Does Sleep Training Actually Improve Infants' Sleep?

In the intense debate over sleep training, even parenting books offer conflicting advice, with about two-thirds advising parents to use sleep training and one-third advising against it. But what does the research actually say? Is sleep training effective?

Research broadly finds that sleep training methods improve sleep for infants according to parent report, yet it is unclear whether they actually improve sleep according to objective measures of actual sleep time. A 2006 review study found that 82 percent of children show clinically significant improvement in sleep with sleep training. However, most of these studies (77 percent) used parent-report measures, which may be biased or inaccurate.

Randomized Controlled Trials of Sleep Training

Fortunately, we have a few randomized controlled trials that use objective measures of sleep, called actigraphy (translation: data from monitors that the infants wear that show whether or not they wake up). One randomized controlled trial of sleep training with 6- to 8-month-olds who had a parent-reported “sleep problem” randomly assigned parents to get instruction on infant sleep or safety. The instructions on infant sleep involved a description of the "checking in" method but also other important information such as the importance of having a bedtime routine. Interestingly, the researchers found that, although parents reported that their babies were waking up less frequently, sleep actigraphy found that the babies in the sleep-training group were waking up just as often. However, the actigraphy found that sleep-trained babies slept longer for their first period before waking—yet, on average, only 16 minutes longer.

After the intervention, 4 percent of the sleep-training group reported their child still had a sleep problem, while 14 percent of the control group did. The parents who received instruction on sleep training also reported improved mood, sleep quality, and fatigue.

This study had some important limitations. Although the intervention group received information about sleep training, it is unclear whether they actually followed these instructions, and researchers did not determine whether any parents in the control group used sleep training. It is also unclear whether it was the sleep training or learning other important sleep information such as bedtime routines that resulted in slightly longer sleep times.

Another randomized controlled trial assigned parents to one of three groups: (1) checking in, (2) bedtime fading, or (3) sleep education. The researchers found that both children in the "checking in" and "bedtime fading" groups slept better after treatment according to parent report but found no differences according to objective measures of sleep.

Although parent report may be biased, these data are still important because it may mean that the parents were sleeping better. There did not seem to be a significant difference in the sleep training group at the end of the intervention for wake after sleep onset, number of awakenings, or total sleep time. Sleep training also did not result in infants “sleeping through the night.” However, the lack of differences among groups may have been because some percentage of parents in the intervention groups did not use the sleep-training methods advised by researchers, and some percentage of the "sleep education" group might have used sleep training even when they were not advised to do so.

A question commonly asked with sleep training methods is whether they lead to any long-lasting changes. One randomized controlled trial followed children for 5 years and focused only on 8- to 10-month-olds with reported “sleep problems.” They allowed mothers to choose either the "checking in" or "camping out" method and taught parents how to use the strategy of their choice as well as positive sleep strategies, such as bedtime routines.

The first report of this study found that mothers were less likely to report their child still had a sleep problem by 10 and 12 months. Yet, 56 percent of mothers at 10 months and 39 percent of mothers at 12 months still reported a sleep problem even after sleep training. A follow-up study found that the difference between the intervention and control groups faded by age two. In the final follow-up study at 6 years, the researchers found no difference in sleep between the two groups, including whether parents reported sleep problems or differences in children’s sleep habits. These findings suggest that the benefits of sleep training may be temporary and not necessary for a child to develop independent sleep skills.

It is very important to mention that we do not have sufficient evidence that sleep training is effective before 6 months. One of the only randomized trials of young infants found no impact of educating parents about infant sleep before 12 weeks. Another larger randomized controlled trial of educating parents about infant sleep before 6 months found that the intervention only decreased night awakening by 8 percent and that the babies only slept for 6 minutes longer during the day.

Translation

Sleep training may provide moderate sleep improvements but is unlikely to result in long-term improvements in sleep or guarantee a child will be a “good sleeper” as a toddler or a preschooler. Two randomized controlled trials suggest that sleep training increases the first stretch of sleep by about 15 minutes when compared to a control group. Some parents may conclude that parent-reported improvements in sleep and the possibility of 15 minutes of extra sleep for their baby is very significant to their quality of life and decide to attempt sleep training. However, other parents may conclude that sleep training may not be worth it without consistent evidence for objective and long-lasting improvements in sleep.