There is good empirical evidence that behavioural factors play a role in the onset/maintenance of many childhood sleeplessness problems and that behaviour therapy can be used as an effective form of management. There is a smaller, but growing, literature supporting the idea that behavioural interventions may also play a significant role in the management of other types of sleep disorder (eg, rhythmic movement disorders, arousal disorders, nightmares and circadian rhythm sleep disorders), and this is an area ripe with research possibilities. This review outlines the nature of behavioural aspects of children’s sleep and how these might be addressed by behaviour therapy. Clinical considerations concerned with the use of behavioural therapy are also highlighted (eg, the role of behaviour therapy as an adjunct to other treatments, its use with special populations, and how it might be delivered to families).
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Sleep patterns are regulated by both biological and psychological processes,1 2 and disorders of sleep may arise because of problems with one, or both, of these processes, to varying degrees. Biological factors include brain areas and neurotransmitters involved in the promotion, maintenance and timing of sleep, as well as physiological abnormalities which may interfere with normal sleep-related processes—for example, large tonsils and adenoids may cause airway obstruction during sleep, resulting in arousal and disrupted sleep. Psychological factors include cognitions, cultural norms and behavioural aspects. This review will consider behavioural aspects (ie, learned behaviours) and their relevance for our understanding of children’s sleep patterns and management of their sleep disorders.
In primary care, advice for parents of sleepless infants is commonly based on behavioural principles, delivered by either health professionals or self-help leaflets/booklets for parents. Such an approach can be successful for many families, but other families will require more intensive and specialised input to fully delineate the behavioural factors maintaining their infants’ sleeplessness, to develop intervention strategies and to be supported through the implementation of any intervention. Behavioural factors may also be relevant for other types of sleep disorder and other groups of young people. Therefore, an understanding of behavioural conceptualisations of young people’s sleep disorders and their clinical application is important for paediatric health professionals in primary and tertiary services so that, where appropriate, advice can be given to families, and/or clinicians can identify families who may benefit from more specialised input.
This review aims to orientate the reader by outlining key aspects of behaviour therapy commonly advised for infant sleeplessness, but also to suggest that such an approach has a wider application. By directing the reader to reviews of current evidence and fuller explanations of behavioural interventions for various sleep disorders and various patient groups, the intention is to increase awareness of how behavioural factors may play a role in the development and treatment of wide-ranging paediatric sleep disorders and to discuss clinical considerations relevant to management planning and decisions about whether to refer a child for specialist behavioural therapy.
As indicated, the relevance of behavioural factors for disordered child sleep is most commonly considered to exist in association with child sleeplessness problems. Indeed, in the recently revised International Classification of Sleep Disorders,3 “behavioural insomnia of childhood” has been introduced as one of the listed disorders. This insomnia may take the form of “inappropriate sleep onset associations” (ie, where the child has not learnt to fall asleep without a set of problematic or demanding conditions such as parents’ being present), “limit-setting sleep disorder” (ie, where the care giver demonstrates insufficient or inappropriate limit-setting to establish appropriate sleep behaviour in the child) or a “combined” subtype where these two problems co-exist. These sleep disorders present with difficulty settling to sleep, nightwaking and/or early waking difficulties.
Sleeplessness difficulties of these types are some of the most commonly reported behaviour problems affecting children from the general population, with estimated overall prevalence rates of ∼30%.4 5 About 25–50% of 6–12-month olds have difficulty settling to sleep or waking in the night, and these figures do not decrease with age: by age 3 years, 25–30% have sleeplessness problems, with similar percentages reported for the 3–5-year age group,6 7 43% of 8–10-year olds8 and ∼23% of 10–17-year olds.9 These problems are not transient; an epidemiological study of a cohort of 5-year olds suggested that sleeping problems at age 5 years were significantly associated with sleeping difficulties at age 6 months (or before) and that children with sleep problems at age 5 years were more likely to have sleeping problems at 10 years of age.10
These high rates, and the persistence of these problems if untreated, are particularly significant in view of the fact that childhood sleep disturbances have been noted to be associated with a range of adverse effects on child and family functioning2 and even may be predictive of poor psychiatric state in adulthood.11 12
Conceptualising some aspects of child’s sleep as behavioural implies that there are elements of sleep that are learnt and therefore that behavioural interventions (which aim to help children to “learn” an appropriate set of sleep behaviours/habits and/or to unlearn inappropriate sleep behaviours) may have a role in managing sleep disturbances with a behavioural aetiology. Interventions, based on learning theory, typically involve both classical and operant conditioning. Classical conditioning is a form of associative learning whereby a neutral stimulus is paired with a naturally occurring stimulus, which evokes the desired behavioural response until, after multiple pairings, the neutral stimulus alone is sufficient to elicit the desired behaviour; thus behaviours are conditioned to be elicited by antecedent conditions. Operant conditioning involves the use of consequences to modify the occurrence and form of behaviour. Interventions commonly include some basic principles, although the precise nature of any intervention and the particular intervention strategy used to implement each principle will vary depending on family and child factors and the nature of the sleep disturbance one hopes to address. General principles, with an example of how they might be applied in the context of a child sleeplessness problem, include the following.
Behaviour can be encouraged by linking it with an antecedent stimulus, which serves to trigger the desired behaviour. For example, establish a wind-down bedtime routine with a definite end point that the child can learn to associate with sleep onset and have an appropriate sleep environment that is used every time the child settles to sleep so that over time the familiar environment becomes associated with the onset of sleep.
Behaviour is likely to recur if the consequences of the behaviour were reinforcing for the child. Reinforcement can be positive (positive reinforcement occurs when a positive condition comes about as a result of their behaviour) or negative (negative reinforcement occurs when a negative condition is stopped or avoided as a consequence of their behaviour). For example, use strategies such as attention/praise/star charts to “reward” the child for behaviour one is trying to encourage and extinction to remove reinforcement. Extinction, or removing reinforcement (eg, drinks, parental presence, attention) maintaining the undesired behaviour (eg, crying, refusal to settle to sleep without the above) can be achieved gradually or abruptly. Removing reinforcement, especially abruptly, is likely to result in a temporary increase in the undesired behaviour (post-extinction response burst) before a decrease in the behaviour. Awareness of this phenomenon is important so that treatment is not prematurely abandoned and erroneously considered ineffective. Parental behaviour during child sleeplessness episodes needs also to be considered in terms of the reinforcement maintaining their own behaviour (eg, parental strategies of staying with the child until they are asleep may continue because this behaviour provides negative reinforcement for the parents by stopping the child’s upset).
A new behaviour can be encouraged by rewarding a series of responses that more and more closely resemble the desired behaviour (shaping). For example, over successive days, reward the child for going up to their bedroom, then for taking part in the bedtime routine, then for staying in their bed with parent present, then with parent absent, etc.
Behaviour is less likely to occur if followed by a punishing consequence. Punishment can be negative or positive. Negative punishment could take the form of a “response cost” (ie, removing a pre-specified amount of a valued item after undesirable behaviour). Positive punishment (ie, providing aversive consequences following the behaviour), although often associated with rapid effects, may also be associated with unpredictable and unwanted side effects. The mechanism of behaviour change associated with positive punishment appears to be complex and not adequately delineated, so it should only be used with caution (see Lerman and Vorndran13 for a review of basic and applied findings).
Rewards are defined by their effects on behaviour. Monitor the relationship between rewards and child behaviour to determine what is reinforcing for any particular child and family.
The more consistently these principles are applied (eg, over successive nights, between carers, at different times of the night), the easier it will be for the child to learn.
There is a growing body of research which suggests that behavioural interventions can be used successfully to manage childhood sleeplessness (several reviews14–17 provide detailed information about different behavioural programmes used and the first of these14 gives a helpful glossary of terms) and also used preventively to good effect to encourage consolidated sleeping and to discourage sleeplessness problems in children and infants.18 19 Indeed, on the basis of controlled empirical data, behavioural approaches are viewed as the first-line treatment of choice for this type of sleep disorder, because they have none of the potential associated negative side effects of sedative hypnotics,20 and a systematic review suggests that, long term, they are more effective.21 The relative efficacy of specific behavioural strategies or components of behaviour therapy has been less well investigated, and, in view of this, decisions about the precise details of any behavioural plan should be explored in the context of collaborative therapy, considering the practical and emotional resources of individual families. The developmental stage of the child is also an important consideration—for example, some strategies require the child to have verbal abilities of a particular level, and reinforcement programmes are likely to be particularly important for older children.
It is, however, worth keeping in mind that sleeplessness, or insomnia, is only one of the types of sleep disturbance that may be experienced by children and that behavioural factors may play a role, to varying degrees, in the cause or maintenance of sleep disturbances other than insomnia. There are over 80 sleep disorders listed in the International Classification of Sleep Disorders,3 which are divided into six main categories: insomnia, sleep-related breathing disorders, hypersomnia of central origin, circadian rhythm disorders, parasomnias and sleep-related movement disorders. Although empirical data on the use of behavioural interventions in the management of sleep disorders other than insomnia are less convincing, both in quality and quantity, preliminary reports suggest that rhythmic movement disorders such as nocturnal headbanging,22 23 parasomnias such as sleep terrors24 and nightmares,25 and circadian rhythm disorders such as delayed sleep phase syndrome26 may also be helped or resolved by behaviour therapy. Unlike infant sleeplessness problems, where behavioural models have been proposed to explain both development and treatment,27 28 current understanding of these other sleep disorders does not permit us to determine the extent that learning plays in the onset of these disorders, although it is more easily conjectured how learnt aspects may contribute to the maintenance of these problems, whatever the original aetiology. For example, the aetiology of nocturnal headbanging is not known, but this sleep disorder is typically noisy and distressing for parents to witness. Therefore, it draws attention and often elicits attempts to prevent or distract the child from this behaviour (which are usually unsuccessful). Over time, the parental attention associated with the headbanging might be enough to sustain the child’s behaviour. There is a need for development of behavioural models specific to different types of sleep disturbance, and for children of various ages, to further understanding about the development and maintenance of sleep disturbance and to suggest therapeutic strategies for evaluation.
It should, of course, be noted that the effective use of behaviour therapy does not necessarily indicate that the sleep disturbance is behavioural in origin or being maintained by behavioural factors. Behaviour therapy may play a role in the management of disorders that are of definite organic origin. For example, behavioural therapy including planned naps, appropriate sleep routines and well-defined 24 h schedules may have a role in management of narcolepsy, a hypersomnia of central origin,7 or facilitate compliance with continuous positive airways pressure therapy for children who require this form of intervention for sleep-disordered breathing.29
A further consideration for clinicians is that multiple sleep disorders may coexist, and so behaviourally based sleep disorders may be present in children with other sleep disorders of more physiological origin or arise as a secondary problem. For example, high rates of comorbidity have been documented in children with obstructive sleep apnoea.30 Common comorbid sleep disorders include enuresis, bruxism, arousal disorders and sleeplessness problems. It may be necessary to use multiple forms of treatment to address individual sleep disorders.
It may also be appropriate to combine treatments in the management of one sleep disorder. For example, studies of the use of melatonin in the treatment of sleeplessness in children with developmental disorders suggest that melatonin is a promising therapy, although there remain many unknowns about its use, so it should currently be used with caution and a full awareness of the uncertainties and issues; see Stores31 and Phillips and Appleton32 for review of intervention studies and discussion of the clinical issues. Importantly, one of the potential limitations of the existing literature in terms of establishing the efficacy of melatonin is that studies have typically included behaviour therapy, of various nature and quality, as part of the treatment. Although this may be an empirical limitation, it may be that this is a clinical strength, as melatonin therapy is likely to be more successful if there are no behavioural issues contributing to the sleeplessness. Combined pharmacological and behavioural therapy has been used successfully in the context of other drugs used to manage sleeplessness in typically developing children,33 although combined approaches are generally under-researched.
An important point to note is that behavioural therapy has been used successfully with a number of special populations who are at increased risk of severe and long-standing sleep disorders for a range of biological and psychological reasons.34 That this form of intervention does not rely on the use of verbal skills makes it especially appropriate for use with children with intellectual disabilities. Intervention studies generally include heterogeneous samples or lack controlled designs, but, with these limitations in mind, results are encouraging (reviewed by Richdale and Wiggs35). Delivering behavioural therapy in a brief, booklet form (with obvious economic implications) has been found to be as successful as behaviour therapy delivered face-to-face in managing the sleeplessness problems of infants with intellectual disabilities.36
Although behaviour therapy delivered in conventional face-to-face format may have apparent drawbacks in terms of the time and cost of implementing the interventions, the potential to prevent long-standing sleeping difficulties and their associated problems (which, as outlined above, may include adult mental health problems) is likely to far outweigh the limitations in both economic and social terms, at both the societal and personal level. Ensuring that families can access appropriate support and advice should be a service priority, and investigation of how to improve access (eg, with brief forms of treatment, low-cost delivery methods, identifying active therapeutic components and their efficacy for particular groups of children and different sleep disorders) should be key research targets for the future. It is hoped that this review reminds paediatric health professionals of the need to consider, broadly, the behavioural aspects of children’s sleep.
Competing interests: None.