Sleep_October.indd

SLEEP, Vol. 29, No. 10, 2006

1263

1. INTRODUCTION

BEDTIME PROBLEMS AND FREQUENT NIGHT WAKINGS

ARE HIGHLY PREVALENT IN YOUNG CHILDREN, OC-

CURRING IN APPROXIMATELY 20% TO 30% of infants, tod-

dlers, and preschoolers.

1-7

In addition, longitudinal studies have

demonstrated that sleep problems first presenting in infancy may

persist into the preschool and school-aged years and become

chronic.

8-11

Furthermore, the impact of disturbed and inadequate

sleep in young children can be both significant and extensive.

There is increasing evidence that sleep disruption and/or insuf-

ficient sleep has deleterious effects on children’s cognitive de-

velopment (e.g., learning, memory consolidation, executive func-

tion), mood regulation (e.g., chronic irritability, poor modulation

of affect), attention, and behavior (e.g., aggressiveness, hyperac-

tivity, poor impulse control), as well as health (e.g., metabolic

and immune function, accidental injuries) and overall quality of

life.

13-16

In addition, studies have documented secondary effects

on parents

(e.g., maternal depression), as well as on family func-

tioning.

17-19

Finally, the economic burden related to healthcare

costs for sleep problems in infants and young children has been

estimated to be considerable.

20,21

A number of treatment strategies

for bedtime behavior problems and night wakings in children ex-

ist, including behavioral management techniques, parent educa-

tion, and medication. In contrast to the paucity of data that exists

regarding pharmacologic treatment,

22-24

there is now a solid body

of literature supporting empirically based behavioral treatments

of bedtime problems and night wakings in infants, toddlers, and

preschoolers. In addition, studies have also demonstrated that

these strategies, compared to pharmacological treatments, are of-

ten more effective, may be more acceptable to both parents and

practitioners,

25-28

and avoid potential harmful side effects associ-

ated with medication use. Behavioral sleep management strate-

gies have the further advantage of potentially generalizing to the

management of daytime issues.

Given the impact of sleep disturbances in infants and young

children, and the availability of empirically supported treatment

strategies, the development of clinical guidelines for the manage-

ment of bedtime resistance and night wakings in young children is

important and necessary. As the basis for developing those clini-

cal guidelines, and building on several previous thorough reviews

of empirically-based non-pharmacologic treatments of behavioral

insomnias of childhood,

29-31

we present an updated critical sum-

mary of the current literature. A brief discussion of the issues per-

taining to the definition and diagnosis of behavioral insomnia of

childhood is included.

1.1 Definition of Disorder and Prevalence

To clarify the definitions used in this review, it is important

to make a distinction between the clinical diagnoses applicable

to bedtime problems and night wakings in children, and the

re-

search definitions used in studies of children with these sleep

problems. First, within the clinical realm, the 1997 International

Classification of Sleep Disorders

separates bedtime problems

Behavioral Treatment of Bedtime Problems and Night Wakings in Infants and

Young Children

An American Academy of Sleep Medicine Review

Jodi A. Mindell, PhD

1,4

; Brett Kuhn, PhD

; Daniel S. Lewin, PhD

; Lisa J. Meltzer, PhD

; Avi Sadeh, DSc

Department of Psychology, Saint Joseph’s University, Philadelphia, PA;

University of Nebraska Medical Center, Omaha, NE;

Children’s National

Medical Center, George Washington University School of Medicine, Washington, DC;

Children’s Hospital of Pennsylvania, Philadelphia, PA;

Depart-

ment of Psychology, Tel Aviv University, Tel Aviv, Israel

Review of Bedtime Problems in Children—Mindell et al

The term “parents” is used throughout the paper for stylistic reasons to denote

any type of guardian or caregiver (e.g., grandparent).

Disclosure Statement

This was not an industry supported study. Drs. Mindell and Sadeh serve as

consultants for Johnson & Johnson. Drs. Kuhn, Lewin, and Meltzer have

indicated no ﬁ nancial conﬂ icts of interest.

Address correspondence to: Jodi A. Mindell, PhD, Department of Psychol-

ogy, Saint Joseph’s University, 5600 City Avenue, Philadelphia, PA 19131;

Tel: (610) 660-1806; Fax: (610) 660-1819; E-mail: [email protected]

Abstract: This paper reviews the evidence regarding the efﬁ cacy of be-

havioral treatments for bedtime problems and night wakings in young

children. It is based on a review of 52 treatment studies by a task force

appointed by the American Academy of Sleep Medicine to develop prac-

tice parameters on behavioral treatments for the clinical management of

bedtime problems and night wakings in young children. The ﬁ ndings in-

dicate that behavioral therapies produce reliable and durable changes.

Across all studies, 94% report that behavioral interventions were efﬁ ca-

cious, with over 80% of children treated demonstrating clinically signiﬁ -

cant improvement that was maintained for 3 to 6 months. In particular,

empirical evidence from controlled group studies utilizing Sackett criteria

for evidence-based treatment provides strong support for unmodiﬁ ed ex-

tinction and preventive parent education. In addition, support is provided

for graduated extinction, bedtime fading/positive routines, and scheduled

awakenings. Additional research is needed to examine delivery methods

of treatment, longer-term efﬁ cacy, and the role of pharmacological agents.

Furthermore, pediatric sleep researchers are strongly encouraged to de-

velop standardized diagnostic criteria and more objective measures, and

to come to a consensus on critical outcome variables.

Keywords: Bedtime problems, night wakings, behavioral insomnia of

childhood, treatment, behavioral treatment

Citation: A Review by Mindell JA, Kuhn B, Lewin DS et al. Behavioral

treatment of bedtime problems and night wakings in infants and young

children. SLEEP 2006;29(10):1263-1276.

PEDIATRIC SLEEP

Closer to our Customers…

(but don’t just take our word for it.)

” Our relationship with Embla, formerly Medcare, has been

mutually beneﬁcial both educationally and professionally.

Every piece of our future

30 bed facility will be networked

together

and they have helped us towards our pending 1st

CoA PSG Accredited sleep program”.

– Andy Desrosiers, Holy

Family Hospital , Massachusetts, USA

“The Embletta is a powerful diagnostic tool. It

provides a wealth of information and a great

deal of detail. At the same time it is easy to

learn and simple to use with excellent reliability.

Once we received our ﬁrst Embletta, we stored our

other devices on the shelf”. – Paul R. Murphy, RPSGT,

Sahlgrenska University Hospital, Gotaborg, Sweden

“ At Scansleep,

we use

Embla, Embletta,

Somnologica

and Enterprise in all our clinics.

All of the hardware has been extremely stable and user-

friendly. We are also enthusiastic users of the software

applications”. – Soren Berg, M.D. PhD, Copenhagen, Denmark

“ The Xactrace Belts are the best

way to measure breathing effort

and are smoothly interfaced with

the Rembrandt Sleep system”.

– Jim Wilcox, RPSGT/CRT,

All Children’s Hospital,

St. Petersburg, Florida, USA

“I have been using Embla, formerly Medcare, equipment since

2001. Starting out with 2 beds, we are currently running 10

beds. I have used Artisan, Monet and the Embla

N7000, and have been amazed by their

reliability and ease of operation. Scoring

and analysis modules are friendly and

ﬂexible and I would highly recommend

all of the products we have used from Embla”. –

Michael D. McDannold, RPSGT,CRT

Vermont Medical SDC, Vermont, USA

To learn how we can get closer to you too, please call

us at 1.888.662.7632 or visit us at www.embla.com

Global Headquarters:

11001 W. 120th Ave., Broomﬁeld, CO 80021

PH: 303.962.1800 FX: 303.962.1810

www.embla.com www.shopembla.com

a global leader in sleep diagnostic systems

1EM101_ad5_8-15.indd 1 8/15/06 2:31:44 PM

SLEEP, Vol. 29, No. 10, 2006

1264

and night wakings into two distinct diagnostic categories: Sleep

Onset Association Disorder and Limit Setting Sleep Disorder. The

most recent revision of the International Classification of Sleep

Disorders

uses similar terminology, but subsumes both of these

clinical entities under the new clinical diagnostic category of Be-

havioral Insomnia of Childhood, which is further classified as

sleep-onset association type, limit-setting type, or combined type.

From a clinical standpoint, it should also be emphasized that the

diagnostic criteria for a sleep disorder require a specific constel-

lation of symptoms of a defined severity level to be present for

a specified time and to result in some significant impairment in

functioning either in the child or in the parent(s) or family. As

with all psychiatric disorders, mild and transient symptoms do not

necessarily constitute a sleep disorder. Bedtime problems, primar-

ily seen in children 2 years of age and older, include bedtime stall-

ing and bedtime refusal. Bedtime refusal behaviors are typically

described as stalling, verbal protests, crying, clinging, refusing

to go to bed, getting out of bed, attention-seeking behaviors, and

multiple requests for food, drinks, and stories (“curtain calls”).

This constellation of sleep behaviors generally falls within the di-

agnostic category of behavioral insomnia of childhood, limit-set-

ting type, in which parents demonstrate difficulties in adequately

enforcing bedtime limits (e.g., inconsistent or inappropriate bed-

time for the child’s age, conceding to multiple requests for atten-

tion after bedtime). In general, night wakings fall within the diag-

nostic category of behavioral insomnia of childhood, sleep onset

association type, with most children relying on sleep onset as-

sociations (e.g., rocking, feeding, parental presence) to fall asleep

at bedtime. During the course of normal nighttime arousals, these

children are then unable to recreate this sleep association, requir-

ing parental assistance to return to sleep.

Night wakings are typi-

cally viewed as problematic by caregivers only when they involve

“signaling” (e.g., accompanied by crying, protesting, or getting

out of bed), and are frequent and/or prolonged.

It should be noted, however, that essentially no empirical stud-

ies of “sleep problems” in children have utilized these specific

clinical definitions. Rather, intervention studies have employed a

number of different research criteria (see below) that are closely re-

lated to criteria for defining a sleep disorder but do not completely

parallel the diagnostic criteria to define “problematic” sleep onset

and maintenance-related behaviors. In addition, because bedtime

resistance and frequent night wakings commonly coexist, thus are

often “lumped” together for the purposes of defining inclusion

criteria for studies and assessing treatment outcomes.

Further-

more, most studies do not distinguish between bedtime resistance

and delayed sleep onset, which although often associated are not

always interchangeable in terms of etiology or treatment.

In any discussion of research definitions of sleep problems in

children, it should be noted that defining a sleep disorder in chil-

dren, compared to adults, is more complex and challenging for

several reasons. First, virtually all behavioral problems in young

children, including bedtime problems and night wakings, are de-

fined primarily by caregivers, and thus the definition is influenced

by a host of variables, including parent education level, parental

psychopathology, family dynamics, household composition, and

parenting styles. Even those studies that have utilized a strict “re-

search definition” of sleep problems have relied largely on parent-

report data, which are subject to a number of reporting biases. The

definition of these sleep problems may also be developmentally

based, namely transient problems that can be understood in the

context of normal physical, cognitive, and emotional changes oc-

curring at various developmental stages. Furthermore, parental

recognition and reporting of sleep problems in children also var-

ies across childhood, with parents of infants and toddlers more

likely to be aware of sleep concerns than those of school-aged

children and adolescents. In addition, culturally-based values and

beliefs regarding the meaning, importance, and role of sleep in

daily life, as well as culturally-based differences in sleep prac-

tices (e.g., sleeping space and environment, solitary sleep vs. co-

sleeping, use of transitional objects) have a profound effect not

only on how a parent defines a sleep “problem” but on the relative

acceptability of various treatment strategies. However, although

it is clear from the above discussion that bedtime problems and

night wakings are defined by a number of subjective complaints

arising from parent’s perception of behavior as well as the effects

of sleep disruption (e.g., irritability and inattention), it should be

emphasized that this is also the case for other broadly accepted

childhood psychiatric disorders (e.g., oppositional defiant disor-

der, enuresis, attention-deficit/hyperactivity disorder).

While the research criteria used in the literature to define bed-

time problems and night wakings are not consistent across stud-

ies, a number of researchers have attempted to operationalize and

standardize the definition of sleep problems in infants and young

children. These definitions generally include parameters related

to some combination of frequency (e.g., number of wakings per

night, nights per week with bedtime resistance), severity (e.g., du-

ration of night wakings), and chronicity (e.g., weeks to months)

of behaviors. For the purposes of this review, we have attempted

to be consistent with the current existing literature, using the no-

sology of bedtime problems/resistance and night wakings to refer

to “sleep problems” in infants and young children.

1.2 Prevalence

The identified prevalence of “problematic” bedtime resistance

and frequent night wakings is remarkably similar across studies,

even when comparing studies across cultures. It is estimated that

overall 20% to 30% of young children in cross-sectional studies

are reported to have significant bedtime problems and/or night

wakings.

1-5

For infants and toddlers, night wakings are one of

the most common sleep problems, with 25% to 50% of children

over the age of 6 months continuing to awaken during the night.

However, because these 2 sleep complaints frequently co-exist

and similar treatments strategies may be used for both, many

studies do not approach them as separate concerns and thus indi-

vidual prevalence rates are difficult to estimate.

7,34

1.3 Etiology

The etiology of bedtime resistance and night wakings in child

hood involves a multifactorial pathophysiologic mechanism and

represents a complex combination of biological, circadian, and

neurodevelopmental factors that are influenced by, but not solely

attributable to, environmental and behavioral variables (such as

sleeping arrangements and parenting styles).

22,35

Thus, bedtime

resistance and night wakings in childhood may be viewed as aris-

ing within a similar paradigm as psychophysiological insomnia in

adults, and involve predisposing, precipitating, and perpetuating

factors. The predisposing factors for these problems are grounded

in circadian and homeostatic perturbations that form the neuro-

biological substrate upon which these sleep problems are super-

Review of Bedtime Problems in Children—Mindell et al

SLEEP, Vol. 29, No. 10, 2006

1265

imposed. The inability to “sleep through the night” and “settling”

problems at bedtime/failure to “self-soothe” after night wakings

essentially represent a delay in the emergence or a regression of

behaviors associated with the neurodevelopmental processes of

sleep consolidation and sleep regulation, respectively, that occur

over the first few years of life.

3,4,36-38

Although the evolution of

sleep consolidation and sleep regulation in childhood is governed

principally by maturation of neural and circadian mechanisms,

like many other neurodevelopmental processes (e.g., emergence

of language, bowel and bladder control), it is also influenced by

the context and environment in which they occur.

37,39,40

Thus, these

sleep problems by definition involve some elements of learned

behavior that are then amenable to modification by behavioral

strategies.

The precipitating and perpetuating factors associated with bed-

time resistance and night wakings are myriad, and include both

extrinsic (e.g., environmental situations, parental issues) and in-

trinsic (e.g., temperament, medical issues) factors and often rep-

resent a combination of these issues. Bedtime problems are often

associated with child temperament or challenges related to calm-

ing a child.

41-45

For example, “fussy” children may insist on a par-

ticular type of soothing/sleep-inducing technique, resisting any

alternative that is less dependent on the caregiver. Some caregiv-

ers may have problems of their own (e.g., depression, alcoholism,

long work hours) that interfere with their ability to set clear limits

both during the day and at bedtime. Caregivers of children with

current medical issues, or a history of a serious illness, may also

have difficulty setting limits, due to guilt, a sense that the child

is “vulnerable,” or concerns about doing psychological harm to

the child. Furthermore, other sleep disorders such as obstructive

sleep apnea have been shown to be associated with increased bed-

time behavior problems.

46,47

In other cases, there is a “mismatch”

between parental expectations regarding sleep behaviors and the

normal developmental trajectory. Finally, environmental factors,

such as living accommodations that require a child to share a bed-

room with a sibling, parent, or additional family members (e.g.,

grandparents) residing in the home, may also contribute to poor

limit setting or negative sleep onset associations. Caution, though,

must be exercised in the interpretation of some of these factors.

For example, sleep proximity within the home and parent expecta-

tions may be determined by cultural, ethnic, and socio-economic

differences.

1.4 Impact

The clinical impact of bedtime resistance and night wakings

usually involves identifiable alterations in an infant or child

’s

behavior. However, any discussion of the significance of pedi-

atric sleep problems must also underscore the importance of the

relationships between sleep problems and mood, development,

learning, performance, and health. A wealth of empirical evi-

dence clearly indicates that significant performance impairments

and mood dysfunction are associated with daytime sleepiness re-

sulting from insufficient or interrupted sleep.

16,48,49

Higher-level

cognitive functions regulated by the prefrontal cortex, such as

cognitive flexibility and the ability to reason and think abstractly,

appear to be particularly sensitive to the effects of disturbed, in-

sufficient, and/or irregular sleep.

16,50-52

Furthermore, these sleep

problems appear to be an important precursor and potential early

indicator of future anxiety, depression, and substance use disor-

ders.

49,53-55

Sleep problems also place a significant burden on par-

ents and the parent-child relationship. Finally, health outcomes

of inadequate sleep include potential deleterious effects on the

cardiovascular, immune, and various metabolic systems, includ-

ing glucose metabolism and endocrine function.

2.0 PURPOSE

The primary objective of this paper is to provide a review of

the empirical evidence regarding the efficacy of behavioral in

terventions for the clinical management of bedtime problems

and night wakings in infants and children. Secondary objectives

include an evaluation of the impact of behavioral interventions

on the child and parent and the durability of outcomes (short-

term and long-term). The primary interventions reviewed here

are standard behavioral treatment techniques that include: 1) ex-

tinction (unmodified extinction, Graduated Extinction, extinction

with parental presence); 2) positive bedtime routines/faded bed-

time with response cost; 3) scheduled awakenings; and 4) parent

education/prevention.

3.0 METHODS

3.1 Identification and Selection of Treatment Studies

Treatment studies selected for review in this paper were identi

fied through PsycLIT and MEDLINE searches (1970-2005) using

the following keywords: (1) sleep problem-disorder- disturbance-

disruption-patterns-sleeplessness; (2) bedtime problems-resis-

tance-struggles-refusal-tantrums; (3) dyssomnias-insomnia; (4)

limit setting sleep disorder-settling problems; (5) night waking-

nighttime awakenings-sleep onset association disorder; (6) treat-

ment-intervention-management-nonpharmacological-cognitive

behavioral-parent training- parenting- mother-infant interaction-

anticipatory-guidance-prevention-primary-care intervention; (7)

children-infant-toddler-infancy-pediatric-babies-newborn-pre-

school.

The criteria for inclusion of a study were as follows: (a) study

included any child between the ages of 0 - 4 years 11 months (old-

er children included in any study were excluded from the analy-

ses; most studies including older children were case reports and

single-case designs); (b) intervention study of any behavioral or

psychoeducational treatment that involved behavioral principles;

and (c) focus was on bedtime problems, night wakings, or a be-

haviorally-based sleep problem (all other sleep disorders were ex-

cluded, including parasomnias and nightmares). Exclusion crite-

ria included: (a) no behavioral intervention or behaviorally-based

psychoeducational component, (b) sleep problem associated with

a primary medical or psychiatric condition (including known

developmental disabilities), and (c) study was not published in

a peer-reviewed publication, such as a dissertation. All types of

studies, including case studies and single-subject designs, were

included in the analyses.

A total of 3,008 abstracts were considered from the initial

search that included all articles published through January 2005.

The large majority of these were excluded because they did not

meet inclusion criteria, with 92 articles selected for full review.

Following full review, 35 articles were excluded primarily be-

cause the study population included children with developmen-

tal disabilities or the treatment was exclusively pharmacological.

“Pearling,” the process of manually scanning the captured arti-

Review of Bedtime Problems in Children—Mindell et al

SLEEP, Vol. 29, No. 10, 2006

1266

cles’ bibliography for additional relevant references not detected

by Medline, netted an additional 5 citations.

The present paper is based on evidence from 52 individual

studies (n > 2,500 subjects) that met inclusion criteria; these stud-

ies are denoted by an asterisk in the reference list. Each article

was reviewed and rated by 2 task force members. Any disagree-

ments were resolved by discussion and consensus among task

force members.

3.2 Treatment Procedures: Description and rationale

Interventions for bedtime problems and night wakings consist

primarily of time-limited parent training strategies that incorpo

rate behaviorally-based interventions, founded on principles of

learning and behavior (e.g., reinforcement, extinction, shaping).

Parent training typically involves a therapist “coaching” the par-

ents to become the active agents of change to address their child’s

problematic sleep patterns, habits, or sleep-related behaviors.

Among the many forms of behavioral health services for young

children, no other treatment has been more thoroughly investi-

gated or widely applied as parent management training.

Extinction

The first studies that were conducted on the treatment of early

childhood sleep problems focused on the use of extinction

. Un-

modified extinction procedures for sleep problems involve having

the parents put the child to bed at a designated bedtime and then

ignoring the child until a set time the next morning (although par-

ents continue to monitor for illness, injury, etc). Behaviors that are

ignored include crying, tantrums, and calling for the parents. Ex-

ceptions to ignoring the child include any concerns that the child

is hurt, ill, or in danger. The biggest obstacle associated with ex-

tinction is lack of parental consistency. Parents must ignore their

child’s cries every night, no matter how long it lasts. If parents

respond after a certain amount of time, the child will only learn

to cry longer the next time. Parents are also instructed that post-

extinction response bursts may occur. That is, often at some later

date there is a return of the original problematic behavior. Parents

are instructed to avoid inadvertently reinforcing this inappropriate

behavior following such an extinction burst. The common term

used in the media and self-help books to describe unmodified ex-

tinction techniques is the “cry it out” approach.

The major drawback of unmodified extinction procedures is

that it is stressful for parents. Many parents are unable to ignore

crying long enough for the procedure to be effective. As a variant

to unmodified extinction, some studies have utilized extinction

with parental presence. This procedure involves the parents stay-

ing in the child’s room at bedtime but ignoring the child and his/

her behavior. Some parents find this approach more acceptable

and are able to be more consistent.

Graduated Extinction

Rather than having the child cry for extended periods,

Grad-

uated Extinction procedures have been developed. The term

“Graduated Extinction” refers to a variety of techniques. Typi-

cally, parents are instructed to ignore bedtime crying and tantrums

for specified periods. The duration or interval between check-ins

with the child is often tailored to the child’s age and temperament,

as well as the parents’ judgment of how long they can tolerate the

child’s crying. Either parents can employ a fixed schedule (e.g.,

every 5 minutes) or they can wait progressively longer intervals

(e.g., 5 minutes, 10 minutes, then 15 minutes) before checking on

their child. With incremental Graduated Extinction, the intervals

increase across successive checks within the same night or across

successive nights. The checking procedure itself involves the par-

ents comforting their child for a brief period, usually 15 seconds

to a minute. The parents are instructed to minimize interactions

during check-ins that may reinforce their child’s attention-seek-

ing behavior.

The goal of Graduated Extinction is to enable a child to de-

velop “self-soothing” skills in order for the child to fall asleep in-

dependently without undesirable sleep associations (e.g., nursing,

drinking from a bottle, rocking by parent). Once these skills are

established, the child should be able to independently fall asleep

at bedtime and return to sleep following normal nighttime arous-

als. In the popular literature, this type of intervention is often re-

ferred to as “sleep training.”

Positive Routines/Faded Bedtime with Response Cost

Positive routines involve the parents developing a set bedtime

routine characterized by quiet activities that the child enjoys.

Faded bedtime with response cost involves taking the child out

of bed for prescribed periods of time when the child does not fall

asleep. Bedtime is also delayed to ensure rapid sleep initiation

and that appropriate cues for sleep onset are paired with posi

tive parent-child interactions. Once the behavioral chain is well

established and the child is falling asleep quickly, the bedtime is

moved earlier by 15 to 30 minutes over successive nights until a

pre-established bedtime goal is achieved. A scheduled wake time

Review of Bedtime Problems in Children—Mindell et al

Table 1

— Diagnostic Criteria of Behavioral Insomnia of Childhood

A. A child’s symptoms meet the criteria for insomnia based upon

reports of parents or other adult caregivers.

B. The child shows a pattern consistent with either the sleep-onset

association type or limit-setting type of insomnia described be

low:

i. Sleep-onset association type includes each of the following:

1. Falling asleep is an extended process that requires special

conditions.

2. Sleep-onset associations are highly problematic or demand

ing.

3. In the absence of the associated conditions, sleep onset is

signiﬁcantly delayed or sleep is otherwise disrupted.

4. Nighttime awakenings require caregiver intervention for the

child to return to sleep.

ii. Limit-setting type includes each of the following:

1. The individual has difﬁculty initiating or maintaining sleep.

2. The individual stalls or refuses to go to bed at an appropriate

time or refuses to return to bed following a nighttime awak

ening.

3. The caregiver demonstrates insufﬁcient or inappropriate

limit setting to establish appropriate sleeping behavior in the

child.

C. The sleep disturbance is not better explained by another sleep dis-

order, medical or neurological disorder, mental disorder, or medi

cation use.

American Academy of Sleep Medicine. The International Classiﬁ

cation of Sleep Disorders, 2nd ed.: Diagnostic and Coding Manual.

Westchester, IL: 2005.

SLEEP, Vol. 29, No. 10, 2006

1267

is established and daytime sleep is not allowed, with the exception

of age-appropriate naps.

These two strategies are similar in that they match the child’s

bedtime with his/her natural sleep onset time and rely heavily

on stimulus control techniques as the primary agent of behavior

change. Both treatments aim to increase appropriate behaviors

and control of affective and physiological arousal, rather than fo-

cusing on reducing inappropriate behaviors, as is done with the

previously described extinction strategies.

Scheduled Awakenings

Scheduled awakenings involve parents awakening and consol

ing their child approximately 15 to 30 minutes before a typical

spontaneous awakening. This strategy begins with establishing a

baseline of the number and time of spontaneous nighttime awaken-

ings. Preemptive awakenings are then scheduled. Parent-induced

scheduled awakenings are typically followed by the parents’ usual

response to a spontaneous awakening, such as rocking or nursing

the child back to sleep. Scheduled awakenings are then faded out,

by systematically increasing the time span between awakenings.

These scheduled awakenings appear to increase the duration of

consolidated sleep.

Parent Education/Prevention

One approach to treatment of sleep disturbances is to prevent

their occurrence. A number of behavioral interventions have been

incorporated into these parent education programs, with a focus

on early establishment of positive sleep habits. Strategies typical

ly target bedtime routines, developing a consistent sleep schedule,

parental handling during sleep initiation, and parental response

to nighttime awakenings. Almost all programs have incorporated

the recommendation that babies should be put to bed “drowsy but

awake” to help them develop independent sleep initiation skills at

bedtime, and enabling them to return to sleep without intervention

following naturally occurring nighttime arousals.

Many parent education programs have targeted soon-to-be-par-

ents, as well as parents of newborns. For this review, preventive

education was designated for parent education that was conducted

during the prenatal period or during the first 6 months. This strat-

egy focuses on a prevention model rather than an intervention

model, as denoted by the above behavioral treatments. In contrast,

general parent education was defined as occurring after 6 months

of age and involved provision of information about normal sleep.

4.0 SUMMARY OF RESULTS

Table 2 (which can be accessed on the web at http://www.jour

nalsleep.org/), summarizes the 11 studies included in the pres-

ent review that meet Sackett evidence Levels I and II, based on

criteria described in the method section. The following section

summarizes the magnitude of changes obtained on infant/toddler

behavior (e.g., bedtime crying) and sleep parameters, the clini-

cal significance of those changes, the durability of improvements

over time, and the comparative efficacy of single and combined

treatments.

4.1 General findings

More than 2,500 infants and toddlers participated across the 52

selected studies that evaluated behavioral interventions for bed

time struggles and frequent night waking. Nearly half of the sub-

ject pool (n=1,135) participated in the methodologically strongest

studies employing a randomized controlled trial (RCT) design.

In the 40 studies that identified the gender of the subjects, 760

out of 1359 subjects were male (56%). Thirty-six studies provid-

ed the mean age of the subjects. The average age of the subjects

in these studies was 20 months. The age range of the total pool

of participants spanned from 1 week to 10 years (although only

participants under 5 years of age were considered in this paper).

Seven studies indicated race; 67% of the 858 of subjects in

these 7 studies were Caucasian. The remaining 33% of subjects,

in those studies in which race was identified, were African-Amer-

ican, Asian, or the study did not provide this information. Nine

teen studies were conducted in the United States, 10 in England,

7 in Australia, and 4 in New Zealand, with the remaining taking

place in Canada, Iceland, Israel, Scotland, Sweden, Switzerland,

or other European countries.

Of the total participant pool, 731 subjects (29%) across 28

studies were clinically referred by a professional (n=579) or self-

referred (n=152) for sleep problems. A significant percentage of

children (52%) across 14 studies were recruited specifically to

participate in the research study, often during routine medical ap-

pointments or by posting community advertisements. It should be

noted that 7 of the 9 studies in Evidence Level I recruited their

research participants, whereas studies using small “n” multiple

baseline designs were more likely to involve participants who

were clinically referred.

Table 3 summarizes the guidelines by which the classification

of evidence was evaluated, as adapted from Sackett.

Of the 52

selected studies, 8 (15%) represented RCTs that were classified

as Level I.

17,18,20,61-65

Three studies (6%) were classified as Level

II.

66-68

Twenty-six (50%) met criteria as a Level III study,

7,19,21,69-

many of which used a multiple-baseline research design. The

remaining 15 (29%) fell into Levels IV

92-94

or V.

57,95-103

These 52 studies assessed the efficacy of a number of behavior

al interventions that varied greatly in procedural delivery. Despite

these differences, most of the interventions can be placed into the

following categories: extinction and its variants (i.e., unmodified

extinction, extinction with parental presence, Graduated Extinc

tion), positive bedtime routines, scheduled awakenings, bedtime

fading with response-cost, Positive Reinforcement, and parent

education/prevention.

This empirical literature includes a wide range of outcome vari

ables. Many researchers collected data on sleep-related variables

(e.g., sleep onset latency, frequency and duration of awakenings,

total sleep time), whereas others focused more on child behav-

ior outcomes (e.g., duration of crying, frequency of leaving the

bedroom, and callouts to parents). Behavioral and sleep related

variables were both measured in only a few studies; some major

studies included no child outcome variables, choosing instead to

focus on parent sleep or emotional adjustment.

Among the studies summarized, 11% identified bedtime re-

sistance as the primary dependent variable,

17,57,76,86,88,103

whereas

frequent awakenings were the main focus in 27% of the stud

ies.

19,21,63,64,66,72,75,78,79,82,89,92,97,104

Exactly 50% of the studies tar-

geted the “clinical dyad” of bedtime resistance and night wak-

ing.

7,20,61,62,65,67-71,73,77,80,81,85,87,90,91,93-96,98,101,102,105

One study was

unique in that it targeted bedtime fears,

whereas 5 studies ad-

dressed nonspecific “sleep problems.”

17,18,84,99,100

Consistent with

previous reviews (e.g.,

5,29,31

), as discussed above, we chose to

consider bedtime disturbance and frequent night waking together.

The 2 sleep disturbances frequently coexist,

and treatments that

Review of Bedtime Problems in Children—Mindell et al

SLEEP, Vol. 29, No. 10, 2006

1268

target 1 often generalize to the other.

4.2 Specific behavioral interventions

In his 1959 study, Williams

appears to have been the first to

formally apply Unmodified Extinction to problematic bedtime

behavior. Extinction has a strong record of accomplishment, now

having been evaluated in 19 separate research studies involving

552 participants.

With the exception of 2 studies

68,85

, in 17 stud-

ies the procedure has proven highly effective in eliminating bed-

time problems and night wakings, and improving sleep continu-

ity.

57,62,63,67,71,74-76,87-89,96,97,99-101,103

Graduated Extinction was first devised by Rolider and Van-

Houten

as a more parent-friendly alternative to Unmodified

Extinction. The protocol was modified slightly and popularized

by Ferber in his 1985 self-help book

, as well as by Douglas in

1989

106

. This variation on the extinction theme has now been eval-

uated in 14 studies and 748 participants. All 14 reported positive

treatment outcomes as indicated by a reduction in bedtime prob-

lems and/or night wakings.

7,16-19,21,62,72,73,81,84,86,99,105

An additional 5

studies relied on the same underlying behavioral principle (i.e.,

gradual removal of parental attention or physical proximity) with-

out using the formal checking procedure outlined in the original

protocol.

23,68,90,96,100

It appears safe to conclude that Graduated Ex-

tinction, as applied to bedtime problems and night waking, now

stands on equal empirical footing as its predecessor.

Extinction with Parental Presence is a more recent variant of

extinction. The procedure is more popular in England but appears

to be making its way to the U.S. Four research studies involving

290 children found the procedure to be effective.

18,84,98,104

Positive Routines were first used by Milan in 1981

107

to address

bedtime tantrums of three children with severe handicaps. Two

studies

17,77

have since evaluated the protocol with 81 typically de-

veloping children, and both concluded that the procedure is rapid

and effective. Positive Routines provides a positive, albeit less

tested, alternative to extinction that may reduce the undesirable

post-extinction response burst that many parents have difficulty

tolerating.

Scheduled Awakenings was first described by McGarr and

Hovell in 1980

, then more formally evaluated by a series of

three studies by Johnson and colleagues.

63,78,79

Forty-four children

have participated across 4 studies. The outcome data indicate

that Scheduled Awakenings afford another treatment option for

frequent nighttime awakenings. Compared with extinction, the

procedure is slightly more complicated to carry out, and studies

suggest that results may take several weeks rather than several

days. Furthermore, scheduled awakenings are not an appropriate

treatment for young children with bedtime struggles.

Having an infant or young child participate in a nightly Standard-

ized Bedtime Routine has become a universal, “common sense”

recommendation. This intervention component was included in no

fewer than 14 of the selected studies.

28,62,67,68,70,75,84,88,89,96,97,99,101,105

However, it was always included as part of a multi-component

treatment package, and has yet to be systematically evaluated as

a stand-alone intervention. The same can be said for Positive Re-

inforcement, which was included as part of the treatment package

in 15 studies

28,62,67,68,70,83,87-89,95-97,99-101

yet was never evaluated as

the sole intervention.

Finally, outcomes from 5 large-scale studies provide evidence

that Parent Education/Prevention may set the standard as the

most economical and time efficient approach to behaviorally-

based pediatric sleep problems. More than 1,000 parents across

5 studies

20,61,65,66,92

have received sleep education and prevention

strategies during their prenatal period or the first 6 months of

infancy. Results have proven to be not only statistically signifi-

cant, but also clinically meaningful to parents who want to teach

their newborn essential sleep skills, although given that no stud-

ies have done follow-up longer than six months the durability of

effects is not yet established. For example, Pinilla

was able to

teach 100% of infants to “sleep through the night” by 8 weeks of

age, whereas only 23% of control infants accomplished this goal.

Wolfson

used only 4 sessions to help 72% of infants to “sleep

through the night” by 3 weeks post-birth, compared to 48% of

control infants. Prevention strategies afford the ability to impact

large numbers of infants and young children without a great deal

Review of Bedtime Problems in Children—Mindell et al

Table 3—AASM Classiﬁcation of Evidence

Recommendation Evidence Levels Study Design Studies

Grades

A I Large, well-designed, randomized, and blinded 9 studies:

17,18,20,61-65,104

controlled study with statistically signiﬁcant

conclusions on relevant variables

B II Smaller, well-designed, randomized and blinded, 4 studies:

66-68,105

controlled study with statistically signiﬁcant

conclusions on relevant variables

C III Well-designed, non-randomized prospective 26 studies:

7,19,21,69-88,90,91,101

study with control group

C IV Well-designed, large prospective study with 3 studies:

23,92,94

historical controls or careful attention to

confounding effects or small prospective study

with control group

C V Small prospective study or case series without 10 studies:

57,95-103

control groups

Adapted from Sackett

Total: 52 studies

The number of subjects represent the total number participating in a given study,

not necessarily the number of subjects who received that particular intervention

component.

SLEEP, Vol. 29, No. 10, 2006

1269

of monetary or time investment. Adair

was able to reduce fre-

quent night waking by half simply by incorporating written in-

formation regarding sleep habits and behavior management into

2 routine well-child medical visits. One potential disadvantage of

large scale, less personalized interventions is that parents may not

implement the treatment as intended, or at all. St. James-Roberts

incorporated Pinilla’s successful treatment package into an educa-

tional brochure, and attained only a modest increase (10%) in the

number of infants who slept through the night at 12 weeks of age.

Group comparison data indicated that the intervention group did

not implement the essential treatment components.

4.3 Overall Efficacy

Based on authors’ conclusions from their own data, 94% (49

of 52) reported that behavioral interventions produced clinically

significant reductions in bedtime resistance and night wakings.

Three studies reported equivocal findings,

64,68,85

and no study re-

ported detrimental effects. The percentage of participants who

improved on relevant outcome measures was reported in a few

studies. The average percentage of subjects who improved was

82% (range 10% - 100%), however, the timing of this determina-

tion varied considerably.

The 11 studies with the strongest research methodologies (Lev-

els I and II) evaluated the outcomes of 9 different behavioral in-

terventions, either alone, comparatively, or in combination. The

interventions most commonly evaluated in the strongest studies

included Unmodified Extinction (4 studies), Parent Education/

Prevention (4 studies), and Graduated Extinction (3 studies).

Standard Bedtime Routines (2 studies) and Positive Reinforce-

ment (2 studies) were also evaluated, but were always included

as part of a larger treatment package. Nine

17,18,20,61-63,65-67

of the 11

studies found positive intervention effects and 2

64,68

were equivo-

cal. Overall, the weight of the evidence from controlled group

studies supports two behavioral interventions: Extinction and Par-

ent Education/Prevention, with clear support for Graduated Ex-

tinction.

We conclude that infants and toddlers who exhibit bedtime

resistance and nighttime awakenings respond favorably to be

havioral interventions. Unmodified extinction and Parent Edu-

cation/Prevention are the two treatment modalities that have the

strongest empirical support. Graduated Extinction, bedtime fad-

ing/positive routines, and scheduled awakenings were also sup-

ported.

4.4 Comparative Efficacy of Treatment Modalities

The studies selected for this review varied greatly in methodol

ogies, therefore it may be difficult to compare “apples to oranges”

in selecting among available treatments. Studies used different

outcome variables and methods of assessment. Most involved

multi-component treatment packages, therefore few data are

available directly comparing one pure treatment to another. For

example, Reid

published an excellent study comparing Extinc-

tion to Graduated Extinction. Each intervention, however, also

included 2 other treatment components (door closing if the child

came out of the bedroom more than once and praise/rewards for

a successful night), making it more difficult to directly compare

interventions.

Despite the methodological differences, there are a few con-

clusions that can be drawn based on the handful of studies that

conducted head-to-head comparisons of 2 or more treatments. In

drawing these conclusions, only the impact on sleep-related vari-

ables were considered. One clear finding in these studies is that

children participating in an active behavioral sleep intervention

demonstrated more rapid and significant resolution to their sleep

disturbance than those who did not receive treatment.

17,62,63

These

findings support previous work suggesting that pediatric sleep

disturbances often become chronic, with few children outgrow-

ing the problem.

9,11

The direct comparison studies provide little evidence to sug-

gest that any 1 behavioral protocol is vastly superior to another.

Positive Routines, Unmodified Extinction, Graduated Extinction,

Extinction with Parental Presence, and Scheduled Awakenings

were all included in 1 or more comparison studies. All 5 studies

that directly compared behavioral treatments found no apprecia-

ble differences in long-term efficacy.

17,62,63,74,104

There is evidence

that Unmodified Extinction may produce faster improvement

than Scheduled Awakenings

, and that combining sedative medi-

cation (antihistamine) with Extinction may produce a more im-

mediate response with reduced infant distress.

Unmodified Extinction and its recent variants (Graduated;

with Parental Presence) appear to be on level playing ground,

along with Positive Routines. One study concluded that com-

pared to Graduated Extinction, “positive routines produced the

fastest improvement in decreasing the tantrum behavior.”

The

data, however, appeared equivalent until approximately week 4

of treatment when Positive Routines continued to produce ad-

ditional improvement as Extinction reached a plateau. Positive

Routines and a variant, Faded Bedtime, appear to provide prom-

ising alternatives to more traditional extinction-based protocols.

Although the 2 protocols were evaluated in only 3 of the selected

studies,

17,69,77

Faded Bedtime with/without response-cost has been

studied more extensively in children with developmental dis-

abilities. Notably, Positive Routines and Faded Bedtime closely

resemble a combination of 2 behavioral interventions (sleep re-

striction and stimulus control instructions) that have received the

strongest research support in the treatment of adult insomnia.

108

Review of Bedtime Problems in Children—Mindell et al

Table 4 —Frequency and Percent of Studies Reporting Durability of

Sleep Improvements

< 6 6-12 >12 No

months months months follow-up

Extinction 20 (59%) 8 (23%) 3 (9%) 3 (9%)

Methods (n=34)

Standardized Bedtime 11 (79%) 1 (7%) 1 (7%) 1 (7%)

Routine (n=14)

Positive Routines 1 (50%) 1 (50%)

(n=2)

Scheduled 3 (75%) 1 (25%)

Awakenings (n=4)

Bedtime Fading/ 1 (100%)

Response Cost (n=1)

Positive 10 (66.7%) 3 (20%) 2 (13.3%)

Reinforcement (n=15)

Education (n=15) 9 (60%) 1 (7%) 1 (7%) 4 (26%)

General Behavioral 2 (50%) 1 (25%) 1 (25%)

Treatment (n=4)

Other (n=12) 5 (41.6%) 5 (41.6%) 2 (16.7%)

Total (n=101) 61 (60%) 20 (20%) 5 (5%) 15 (15%)

SLEEP, Vol. 29, No. 10, 2006

1270

Three of the five direct comparison studies provided sufficient

original data that Kuhn and Elliott

were able to calculate treat-

ment effect sizes. Unmodified Extinction produced a larger effect

size (d = 2.31) than Scheduled Awakenings (d = 1.11) on number

of awakenings at 6 weeks post-treatment.

Effect sizes for fre-

quency and the duration of bedtime tantrums were comparable

for Graduated Extinction (d =0.75; duration= 1.50) and Positive

Routines (d =0.88; duration=1.83).

At 3 weeks post-treatment,

Unmodified Extinction produced a slightly larger effect size (d =

2.63) than Graduated Extinction (d = 1.93) on “good bedtimes”

(settled alone in less than 10 minutes), but the results were re-

versed for “good nighttimes” (slept through night without sleep-

ing with or waking parents) with Graduated Extinction (d = 2.03)

slightly outperforming Unmodified Extinction (d = 1.29).

Most

importantly, the effect sizes for all four interventions surpassed d

= 0.80, which reflects a large treatment effect.

109

Overall, these ef-

fect sizes indicate that Positive Routines, Unmodified Extinction,

Graduated Extinction, and to a lesser degree Scheduled Awaken-

ings, all represent effective treatment options for the treatment of

pediatric bedtime problems and frequent night waking.

4.5 Source of Outcome Assessment

Nine studies (17%) employed at least 1 objective outcome

measure such as direct observations, videotapes, audiotapes, or

actigraphy data.

7,21,69,84,85,88,97,103,104

Seven studies (13%) used a

standardized rating scale to assess outcomes related to child be-

havior or infant security.

62,66,70,74,83,84,88

However, the overwhelming

majority of intervention studies (77%) relied on parent complet-

ed daily diaries as the primary outcome measure. These diaries

most frequently assessed child sleep (56%), however a few (8%)

tracked bedtime behavior such as crying, tantrums, or leaving the

bedroom. Seven studies (13%) used diaries to collect data on both

child sleep and bedtime behavior.

21,65,70,71,83,86,103

Parent completed sleep diaries typically include daily record-

ings for nightly bedtime, time asleep, the number, timing, and du-

ration of any night wakings, the time of morning waking, and the

duration of any daytime sleep. Although there are some limitations

to parent-report measures, they are the most widely used mea-

sure of sleep in clinical settings and therefore tend to have high

content and face validity. For infants and toddlers, sleep related

complaints come from the parents rather than from the child

110

;

therefore parents are the most obvious source of information for

their child’s sleep behavior.

111

Parent completed sleep diaries

possess reasonable validity, high internal consistency, and good

agreement (> 90%) with video or voice activated recordings, and

actigraphic measures of children’s sleep-wake patterns.

7,42,75,82,112-

115

Parents of sleep disturbed infants have been shown to be good

reporters on sleep schedule measures, but do more poorly on sleep

quality measures.

116

4.6 Secondary Outcomes

A number of studies assessed the effects of sleep interven

tions on secondary outcome variables, such as daytime behav

ior. These studies addressed possible adverse effects of behav

ioral interventions, as well as the potential beneficial effects on

daytime behavior. A total of 13 studies selected for this review

reported results pertaining to child daytime functioning such as

crying, irritability, detachment, self esteem, or emotional well-

being.

17,28,62,66,70,73,74,83,84,88-90,101

Five studies

17,88-90,101

based their

conclusions solely on subjective retrospective parental report,

whereas 8 studies.

28,62,66,70,73,74,83,84

collected formal data such as

standardized rating scales or observations of parent-child interac

tions.

Adverse secondary effects as the result of participating in be

haviorally based sleep programs were not identified in any of the

studies. On the contrary, infants who participated in sleep inter

ventions were found to be more secure

74,117,118c

predictable,

less

irritable,

and to cry and fuss less following treatment.

Mothers

indicated that behaviorally-based sleep interventions had no ef

fect on maintaining the practice of breast feeding or on infant’s

total daily fluid intake.

66,92

In a number of studies, parents of older

children reported improvements in their children’s daytime be

havior after participation.

28,70,83,84,89,101

For example, Seymour

reported that 73% of parents reported positive changes in their

child’s daytime behavior. There are several potential mechanisms

to account for these findings, but 1 likely factor is the increased

total sleep time and improved sleep quality that children and their

parents experience following effective treatment.

It is important to indicate that sleep related behavioral inter

vention also led to improvement in the well-being of the parents

beyond the specific benefits in sleep patterns in the children.

Twelve studies collected outcome measures on parent mood,

stress, or marital satisfaction.

7,17-21,28,62,64,73,74,105

A few studies col-

lected data on fathers, however the majority focused on mothers

who tended to demonstrate elevated levels of depressed mood

and more disturbed sleep at pre-treatment, probably because they

assumed the most responsibility in caring for a sleep disturbed

infant or toddler. The results were remarkably consistent across

studies. Following intervention for their child’s sleep disturbance,

parents exhibited rapid and dramatic improvements in their over

all mental health status,

64,105

reporting fewer symptoms of depres-

sion.

7,18,19,21

They reported an increased sense of parenting effi-

cacy,

enhanced marital satisfaction,

7,17,21

and reduced parenting

stress.

62,117

For instance, Eckerberg

117

reported that following suc-

cessful implementation of a behavioral intervention that led to

significant improvement of their infant sleep, the parents reported

improvements of their own mood, stress level, and fatigue. Simi

larly, Hiscock

reported a 45% decrease in depression scores at

2 and 4 months post-treatment in depressed mothers after par

ticipating in a behavioral infant sleep program. The only factor

that predicted an increase in maternal depression scores was per

sistent infant sleep problems

. Another study reported that 70%

of participating mothers fell above the cutoff score for clinical

depression at baseline, but only 10% were still depressed fol

lowing intervention for their sleep disturbed infant.

Finally, in

a prevention study, Wolfson

provided sleep education to parents

before and after the birth of their infant. Parents who received

the sleep education reported feeling an increased sense of compe

tence, whereas parents in the control group reported higher stress

levels.

Given the strong association between chronic sleep distur

bance and risk for depression,

119,120

it is possible that the observed

reduction in parental depression is mediated by the improved

parental sleep patterns once infant and toddler sleep problems

are ameliorated. Three of the selected studies

7,21,95

collected sec-

ondary outcome data on parent sleep variables following child

Review of Bedtime Problems in Children—Mindell et al

France, 1992 and Eckerberg, 2004 were not selected to be included in this re-

view, however the outcome data from these studies were based on previous stud

ies that were selected.

SLEEP, Vol. 29, No. 10, 2006

1271

participation in a behavioral sleep intervention. One found only

minor improvement in parental sleep variables.

Mindell,

how-

ever, reported an 80% reduction in the frequency of parental night

waking and less time awake at 1 month post-treatment.

4.7 Durability of Sleep Improvements

As seen in Table 4, 85% of the studies reviewed in this paper

examined the maintenance of treatment effects over time. These

studies demonstrated that treatment related changes across most

types of interventions were maintained at short (< 6 months), inter

mediate (6 - 12 months) and long range follow-up (> 12 months).

Of the studies that reported improvements in either bedtime be-

haviors or a decrease in night wakings, 89% reported success in

all of the participants at follow-up; the other 11% reported contin-

ued treatment gains for over two-thirds of their participants.

In the majority of the studies (60%) the follow-up assessment

occurred less than 6 months after treatment ended. In fact, only

5% of the studies reviewed reported the maintenance of treatment

effects more than 1 year after the intervention ended. The durabil-

ity of treatment effects should thus be interpreted with caution.

5.0 TREATMENT INTEGRATION

5.1 Multi-Faceted Interventions

Two or more types of interventions, or multi-faceted interven

tions (MFI), were used in 58% (n=30) of the reviewed studies.

Concurrent implementation of multiple interventions was used in

the majority of these studies whereas multiple baseline designs, or

serial implementation of interventions, were reported in 30% (n

= 9) of the 30 MFI studies. Only 4 of these articles were graded

as evidence levels I or II,

20,62,65,68

and 16 were graded as evidence

level III. The remaining 10 were graded as evidence Levels IV or

The types of intervention strategies varied a great deal across

studies. The most common MFI interventions paired either Posi-

tive Reinforcement (n = 14) and/or Standard Bedtimes Routines

(n = 17) with either Extinction or Graduated Extinction. In 3 MFI

studies, there was reference to a behavioral intervention, but the

type of intervention was not specified. Other behavioral tech-

niques, such as somatic relaxation and deep breathing, time out,

punishment, response cost, and a children’s bedtime story, were

used in 1 or 2 studies each in combination with 1 or more of the

more common interventions.

Some form of parent education, either regarding sleep train-

ing (behavioral techniques, limit setting, elimination of noctur-

nal feeding) or general information about sleep (developmental

changes in sleep across the first several years of life) was used in

11 of the MFI studies. Although these 11 studies explicitly stated

that parent education was a component of the intervention, it is

highly likely that a much larger percentage of studies used some

form of written or verbal education. Several studies were consid-

ered MFI

20,65,72,91,92

because they explicitly stated that parents were

instructed in written materials to implement multiple techniques.

Finally, in 4 of these studies

20,65,72,92

written material alone was

compared to therapist-guided interventions, and as discussed be-

low, head-to-head comparisons tended to favor therapist interven-

tions.

Eight of the MFI studies reported tailoring treatments for the

individual child and family. In each of these studies, the types

of interventions were discussed, but the specific approach to in

dividualizing the treatments was not specified. Tailoring a treat-

ment to a specific patient’s needs is the norm in clinical settings,

but this approach establishes several confounds that can limit the

ability to generalize study findings, as well as the claims that the

treatment is efficacious.

In a similar vein, the strength of the MFI studies lies in their

high ecological validity, namely, most clinicians are likely to

combine intervention strategies with their patients rather than

rely on a single approach. The obvious weakness of the studies

using MFI is that the efficacy of individual interventions cannot

be analyzed. Multiple baseline or ABAB designs pose even more

complexity in testing efficacy as there are carry over effects of

the initial intervention strategy. While counter-balancing different

interventions helps to control for carry over effects, only 1 study

reported the sequence of the intervention and counterbalancing

was not reported.

The reported efficacy of the MFI interventions was high, be-

tween 50% and 100% of subjects had improved partially or com-

pletely, with all but 4 studies reporting between 75% and 100%

improvement in bedtime behavior problems and nighttime awak-

enings. Thus, while there are several weaknesses and confounds

inherent in this subgroup of studies that used MFI and tailored

interventions, taken as a whole there was generally a large mag-

nitude of positive change in all but 2 studies.

65,68

The strength of

these studies lies in the high ecological validity of both tailored

and MFI treatment approaches.

5.2 Combining Behavioral and Pharmacological Therapies

Only 1 of the articles

reviewed here combined behavioral in-

tervention with pharmacotherapy. Although there are numerous

studies of combined behavioral and pharmacologic treatments of

adult insomnia, this is the only identified publication in the Eng-

lish language literature on children. The study involved a double

blind, placebo controlled trial of trimeprazine (a sedating antihis-

tamine). Thirty-five children aged 7 - 27 months were assigned

to groups receiving training in extinction and were administered

trimeprazine or placebo. The group receiving active medication

improved more quickly, but relapsed slightly upon withdrawal,

resulting in no group differences at follow-up. This finding is con-

sistent with studies in adults

121

showing a faster response when

pharmacologic agents are combined with behavioral treatments.

While the focus of this paper is on behavioral interventions and

not pharmacotherapy it is important to address the issue of medi-

cations because of their widespread use in clinical practice

and

there may be specific cases in which it is justifiable to initiate a

combined behavioral and pharmacological therapy.

122

6.0 TREATMENT RESPONSE AND MODERATING VARIABLES

The role of key factors such as the length of the therapy and

patient and parent characteristics has not been systematically as

sessed. However, a number of studies have assessed differences

in therapist discipline (e.g., psychologist versus nurse practitio-

ner) and the manner in which the interventions were delivered

(written materials versus direct patient therapist contact).

6.1 Child and parent characteristics

There are no systematic reports on patient and parent charac

Review of Bedtime Problems in Children—Mindell et al

SLEEP, Vol. 29, No. 10, 2006

1272

teristics vis-à-vis the outcomes of the interventions. Carpenter

in a study of group intervention found that 73% of the parents re-

ported improvement and suggested that marital problems, parental

depression, and similar problems accounted for the failure of the

intervention in the other parents. Similarly, Jones and Verduyn

reported 84% success in resolving sleep problems using a behav-

ioral management program, and indicated that the sleep problems

were less likely to resolve if marital discord was involved or if

only 1 parent attended therapy. In another study,

positive out-

come of 1 of the interventions (standard ignoring) was associated

with maternal characteristics, namely, mothers who were less de-

pressed, less distressed about parenting, and made less disciplin-

ary mistakes were more likely to achieve better outcomes.

6.2 Treatment format

The format of the interventions for sleep problems in early

childhood has varied considerably across studies. Although most

studies have been based on therapist-parent sessions as the main

mode of delivery, studies have explored other more economic

modes of delivery, such as interventions by para-professionals or

interventions based on an information booklet only.

Eckerberg

compared the effects of interventions based on ad-

vice and support to interventions based on written information

only. The therapeutic approach in both interventions was based on

Graduated Extinction. Both interventions reduced protesting and

sleep latency, reduced the number of night wakings, and extended

sleep duration. The results failed to support differential effects of

the treatment format. Scott and Richards

compared 3 types of

interventions: advice, advice and support, and a booklet group.

All 3 groups gradually improved with time.

St. James-Roberts

assessed the effects of 3 intervention for-

mats: (1) a behavioral group that received written material and

discussed the topic with a clinician; (2) an educational group

that received written guide with general guidelines but no spe-

cific behavioral instructions; (3) and a control group that received

normal health services that were available to the other 2 groups.

The behavioral intervention led to a modest (10%) increase in

the number of infants who met the criteria for sleeping through

the night (5 hours or more) at 12 weeks of age. The educational

intervention produced no noticeable differences compared to the

control group. It is impossible to determine if the contact with the

clinicians or the specific behavioral instructions led to the limited

outcome differences between the groups.

Seymour

compared the effects of written information with

and without therapist contact. A waiting list group served as an ad-

ditional comparison group. Both treatments (written information

with or without therapist contact) led to a significant improvement

after 4 weeks of treatment. The results were achieved faster in the

group with therapist contact. However, after 4 weeks of treatment

there were no significant differences between the 2 interventions.

The positive outcomes were maintained at a 3-month post treat-

ment follow-up.

Finally, Weymouth

performed 3 studies with different modes

of delivery. In study 1, the intervention included a booklet, clini-

cal support, and clinical support with therapists. In study 2, the

intervention included a booklet and reduced contacts with the

therapist, and in study 3 the intervention included only the book-

let. The author concluded that some parents could succeed with a

booklet alone, whereas others require additional clinical support.

The results of these studies provide limited support for the

cost-effectiveness of using clinical sessions as part of the inter-

vention model for sleep problems in early childhood. In a tele-

phone survey of parents of 12 to 35 month old children

123

it was

reported that many parents have used interventions methods

based on information provided by the media (e.g., books, parent-

ing magazines, TV) with high rates of success (above 70% for

some popular interventions). These results also suggest that many

parents can successfully utilize information on sleep related be-

havioral interventions with no need for professional help.

Another consideration of cost-effectiveness is the potential use

of group rather than individual sessions. Only 3 studies assessed

the use of parent group sessions and none of them compared

group versus individual session format. Reid,

Carpenter,

and

Szyndler

reported positive outcomes for group interventions.

Research comparing the outcomes of individual versus group ses-

sion format is needed to assess the possible advantage over more

costly individual sessions.

In summary, the mode of delivery varied across studies, with

some studies finding little increased benefit for face-to-face inter-

ventions. Several factors, such as symptom chronicity and sever-

ity, parental mental health and coping skills, are likely moderat-

ing factors. The quality and content of the interventions is also a

key consideration that requires further assessment.

6.3 Treatment duration

The duration of the interventions varied considerably among

published studies. However, there are no published studies com

paring structured treatment programs of different durations. Most

interventions ranged between 2 weeks and 2 months. The findings

suggest that even relatively short interventions (1 - 3 sessions)

can be very effective in improving sleep in early childhood. More

research is needed to assess the value of more extended treatment

programs in terms of short- versus long-term effects on the child’s

evolving sleep patterns.

7.0 CONCLUSIONS

7.1 General conclusions

This review of 52 treatment studies indicates that several

well-defined behavioral approaches produce reliable and durable

changes in bedtime problems and night wakings in infants and

young children. Across all studies, 94% report that behavioral

interventions produced clinically significant improvements in

bedtime problems and/or night wakings. Approximately 82% of

children benefit from treatment and the majority maintain these

results for 3 to 6 months. Empirical evidence from controlled

group studies strongly supports unmodified extinction, Gradu-

ated Extinction, and preventive parent education about sleep. In

addition, the majority of studies also included a consistent bed-

time routine, Positive Reinforcement, and general parent educa-

tion about sleep.

These findings are consistent with the conclusions of 2 pre-

vious reviews

5,29

that used previously established criteria in the

field of clinical psychology

124

to evaluate the empirical support

for behavioral interventions. Mindell

, in 1999, found that extinc-

tion and parent education on the prevention of sleep problems to

be well-established treatments. Furthermore, Graduated Extinc-

tion and scheduled awakenings were probably efficacious, with

Review of Bedtime Problems in Children—Mindell et al

SLEEP, Vol. 29, No. 10, 2006

1273

positive routines a promising intervention. An updated review by

Kuhn and Elliott

in 2003 found extinction, Graduated Extinction,

and early intervention/parent education to be well-established in-

terventions. Scheduled awakenings were considered probably ef-

ficacious, whereas extinction with parental presence and positive

routines/faded bedtime with response cost were promising inter-

ventions.

7.2 Methodological issues

The outcomes of the research on the efficacy of clinical in

terventions for early childhood sleep problems have been very

positive. However, clearly some notable methodological limita-

tions need to be considered. The lack of standard definitions and

criteria for sleep problems in early childhood limits the possibility

of comparisons between studies and sometimes even for differ-

ent interventions within studies. A similar problem is the lack of

standardized outcome measures that would enable comparisons

between studies.

Another potential concern in this area is the inclusion of single-

case design studies, rather than sole reliance on RCTs as empirical

evidence for these behavioral interventions. The primary limita-

tion in studying a single-case is that the results from that partic-

ular case may not be relevant to other cases (external validity).

However, larger samples producing statistically significant find-

ings do not necessarily mean that such effects are more power-

ful or clinically significant.

125

Experimental single-case research

designs (e.g., ABAB, multiple baseline) are stronger than large

group designs at isolating mechanisms of change (internal valid-

ity), and are therefore used more commonly in applied behavioral

research, thus highly applicable to the question at hand.

Advances in technology have led to new objective methods

to assess sleep in young children. These relatively non-intrusive

techniques (e.g., time-lapse video, actigraphy, see Thoman and

Acebo

126

for review) may provide clinicians an opportunity to

objectively assess target symptoms or problems in addition to

parental subjective reports. It has been suggested that inflated

improvement effects could result from parental fatigue when par-

ents are asked to document each night-waking on a daily basis for

extended periods.

104

However, there are clear benefits to parental

report and the combination of subjective and objective measures.

Parental subjective experience of the sleep problem is clearly

valuable. Furthermore, objective measures may capture nighttime

awakenings that are not indicative of sleep disruption, providing

a better understanding of children’s sleep in general. A combina-

tion of these measures is necessary to identify those children with

clinically significant sleep problems.

The scarcity of studies comparing different delivery methods

(e.g., clinical session versus booklet information) and their con-

flicting results makes it difficult to assess the essential compo-

nents needed for an effective intervention. Some of these questions

could be answered by traditional outcome research (comparisons

between groups). Another approach is the use of process research

to assess the contributions of specific elements of interventions

(e.g., discussing parental fears and anxieties prior to the behav-

ioral coaching). The complementary role of process research has

not been well recognized and implemented in the study of behav-

ioral interventions for sleep problems (see Shirk and Russell

127

for

a review of these methodological issues).

Another crucial issue is the assessment of the long-term effi-

cacy of the interventions. Most studies reviewed here had a fol-

low-up period of 6 months or shorter. Recently, the long-term

maintenance of positive outcomes of cognitive-behavioral inter-

ventions has been questioned in different areas of psychopathol-

ogy in adults.

128

Future research should include longer follow-up

periods than those that have been traditionally used.

7.3 Future research

It is clear that there are many crucial questions that remain to be

answered regarding the treatment of bedtime problems and night

wakings in young children. For instance, in light of the wide-

range efficacy demonstrated by different intervention methods,

what are the actual curative factors or the essential ingredients of

these interventions? Other intriguing questions include: What are

the outcome changes in actual sleep patterns as opposed to those

reported sleep patterns? How long are these positive outcomes

maintained? What are the negative side effects, if any?

Additional research is also needed on the impact of interven

tions on mood, behavior, and development. Specific child and

parent characteristics need further study, such as child (e.g., tem-

perament, age) and parent (e.g., depression, parenting style) vari-

ables related to treatment success.

In addition to the above methodological concerns, future re-

search should move toward the use of standardized research diag-

nostic criteria, as well as standardized assessment measures. The

use of standardized diaries and questionnaires would allow com-

parison across studies and their outcomes, enabling meta-analytic

studies in this area. Furthermore, the addition of objective assess-

ment tools, such as actigraphy, would be highly beneficial.

Another primary area in need of further research is the role

of pharmacological agents, either alone or in combination with

behavioral interventions, in the treatment of sleep issues in young

children. These agents are frequently prescribed by pediatricians

and child psychiatrists

26,129

, however, there is limited research on

their efficacy, risks, benefits, and limitations. With the advent of

many new hypnotics and the potential risks associated with medi-

cations in young children, this research becomes even more cru-

cial. Finally, research is needed to evaluate the efficacy of alterna-

tive treatment modalities such as infant massage and nutritional

therapy/supplements.

7.4 Summary

After an extensive review of the pediatric sleep literature, we

found that two behavioral interventions for bedtime problems and

night wakings in young children, specifically Unmodified Extinc

tion (including Extinction with Parental Presence) and Preventive

Parent Education, have received strong empirical support across

the highest-level of studies. In addition, support is provided

for graduated extinction, bedtime fading/positive routines, and

scheduled awakenings. An overwhelming majority of children

respond favorably to these behavioral techniques, resulting in

not only better sleep, but also improvements in child and family

well-being. Although significant advances have been made in the

behavioral management of these common sleep problems, clearly

additional research is necessary and there are more questions to

be answered. It is essential that future studies use standardized re-

search diagnostic criteria, include more objective measures, and

that pediatric sleep researchers develop a consensus on critical

outcome variables

Review of Bedtime Problems in Children—Mindell et al

SLEEP, Vol. 29, No. 10, 2006

1274

REFERENCES

1. Lozoff, B, Wolf, AW, and Davis, NS. Sleep problems seen in pediat-

ric practice. Pediatrics 1985;75:477-483

2. Armstrong, KL, Quinn, RA, and Dadds, MR. The sleep patterns of

normal children. Med J Aust 1994;161:202-206

3. Burnham, MM, Goodlin-Jones, BL, Gaylor, EE, and Anders, TF.

Nighttime sleep-wake patterns and self-soothing from birth to one

year of age: a longitudinal intervention study. J Child Psychol Psy-

chiatry 2002;43:713-725

4. Goodlin-Jones, BL, Burnham, MM, Gaylor, EE, and Anders, TF.

Night waking, sleep-wake organization, and self-soothing in the

first year of life. J Dev Behav Pediatr 2001;22:226-233

5. Mindell, JA. Empirically supported treatments in pediatric psychol-

ogy: Bedtime refusal and night wakings in young children. Journal

of Pediatric Psychology 1999;24:465-481

6. Bixler, EO, Kales, JD, Scharf, MB, Kales, A, and Leo, LA. Inci-

dence of sleep disorders in medical practice: A physician survey.

Sleep Research 1976;5:62

7. Mindell, JA, and Durand, VM. Treatment of childhood sleep disor

ders: Generalization across disorders and effects on family mem

bers. Special issue: Interventions in pediatric psychology. J Ped

Psychol 1993;18:731-750

8. Zuckerman, B, Stevenson, J, and Bailey, V. Sleep problems in early

childhood: Continuities, predictive factors, and behavioral corre

lates. Pediatrics 1987;80:664-671

9. Kataria, S, Swanson, MS, and Trevathon, GE. Persistence of

sleep disturbances in preschool children. Behavioral Pediatrics

1987;110:642-646

10. Pollock, JI. Predictors and long-term associations of reported sleep

difficulties in infancy. Journal of reproductive and infant psychol-

ogy 1992;10:151-168

11. Pollock, JI. Night waking at five years of age: Predictors and prog-

nosis. Journal of Child Psychology and Psychiatry and Allied Disci-

plines 1994;35:699-708

12. Beebe, DW, and Gozal, D. Obstructive sleep apnea and the prefron

tal cortex: towards a comprehensive model linking nocturnal upper

airway obstruction to daytime cognitive and behavioral deficits. J

Sleep Res 2002;11:1-16

13. Keren, M, Feldman, R, and Tyano, S. Diagnoses and interactive pat

terns of infants referred to a community-based infant mental health

clinic. J Am Acad Child Adolesc Psychiatry 2001;40:27-35

14. Gais, S, Plihal, W, Wagner, U, and Born, J. Early sleep triggers mem

ory for early visual discrimination skills. Nat Neurosci 2000;3:1335-

1339

15. Lavigne, JV, Arend, R, Rosenbaum, D, Smith, A, Weissbluth, M,

Binns, HJ, and Christoffel, KK. Sleep and behavior problems among

preschoolers. J Dev Behav Pediatr 1999;20:164-169

16. Sadeh, A, Gruber, R, and Raviv, A. Sleep, neurobehavioral func-

tioning, and behavior problems in school-age children. Child Dev

2002;73:405-417

17. Adams, LA, and Rickert, VI. Reducing bedtime tantrums: Compari-

son between positive routines and Graduated Extinction. Pediatrics

1989;84:756-761

18. Hiscock, H, and Wake, M. Randomised controlled trial of behav

ioural infant sleep intervention to improve infant sleep and maternal

mood. Bmj 2002;324:1062.

19. Leeson, R, Barbour, J, Romaniuk, D, and Warr, R. Management of

infant sleep problems in a residential unit. Child Care Health Dev

1994;20:89-100

20. Wolfson, A, Lacks, P, and Futterman, A. Effects of parent training

on infant sleeping patterns, parents’ stress, and perceived parental

competence. J Consult Clin Psychol 1992;60:41-48

21. Durand, VM, and Mindell, JA. Behavioral treatment of multiple

childhood sleep disorders: Effects on child and family. Behav Mod

1990;14:37-49

22. France, KG, Blampied, NM, and Wilkinson, P. A multiple-baseline,

double-blind evaluation of the effects of trimeprazine tartrate on in

fant sleep disturbance. Exp Clin Psychopharmacol 1999;7:502-513

23. Richman, N. A double-blind drug trial of treatment in young children

with waking problems. J Child Psychol Psychiatry 1985;26:591-

598

24. Simonoff, EA, and Stores, G. Controlled trial of trimeprazine tar-

trate for night waking. Arch Dis Child 1987;62:253-257

25. Russo, R, Gururaj, V, and Allen, J. The effectiveness of diphenhydr

amine HCL in pediatric sleep disorders. Journal of Clinical Pharma

cology 1976;16:284-288

26. Owens, JA, Rosen, CL, and Mindell, JA. Medication use in the

treatment of pediatric insomnia: results of a survey of community-

based pediatricians. Pediatrics 2003;111:e628-635

27. Kales, A, Allen, C, Scharf, MB, and Kales, JD. Hypnotic drugs and

their effectiveness. Archives of General Psychiatry 1970;23:226-

232

28. Richman, N, Douglas, J, Hunt, H, Lansdown, R, and Levere, R. Be

havioural methods in the treatment of sleep disorders: A pilot study.

J Child Psychol Psychiatr 1985;26:581-590

29. Kuhn, BR, and Elliott, AJ. Treatment efficacy in behavioral pediatric

sleep medicine. Journal of Psychosomatic Research 2003;54:587-

597

30. Mindell, JA, and Owens, J. A Clinical Guide to Pediatric Sleep:

Diagnosis and Management of Sleep Problemsed). Philadelphia:

Lippincott Williams & Wilkins, 2003:

31. Ramchandani, P, Wiggs, L, Webb, V, and Stores, G. A systematic re

view of treatments for settling problems and night waking in young

children. British Medical Journal 2000;320:209-213

32. American Academy of Sleep Medicine. International Classification

of Sleep Disorders, Reviseded). Rochester, MN: American Acad

emy of Sleep Medicine, 1997:

33. American Academy of Sleep Medicine. International Classification

of Sleep Disorders, Second Edition (Second ed). Westchester, IL:

American Academy of Sleep Medicine, 2005:

34. Richman, N. Sleep problems in young children. Arch Dis Child

1981;56:491-493

35. Blampied, NM, and France, KG. A behavioral model of infant sleep

disturbance. J Appl Behav Anal 1993;26:477-492

36. Anders, TF, Halpern, LF, and Hua, J. Sleeping through the night: a

developmental perspective. Pediatrics 1992;90:554-560

37. Sadeh, A, and Anders, TF. Infant sleep problems: Origins, assess-

ment, interventions. Infant Mental Health Journal 1993;14:17-34

38. Sadeh, A, Klitzke, M, Anders, TF, and Acebo, C. Case study: sleep

and aggressive behavior in a blind, retarded adolescent. A concomi

tant schedule disorder? Journal of the American Academy of Child

and Adolescent psychiatry 1995;34:820-824

39. Mirmiran, M, Maas, YGH, and Ariagno, RL. Development of fetal

and neonatal sleep and circadian rhythms. Sleep Medicine Reviews

2003;7:321-334

40. Thoman, EB. Sleeping and waking states in infants: a functional

perspective. Neuroscience and Biobehavioral Reviews 1990;14:93-

107

41. Carey, WB. Night waking and temperament in infancy. Journal of

Pediatrics 1974;84:756-758

42. Keener, MA, Zeanah, CH, and Anders, TF. Infant temperament,

sleep organization, and nighttime parental interventions. Pediatrics

1988;81:762-771

43. Owens-Stively, J, Frank, N, Smith, A, Hagino, O, Spirito, A, Ar-

rigan, M, and Alario, A. Child temperament, parenting discipline

style, and daytime behavior in childhood sleep disorders. Journal of

Developmental and Behavioral Pediatrics 1997;18:314-321

44. Sadeh, A, Lavie, P, and Scher, A. Maternal perceptions of tempera

ment of sleep-disturbed toddlers. Early education and development

1994;5:311-322

45. Van Tassel, EB. The relative influence of child and environmental

characteristics on sleep disturbances in the first and second years of

life. Journal of Developmental and Behavioral Pediatrics 1985;6:81-85

Review of Bedtime Problems in Children—Mindell et al

SLEEP, Vol. 29, No. 10, 2006

1275

46. Owens, J, Spirito, A, Marcotte, A, McGuinn, M, and Berkelham

mer, L. Neuropsychological and Behavioral Correlates of Obstruc

tive Sleep Apnea Syndrome in Children: A Preliminary Study. Sleep

Breath 2000;4:67-78

47. Lewin, DS, Rosen, RC, England, SJ, and Dahl, RE. Preliminary

evidence of behavioral and cognitive sequelae of obstructive sleep

apnea in children. Sleep Med 2002;3:5-13

48. Bates, JE, Viken, RJ, Alexander, DB, Beyers, J, and Stockton, L.

Sleep and adjustment in preschool children: sleep diary reports by

mothers relate to behavior reports by teachers. Child Development

2002;73:62-75(14)

49. Lam, P, Hiscock, H, and Wake, M. Outcomes of infant sleep prob

lems: A longitudinal study of sleep, behavior, and maternal well-be-

ing. Pediatrics 2003;111:

50. Randazzo, AC, Muehlbach, MJ, Schweitzer, PK, and Walsh, JK.

Cognitive function following acute sleep restriction in children ages

10-14. Sleep 1998;21:861-868

51. Blunden, S, Lushington, K, and Kennedy, D. Cognitive and be-

havioural performance in children with sleep- related obstructive

breathing disorders. Sleep Medicine Reviews 2001;5:447-461

52. Sadeh, A, Gruber, R, and Raviv, A. The effects of sleep restriction

and extension on school-age children: What a difference an hour

makes. Child Development 2003;74:444-455

53. Dahl, RE. The regulation of sleep and arousal: Development and

psychopathology. Development and Psychopathology 1996;8:3-27

54. Gregory, AM, and O’Connor, TG. Sleep problems in childhood: A

longitudinal study of developmental change and association with

behavioral problems. Journal of the American Academy of Child

and Adolescent psychiatry 2002;41:

55. Thunstrom, M. Severe sleep problems in infancy associated with

subsequent development of attention-deficit/hyperactivity disorder

at 5.5 years of age. Acta Paediatrica 2002;91:584-592

56. Kazdin, AE. Parent management training: Treatment for opposition-

al, aggressive, and antisocial behavior in children and adolescents.

ed). Oxford University Press, 2005:

57. Williams, CD. The elimination of tantrum behavior by extinction

procedures. J Abnorm Soc Psychol 1959;59:269

58. Ferber, R. Solve your child’s sleep problemsed). New York: Simon

& Schuster, 1985:

59. Mindell, JA. Sleeping through the night: How infants, toddlers, and

their parents can get a good night’s sleep (revised). New York: Harp-

erCollins, 2005:

60. Sackett, DL. Rules of evidence and clinical recommendations for

the management of patients. Can J Cardiol 1993;9:487-489

61. Kerr, SM, Jowett, SA, and Smith, LN. Preventing sleep problems

in infants: A randomized controlled trial. J Adv Nurs 1996;24:938-

942

62. Reid, MJ, Walter, AL, and O’Leary, SG. Treatment of young chil-

dren’s bedtime refusal and nighttime wakings: A comparison of

“standard” and graduated ignoring procedures. J Abnorm Child

Psychol 1999;27:5-16

63. Rickert, VI, and Johnson, CM. Reducing nocturnal awakening and

crying episodes in infants and young children: A comparison be

tween scheduled awakenings and systematic ignoring. Pediatrics

1988;81:203-212

64. Scott, G, and Richards, MP. Night waking in infants: Effects of pro-

viding advice and support for parents. J Child Psychol Psychiatr

1990;31:551-567

65. St James-Roberts, I, Sleep, J, Morris, S, Owen, C, and Gillham, P.

Use of a behavioural programme in the first 3 months to prevent

infant crying and sleeping problems. Journal of Paediatrics & Child

Health 2001;37:289-297

66. Pinilla, T, and Birch, LL. Help me make it through the night: Be-

havioral entrainment of breast-fed infants’ sleep patterns [see com

ments]. Pediatrics 1993;91:436-444

67. Seymour, FW, Brock, P, During, M, and Poole, G. Reducing sleep

disruptions in young children: Evaluation of therapist-guided and

written information approaches: A brief report. J Child Psychol Psy-

chiatr 1989;30:913-918

68. Weir, IK, and Dinnick, S. Behaviour modification in the treatment of

sleep problems occurring in young children: A controlled trial using

health visitors as therapists. Child Care, Health Dev 1988;14:355-

367

69. Ashbaugh, R, and Peck, S. Treatment of sleep problems in a toddler:

A replication of the faded bedtime with response cost protocol. J

Appl Behav Anal 1998;31:127-129

70. Burke, RV, Kuhn, BR, and Peterson, JL. Brief Report: A “Story-

book” Ending to Children’s Bedtime Problems--The Use of a Re-

warding Social Story to Reduce Bedtime Resistance and Frequent

Night Waking. J. Pediatr. Psychol. 2004;29:389-396

71. Chadez, LH, and Nurius, PS. Stopping bedtime crying: Treating the

child and the parents. J Clin Child Psychol 1987;16:212-217

72. Eckerberg, B. Treatment of sleep problems in families with small

children: is written information enough? Acta Paediatr 2002;91:952-

959.

73. Fisher, J, Feekery, C, and Rowe, H. Treatment of maternal mood

disorder and infant behaviour disturbance in an Australian private

mothercraft unit: a follow-up study. Arch Women Ment Health

2004;7:89-93

74. France, KG, Blampied, NM, and Wilkinson, P. Treatment of infant

sleep disturbance by trimeprazine in combination with extinction. J

Dev Behav Pediatr 1991;12:308-314

75. France, KG, and Hudson, SM. Behavior management of infant

sleep disturbance. J Appl Behav Anal 1990;23:91-98

76. Friman, PC, Hoff, KE, Schnoes, C, Freeman, KA, Woods, DW,

and Blum, N. The bedtime pass: an approach to bedtime crying and

leaving the room. Arch-Pediatr-Adolesc-Med 1999;153:1027-1029

77. Galbraith, L, and Hewitt, KE. Behavioural treatment for sleep dis-

turbance. Health Visit 1993;66:169-171

78. Johnson, CM, Bradley-Johnson, S, and Stack, JM. Decreasing the

frequency of infants’ nocturnal crying with the use of scheduled

awakenings. Fam Pract Res J 1981;1:98-104

79. Johnson, CM, and Lerner, M. Amelioration of infant sleep distur-

bances: II. Effects of scheduled awakenings by compliant parents.

Infant Ment Health J 1985;6:21-30

80. Largo, RH, and Hunziker, UA. A developmental approach to the

management of children with sleep disturbances in the first three

years of life. Eur J Pediatr 1984;142:170-173

81. Lawton, C, France, KG, and Blampied, NM. Treatment of infant

sleep disturbance by Graduated Extinction. Child Fam Beh Ther

1991;13:39-56

82. McGarr, RJ, and Hovell, MF. In search of the sand man: Shaping an

infant to sleep. Educ Treat Child 1980;3:173-182

83. McMenamy, C, and Katz, RC. Brief parent-assisted treatment for

children’s nighttime fears. J Dev Behav Pediatr 1989;10:145-148

84. Minde, K, Faucon, A, and Falkner, S. Sleep problems in toddlers:

Effects of treatment on their daytime behavior. Journal of the Amer-

ican Academy of Child and Adolescent psychiatry 1994;33:1114-

1121

85. Rapoff, MA, Christophersen, ER, and Rapoff, KE. The manage

ment of common childhood bedtime problems by pediatric nurse

practitioners. J Ped Psychol 1982;7:179-196

86. Rolider, A, and Van Houten, R. Training parents to use extinction to

eliminate nighttime crying by gradually increasing the criteria for

ignoring crying. Educ Treat Child 1984;7:119-124

87. Ronen, T. Intervention package for treating sleep disorders in a

four-year-old girl. Journal of Beh Ther and Experimental Psychiatry

1991;22:141-148

88. Sanders, MR, Bor, B, and Dadds, M. Modifying bedtime dis-

ruptions in children using stimulus control and contingency man-

agement techniques. Behavioural Psychotherapy 1984;12:130-141

89. Seymour, FW, Bayfield, G, Brock, P, and During, M. Management

of night-waking in young children. Aust J Fam Ther 1983;4:217-

SLEEP, Vol. 29, No. 10, 2006

1276

223

90. Skuladottir, A, and Thome, M. Changes in infant sleep problems

after a family-centered intervention. Pediatr Nurs 2003;29:375-378

91. Szyndler, J, and Bell, G. Are groups for parents of children with

sleep problems effective? Health Visit 1992;65:277-279

92. Adair, R, Zuckerman, B, Bauchner, H, Philipp, B, and Levenson,

S. Reducing night waking in infancy: A primary care intervention.

Pediatrics 1992;89:585-588

93. Richman, N. A double blind drug trial of sleep problems in young

children. Journal of Child Psychology and Psychiatry and Allied

Disciplines 1985;26:591-198

94. Weymouth, J, Hudson, A, and King, N. The management of chil

dren’s nighttime behaviour problems: Evaluation of an advice book-

let. Behavioural Psychotherapy 1987;15:123-133

95. Bidder, RT, Gray, OP, and Pates, RM. Brief intervention therapy

for behaviourally disturbed pre-school children. Child Care, Health

Dev 1981;7:21-30

96. Carpenter, A. Sleep problems: a group approach. Health Visit

1990;63:305-307

97. Didden, R, Moor, JD, and Kruit, IW. The effects of extinction

in the treatment of sleep problems with a child with a physi-

cal disability. International Journal of Disability, Development

& Education 1999;46:247-252

98. Jones, DP, and Verduyn, CM. Behavioural management of sleep

problems. Archives of Disease in Childhood 1983;58:442-444

99. Roberts, S. Tackling sleep problems through clinic-based approach.

Health Visit 1993;66:173-174

100. Sanger, S, Weir, K, and Churchill, E. Treatment of sleep problems:

the use of behavioural modification techniques by health visitors.

Health Visit 1981;54:421-424

101. Seymour, FW. Parent management of sleep difficulties in young

children. Behaviour Change 1987;4:39-48

102. Weissbluth, M. Modification of sleep schedule with reduction of

night waking: A case report. Sleep 1982;5:262-266

103. Wright, L, Woodcock, J, and Scott, R. Treatment of sleep distur

bance in a young child by conditioning. South Med J 1970;63:174-

176

104. Sadeh, A. Assessment of intervention for infant night waking: Pa-

rental reports and activity-based home monitoring. J Consult Clin

Psychol 1994;62:63-68

105. Pritchard, A, and Appleton, P. Management of sleep problems in

pre-school children. Early Child Dev Care 1988;34:227-240

106. Douglas, J (1989) Training parents to manage their child’s sleep

problem. In Handbook of parent training: Parents as co-therapists

for children’s behavior problems. (Schaefer, CE, and Briesmeister,

JM, eds) pp. 13-37, John Wiley and Sons, New York

107. Milan, MA, Mitchell, ZP, Berger, MI, and Pierson, DF. Positive rou-

tines: A rapid alternative to extinction for elimination of bedtime

tantrum behavior. Child Behavior Therapy 1981;3:13-25

108. Morin, CM, Culbert, JP, and Schwartz, SM. Nonpharmacological

interventions for insomnia: a meta-analysis of treatment efficacy.

Am J Psychiatry 1994;151:1172-1180

109. Cohen, J. Statistical power analysis for the behavioral sciences.ed).

Hillside, NJ: Lawrence Erlbaum Associates, 1977:474

110. Ferber, R, and Kryger, M. Principles and practice of sleep medicine

in the child). Philadelphia: Saunders, 1995:

111. Klackenberg, G (1987) Incidence of parasomnias in children in a

general population. In Sleep and its disorders in children. (Guil

leminault, C, ed) pp. 99-113, Raven Press, New York

112. Anders, TF, Carskadon, MA, and Dement, WC. Sleep and sleepiness

in children and adolescents. Pediatr Clin North Am 1980;27:29-43

113. Corkum, P, Tannock, R, Moldofsky, H, Hogg-Johnson, S, and

Humphries, T. Actigraphy and parental ratings of sleep in chil-

dren with attention-deficit/hyperactivity disorder (ADHD). Sleep

2001;24:303-312

114. Tikotzky, L, and Sadeh, A. Sleep patterns and sleep disruptions in

kindergarten children. J Clin Child Psychol 2001;30:581-591

115. Wiggs, L, and Stores, G. Children’s sleep: how should it be as

sessed? Association of Child Psychology and Psychiatry Review

and Newsletter 1995;17:153-157

116. Sadeh, A. Stress, Trauma, and Sleep in Children. Child & Adoles-

cent Psychiatric Clinics of North America 1996;5:685-700

117. Eckerberg, B. Treatment of sleep problems in families with young

children: effects of treatment on family well-being. Acta Paediatr

2004;93:126-134

118. France, KG. Behavior characteristics and security in sleep-disturbed

infants treated with extinction. J Pediatr Psychol 1992;17:467-475

119. Holsboer-Trachsler, E, and Seifritz, E. Sleep in depression and sleep

deprivation: a brief conceptual review. World J Biol Psychiatry

2000;1:180-186

120. Morawetz, D. Depression and insomnia: Which comes first? Aust J

Counseling Psychology 2001;3:19-24

121. Smith, MT, Perlis, ML, Park, A, Smith, MS, Pennington, J, Giles,

DE, and Buysse, DJ. Comparative meta-analysis of pharmacother

apy and behavior therapy for persistent insomnia. Am J Psychiatry

2002;159:5-11

122. Younus, M, and Labellarte, MJ. Insomnia in children: when are

hypnotics indicated? Paediatr Drugs 2002;4:391-403

123. Johnson, CM. Infant and toddler sleep: a telephone survey of par-

ents in one community. Journal of Developmental and Behavioral

Pediatrics 1991;12:108-114

124. Chambless, DL, Sanderson, WC, Shoham, V, Bennet Johnson, S,

Pope, KS, and Crits-Christoph, P. An update on empirically vali-

dated therapies. Clin Psychol 1996;49:5-18

125. Drotar, D. Enhancing reviews of psychological treatments with

pediatric populations: thoughts on next steps. J Pediatr Psychol

2002;27:167-176

126. Thoman, EB, and Acebo, C (1995) Monitoring of sleep in

neonates and young children. In Principles and Practice of

Sleep Medicine in the Child (Ferber, R, and Kryger, M, eds)

pp. 55-68, W.B. Saunders, Phildelphia

127. Shirk, SR, and Russell, RL. Change processes in child psychother

apy: Revitalizing treatment and research.ed). New York: Guilford

Press, 1996:395

128. Westen, D, Novotny, CM, and Thompson-Brenner, H. The empiri-

cal status of empirically supported psychotherapies: assumptions,

findings and reporting in controlled clinical trials. Psychological

Bulletin 2004;130:631-663

129. Rosen, CL, Owens, JA, and Mindell, JA. Use of pharmacotherapy

for insomnia in children and adolescents: A national survey of child

psychiatrists. Sleep 2005;28:A79

Reference;

Evidence

Level

Study

Design

(control)

N (com-

pleted); M,

F (com-

pleted); Age

(range)

Treatment Strategy;

Treatment Condi

tions (% completed);

Description of Treat

ment Conditions

Problem Deﬁnition Target

Behav-

ior

Outcome Measures

– Child Sleep and Other

Major Findings

Limitations

Adams &

Rickert

; I

Random-

ized control

trial

36; 16M,

20F; 18-48

months

Graduated extinction

Positive routines

Control group

5 bedtime tantrums

per week. No addi-

tional information.

Bedtime

prob-

lems

Child sleep variables:

time to bed, tantrum

behaviors, time asleep;

parent daytime mood and

behavior: dyadic adjust-

ment scale

Treatments effective- < # of

tantrums and tantrum duration at

weeks 2, 3, 4, 5, 6 and follow-up (3

and 6 weeks) compared to controls.

No differences between Positive

Routines and Graduated Extinction.

Positive routines produced more

change in marital satisfaction on

the DAS.

No objec-

tive mea-

surement.

Hiscock &

Wake

; I

Random-

ized control

trial-

control

condition

placebo

146 (com-

pleted);

89M, 79 F

(enrolled);

6-12 months

Graduated extinction

(with parental pres-

ence)

Control group

2 weeks of a sleep

problem, including

at least one of the

following: 1. Waking

> ﬁve nights a week,

2. Waking > 3 times a

night, 3. > 30 minutes

to fall asleep, 4. re-

quiring parental pres-

ence to fall asleep.

General

sleep

problem

Child sleep variables:

sleep problem (yes/no),

sleep diary; parent sleep

variables: maternal sleep

quality/quantity; maternal

depression

Behavioral intervention signiﬁ-

cantly reduced sleep problems and

maternal depression at 2 months,

but not at 4 month follow-up.

No inter-

reliability

check for

diagno-

sis. No

treatment

ﬁdelity

check. No

objective

measure-

ment.

Kerr,

Jowett, &

Smith

; I

Random-

ized control

trial

169; NS;3

months

Parent education/pre-

vention (3 months of

age)

Control group

Prevention of bedtime

problems and night

wakings

Bedtime

prob-

lems,

night

wakings

Child sleep variables:

structured interview

Preventive education resulted in

fewer settling difﬁculties, fewer

night wakings per week, and fewer

night wakings per night.

No inter-

reliability

check for

diagno-

sis. No

treatment

ﬁdelity

check. No

objective

measure-

ment.

Pinilla &

Birch

; II

Random-

ized control

trial

26; 12M,

14F; NS

Parent education/pre-

vention (prenatal)

Control group

Not speciﬁed

Night

wakings

Child sleep variables:

sleep diaries (total sleep

time, longest sleep

episode, average length

of sleep per episode,

feedings); child daytime

mood and behavior:

Bates’ Infant Characteris-

tics Questionnaire

Prenatal parent education led to de-

creased night wakings. At 8 weeks

of age, 100% in treatment group

sleeping from midnight to 5 AM

compared to 23% of control infants.

No treat-

ment

ﬁdelity

check. No

objective

measure-

ment.

Reid,

Walter &

O’Leary

;

Random-

ized control

trial

43; 23M,

20F; 16-48

months

Extinction

Graduated extinction

Wait-list control

Bedtime problems (>

30 min to settle or not

settle alone) and/or

4 difﬁcult nighttimes

(night wakings or

sleep in parent bed) at

least 4 days per week

Bedtime

prob-

lems,

night

wakings

Child sleep variables:

subjective parents assess-

ment, settle<10 minutes,

# night waking; child

behavior: CBCL, BDI;

parent mood and behav-

ior: PSI, DAS, Parenting

Scale; parent expectan-

cies, parent satisfaction

Treatment more effective for

bedtime problems and night wak-

ings than control and maintained

2 months; no difference between

standard ignoring and graduated ig-

noring, although increased compli-

ance and less stress with graduated

ignoring for nighttimes

No inter

reliability

check for

diagno-

sis. No

objective

measure-

ment. Small

sample.

Reference;

Evidence

Level

Study

Design

(control)

N (com-

pleted); M,

F (com

pleted); Age

(range)

Treatment Strategy;

Treatment Condi

tions (% completed);

Description of Treat-

ment Conditions

Problem Deﬁnition Target

Behav-

ior

Outcome Measures

– Child Sleep and Other

Major Findings

Limitations

Rickert &

Johnson

; I

Random-

ized control

trial

33 (27

families,

11sibling

pairs); 18M,

15F; 6-54

months

Extinction

Scheduled awaken-

ings

Control group

Wakes at least 1x/

night for 4 weeks

Night

wakings

Child sleep variables:

daily schedule, night

wakings; Reliability

check

Systematic ignoring and scheduled

awakenings decreased night wak-

ings and crying episodes more than

controls. SI group showed fastest

improvement. Reliabilities between

parents 82-100%

No inter-

reliability

check for

diagno-

sis. No

treatment

ﬁdelity

check. No

objective

measure-

ment.

Multiple

children per

family.

Scott &

Richards

;

Random-

ized control

trial

120; 76M,

44F; 1

month -18

months

Parent education

/prevention

Night waking problem Night

wakings

Child sleep variables:

number of nights with

night wakings, number

of night wakings per

night, duration of night

wakings; parent daytime

mood and behavior:

maternal well-being

No support for efﬁcacy of interven

tion. All three groups somewhat

improved.

No objec-

tive mea-

surement.

80 Sey-

mour, et al.

; II

Random-

ized control

trial

45; 28M,

17F; 9

months – 5

years

Parent education /pre-

vention (extinction)

Wait list control

Not speciﬁed

Bedtime

prob-

lems,

night

wakings

Child sleep variables:

minutes awake each

night, number of night

wakings, settling time

Both interventions led to signiﬁcant

improvement 4 weeks post-treat-

ment (signiﬁcant reduction in wake-

fulness during the night, night-wak-

ings and settling time. Faster results

with therapist contact. After 4

weeks post-treatment no signiﬁcant

differences between interventions.

Positive outcomes were maintained

at 3-months.

No treat-

ment

ﬁdelity

check. No

objective

measure-

ment.

St. James-

Roberts, et

al.

; I

Random-

ized control

trial

478; NS; NS Parent education /pre-

vention (behavioral

intervention)

General education

Prevention Bedtime

prob-

lems,

night

wakings

Child sleep variables:

TST, sleep bout length,

minutes of crying, cry

bout length, % of infants

with nighttime sleep

period > 5hr

Education intervention led to lim-

ited changes parents’ behavior (was

not implemented); only ﬁnding was

a 10% increase sleep consolidation

for behavioral group at 12 weeks.

No objec-

tive mea-

surement.

Weir &

Dinnick

;

NS 43; NS; 4

months- 4.5

years

Extinction (individu-

alized)

Control group

Settling and night

waking problems

Bedtime

prob-

lems,

night

wakings

Child sleep variables:

clinical ratings of sever-

ity, # of children with

settling problems and

waking problems

Marked improvement in interven-

tion and control groups. No differ-

ences between groups.

No treat-

ment ﬁdel-

ity check.

No objec-

tive mea-

surement.

Differential

attrition

across

groups.

Reference;

Evidence

Level

Study

Design

(control)

N (com-

pleted); M,

F (com-

pleted); Age

(range)

Treatment Strategy;

Treatment Condi

tions (% completed);

Description of Treat

ment Conditions

Problem Deﬁnition Target

Behav-

ior

Outcome Measures

– Child Sleep and Other

Major Findings

Limitations

Wolfson, et

al.

; I

Random-

ized control

trial

60; 29M,

31F; 6-9

weeks

Parent education/

prevention (pre and

postnatal)

Control group

Prevention Bedtime

prob-

lems,

night

wakings

Child sleep variables:

TST, number of sleep

episodes, longest sleep

episode, number of wak-

ing episodes, number of

feedings; parent daytime

mood and behavior:

hassles and uplifts scale,

parental efﬁcacy

Intervention led to improved infant

sleep, parental sleep, parental com-

petency. Control reported increased

stress over time.

No objec-

tive mea-

surement