Sleep Training
Helping your baby sleep regularly—and, ideally, through the night—has tremendous benefits for both of you. But what’s the best way to make that happen? Some people swear by sleep training, and others worry that it will cause harm to the infant, the parent-child relationship, or both. How do you maximize sleep for your whole family while building trust and strengthening your relationship with your baby?
Competing Opinions
Perspective #1: Sleep training offers numerous benefits, the primary one being increased sleep for the whole family—which is important for everyone’s physical and emotional health. There’s no evidence that, assuming parents are emotionally attuned, their babies will suffer any negative effects of sleep training.
Perspective #2: When babies cry, it’s a signal of distress or discomfort, letting us know they need help. When you don’t respond to your infant’s request for help or comfort, you’re communicating that when she needs you, she can’t count on your being there.
What the Science Says
AS YOU TURN to this entry in the book, you’re likely wondering whether it’s OK to sleep train. You might be hoping that the science “allows” it, or maybe you’re dreading that research will tell you that it’s the best thing for your baby and your family. But what is it that we’re really talking about? What do we mean when we talk about sleep training?
Let’s start by getting clear on terms. Sleep training is not synonymous with the harsh “cry-it-out” approaches, where babies are fed and diapered, then placed in their cribs and told good night, not to see their parents again until morning. The idea behind this strategy is that infants are allowed to “cry it out” and thereby become trained to sleep on their own. The most famous proponent of this method is Richard Ferber, who once wrote that even if a baby becomes so upset that she vomits in the crib, parents should still resist picking her up. Instead, they should quietly clean up the mess and quickly leave the room so that the training can continue. As one leading researcher on infant sleep puts it, this more severe cry-it-out approach “is not usually recommended nowadays because of the distress it causes parents and infants.” (For the record, Ferber later revised and clarified some of his more controversial positions, explaining that he does not endorse “simply leaving a child in a crib to cry for long periods alone until he falls asleep.” He also softened his position on co-sleeping and took a more flexible stance on other child-rearing subjects.)
So if modern, attuned sleep training isn’t about setting the baby in the crib then turning off the light and listening to her scream until she wears herself out, what is it? And what does research say about it?
Among current authorities and experts, “sleep training” is often used as an umbrella term for many different types of strategies for helping babies learn to fall asleep on their own. Just by keeping lights dim while we change a newborn’s diaper before bed, we’re beginning the process of sleep training, teaching the baby to understand that we sleep at night. Other strategies call for slightly rousing the infant after he’s nursed or bottle-fed to sleep so he has to complete the process of dozing off in the crib on his own. Or the parent might remain close to the crib, possibly even placing a hand on the baby, or sleeping in the room near the infant, slowly moving farther away over time (a process sometimes referred to as “camping out”) while helping the child learn to relax and go to sleep on his own. Some even consider co-sleeping a type of sleep training. (See the entry “Co-sleeping.”)
For our purposes here, when I discuss sleep training, I’ll be referring to an approach that involves putting the baby down while he’s still awake but growing sleepy, then waiting a period of time before responding to his cries. After an interval, the parent responds and offers reassurance and comfort. Then the interval is extended a bit more, and the process is repeated until the baby falls asleep. The goal is that after a few nights of this progression, the baby becomes comfortable enough with the new arrangement to go to sleep on his own, and then sleeps better when he does. This approach is still known by many as “Ferberizing,” and there are more and less gentle and nurturing ways to practice it. Since the harsh and rigid versions, again, aren’t recommended by most experts and authorities on infant well-being, this discussion will examine the science surrounding the gentler versions, where parents do allow their babies to cry as they develop the ability to fall asleep by themselves, but the parents remain emotionally connected along the way.
(As a caveat, by the way, I’ll offer a quick reminder that this isn’t a how-to book. If you’re looking for specific strategies and instructions, you’ll find many valuable books and websites that will walk you through various approaches in a step-by-step manner. Here, I’m simply offering a broad, global perspective on how this particular approach to infant sleep is currently viewed among researchers who study the subject. My hope is that it will help you get an overall feel for whether you want to sleep train your baby, and help you explore many of the relevant variables as you make your decision.)
Research on sleep training exists, but not as much as you might expect, given its importance. There aren’t more than twelve to fifteen strong, high-quality studies available on sleep training and its outcomes, and even in these, researchers are often examining multiple strategies in the same study, typically relying heavily on parental report, which is quite open to bias. Further, studies that compare various sleep strategies to one another are rare. As a result, we should be careful about drawing conclusions too confidently. Still, we can discuss the scientific opinions and findings regarding the primary questions surrounding sleep training: Is it effective, will it harm the child, and will it damage the parent-child relationship?
In response to the first question, research responds in the affirmative. Sleep training does appear to improve sleep for both child and parent. This enhanced sleep leads to reduced parental fatigue and improvements in mood, including diminished parental depressive symptoms and marital conflict, all of which play significant roles in child outcomes. It’s better for the baby’s sleep as well. One study looked at caregivers who used gentler sleep-training interventions and found that five months after implementing the strategies, the number of parents reporting sleep problems had dropped by nearly 30 percent. The conclusion of another study emphasized emotional responsiveness on the part of the parent, explaining that “parents’ emotional availability to children in sleep contexts promotes feelings of safety and security and, as a result, better-regulated child sleep.” The news isn’t all good, though. Other studies have found that many of the sleep-related gains experienced early on as a result of sleep training will wax and wane over the following months. Science warns us, in other words, not to expect that sleep training is a once-and-for-all moment in our child’s life, not to believe that we’re somehow pressing the “perfect sleep” button that results in nighttime bliss for the remainder of childhood. Still, at least on a short-term basis, sleep-training techniques do seem to offer sleep benefits for a majority of babies.
But does it harm the child in some way, to allow her to experience moments of stress or being upset as she falls asleep? One thing seems clear: any improved sleep helps in the short term. Research has found that babies who sleep better can be more adaptable, better emotionally regulated, and less easily distractible.
Still, some parents worry about the stress caused by sleep training, even when it’s done in a nurturing fashion (as in the “camping out” method or other gentler sleep-training approaches). For most babies, short-term, mild-to-moderate stress can actually be positive in terms of building resilience, and tolerable stress experiences lead to what’s called stress inoculation. The fear, however, is that allowing especially upset babies to cry too long will establish a pattern of stress activation that leads them, especially those who are more sensitive and easily dysregulated, to develop less resilience over time. There’s concern that for babies who have a stress response that’s extreme, frequent, and long-lasting, the physiological and psychological distress may be harmful.
The argument is that a cessation of crying might not be evidence that the baby has learned to “self-soothe,” but instead that she has instead adapted to the stress by shutting down, even possibly dissociating to some extent, employing a defensive strategy that triggers what one researcher calls “an enduring hypometabolic state that might be problematic for the high-energy needs of the developing sleeping brain.” Indeed, research has demonstrated higher nighttime levels of the stress hormone cortisol in very young infants (one to three months) when mothers are less “emotionally available” during the times associated with sleep. (This study looked at mothers, but, likely, the same would apply to fathers as well.)
A related study measured the levels of the stress hormone cortisol in both babies and mothers during a sleep-training experiment. The researchers found that after two nights of crying it out, infants stopped expressing their distress by crying, but their cortisol levels remained at the same elevated level. Without the infant crying, the maternal cortisol levels decreased, even though the babies were experiencing the same amount of distress. The worry is that during the sleep-training process, parents are being conditioned to believe that everything is all right, when in actuality babies have simply learned to stop crying and go to sleep, even though they’re still upset. Other research, by the way, has shown that children in distress don’t always cry, depending on the circumstances and the child’s temperament.
The cortisol study has been criticized for certain methodological limitations—chief among them that baseline cortisol levels weren’t clearly established—and we should be careful about interpreting the results as evidence that sleep training is harmful for babies. But it makes sense that this study’s results would at least lead to further and improved research on the question, since it raises the possibility that a sleeping baby isn’t foolproof evidence that sleep training is successful across the board. After all, waking through the night is part of typical development for infants and young children, as is the need for the comfort and support of their parents. Those who challenge the efficacy of sleep training therefore raise the concern that children who learn not to cry at night have actually learned not to ask for or expect comfort, even when they’re internally distressed and need help.
Just how much stress an infant experiences during sleep training has not been well studied. But we do know that babies have a wide range of hardiness, sensitivity, and need, and that the intensity of the stress will be different for different babies. In other words, some babies might get highly distressed being left to cry for five minutes, whereas other babies might experience only mild distress.
Certain researchers have attempted to examine the question of sleep training’s effect on overall well-being by looking at children years after the fact. Some studies have found no negative outcomes resulting from sleep training in terms of overall child behavior, even up until age five or six. In one of the most robust studies to date, researchers followed up five years later on children who had undergone some sort of “behavior-modification program designed to improve infant sleep.” The study concluded that there was “no evidence of longer-term adverse effects” on children’s mental health, including their behavior, stress levels, sleep, or their relationship with their parents.
For what it’s worth, by the way, this study also didn’t find any longer-term positive effects, either. Put differently, it’s not an endorsement of sleep training, and for those uncomfortable with the idea, the study doesn’t argue that all parents should sleep train. It simply found that caregivers don’t have to worry that gentler sleep-training techniques will do long-lasting damage.
What about the parent-child relationship? Will a baby who’s been sleep trained lose overall trust that her parent will attend to her needs? Some studies have shown that sleep training either has no effect on the parent-child relationship or can even improve infant security and attachment. The key, according to many researchers, is the emotional availability of parents in the process, which can be correlated with improved sleep in infants. At the moment, there’s no evidence that gentle sleep training, when approached with an emphasis on honoring the child-parent bond, will negatively impact the degree of closeness in that relationship.
The Bottom Line
TO SUM UP: Does the science say it’s OK to sleep train? Yes, as long as you remain emotionally attuned to your child. Does research say you should do it? No. Just as there’s no definitive research establishing that sleep training is directly harmful to babies, neither is there proof that it’s beneficial. The truth is that, as far as we can tell given the body of research at this point, it doesn’t make much difference at all in terms of long-term effects on a child. As is so often the case, this decision comes down to your feelings about what you, your child, and your family need right now.
To explain a bit more fully: You might choose to sleep train simply because it makes the most sense to you and appears to be the best choice for your family. Your situation may leave you with little choice, either because of a work schedule or some other family dynamic. If you do sleep train, make sure to educate yourself on the least-stressful and most relationship-honoring methods available. For one thing, most experts agree that you shouldn’t start too early, suggesting waiting until your infant is four to six months old, once good feeding habits and sleep patterns have been established. Also, keep in mind that babies should never be left in highly stressed states for long periods of time. The “rule,” according to whichever expert/approach you’re following, might call for allowing your child to cry for three minutes (or five or ten or more). But there’s no scientific data about the amount of time a caregiver should wait before responding to a cry. Some infants might calm down and be able to fall asleep more quickly with shorter intervals of parental comfort, and some might calm down more quickly with a bit longer. As you’ve heard me say throughout this book, you should consider the information at hand, then, based on your knowledge of your own child’s needs—which change all the time throughout development—let your instincts guide you to make the best decision possible.
Keep in mind, too, that there’s a big difference between an intense “attachment cry” from a highly stressed baby who is in distress for long periods of time, and a mild “fussy cry” from a baby who isn’t really stressed but is just trying to get settled and who falls asleep fairly quickly. When stress ceases being mild and tolerable and turns intense and toxic, it can create distrust and fear rather than building resilience. As a result, the baby can be left unsure about whether, when she is in distress, anyone will come. So if your baby fusses a bit, doesn’t get too upset, and then puts herself to sleep, fantastic. That’s likely a valuable resilience-building experience. But every baby is different, and they have different needs at different times. I believe that if a baby is in intense distress for long periods of time, that’s not healthy. In that case, she needs a parent to help her feel soothed and safe enough to go to sleep, at least at that time in her development.
One key point if you decide to sleep train: be sure to check your expectations. Babies typically aren’t great sleepers. They have raw little nervous systems and small stomachs, and as a rule, they just can’t sleep through the night early on. Plus, even when a baby begins sleeping well, illness, developmental spurts, separation anxiety, teething, travel, or another change can crop up to affect sleep, and you may have to start over. Remember that sleep training isn’t a one-time event that means your baby will always sleep through the night. That likely won’t happen for years. Even healthy, well-adjusted school-age kids can regularly wake in the middle of the night and need to be comforted by their parents. You’ll likely be working on sleep for a few years, any which way you go. Plus, for many babies—some estimate as high as 20 percent—no amount of sleep training will be effective. So be careful with any assumptions that you’ll be able to find the “right” program that leads to blissful slumber throughout the household.
If you decide against any formal sleep training, take care of yourself so that you can be the parent your child needs you to be. It’s always a difficult balance to strike, deciding just how much of ourselves and our own needs to sacrifice for the good of our baby. If you determine that it just doesn’t feel right to let your infant cry at all, and you choose to forgo sleep training, find a way to get enough rest so that you’ll be capable of showing up in the various ways your infant needs.
My final bottom-line reminder on this subject is that every child is different, as is every family. Each child will go through different phases, and a given approach might be effective during one stage but not in another. It’s therefore essential that you remain attuned to your baby’s needs and prepared to flexibly adjust to what’s called for in each new moment. Your needs matter, too, and your sleep is important. Make decisions based on what your child requires, considering your own needs and those of other family members as well. Listen to experts, but follow your own instincts and be willing to adjust when circumstances change and call for a shift in terms of what it means to best care for your child and yourself.
On a Personal Note
I HEARD ARGUMENTS in favor of various cry-it-out methods when I was a young mother, but as someone who had studied physiological states of stress and the role of the attachment system in modulating stress, I just couldn’t get comfortable with the idea. I had lots of friends who believed in and practiced some version of crying-based sleep training (some harsh, some gentler.) They were loving, attuned, intentional parents whom I respected. Some of their stories made me jealous. (“My baby fussed just a little for the first night, almost not at all the second night, and now we’re all sleeping! It’s changed our lives!”) Some of their stories, I’ll acknowledge, made me feel judgmental. (“My baby screamed her head off for over an hour until I finally went in, and she had vomited all over her bed. I’m not sure if she’s actually sick or just got that upset.”) One friend said, “She needs to learn that if she cries, I’m not coming.” And I felt deeply that I wanted my baby to learn the opposite of that message. So, I read about every version of sleep training, from the strictest to the gentlest, but I just couldn’t feel at ease with any approach where I had to let my babies cry without responding. The way I saw it, if my priority during daytime hours was to respond when my children needed me, I wanted to do the same at night. So for the most part, my husband and I chose not to use any type of approach that involved letting our babies cry.
Looking back now, was it the correct decision? I think it was, for me and for our family. From one perspective, I took the easy road, at least in the short term. I nursed my babies to sleep because it was the quickest avenue to freedom each evening and the fastest way back to sleep in the middle of the night. Plus, our firstborn was a sensitive baby who got stressed easily, so I felt that allowing him to cry would overwhelm him.
But I’ll also admit that my babies weren’t great sleepers, so I was exhausted a lot, and at times I was pretty frustrated and tired, even resentful about having to nurse at 3 a.m. (and again at 4:30). It’s also the case that I was a stay-at-home mom during most of that time, and I had a husband with a job that allowed him to get up with our kids first thing in the morning so I could sleep a bit more. I tried some gentler methods and they didn’t seem to make much of a difference, but I didn’t know about the “camping out” method, and I think I would have been comfortable trying that. But I would have also followed my instinct to go to my babies if I felt that they were needing my presence to feel safe, and to settle.
In the end, I don’t regret responding to my babies when they called, day or night, and I’m confident that knowing myself, I would have regretted letting them cry without support. I wanted them to know that if they were communicating “I need you,” I would come. But it was a big sacrifice, and my husband and I paid a price for it. Who knows—maybe our children did, too, in that they might have slept better if I’d sleep trained, and they would have had a more rested, more patient mother. For what it’s worth, my boys were all good sleepers by the time they were two or so—and they’re making up for all those lost hours of sleep now that they’re teenagers who like to sleep a lot! And, my friends who used cry-it-out methods also have awesome kids and great relationships with them. So, every family has to do what feels right to them and what works best for everyone involved. If I had another baby now, I’d still be pretty uncomfortable not responding when she cried. But after looking back, I’d be fine with allowing her to mildly and briefly fuss, to give her an opportunity to fall asleep on her own.