The ageless debate over behavioral sleep interventions for infants, particularly those extinction-based techniques that allow for some period of “crying it out,” will likely be revived with the recent publication in the journal Pediatrics of a new research study.
While I’ve covered this topic before, including some back and forth with other PT bloggers, this new study is a bit different in that they compared a graduated extinction group not only to a control group but also to a supposedly kinder and gentler behavioral technique called bedtime fading. These sort of Ferber-lite approaches have sprouted up as an alternative to the classic model of graduated extinction because some parents find extinction just too difficult to do while others are concerned that the technique might be harmful to a child’s long term well-being.
Basically, a group of infants around 10 months of age, all with some level of sleeping problems, were randomized to one of three groups. One was a control group that received some educational materials about sleep. Another was the graduated extinction group in which parents were taught to put an awake infant in his or her crib and then, if the child is fussy, gradually increase the interval of time in returning to the crib, at which point a parent is instructed to provide comfort but not pick up the baby. For this study, this interval on the first night was 2 minutes for the first awakening, 4 minutes for the second, and 6 minutes after that. As the week progressed, this interval was increased further, if needed.
The third group was randomized to the bedtime fading condition. Here, the parent puts the infant in the crib awake and sees how long it takes for her to fall asleep. If it takes longer than 15 minutes, then the next night she goes to bed 30 minutes later. If it takes less, the infant goes to bed 30 minutes earlier. This procedure repeats until a bedtime is found that consistently results in a sleep “latency” of 15 minutes of less. If the child wakes up later, parents in this group were instructed to do whatever they do usually.
Of interest were a number of sleep parameters (number of night awakenings, total length of sleep, time to falling asleep, etc.) as assessed by sleep diaries that the parents completed as well as ankle monitors that recorded movement. To look at the question of whether these sleep interventions caused harm, levels of saliva cortisol (which look at stress hormones) were gathered, along with a measure of parent-child attachment and different levels of emotional-behavioral problems.
In terms of results, significant improvements in sleep were found generally for both treatment conditions relative to controls. For example, the number of night awakenings for infants in the graded extinction condition dropped in half (2.8 to 1.4) after three months while for the control condition it did not significantly change (2.6 to 2.1). While both techniques were helpful, there was some indication for some of the measures that graded extinction might work a little better. Another hopeful finding, similar to previous research, was that many of these sleep variable improved quickly, like within a week.
Further, levels of maternal stress were found to decrease for the two behavioral treatment groups relative to controls, although maternal mood improved in all three groups.
Evidence for harmful effects on the child were not found. Afternoon infant cortisol levels were found actually to decrease for the two behavioral treatment groups relative to controls. In addition, no group differences were found related to how securely attached the infant was or to levels of behavioral problems when measured a year later.
The authors concluded that their data support the hypothesis that brief behavioral sleep interventions are effective and do not lead to later child behavioral problems or disrupted attachment.
Will this study end the debate on behavioral sleep interventions? Probably not. The study is not without some problems including a quite small sample size and some inconsistency in the findings (such as finding improvements for the behavioral techniques with the parent sleep diaries but not according to the ankle monitors).
What might be more surprising to some is the idea that there actually are some research-based alternatives to what many parents see as two suboptimal choices, namely staying sleep deprived versus what seems to some as overly harsh cry it out methods. This research, similar to another important study, supports the idea that graduated extinction actually does not traumatize infants or disrupt the parent-child bond but nonetheless some parents might be more comfortable with at least trying other methods first.
@copyright by David Rettew, MD
David Rettew is author of Child Temperament: New Thinking About the Boundary Between Traits and Illness and a child psychiatrist in the psychiatry and pediatrics departments at the University of Vermont College of Medicine.