Introduction
The main purpose of this chapter is to highlight the various factors that contribute to the development of early childhood insomnia in the context of the transactional model of infant sleep, as described in Fig. 11.1 . , This model reflects the idea that the development of insomnia in infants results from a dynamic, complex, and interactive process between the infant and various proximal and distal environmental factors, mediated mainly through sleep-related interactions between infants and their parents. Understanding the role of these different factors is crucial for the assessment and treatment of sleep problems in infancy.
Specifically, the current chapter will review research on the role of parental factors and the parent-infant relationship in early childhood insomnia. The chapter will not cover other important parts of the model, such as physiological and constitutionally based factors (e.g., temperament, medical problems) or sociodemographic and cultural aspects that are related to infant sleep development. These domains are covered and discussed in other chapters of this book.
Throughout the chapter, we will refer to various sleep variables that reflect the behavioral aspects of sleep. These aspects include both sleep quantity variables, such as sleep duration and time in bed, and sleep quality variables. Sleep quality refers to both sleep initiation (sleep onset latency) and sleep consolidation (e.g., number and duration of night-awakenings, sleep percent, and longest period of uninterrupted sleep). Sleep regulation and self-soothing skills will also be covered, given their role in sleep initiation and maintenance.
Phenomenology of insomnia in infancy
The evolution of sleep consolidation and regulation is one of the main developmental processes during infancy. During the first months of life, infants wake frequently during the night, and these awakenings are usually accompanied by feeding and other types of external regulation (e.g., holding, rocking). However, the associations between feeding, external regulation, and falling asleep gradually weaken as the physiological sleep-wake system matures (i.e., circadian and homeostatic processes); the ability to retain calories increases; and adaptive sleep associations are learned. Throughout the second half of the first year of life, most infants develop consolidated and regulated sleep. Whereas sleep consolidation refers to the reduction in the length and number of night-wakings, sleep regulation refers to the capacity of the infant to independently resume sleep following natural nocturnal night-wakings.
Although most infants attain the ability to “sleep through the night” during the first year of life, there is substantial variability between infants. , In fact, bedtime problems and difficulties in the process of nighttime sleep consolidation and regulation affect 15% to 30% of all young children between the ages of 6 months and 3 years. These problems are among the most common concerns brought to the attention of pediatricians and other child-care professionals. , If not treated, they may persist in 20% to 50% of children , and may have negative consequences for children’s socioemotional, behavioral, and cognitive development, as well as for parental sleep and mental health.
Early childhood bedtime and night-waking problems are classified in the International Classification of Sleep Disorders (3rd ed., 2014) under the general diagnostic category of Chronic Insomnia Disorder, which refers to both adults and children. The main criteria (criteria A) specifies that the patient reports or the patient’s parents or caregiver observes one or more of the following: (1) Difficulty initiating sleep; (2) Difficulty maintaining sleep; (3) Waking up earlier than desired; (4) Resistance going to bed on appropriate schedule; and (5) Difficulty sleeping without parent or caregiver intervention. In addition to criteria A, associated daytime dysfunction (e.g., fatigue, sleepiness, mood disturbance/irritability) is observed or reported (criteria B). Moreover, the sleep problems exist despite adequate sleep opportunities. Finally, disturbances occur at least 3 times per week, have been present for at least 3 months, and cannot be explained by another sleep disorder (e.g., obstructive sleep apnea). Similarly, the diagnostic criteria for an Insomnia Disorder according to the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) specify that in children, sleep onset and night-waking problems may manifest as difficulies initiating sleep and returning to sleep without caregiver intervention . Thus, both the ICSD and the DSM emphasize that early childhood insomnia involves self-regualtion difficulties. These difficulties are manifested by infant signaling (e.g., by crying, fussing, refusing to lie down, etc.) at bedtime and upon waking at night, and by infant reliance on caregiver assitance to transition from wakefulness to sleep.
It is important to emphasize that the diagnosis of insomnia in infants depends largely on parents’ experiences and perceptions. Parents may perceive their infant as having a sleep problem to the extent that the infant’s night-wakings exceed the parents’ expectations of “normal” awakenings with respect to the infant’s age, the awakenings cause disruption to their own sleep, and/or cause considerable emotional distress. , , However, not all parents of “signalers” are troubled by their infants’ awakenings. For instance, they may not consider their infant to have a sleep problem when the night-wakings are short and require only short parental involvement, and the parents have no difficulty resuming their own sleep. A certain frequency of awakenings per night could thus be seen as problematic by some parents, and not by others. Likewise, a similar extent of sleep fragmentation or resistance could be perceived as a problem at age of 18 months, despite being perceived as normal by the same parents only a few months earlier. Thus, the subjective experience of the parent is a critical factor to consider as part of the assessment process.
One term which is frequently used to describe the dependency of infants on external help to fall asleep is “sleep-onset associations”—a certain set of conditions that the child learns to rely on at the time of sleep onset in order to fall asleep. Usually, this term refers to the caregiver’s help and involvement, but it can also refer to other external objects that the infant associates with falling asleep, such as music or media screens. In the progress of their development, most infants acquire sleep-onset associations that allow for self-regulation and independent sleep initiation. These sleep-onset associations may include engaging with objects within their control (e.g., baby blanket, pacifier) or being in the same external conditions (e.g., dark, quiet room) at sleep onset and during the night. Infants who are capable of using these sleep-onset associations (i.e., “self-soothers”) tend to experience brief natural arousals rather than prolonged night-wakings and resume sleep without signaling or requiring their parents’ help. In contrast, “signalers” are usually unable to reastablish sleep on their own and typically present frequent and/or prolonged awakenings. , ,
A major question in the field of early childhood insomnia is how could these differences between infants be explained? Why do some infants become “self-soothers” while others become “signalers?” The following sections of this chapter will describe the role of parent and parent-infant relationship factors in the development of early childhood insomnia.
The various sections of this chapter are organized according to the different levels of influence, as suggested in the transactional model ( Fig. 11.1 ). , First, the “parent-child interactive context” (interactive behaviors and interpersonal system) will be described, followed by “parenting factors” (parental mental health, cognitions, and cry tolerance), ending with the role of the broader “family” context (coparenting, chaos).
Parental sleep-related behaviors and infant sleep
Parental bedtime and nighttime involvement
Based on the transactional model, the link between parental sleep-related behaviors and infant sleep is both direct and bidirectional. Infants for whom it is more difficult to self-soothe are more likely to elicit excessive parental nighttime help. At the same time, when parents actively settle their infant to sleep, infants learn to associate falling asleep with parental assistance and this may hinder the development of nocturnal self-soothing skills that are necessary for consolidated sleep. , ,
A few early studies examined the links between parental bedtime involvement and infant sleep. , In a seminal longitudinal study on this topic conducted by Anders, Halpern, and Hua, 21 parents and their infants were assessed with videosomnography, so that both parental behaviors at bedtime and infant sleep were observed. The findings demonstrated significant concurrent associations between parent-infant interactions at bedtime and infant self-soothing behaviors at 3 and 8 months. Infants who were put into the crib awake and fell asleep on their own at bedtime were more likely to resume sleep on their own following awakenings later in the night, whereas infants who were settled to sleep by their parents at the beginning of the night were significantly more likely to receive similar help from their parents following night-wakings. However, no longitudinal links were found between parent-infant bedtime interactions at 3 months and infant sleep and self-soothing behaviors at 8 months, and hence the directionality of the links could not be established.
During the last decades, a vast body of empirical research around the world has consistently demonstrated that parental presence and active involvement in settling the infant to sleep (e.g., by feeding, holding, rocking) are strongly associated with frequent and/or prolonged infant night-wakings and with lower nighttime self-soothing. In contrast, infants who are placed into their cribs awake at the beginning of the night and/or use a sleep aid, show less nighttime wakefulness and are more likely to self-soothe during the night. , However, most of the research on this topic has been correlational, making it difficult to infer about the directionality of effects.
Overall, research on the effectiveness of clinical sleep interventions (see Chapter 14 ) and developmental-longitudinal studies conducted during the last two decades support the assumption that parental bedtime and nighttime involvement influence the development of infant sleep. For example, in a study of 80 infants, in which parent-infant nighttime interactions and infant sleep were studied through the use of videosomnography at five time points across the first year of life, one of the main factors that significantly predicted infant nighttime self-soothing at 12 months was longer parental response delays to infant awakenings at 3 months (the other two significant factors were time out of crib and percentage of quiet sleep). A recent study similarly found that when parents introduced a short interval before feeding in the first weeks of life, infants had longer nighttime sleep periods at 3 months, based on both video and parent-report measures. However, no significant links were observed with infant sleep at 6 months. In another recent longitudinal study, parental presence at sleep onset and the overall frequency of parent settling activities at 1 month of age predicted worse infant sleep at 6 and 12 months, based on sleep diaries.
Most studies on parenting and infant sleep have focused on whether parents are present or not at bedtime, and on their level of involvement and response latency to infant awakenings. Recent research has examined whether beyond these factors, the type or quality of parental intervention might also be an important factor. For example, Voltaire and Teti examined whether the manner in which parents intervene with their infants during the first 3 months of life predicts infant night-wakings (as reported by the parents) across the first 9 months in a group of cosleeping and solitary sleeping infants. Through direct observations these authors differentiated between non-distress-initiated interventions (i.e., interventions initiated by parents in response to a calm infant awakening or during infant sleep) and distress-initiated interventions (e.g., in response to infant crying). The authors found that only non-distress-initiated parent interventions predicted worse infant sleep outcomes, but this was only significant for infants who were sleeping in a separate room. Higher frequencies of distress-initiated interventions, on the other hand, predicted a sharper decrease in infant diary-based night-wakings in both cosleeping and solitary sleeping infants.
The links between parental quality and infant sleep were further studied by Philbrook and Teti who examined the role of maternal emotional availability at bedtime in predicting observed infant sleep at 1, 3, and 6 months of age. Maternal emotional availability refers to the overall quality of emotional attunement with the infant, expressed through aspects such as sensitivity, structuring, and low hostility. The findings demonstrated significant within-subject effects. That is, when mothers were more emotionally available than usual, infants slept more and displayed less distress during the night. Moreover, an interaction between maternal bedtime practices and emotional availability was found in the prediction of infant sleep; infants whose mothers were more emotionally available at bedtime and used fewer arousing activities developed consolidated sleep faster. In addition, higher maternal emotional availability in combination with less close contact at bedtime was associated with more infant sleep across the night.
Altogether, these longitudinal studies suggest that parental difficulty in withholding their response to infant awakenings and low maternal emotional availability may hinder the acquisition of infant self-soothing abilities that promote better sleep consolidation (i.e., parent-driven effect). However, the question of whether infant sleep problems elicit more parental presence and intervention at bedtime and nighttime remains quite open, as this line of influence (i.e., infant-driven effect) has seldom been studied directly. In one study, both mother-driven and infant-driven mediational models were assessed, but only the mother-driven model showed significance. However, this was a cross-sectional study and therefore directionality could not be accurately assessed. In a longitudinal study, neither infant night-wakings nor infant distress (e.g., crying) predicted later parenting practices (e.g., nursing, physical contact), although higher infant distress did predict lower maternal emotional availability at the following time point. Thus, these findings provide partial support for infant-driven effects. As studies on infant-driven effects are extremely limited, additional longitudinal research is needed to examine the bidirectional paths of influence between parental behaviors and infant sleep-wake behaviors.
Breastfeeding and cosleeping
Two specific and interrelated parental nighttime practices, which have been investigated in relation to infant sleep development, are breastfeeding and cosleeping. Breastfeeding provides a myriad of benefits for the development of infants, making it a highly recommended practice. Nevertheless, in the context of sleep, breast milk is more easily digested and thus breastfed infants need to be fed more frequently. Around the age of 6 months, when nighttime feeding is usually no longer physiologically necessary to satisfy hunger, breastfed infants who are nursed to sleep at bedtime are more likely to become dependent on nursing to resume sleep during the night. , In fact, both breastfeeding and cosleeping allow the infant to spend prolonged periods of time in close contact to the mother and may therefore limit the infant’s opportunity to practice nighttime self-soothing. Indeed, research demonstrates that both breastfeeding and cosleeping are associated with more night-wakings, shorter sleep duration, and less self-regulated sleep. ,
With regards to cosleeping (roomsharing and/or bedsharing), new studies are consistent with previous findings , of mothers reporting more disturbed infant sleep (more night-wakings, shorter sleep duration) in cosleeping arrangements. , , , , However, recent longitudinal studies comparing objectively measured infant sleep patterns (e.g., via actigraphy) of roomsharing versus solitary sleeping infants found no significant differences between groups. , Thus, it could be that mothers of roomsharing infants report more night-wakings as they are more aware of infant awakenings because of their physical proximity to the infant, and/or that infants who sleep in a separate room do not wake less than roomsharing infants but are more capable of self-soothing.
Notably, the research on cosleeping and infant sleep problems is mostly correlational and thus it is impossible to attribute causality to the findings. It could be that parents of cosleeping infants are more likely to intervene faster when the infant wakes up, and this may delay the acquisition of self-soothing skills (parent-driven effect). However, it is also possible that parents of infants with sleep problems are more likely to bring their infants to their room/bed as a way to cope with the infant’s awakenings (i.e., reactive cosleeping; infant-driven effect). Moreover, decisions regarding sleeping arrangements are heavily influenced by cultural and societal norms, and therefore findings concerning the links between infant sleep quality and cosleeping that are based on Western samples should be considered in the context of Western norms that overall encourage solitary sleeping arrangements.
Bedtime routines
Another aspect of parental bedtime behavior that is linked with children’s bedtime behaviors and nighttime sleep patterns is parental consistency during bedtime routines. Bedtime routines are defined as the predictable and pleasant/calming activities that occur in the time preceding lights out, such as parent-child interactive activities (e.g., reading, singing) and hygiene-related activities (e.g., bathing). Parental soothing behaviors that are aimed at helping the infant fall asleep, such as rocking or feeding to sleep, are not conceptualized as part of the bedtime routine, and as described above, these are associated with delayed self-soothing capacities and with more problematic sleep. Bedtime routines are thought to exert their positive influence on sleep by providing the infant with a sense of predictability, calmness, and security. These aspects promote down-regulation, facilitating the infant’s transition between wakefulness and sleep. Indeed, evidence from both cross-sectional and longitudinal studies, , conducted on infants and preschool children, suggests that a consistent bedtime routine is associated with earlier bedtimes, shorter sleep latency, reduced night-awakenings, less bedtime resistance, longer nighttime sleep duration, and better caregiver-reported sleep quality. For example, a large cohort study including more than 10,000 children (0–5 years old) from 14 different countries found that having a bedtime routine was associated with better sleep outcomes as reported by the parent. Moreover, within both predominantly Asian and predominantly Caucasian cultural regions, there was a dose-dependent relationship with better sleep outcomes associated with more frequent use of having a bedtime routine, underscoring the importance of bedtime routine consistency for both infants and preschoolers across countries and cultures. In a longitudinal study, parents of 468 children completed a questionnaire asking about bedtime and bedtime routines, their child’s sleep duration, nighttime waking, sleep latency, and sleep problems at 3, 12, 18, and 24 months of age. The findings demonstrated a few longitudinal associations between more bedtime routine consistency and better sleep outcomes at later assessment points. Specifically, consistency of bedtime routines at 12 months predicted fewer nighttime waking and sleep problems at 18 months, and an earlier bedtime at 18 months predicted consistency of bedtime routines at 24 months. Moreover, bedtime routine consistency was concurrently associated with longer sleep duration and earlier bedtimes.
However, not all studies support an association between bedtime routine consistency and sleep quality in infants. Recently, Adams, Savage, Master, and Buxton examined the associations between bedtime routine consistency and actigraphic sleep in infants aged 6 to 24 weeks. Significant associations between more consistent bedtime routines and longer true sleep time (excluding nighttime wakefulness) were found only at 6 weeks, but not at later ages, and no significant associations were found with sleep percent. These restricted findings, in comparison to other studies on bedtime routines, might be related to the young age of the infants and to the fact that sleep was measured using actigraphy, whereas most research in the field so far has been based on parental reports of both bedtime routines and child sleep.
Attachment and infant sleep
The attachment relationship refers to the enduring bond of the infant to the caregiver. A prominent factor in the development of a secure attachment relationship is the ability of the primary caregiver to provide physical proximity and emotional security to the child in threatening or stressful situations. , Going to sleep represents a separation from the ongoing interactions with the attachment figure and is therefore a potentially stressful situation for the infant. Theoretically, a secure attachment relationship facilitates feelings of emotional safety in the child, which in turn is thought to enhance the infant’s capacity to separate from the caregiver when going to sleep leading to better infant sleep regulation. On the other hand, an insecure attachment that is characterized by a high level of anxiety and worry, might manifest in difficulties separating from the caregiver at bedtime and in heightened vigilance, which is contrary to the physiological and psychological state of relaxation required for falling asleep. , In light of these assumptions, the links between attachment and infant sleep-wake regulation have been a topic of theoretical and empirical interest. Studies on attachment and infant sleep have theoretically posited that insecure attachment styles, and in particular ambivalent attachment, would lead to more infant sleep problems. Consistent with the transactional model, another possibility is that sleep problems would compromise the attachment relationship, through their effect on parental sensitivity or emotion regulation. However, most research on the links between attachment and sleep has been cross-sectional, preventing the possibility to infer about directionality. Moreover, the empirical findings in this area are rather mixed, with some studies finding significant associations between insecure attachment and sleep problems, and others not. , For example, in a study of 94 nonrisk 12-month-olds, the prevalence of infants who were defined by their mothers as “night-wakers” was high in the majority of cases and was only marginally different between the secure and insecure attachment groups (based on the Strange Situation Procedure). In a subgroup of 37 infants, actigraphic sleep measures were not related to attachment security, revealing similar levels of sleep efficiency and awakening in secure and insecure groups. In a subsequent study with 57 low-risk 12-month-old infants, sleep was assessed using actigraphy and maternal reports, and the Attachment Q-Set procedure was used to assess the infant’s level of attachment security and dependency. Findings showed that only mothers’ reports of infant sleep problems (and not actigraphy) were associated with the child’s dependency score, but not with the security score. The authors suggested that dependency might be a more relevant factor than security in explaining sleep regulation problems. On the contrary, another study found that mothers of infants with an insecure-ambivalent/resistant attachment style reported more frequent and prolonged infant night-wakings than mothers of infants with an insecure-avoidant attachment style. Morrell and Steele compared attachment (assessed with the Strange Situation Procedure) between 14- and 16-month-old infants with and without reported sleep problems. There was a higher percentage of ambivalent attachment in the sleep problems group (12.5%) as compared to the good sleepers (1.7%). A follow-up assessment indicated that ambivalent attachment was predictive of persistent sleep problems a year later.
More recent studies that examined the longitudinal associations between attachment security and infant sleep have also provided mixed results. , , , For example, Simard, Bernier, Bélanger, and Carrier investigated the relations between attachment assessed with the Strange Situation Procedure and infants’ sleep quality, using both objective and subjective measures. Insecure-ambivalent attachment, assessed at 18 months, was not associated with actigraphy-assessed sleep, but was associated with longer nocturnal wake duration as reported by mothers at the age of two years. In another study by the same group attachment security at 15 months (assessed with the Attachment Q-set) was prospectively associated with more actigraphic minutes of sleep and with greater sleep efficiency at 2 years. In contrast, in a study assessing family predictors of reported infant sleep problems at the ages of 6, 15, 24, and 36 months, attachment security, measured with the Strange Situation Procedure, was not predictive of infant sleep trajectories. In another longitudinal study, Pennestri et al. examined the relations between maternal report of infant sleep at four different time points (6, 12, 24, and 36 months) and attachment (assessed with Strange Situation Procedure at 36 months). Children with a disorganized attachment style went to bed later and had shorter sleep duration and more night-wakings, compared to both insecure-ambivalent and secure children. However, no significant differences were found between the secure and ambivalent groups for any of the sleep parameters.
As can be noticed, most studies in the field examined whether attachment security predicts infant sleep. A recent longitudinal study examined the opposite direction of prediction by assessing whether infant sleep predicts later infant-father and infant-mother attachment security. Opposite association patterns were found for mothers compared to fathers; longer infant nighttime awakenings at 3 months were predictive of lower infant-mother attachment security at 24 months, but a higher number and longer duration of infant nighttime awakenings at 3 months were unexpectedly associated with higher infant-father attachment security at 24 months. The authors examined whether the surprising findings regarding fathers could be explained by paternal involvement in nighttime caregiving, but the data did not support this possibility. They recommended that future research would include a more comprehensive measure of paternal nighttime caregiving to further examine its role in the links between infant sleep and infant-father attachment.
In summary, although there is a solid theoretical basis to assume that attachment security and sleep regulation will be strongly associated, the empirical findings in this area are overall inconclusive. The mixed findings based on the Strange Situation Procedure might be related to the small variance this procedure tends to yield in low-risk samples. However, inconclusive findings arose also from the few studies using the continuous, and thus more variable, scales of the Q-sort procedure. Additional research is needed that would examine the bidirectional links between sleep and attachment in more diverse samples. Moreover, as previous research has demonstrated significant links between maternal parenting quality (e.g., sensitivity, emotional availability) and infant attachment security, it would be interesting to explore how both domains interact to predict infant sleep development.
Parenting factors and infant sleep
Parental psychopathology
Parental psychopathology plays a major role in the way parents perceive their children and interact with them. Research on maternal depression and child development has clearly demonstrated that depressed mothers are more likely to show poorer parenting practices and compromised mother-infant interactions, which may be responsible for long-term negative outcomes for the child. , According to the transactional model, parents who suffer from mental health problems may engage in less adaptive bedtime and nighttime behaviors, which may lead to the development of sleep disturbances. , On the other hand, persistent infant night-wakings that require intensive parental intervention may elicit parental stress and chronic sleep disturbances, that over time may lead to more parental emotional distress. , ,
Empirical research on the links between infant sleep and parental psychopathology has so far focused mainly on the role of maternal prenatal or postnatal depression/depressive symptoms. A few studies have demonstrated that newborns and young infants born to mothers with clinical prenatal depression were more likely to show less efficient and more fragmented sleep as assessed with actigraphy and with polysomnography, compared to infants born to nondepressed mothers. , Moreover, newborns of depressed mothers who were observed during sleep spent less time in deep sleep and more time in indeterminate (disorganized) sleep. It has been suggested that because these changes in sleep-wake organization are observed so early in life, they might be influenced by biological factors. Specifically, genetic factors or fetal exposure to maternal hormonal abnormalities and changes in the maternal hypothalamic pituitary adrenal axis may put the infant at risk of developing regulatory problems, including sleep problems. , However, not all studies on sleep in infants of depressed pregnant women demonstrate significant changes. For instance, Galbally et al. compared sleep of 6- and 12-month-old infants who were born to depressed and control women and found no significant differences in reported sleep problems between the groups.
Interestingly, there are only a few studies that have investigated the links between maternal postpartum clinical depression and infant sleep problems. This is surprising considering the high prevalence of postpartum depression, which is estimated to stand at 10% to 15%. , In one study that examined the relations between maternal depression in the postpartum and maternal reports of infant sleep problems, the rate of clinically significant depression scores (based on the clinical cutoff of the Center for Epidemiological Studies Depression Scale—CES-D) was about double in mothers of night-waking infants than in mothers whose infants did not usually wake up during the night at 6 months postpartum. Furthermore, in a study of women with a history of clinical depression, disturbed infant sleep at 6 weeks, as reported by the mother, was associated with a higher depression score both concurrently at 6 weeks postpartum and longitudinally at 16 weeks postpartum. A recent large longitudinal study, based on data from 2222 mothers and infants, examined the predictive links between maternal depression in the perinatal period (pregnancy and 3 months postpartum) and infant actigraphic and reported sleep at 12 months. Mothers who scored ≥13 points on the Edinburgh Postnatal Depression Scale (EPDS) during pregnancy and/or at 3 months postpartum were considered perinatally depressed. Whereas depressed mothers were more likely to consider their infant’s sleep as problematic and to report >3 night-wakings per night than nondepressed mothers, no significant differences were found for the actigraphic sleep measures. Moreover, there were no differences in soothing techniques employed by depressed versus nondepressed mothers.
The associations between maternal prenatal and postnatal depressive symptoms (distinct from clinical depression) and infant sleep have been studied quite extensively, and most show small to medium correlations. , , Most of these studies are based on maternal reports of infant sleep. The few studies that assessed the links between maternal depressive symptoms and objective measures of infant sleep found only limited support for significant findings. In a study assessing infant sleep-wake patterns with videosomnography, mothers with higher depressive symptoms at 1-month postpartum were more likely to have infants who fell asleep independently at 12 months. The authors suggested that these mothers might have waited longer before responding to their infants’ awakenings, which may have led to the development of self-soothing skills. Another study found that greater postpartum maternal depressive symptoms were associated with more inconsistent total sleep duration, as assessed by actigraphy. However, sleep quality measures, such as infant awakenings, were not included in this study.
Most research on parental depression and infant sleep has focused on mothers, but a few studies have also found significant concurrent associations between paternal depressive symptoms and infant sleep. For instance, in one study, fathers of infants who perceived their infant’s sleep as problematic at 4 months of age had increased depressive symptoms concurrently and also at 6 months of age. Another study found significant associations between paternal emotional distress symptoms and 8- to 12-month-old infants’ bedtime difficulties, as perceived by mothers.
Clearly, most research on parental psychopathology and infant sleep has focused on maternal depression or depressive symptoms. Recent studies have contributed to this growing field by examining other aspects of parental emotional distress, such as maternal anxiety and stress. , Probably, the only study so far which examined mothers with a clinical prenatal and postpartum anxiety disorder did not find support for a significant association with infant sleep. Similarly to studies on maternal depressive symptoms, studies on nonclinical samples found that prenatal anxiety symptoms and postnatal anxiety, stress, and PTSD symptoms were moderately associated with more maternal reports of infant sleep problems. , , , Some of these studies found significant associations with fathers’ stress symptoms as well. , , A recent longitudinal study of 225 mothers and their infants examined the links between maternal depressive, anxiety, and stress symptoms and infant actigraphic and reported sleep. The findings demonstrated that mothers with higher emotional distress symptoms—and especially those with parenting-stress symptoms—were more likely to experience their infant’s sleep as problematic. However, no significant correlations were found between maternal emotional distress symptoms and objective-actigraphic sleep measures. Moreover, trajectory analyses indicated no significant effects of changes in maternal emotional distress variables on changes in infant subjective or objective sleep. These findings support growing evidence suggesting that mothers who suffer from emotional distress symptoms are more likely to experience the sleep of their infants as problematic, though there is little evidence to suggest that maternal emotional distress leads to the development of objective sleep difficulties.
Although the transactional model postulates that the links between maternal emotional distress and infant sleep problems are mediated through maternal bedtime and nighttime behaviors, underlying mechanisms of the links between maternal depression and infant sleep have hardly been investigated. Teti and Crosby examined such a mediation model in 45 infants (1–24 months old). Mothers’ depressive symptoms were associated with more maternal presence and close physical contact with infants during the night, though no significant associations were found between depressive symptoms and bedtime practices. Furthermore, the mediation analyses provided support for a mother-driven model in which maternal presence with infants during the night mediated the links between maternal depressive symptoms and infant night-waking. However, the authors emphasized that no causal relations from these results could be inferred, given the cross-sectional design of the study. One longitudinal study of 5568 mothers looked at infant temperament as a possible mediator and found that maternal antenatal depression, measured with the EPDS, predicted infant negative affectivity at 9 months which in turn predicted more reported infant night-wakings at 2 years of age.
Overall, only a few studies examined whether infant sleep disturbances predict maternal emotional distress (i.e., infant-driven effects). In the study of Teti and Crosby described above, no statistical support was obtained for the infant-driven model, in which infant night waking predicted maternal depressive symptoms via maternal nighttime presence. However, since the mediated paths in the infant-driven models approached significance, the authors concluded that it is possible that both mother- and infant-driven influences are relevant in explaining the links between maternal depressive symptoms and infant sleep.
Maternal sleep disturbances have been suggested to underlie the predictive links between infant sleep problems and maternal emotional distress. , Consistent with this asumption, links between poorer maternal sleep and increased severity of depressive symptoms have been reported, and infant sleep has been shown to be strongly associated with maternal sleep. , However, to the best of our knowledge, there are no longitudinal studies that examined the role of maternal sleep as a mediator of the link between infant sleep and maternal emotional distress. Nevertheless, indirect support for this notion comes from infant sleep intervention studies which documented, in addition to improvements in infant sleep, alleviation in maternal sleep and mood.
A few studies have tried to identify moderators of the links between maternal emotional distress and infant sleep. , , , These studies may shed light on the mixed findings regarding the direct links between these domains. For instance, in a study with two large longitudinal cohorts, infant reactivity and gender were examined as moderators of the association between maternal symptoms of antenatal depression and infant sleep. The findings demonstrated that reactive boys had a higher number of reported awakenings and shorter sleep duration when previously exposed to maternal symptoms of antenatal depression in comparison to girls and to infants with lower reactivity. Another study examined family structure as a possible moderator of the link between maternal depressive and anxiety symptoms and infant sleep, by comparing the strengths of these links between two-parent families and solo-mother families (i.e., mothers who decided to parent alone). The findings demonstrated that only in solo-mother families higher maternal emotional distress was associated with more infant diary-based night-wakings. The authors suggested that paternal involvement in two-parent families may mitigate the association between maternal poor sleep quality and emotional distress, explaining the stronger links found in solo-mother families.
To summarize, research based on clinical samples has demonstrated that mothers’ prenatal clinical depression is a risk factor for the development of infant sleep problems in the beginning of the infant’s life. Moreover, findings consistently show that mothers with elevated depressive symptoms are more likely to perceive their infants’ sleep as problematic. However, the few studies that are based on objective sleep measures are scarce and do not support the notion that maternal emotional distress contributes to the development of poor objective infant sleep quality. Future studies are needed to further examine these links within longitudinal designs using objective and subjective measures of sleep. These studies might shed light on the directionality of the effects and may clarify whether these links are restricted to the subjective experience of the mother. Also, studies examining various mediators and moderators of these links are needed to elucidate possible underlying mechanisms.
Parental cognitions, cry reactivity, and infant sleep
According to the transactional model, parental cognitions regarding infant sleep may impact their sleep-related behaviors and levels of nighttime involvement, which as described above, directly influence infant sleep. Parental cognitions refer to perceptions, attitudes, attributions, expectations, interpretations, and beliefs parents have regarding their children. Parental cognitions regarding child behavior have been significantly associated with the way parents respond to their children and have been associated with child development and parent-child interactions. Increasing evidence suggests that parental cognitions are an important factor to consider in the development of infant sleep problems. , , , For instance, in one of the first studies on this topic, mothers of infants with sleep problems reported more cognitions related to difficulty with limit-setting, increased doubts about parenting competence, and increased anger at the infant’s demands. A follow-up study demonstrated that the most relevant factors for concurrent sleeping problems were maternal cognitions reflecting concerns about setting limits and fussy-difficult temperament. Moreover, these variables explained the degree to which parents used active physical soothing to settle their infants to sleep, which in turn predicted the persistence of sleeping problems. In a study assessing the links between infant sleep and parental sleep-related cognitions in clinical and control samples, parents of sleep-disturbed infants reported more concerns and difficulties with limiting their nighttime involvement than did control parents. Furthermore, significant differences in sleep-related cognitions were found between fathers and mothers. Given hypothetical examples of infants with sleep problems, fathers were more likely than mothers to endorse an approach that encourages infant self-soothing (“limit-setting”). Moreover, the findings revealed that fathers’ cognitions reflecting difficulties in limit setting were linked to more infant night-wakings, in addition and independently of maternal cognitions, implying that the likelihood of infant sleep problems increases when both parents have trouble in limiting their involvement. In a longitudinal study (from pregnancy through the first year) aimed at assessing the prospective links between maternal sleep-related cognitions and infant sleep, significant predictive and concomitant links were demonstrated. Specifically, maternal cognitions which emphasized the possibility that infants experience distress upon awakening and that parents should therefore help them, predicted and were associated with more disturbed infant sleep at 6 and 12 months of age. On the other hand, maternal cognitions emphasizing the importance of limiting parental nighttime involvement predicted and were associated with more consolidated sleep. In addition, parental soothing techniques mediated the links between maternal cognitions and infant sleep. Mothers who tended to interpret infant night-wakings at 6 months as a sign of distress that requires immediate attention were more actively involved in bedtime soothing at 12 months, and this, in turn, was related to more infant night-wakings. Consistent with transactional perspectives on this topic, infant sleep also predicted a change in maternal cognitions, although this link seemed to be weaker than the direction of prediction from maternal cognitions to infant sleep. Similarly, Teti and Crosby also found support for a mediation model showing that mothers’ nighttime presence with the infant mediated the relations between maternal worries about infant nighttime needs, and infant night waking. Relatedly, maternal separation anxiety (i.e., feelings of guilt, worry, and sadness that accompany short-term separations from the child) has been studied in relation to infant sleep. Higher levels of separation anxiety were found to be associated with greater maternal physical proximity at bedtime, more involvement during the night, and more actigraphic nighttime awakenings. ,
Another related parenting factor that seems to be associated with parental cognitions and infant sleep and has been the focus of recent research is parental cry tolerance (PCT). In the first study on this topic, Sadeh and colleagues found that parents of infants with night-waking problems were less tolerant to infant crying compared to parents of infants without sleep problems and to childless controls. PCT in this study was measured using parental responses to audio and video recording of infant crying. In the video procedure, participants were presented with a 2-minute video clip of a crying infant, with gradually increasing crying intensity. Prior to watching the video, a written cover story was presented to the participants with a rationale that delayed response is recommended because the infant they observed is very demanding and his parents are trying to encourage him to self-soothe. The delay to intervene was used as a measure of PCT. The findings of this study, showing lower cry tolerance in parents of sleep-disturbed infants, may suggest that PCT may be an important factor in the development of infant sleep. It could be that parents with lower PCT show faster responses and more active involvement in soothing their infants to sleep, which may lead to less consolidated infant sleep. However, consistent with the transactional model, it could also be that parents develop lower cry tolerance in reaction to their infant’s sleep problems. To further clarify the direction of these links, a longitudinal study was conducted in which PCT was measured at pregnancy and 6 months, and infant sleep was assessed at 3 and 6 months. Concurrent associations were found between lower maternal cry-tolerance and poorer actigraphic infant sleep at 6 months, as well as between lower PCT and more active nighttime soothing. Furthermore, lower cry-tolerance at pregnancy predicted better infant sleep at 3 months, whereas more disrupted sleep at 3 months predicted lower cry-tolerance at 6 months. The authors suggested that this shift in directionality may be explained by the evolving needs of the infant throughout the first months of life; lower PCT during late pregnancy may represent a parenting attitude more attuned to the infant needs and regulatory capacities. Later, parents may become more sensitive to infant crying when they need to continuously take care of a night-waking infant, as manifested in the reduction in cry tolerance at 6 months. In line with these notions, decreases in both PCT and cognitions attributing distress to nighttime awakening have been demonstrated following behavioral interventions for infant sleep problems. These decreases were associated with parent-reported and actigraphic improvements in infant sleep following treatment, providing further evidence for the mutual evolution of these constructs.
Taken together, the findings of these studies highlight the role of parental sleep-related cognitions and cry-tolerance in infant sleep development and, consistent with the transactional model, demonstrate that these links are mediated through parental bedtime practices.
Family factors and infant sleep: The role of fathers, couple relationship, and family chaos
At the distal level of influences on child sleep, the transactional model stipulates that family stressors, such as low-quality marital relationships, will influence infant sleep through their impact on parental sleep-related behaviors. , In practice, most studies on parenting and infant sleep problems have focused primarily on mothers, paying little attention to the role of other family variables such as the couple’s relationship and coparenting.
A few studies have demonstrated the importance of the couple relationship as a factor that may contribute to infant sleep development. , , For example, lower coparenting quality and lower marital adjustment as perceived by the mother have been found to predict continuing infant cosleeping arrangements through the first year of life. , Moreover, Bernier and colleagues reported that greater paternal marital satisfaction as well as mothers’ perceived social support were predictive of better maternal reports of child sleep consolidation at 2 years. These relations were stronger in families from lower SES backgrounds. Evidence for infant-driven effects on coparenting has also been found. Reported infant night-wakings predicted coparenting quality, with parent sleep quality mediating these links. Another line of research has demonstrated better infant sleep quality and less bedtime resistance in families where both parents shared caregiving. , For example, in two studies, a relatively higher involvement of fathers in overall infant caregiving (as reported by mothers and fathers) when infants were 1 and 3 months old was predictive of more consolidated infant sleep, as assessed with actigraphy, at 6 months of age. , Whereas father involvement and the marital relationship seem to play a role in infant sleep development in two-parent families, a recent study suggests that the absence of a father does not seem to play a role when the mother has decided in advance to bring a child to the world on her own; in a study comparing infant sleep in families of solo-mothers (single women who have decided to parent alone) compared to two-parent families, there were no significant differences between the groups in mother or infant sleep quality, though solo-mothers were more likely to cosleep with their infants.
Another family variable that seems to be associated with infant sleep problems is family chaos, although its impact has mainly been studied in families with preschool and older children. One longitudinal study of 167 families, with 5 assessment points during the first year, examined the role of family chaos in infant sleep. Infants in homes characterized by higher chaos demonstrated delays in actigraphic sleep consolidation, manifested in greater sleep fragmentation and more variable sleep duration. However, chaos was also associated with longer sleep duration, maybe as a way to compensate for fragmented sleep.
Overall, these studies suggest that exposure to family chaos, parent conflict, and low paternal involvement in two-parent families acts as a risk factor for the development of infant sleep problems. These couple and family variables may exert their influence through undermining the infant’s sense of security. It is also possible that they lead to less adaptive parental bedtime and nighttime behaviors which in turn negatively impact infant sleep. ,
Conclusions and research limitations
The research reviewed in this chapter fits well into the conceptual framework of the transactional model describing bidirectional and dynamic links between various parenting and parent-infant relational factors and the development of early childhood insomnia. Recent findings, especially those that are based on longitudinal designs, have highlighted the important role of parenting factors, such as parental behaviors, emotional availability, cognitions, cry tolerance, mental health, and coparenting in the development of infant sleep.
Nevertheless, some significant research gaps still exist. Although the number of longitudinal studies has increased considerably during the last decade, most research is still based on cross-sectional designs, and thus it is difficult to ascribe causality or even directionality to the links. Moreover, most longitudinal studies have explored the role of parental factors in infant sleep development, but research regarding the influence of infant sleep problems on parental behavior and mental health has surprisingly been very limited. Thus, the relative influence of parent-driven effects versus infant-driven effects should be further explored. In addition, most studies are based on parental reports of infant sleep, and the restricted research that included objective methods, such as actigraphy and videosomnography, often shows less robust findings. It is unclear if this discrepancy is a result of inflated correlations due to shared method variance in studies using parent reports alone. More research that is based on objective methods is needed to clarify this question. Furthermore, although the transactional model assumes complex relations between the different levels of parental influences, most studies to date have examined the effect of parenting variables on infant sleep separately, and there has been very little research on the interaction between these variables, or between parenting factors and infant intrinsic variables (i.e., moderation analysis). Likewise, only a few studies have so far examined the mediating role of parent-infant bedtime and nighttime interactive factors (e.g., bedtime soothing), in the links between distal parenting factors (e.g., mental health, coparenting) and infant sleep problems, within the context of longitudinal research. Lastly, although accumulating evidence has started to underscore the role of fathers and the broader family context (e.g., family chaos), this area of research is still in its infancy. In sum, future research should examine more complex models taking into account the bidirectional prospective links between parenting and different facets of infant sleep and the possible mediators (e.g., parent-infant relationship/interaction) and moderators (e.g., interaction between parental depression and infant temperament) relating to the different distal and proximal levels of the transactional model.
Clinical implications
Early childhood insomnia is a major concern in infancy and a source for parental concern and distress that can significantly impact the well-being of the family. The findings reviewed in this chapter linking various parenting and parent-infant relationship factors to infant sleep problems imply that all these variables should be comprehensively evaluated during clinical assessment of early childhood insomnia, in addition to the assessment of potential medical and physiological sources for the sleep problem. Naturally, the role of these variables will vary from one family to the other. Whereas in some families, parental sleep-related cognitions may play a significant role in influencing parent-infant interactions around bedtime, in other families it might be low coparenting or mental health problems that may be more relevant in explaining the dynamics around sleep, and yet in other families the sleep problems might be more directly and specifically related to bedtime practices such as extensive feeding or inconsistent bedtime routines. Hence, it is important that health care professionals who conduct clinical assessments of infant sleep problems will be aware of the various factors that may contribute to the maintenance of the sleep problem and assess their possible influence in every single family. Accordingly, infant sleep interventions could be adjusted and tailored to the specific and unique characteristics of the infant and the family. The different parenting factors that were described in this chapter are all modifiable. Targeting the specific factors that contribute to the sleep problem in an individualized manner may increase the likelihood of successful intervention.
References
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
