The study opens with this paragraph:
“Behavioral techniques effectively reduce infant sleep problems and associated maternal depression in the short- to medium-term (4–16 months’ postintervention). Despite their effectiveness, theoretical concerns persist about long-term harm on children’s emotional development, stress regulation, mental health, and the child-parent relationship. “
Behavioral sleep techniques did not cause long-lasting harms or benefits to child, child-parent, or maternal outcomes. Parents and health professionals can feel comfortable about using these techniques to reduce the population burden of infant sleep problems and maternal depression.”
Potential Methodology Issues Not addressed in the Study:
- Sleep problems were ‘parent reported’, at 7mths a questionnaire was given with the question “Over the last 2 weeks has your child’s sleep generally been a problem for you?” 47% of respondents answered ‘yes’ and thus became eligible for trial inclusion. The ‘problem’ was not elaborated upon and I would have liked to have seen more questioning to ascertain exactly what the problems were, what had been tried until that point, what information the parents had received and what support had they received up until that point in those first important 7months as this time period could have a big effect.
- The sample (out of those reported sleep problems) was selected by the researchers, I can find no mention of how they were selected and what other criteria was looked into in terms of the infant’s/family history/psychological wellbeing until that point.
- Parents must have been open to the idea of sleep training to agree enrollment in the study, would this therefore mean that their opinions/beliefs were reflected in other parenting practices?
- Training was carried out on babies 8-10months of age so cannot be applied to any infants younger than this. It also does not tell us the effects of training on older children/toddlers.
- Parents were able to choose the type of sleep training used – either gradual extinction (what you and I know as controlled crying) or gradual withdrawal (parent starting out sitting with child and moving further away but involving no crying) – though no distinction was made between the type of sleep training used and later impact in the results.
- Control/Intervention group allocation was blinded only to researchers not parents (understandably it would be tricky to be blinded to parents – but I wonder how knowledge that they were in the intervention group affected parental perception).
- The control group visited the same nurses and were free to ask for sleep advice, but the nurses in these control groups were not trained to offer specific sleep training advice, however what advice did they give? Pretty much any parent I know could give controlled crying instructions without special training, do we know that they didn’t give similar advice to the intervention group? I cannot find this information out and to me from this point in the study loses all credibility for me – Do we know what the control group did sleep wise? We know they reported that they had had a problem with their child’s sleep in order to enter the study, so it’s pretty likely that they would want to do *something* – were they followed up and questioned and asked exactly what they did do? Again I can find no mention – How do we know that a large majority of the control group DIDN’T sleep train?
- Outcomes were measure by cortisol samples, taken at 6yrs of age (why on earth would they take them at 6year of age?! I find this very confusing and not at all relevant? From what I can see the first follow up – and potential cortisol testing was at 10 months, yet the concerns over stress to infants from sleep training is during and in the immediate aftermath of the training, a sample taken 2 months up to 5 years later seems bizarre? Where is the cortisol sample DURING and IMMEDIATELY after the training? This is the one that matters IMO).
- Child emotional questionnaires and ‘quality of life’ questionnaires were parent reported (and occasionally child reported) – meaning that parents filled in the questionnaires giving their own opinions. We know questionnaire reporting is notoriously unreliable with two main factors – 1) wanting to tell the investigator what they want to hear and 2) sticking to ‘middle of the road’ answers, i will add in 3) here – what parent will want to report that they feel their child has an emotional/behavioural difficulty? Especially not when they have been enrolled into a trial looking into the effects their early parenting may have had upon these. I find it very hard to determine whether the measures of psychological wellbeing were all parent reported as the results are very vague, but if I were to make an assumption I would guess that the majority were parent reported. Child reporting surely would include bias – what child would say negative things about their parents in front of said parents?
- In terms of parenting style (and thus I presume eluding to bonding too?) the researchers appear to feel that “authoritative parenting” (high warmth, high control) is optimal parenting whereas what they call permissive parenting (high warmth, low control) is sub-optimal. I would argue that having “high control” over children is NOT optimal parenting, or indeed respectful parenting and am not convinced at all that highly controlling parents are those exhibiting the most healthy parenting style. I would like to see how they define “high and low control” and look to see how many initial “permissive parents” morphed into “authoritative parents” from the intervention group as a result of the training undertaken, this information though is not provided.
- Nearly a third of the sample were lost to follow up (31%) – that’s a BIG number and a major limitation – what if they didn’t agree/respond to follow up because they found the intervention traumatic/it didn’t work for them/it went against their instinct?
Perhaps the most interesting part of this research is this:
“There was no evidence that a population based targeted intervention that effectively reduced parent-reported sleep problems and maternal depression during infancy had long-lasting harmful or beneficial effects on child, childparent, or maternal outcomes by 6 years of age. Thus, this trial indicates that behavioral techniques are safe to use in the long-term to at least 5 years.”
Read that again, I’m pretty certain I dispute their claim that it has no long lasting harmful effects given the parental reporting, strange timing of cortisol testing, lack of information on what techniques the control group used, promotion of authoritative/controlling parenting as the optimum type, lack of information of life before 7 months of age (or in fact not much about life afterwards aside from financial questioning – what about childcare for a start) and lack of in-depth information about bonding (why no Ainsworth ‘Strange Situation’ type testing?). No the most interesting part to me is even with all of the above limitations in mind this research tells us there are NO LASTING BENEFITS to sleep training……Now which paper picked up on that then?! No, thought not.
So what DOES the research tell us?
- There is still NO evidence that shows controlled crying under 8 months of age has no ill effects
- The results did NOT differentiate between the different types of sleep training used (e.g: No Cry V Crying related options) in the results, therefore we do not know about the individual methods and their outcomes, only ‘sleep training’ as a broad label of many different types of training.
- That sleep training does NOT have lasting positive effects on a child’s sleep behaviour
- That nearly 50% of parents still have problems with their baby’s sleep by 7months of age (hey, perhaps that’s because it is NORMAL infant sleep and our expectations are incorrect)
- That 31% of parents for some reason did not agree to follow up with the researchers
- That parents tell researchers that they did not feel that sleep training affected their child negatively (tell me something new).
- That controlled crying DOES work in the short term (I don’t think this has ever been disputed?).
- That concerns over the effect of controlled crying on babies are still very valid, particularly in light of the recent Middlemiss study (that measured cortisol levels DURING training, not 5yrs later!) – A good summary of the concerns of sleep training can be read HERE.
- That a whole lot more health professionals and ‘baby experts’ are going to use this as arsenal to tell parents that there are no concerns with sleep training involving baby crying, that it’s a good thing to do, even if it feels wrong to them, as parents, to do it.
Sarah (Founder of BabyCalm)
You can read more of Sarah’s articles HERE.
Price. A, Wake. M, Ukoumunne. O and Hiscock. H. ‘Five-Year Follow-up of Harms and Benefits of Behavioral Infant Sleep Intervention: Randomized Trial’ Pediatrics; September 10, 2012;