Why the headline "It's OK to let your baby cry himself to sleep" doesn't help parents with baby-sleep problems
The headlines read: “It's OK to let your baby cry himself to sleep,” i and the reports are accompanied by images of babies in the first months of life.
Although a small study like this grabs attention, the media fuss is only likely to confuse parents who are struggling through the misery of baby sleep problems. This latest South Australian trial applies to infants aged between six and 16 months (average age about 11 months).1
The data is one thing; the way people interpret their research data is a different matter altogether. Parents know that there is a big difference between the needs of a six month old baby and a 16 month old toddler. Unfortunately, researchers regularly fail to distinguish between the developmental needs of infants at different ages, and extrapolate findings relevant to older children back to much younger babies. This is hard on families who are trying to find their way.
I find it an interesting parent-baby sleep study because, very unusually, an entirely new approach (which we use in our clinic) - 'bedtime fading' - is compared with 'graduated extinction.' Associate Professor Michael Gradisar and his team advertised for parents with infant sleep problems and then randomised them to three groups.1 The first group (n=14) were advised to apply ‘graduated extinction,’ a first wave behavioural (FWB) approach,ii also known as controlled comforting or responsive settling or sleep training. I’ll refer to FWB as sleep training in this blog. The second group (n=15) applied ‘bedtime fading,’ which requires setting a regular wake-up time in the morning, and making bedtime incrementally later (importantly, not a FWB strategy). The third group (n=14), ‘education control,’ were directed to a government website that again suggested various FWB sleep training strategies, such as putting the child in the cot awake to self-settle.
The media has focussed on the findings of a drop in night-waking by the sleep training group of 2.8 to 2.2 night-time wakes at seven days after the interventions, when both the education control (2.6 to 2.6) and the bedtime fading (2.0 to 2.2) show no change. But we wouldn't expect bedtime fading to have an effect until two weeks or so afterwards - and after a few disrupted nights of crying due to graduated extinction, it would not be surprising to find a child having better catch-up nights by the end of the week. What really matters to parents is whether or not these strategies result in signficant changes that stick. One month later, the graduated extinction group were waking 1.4 times at night; the bedtime fading 1.8 times. By three months, the bed-time fading group had dropped from 2 wakes to one at night; the graduated extinction group from 2.8 to 1.4 wakes – much the same change, practically speaking; night-waking in education control fell from 2.5 times to twice. How much of this is due to the inevitable developmental shift towards self-settling as the months pass? We can't tell from this study. Were the changes in night-waking more likely to occur in the toddlers than in the younger babies? We can’t tell this, either. Trying to create subsamples in such a small study would not offer reliable data. The best we can say is that in this study, where the numbers are too small to give us information that we can rely on for clinical decision-making, the difference between the graduated extinction and bedtime fading groups is about one less night-time waking episode every other night at three months (and the length of time taken as 'night-time' remains undefined).
If we are really going to help parents, we have to reach for high level evidence. We have to analyse all the studies that have been performed on the same topic, assessing them for rigour and synthesising the results in systematic reviews and meta-analyses. This way, we start to get a better picture of what works.
In 2013 my co-author and I published a meta-narrative systematic review investigating the effects of sleep training in the first 6 months (43 studies).2 In 2016 Kempler et al published a quantitative systematic review of randomised controlled trials investigating the effects of sleep training in babies in the first year of life (2500 babies).3, 4 The key findings of these two recent systematic reviews are the same: sleep training does not improve the number of night wakings. Sleep training results in only small changes in the total amount of time a baby spends in the cot without signalling (e.g. 29 minutes more at night). Sleep training is not associated with reliable evidence of improvement in maternal mental health.
At Gradisar et al’s three month follow-up, children in the bedtime fading and graduated extinction groups took ten or 12.7 minutes less, respectively, to go to sleep at bed-time then pre-treatment. The education control group took two minutes longer. Again, we can't tell if these strategies were more effective with the toddlers rather than the young babies, although we'd expect so. In the first week there was a moderate stress reduction for mothers in the graduated extinction group, no change in stress reduction in the control group, and a large reduction in mothers’ stress in the bedtime fading group. This makes sense, since bedtime fading requires responding to the child’s cues of sleepiness, rather than the clock, and avoids the stress of delaying response to or ignoring a baby’s cues (e.g. crying). But there were no improvements in maternal mental health in any group overall.
If, then, we are still not demonstrating improved maternal mental health or significantly decreased night waking, can we justify using sleep training as a key public health strategy for mothers and babies in the first 6-12 months post-birth, the way we are in Australia today?5
In our parent-baby sleep work at Possums, we understand how confusing it is for families to be confronted by so much conflicting advice, and work with parents exactly where they are at. If something is working, there's no reason to change it. We use bedtime fading as one part of a much broader program to help achieve healthy parent-baby sleep.6 We don’t call it ‘bedtime fading:’ we discuss putting bedtime back incrementally, and getting up at the same time each morning - alongside a range of other strategies that support healthy function of the sleep-wake homeostat and circadian clock. With older infants, once we've done the foundational work, we think through whether the family believes their child is ready to learn something new in the night.
Health professionals often tell parents that they must ‘teach their baby to sleep’ if they want him or her to develop optimally and have healthy sleep later in childhood. This frightens parents into compliance with FWB sleep training, and exacerbates everyone’s sleep anxiety (which worsens sleep). Yet large studies show this claim isn’t true.7-9 Gradisar et al also find no demonstrated improvement in the children’s development and attachment in any of the groups, twelve months on.
The most important thing is to look for joy in life with the baby as best we can, to get to know our baby through experimentation and responsive interaction, to remove those obstacles that get in the way of healthy sleep for both parents and baby and which do result in excessive night-waking - and to have strategies up our sleeve for managing the difficult thoughts and feelings that arise whenever we undertake a challenging task, let alone the challenging journey of raising a child.
That’s the aim of our parent-baby sleep work in the Possums Clinic.
ii First wave behavioural strategies include advice to delay or ignore your responses to your baby’s cues; use feed-play-sleep cycles; avoid bad habits by not allowing your baby to associate feeds with sleep; teach self-settling; put the baby down to sleep in the cot while still awake; aim for certain durations of wake and sleep times; encourage second sleep cycles since ‘sleep breeds sleep’; avoid overstimulation and overtiredness; and watch for tired cues (usually listed for parents). Since the advent of first wave behaviourism in the 1950s and 1960s, behaviourism has developed into what are referred to as second and third wave behaviourisms. First wave behaviourism remains a dominant school of psychological approach only in the care of parents and babies.
1. Gradisar M, Jackson K, Supurrier N, Gibson J, Whitham J, Williams AS, et al. Behavioral interventions for infant sleep problems: a randomised controlled trial. Pediatrics. 2016;137:e20151486.
2. Douglas P, Hill PS. Behavioural sleep interventions in the first six months of life do not improve outcomes for mothers or infants: a systematic review. J Dev Behav Pediatr. 2013;34:497–507.
3. Kempler L, Sharpe L, Miller CB, Bartlett DJ. Do psychosocial sleep interventions improve infant sleep or maternal mood in the postnatal period? A systematic review and meta-analysis of randomised controlled trials. Sleep Medicine Reviews. 2016;29:15-22.
4. Douglas PS. High level evidence does not support first wave behavioural approaches to parentinfant sleep. Sleep Medicine Reviews. 2015:10.1016/j.smrv.2015.1010.1007.
5. Douglas PS, Whittingham K. Response to 'Sleeping like a baby? Infant sleep impact on care givers and current controversies'. Journal of Paediatrics and Child Health. 2015;51:234.
6. Whittingham K, Douglas PS. Optimising parent-infant sleep from birth to 6 months: a new paradigm. Infant Mental Health Journal. 2014;35:614-623.
7. Price AM, Wake M, Ukoumunne OC, Hiscock H. Outcomes at six years of age for children with infant sleep problems: longitudinal community-based study. Sleep Medicine. 2012;13 991-998.
8. Mindell J, Lee C. Sleep, mood, and development in infants. Infant Behaviour and Development. 2015;41:102-107.
9. Price A, Quach J, Wake M, Bittman M, Hiscock H. Cross-sectional sleep thresholds for optimal health and well-being in Australian 4-9-year-olds. Sleep Medicine. 2015:doi:10.1016/j.sleep.2015.1008.1013.