New Infant Sleep Research: Have You Gotten the Whole Story?


Traditional media and social media feeds have been all a buzz this week, sharing the results of a new study published about infant sleep training. The results of the study claim that both sleep training methods tested “provide significant sleep benefits above control, yet convey no adverse stress responses or long term effects on parent-child attachment or child emotions and behaviour.”

Sounds intriguing, right? Sure does – even more so when you are an exhausted parent, whose infant wakes often through the night.

My primary focus in supporting families is informed, intuitive choices – using quality information and connecting with their own inner wisdom, to make the health and parenting choices that are right for their families.

In the spirit of being fully informed, here are some points to consider, that weren’t included in a lot of the media buzz about this recent infant sleep training study:

  • When an infant wakes at night, he has a need. Suggesting a parent’s response to that need is more reinforcing than sleep, creating what behavioural psychology has termed a ‘coercive behaviour trap’, undermines the parent’s instinct to fulfill baby’s needs – whether nutrition, soothing or reduction of discomfort.
  • There were only 43 infants in the study, divided into three test groups (14 or 15 infants per group). Over 90% were from two parent families, where the parents had higher educational qualifications and were middle to high income earners. This small sample size, with all infants of similar demographics, is not representative of the entire population and could not be representative of all possible responses to each type of intervention (whether sleep training or education).
  • While infants were identified as healthy, with typical infant development, it was the parents identifying that their child had a “sleep problem”. The level of parent knowledge around normal infant sleep development and its influencers and the potential underlying causes (beyond infant behaviour) for the “sleep problems” were not reported.
  • The study included infants age 6 months to 16 months (at the beginning of the study) – a wide range of developmental levels and needs.
  • The method of infant feeding, which can affect sleep patterns, was not reported for each infant. We don’t know whether the infants were receiving breast milk or a breast milk substitute and what their solid food intake included (healthy whole foods or processed convenience foods).
  • All of the babies slept in cribs, but we don’t know whether they were in a separate room from their parents or whether they were rooming in. Bed-sharing was excluded from the study due to safety concerns. Rooming in (baby & parents sleeping on separate surfaces, in the same room) can be classified as co-sleeping. And, co-sleeping, when done safely, has many benefits (babies & parents get more sleep, nighttime breastfeeding is easier, sleeping in the same room reduces the risk of SIDS, absence of nighttime separation anxiety) (More information about this here and here.)

The Results

The results reported that both sleep training methods studied provide significant sleep benefits over the control group (provided with education about sleep). Let’s take a look at the details:

(**Numbers are approximate as they are extrapolated from graphs in the published article, where actual values were not reported.)

Length of Time to Fall Asleep: At the start of the study, infants took an average of about 18 minutes to fall asleep. At the end of the 3 month treatment period, infants whose parents used one of the sleep training methods took 5-9 minutes to fall asleep, while the infants in the control group took 20 minutes to fall asleep. Is there a difference? Absolutely. But – is taking 20 minutes to fall asleep problematic? Absolutely not.

Group Before Study Onset After 3 Months of Treatment
Graduated Extinction
(AKA Controlled Crying)
(gradual delay in parent’s response)
17.5 minutes 5 minutes
Bedtime Fading
(gradually limiting opportunity to sleep)
17.5 minutes 9 minutes
Control
(Sleep Education)
18 minutes 20 minutes

Total Amount of Time Awake After Sleep Onset: Infants in all three groups experienced a reduction in the amount of time they were awake through the night. The largest drop was found in the Graduated Extinction group, but critics would question whether this is indicative of a positive outcome or whether the infants in that group were still waking, but not signaling their needs with crying because they’d learned that crying won’t bring help from a parent.

Group Before Study Onset After 3 Months of Treatment
Graduated Extinction
(AKA Controlled Crying)
(gradual delay in parent’s response)
58 minutes 12 minutes
Bedtime Fading
(gradually limiting opportunity to sleep)
44 minutes 20 minutes
Control
(Sleep Education)
60 minutes 29 minutes

Average Number of Awakenings: At the end of the 3 month treatment period, infants in the sleep training groups were waking once through the night, while the infants in the control group were waking twice, through the night. At the ages of infants in the study, waking 1-2 times per night and signalling for a parents help, can be completely normal, especially with the variety of sleep influencers that may be having an effect – teething, growth spurts and mental leaps, to name a few.

Group Before Study Onset After 3 Months of Treatment
Graduated Extinction
(AKA Controlled Crying)
(gradual delay in parent’s response)
2.75 1.4
Bedtime Fading
(gradually limiting opportunity to sleep)
2.0 1.0
Control
(Sleep Education)
2.6 2.1

Total Sleep (Hours): The average amount of sleep increased in all groups and the laregest increase was in the control group.

Group Before Study Onset After 3 Months of Treatment
Graduated Extinction
(AKA Controlled Crying)
(gradual delay in parent’s response)
9.95 increased by 0.32 hours
Bedtime Fading
(gradually limiting opportunity to sleep)
9.95 increased by 0.09 hours
Control
(Sleep Education)
9.75 increased by 0.36 hours

Infant Stress: Cortisol (a key stress hormone) levels were measured in the morning and afternoon before the study began and at all follow up times. While the results reported state that there were moderate to large declines in in the afternoon cortisol measurements of the infants in the sleep training group, the infants’ stress levels during sleep training were not measured and other sources of stress, beyond “sleep problems” were not accounted for, in any of the groups.

Maternal Stress: Measures of maternal stress and mood improved from pretreatment to the 12 month follow-up, for all groups. (Hooray for the Mamas!)

Child Attachment at 12-Month Follow-Up: This measure was included as an apporach to evaluating the effect of the sleep training methods on parent-infant attachment. The results reported that there were no significant differences found between secure and insecure attachment styles, between groups. The challenge to these results? There was no baseline measurement of attachment done before the study began, so an understanding of any shifts in attachment related to the sleep training is not available. In addition, secure attachments were exhibited in only 58.67% of the infants studied – an alarmingly low number that leaves me wondering what else has been happening with these families and how they’ve been supported.

The moral of the story? We can’t judge a book by its cover.

How studies are presented in the media, often gives only a sensationalized, surface look. When you encounter those that turn your stomach or put knots in your throat (like this one did for me!), trust those intuitive responses in your body and dig a little deeper – whether on your own or with the support of a trusted and knowledgeable health care provider.

If you are local to the Halifax area and would like to learn more about normal infant sleep – its development and influencers – join me for one of my upcoming Normal Infant Sleep: Myths & Realities sessions. (More information & a list of upcoming events are here.)

With love & gratitude,

Dr. Sarah