Cry it out, or not

It doesn't take much googling to work out that sleep is one of the biggest challenges for new parents.  There is a huge market for products that claim to help babies sleep (noise devices, self-help books, baby wraps), however, according to some media coverage from a few years ago, research shows that it is best to let your baby "cry it out". Media headlines like "Let crying babes lie: Study supports notion of leaving infants to cry themselves back to sleep" and  "Study shows 'crying it out' is best for babies"  for some reason didn't sit right with me. I looked into it and found that this research was published in Developmental Psychology, which is a really reputable journal. But the journalist went a bit too far. 

In the media coverage of the story, the journalists report that at 6 months, two thirds of infants slept through the night . Those that didn't, however, the transitional sleepers, were more likely to be boys, their mothers were more likely to describe their temperament as difficult, they were more likely to be breastfed (this didn't sit well with breastfeeding advocates) and their mothers were more likely to be depressed. The author Marsha Weinraub from Temple University is quoted as suggesting that ...

"The best advice is to put infants to bed at a regular time every night, allow them to fall asleep on their own and resist the urge to respond right away to awakenings."

Is this the take home message from the research? Or is this what Dr Weinraub responded when the journalists bullied her into telling them some advice for parents? I can just hear it now...

"So Dr Weinraub,  what does your research suggest parents should do about their baby who wakes in the night?"

The title of the original article gives it away. 

Weinraub et al., (2012). Patterns of Developmental Change in Infants' Nighttime Sleep Awakenings from 6 through 36 months of age. Developmental Psychology, 48, 1511-28

I don't even have to read the methodology to know that this research has nothing to do with the pros or cons of controlled crying. It is obviously a descriptive developmental study about age-related changes in sleep behaviour across infancy, not about parenting or sleep training. There is nothing in this article that suggests leaving infants to cry themselves back to sleep is either good or bad. This is a case of a quote from the author being used to mislead readers about the true nature of a research finding and is an example of terrible science journalism. 

So what does this study tell us about infant sleep development?

In a large longitudinal study, parents were questioned about their infants' sleep behaviour when the babies were 6 months, 15 months, 24 months and 36 months of age. In addition, measures of attachment, temperament, breastfeeding, maternal depression, maternal sensitivity, father presence, illnesses, family size, poverty, childcare, maternal education, and martial conflict were taken. 

Models of developmental change in sleep behaviour identified two groups of infants; SLEEPERS were those who slept through the night from 6 months and whose sleep behaviour was didn't change across development.  These children represented 66% of the group. In contrast, the TRANSITIONAL SLEEPERS were those children who woke several times during the night at 6 months, however, their night waking improved with age, such that by 18 months their sleeping did not differ from the SLEEPERS group. Those in the TRANSITIONAL group were more likely to be boys, breastfed at 6 and 15 months, and had more difficult temperaments. Mums of transitional sleepers were also more likely to be depressed and have partners with poor health. Children were more likely to come from large families and spent less time in day care at 9 months.  

The authors concluded that there are two distinct developmental profiles in sleep development but note that it is unclear what the implications of these patterns of change are. Given that the transitional group do not differ from the sleeper group at 18 months, these findings may be absolutely inconsequential. All children learn to sleep through the night eventually, and it is possible that the length of time that it takes them to do so doesn't make a difference. Alternatively, it may be that sleep in transitional children may continue to differ qualitatively. For example, the effects of anxiety-producing events on children's ability to sleep may differ during childhood. I think that this is an interesting empirical question. 

The issue of breastfeeding is one that the authors do speculate on in the discussion. While they suggest that breastfed infants were more likely to wake in the night past 6 months of age, they acknowledge that their data does not speak to whether that was because of hunger or because of breastfeeding interfering with learning to self soothe. 

I think this is a really interesting descriptive study and it is a pity that was butchered by the media. I was surprised that as many as 67% of infants were waking one night per week at 6 months and wonder whether parent report is the most accurate way to measure this. Perhaps actigraphy would be useful in getting an objective measure of age-related changes in sleep behaviour. Letting your baby cry it out is one way of teaching them to sleep through the night, however, this particular piece if research certainly does not speak to whether this is a good or bad thing to do.  

Jenny Richmond

Lab Director