but they’re also expensive.  Not every new family can afford to have that person come in and help (though if someone also requires breastfeeding help or other forms of specific advice, I would highly recommend it).  Some might argue that it should be a government program, like they have in France, in which a nurse comes to do home visits and stays for a bit to help new moms out.  Sure, it’s great, but it’s also expensive (in the form of taxes) and it still doesn’t build the bonds between individual that can come if acts like this are done out of kindness and friendship.  Frankly I would love to see a charity that just matched new moms with “old” moms who were willing to serve as friend and mentor.  Yes, it would involve watching another baby sometimes, maybe even bringing a meal now and again, but I don’t know that there’s much more rewarding than helping another mom out and knowing you can be contributing to them being a happy dyad.  And I’d like to think you’d have a friend for life from it and perhaps help that mom do the same for another mom down the line.  In the meantime, I would suggest that if you have it in you, help a new mom out.  Offer to hold her baby for a couple hours if she’s sleep-deprived and let her nap (if you’re already friends, this will probably be an easier sell), bring her a meal once in a while, even offer to help with laundry or grocery shopping if you can.  I also firmly believe that if we cut out the idea that moms need to be isolated from everyone else and kill the ridiculous expectations that new moms are supposed to “do it all”, we would find that we could help people out enough to stop forcing moms to choose between their own sanity and the well-being of their infants.



[1] Owens JL, France KG, Wiggs L.  Behavioural and cognitive-behavioural interventions for sleep disorders in infants and children: a review.  Sleep Medicine Reviews 1999; 3: 281-302.

[2] Reid MJ, Walter AL, O’Leary SG.  Treatment of young children’s bedtime refusal and nighttime wakings: a comparison of “standard” and graduated ignoring procedures.  Journal of Abnormal Child Psychology 1999; 27: 5-16.

[3] France KG.  Behavior characteristics and security in sleep-disturbed infants treated with extinction.  Journal of Pediatric Psychology 1992; 17: 467-475.

[4] Mindell JA, Kuhn B, Lewin DS, Meltzer LJ, Sadeh A.  Behavioral treatment of bedtime problems and night wakings in infants and young children.  Sleep 2006; 29: 1263-1276.

[5] Lupien SJ, McEwan BS, Gunnar MR, Heim C.  Effects of stress throughout the lifespan on the brain, behavior, and cognition.  Nature Reviews 2009; 10: 434-445.

[6] Gunnar, M. R. (2006). Social regulation of stress in early childhood. In K. McCartney & D. Phillips (Eds.), Blackwell Handbook of Early Childhood Development (pp. 106-125). Malden: Blackwell Publishing.

[7] Halligan SL, Herbert J, Goodyer I, Murray L.  Disturbances in early morning cortisol secretion in association with maternal postnatal depression predict subsequent depressive symptomology in adolescents.  Biological Psychiatry 2007; 62: 40-46.

[8] Jones NA, Field T, Davalos M.  Right frontal EEG asymmetry and lack of empathy in preschool children of depressed mothers.  Child Psychiatry Hum Dev 2000; 30: 189-204.

[9] Middlemiss W, Granger DA, Goldberg WA, Nathans L.  Asynchrony of mother-infant hypothalamic-pituitary-adrenal axis activity following extinction of infant crying responses induced during the transition to sleep.  Early Human Development in press.

[10] Feldman R.  From biological rhythms to social rhythms: physiological precursors of mother-infant synchrony.  Developmental Psychology 2006; 42: 175-188.

[11] Feldman R.  Parent-infant synchrony and the construction of shared timing: physiological precursors, developmental outcomes, and risk conditions.  Journal of Child Psychology and Psychiatry 2007; 49: 329-354.

[12] de Weerth C, Zijl RH, Buitelaar JK. Development of cortisol circadian rhythm in infancy. Early Human Development 2003;7: 39-52.

[13] Lyons-Ruth K, Connell DB, Grunebaum HU, Botein S.  Infants at social risk: maternal depression and family support services as mediators of infant development and security of attachment.  Child Development 1990; 61: 85-98.