This material is protected by U.S. copyright law. Unauthorized reproduction is prohibited. To purchase quantity reprints,
please e-mail reprints@ons.org or to request permission to reproduce multiple copies, please e-mail pubpermissions@ons.org.
Sleep/Wake Disturbances in People With Cancer
and Their Caregivers: State of the Science
Ann M. Berger, PhD, RN, AOCN®, Kathy P. Parker, PhD, RN, FAAN,
Stacey Young-McCaughan, PhD, RN, AOCN®, Gail A. Mallory, PhD, RN, CNAA,
Andrea M. Barsevick, RN, DNSc, AOCN®, Susan L. Beck, PhD, APRN, FAAN, AOCN®,
Janet S. Carpenter, PhD, RN, Patricia A. Carter, PhD, RN, CNS, Lynne A. Farr, PhD,
Pamela S. Hinds, PhD, RN, Kathryn A. Lee, RN, PhD, FAAN, Christine Miaskowski, RN, PhD, FAAN,
Victoria Mock, DNSc, FAAN, Judith K. Payne, PhD, RN, AOCN®, and Martica Hall, PhD
Key Points . . .
Purpose/Objectives: To review the state of the science on sleep/wake
disturbances in people with cancer and their caregivers.
Data Sources: Published articles, books and book chapters, confer-
Sleep/wake disturbances in people with cancer and their care-
ence proceedings, and MEDLINE®, the Cumulative Index to Nursing
givers can occur alone or as part of symptom clusters but have
and Allied Health Literature®, PsycINFO, and the Cochrane Library
received little attention from healthcare providers.
computerized databases.
Data Synthesis: Scientists have initiated studies on the prevalence of
Several models offer approaches to examine the numerous fac-
sleep/wake disturbances and the etiology of sleep disturbances specific to
tors that may underlie sleep/wake disturbances in people with
cancer. Measurement has been limited by lack of clear definitions of sleep/
cancer and their caregivers.
wake variables, use of a variety of instruments, and inconsistent reporting
A need exists to screen and assess sleep and wakefulness and
of sleep parameters. Findings related to use of nonpharmacologic interven-
to establish evidence for the reliability and validity of instru-
tions were limited to 20 studies, and the quality of the evidence remains
ments used in this population.
poor. Few pharmacologic approaches have been studied, and evidence
for use of herbal and complementary supplements is almost nonexistent.
Randomized clinical trials to test interventions to promote sleep
Conclusions: Current knowledge indicates that sleep/wake distur-
in this population are needed, but nonpharmacologic approaches
bances are prevalent in cancer populations. Few instruments have been
found to be effective in people with insomnia can be included in
validated in this population. Nonpharmacologic interventions show
patient-teaching materials.
positive outcomes, but design issues and small samples limit gener-
alizability. Little is known regarding use of pharmacologic and herbal
Future directions for research on sleep/wake disturbances have
and complementary supplements and potential adverse outcomes or
been identified and are ready for study.
interactions with cancer therapies.
Implications for Nursing: All patients and caregivers need initial
and ongoing screening for sleep/wake disturbances. When disturbed
sleep/wakefulness is evident, further assessment and treatment are
RN, DNSc, AOCN®, is an associate member and director of nursing
warranted. Nursing educational programs should include content regard-
research at Fox Chase Cancer Center in Philadelphia, PA; Susan
ing healthy and disrupted sleep/wake patterns. Research on sleep/wake
L. Beck, PhD, APRN, FAAN, AOCN®, is interim associate dean for
disturbances in people with cancer should have high priority.
academic programs in the College of Nursing at the University of
Utah in Salt Lake City; Janet S. Carpenter, PhD, RN, is an associate
professor in the School of Nursing at Indiana University in India-
napolis; Patricia A. Carter, PhD, RN, CNS, is an associate professor
Ann M. Berger, PhD, RN, AOCN®, is an associate professor, an
and assistant dean for student and clinical affairs in the School of
advanced practice nurse, and the Niedfelt Nursing Professor in the
Nursing at the University of Texas at Austin; Lynne A. Farr, PhD, is
College of Nursing at the University of Nebraska Medical Center in
an emeritus professor in the College of Nursing at the University of
Omaha; Kathy P. Parker, PhD, RN, FAAN, is the Edith F. Honeycutt
Nebraska Medical Center; Pamela S. Hinds, PhD, RN, is the direc-
professor in the Nell Hodgson Woodruff School of Nursing at Emory
tor of nursing research at St. Jude Children's Research Hospital in
University in Atlanta, GA; Stacey Young-McCaughan, PhD, RN,
Memphis, TN; Kathryn A. Lee, RN, PhD, FAAN, is a professor and
AOCN®, is a colonel in the U.S. Army Nurse Corps and chief of the
(Continued on next page)
Department of Clinical Investigations in Fort Sam Houston, TX;
Gail A. Mallory, PhD, RN, CNAA, is the director of research at the
Oncology Nursing Society in Pittsburgh, PA; Andrea M. Barsevick,
Digital Object Identifier: 10.1188/05.ONF.E98-E126
son, Edwards, & Sepion, 2005; Hockenberry-Eaton et al.,
the Livingston Chair in Nursing in the Department of Family Health
Care Nursing in the School of Nursing at the University of California,
1998; Vena, Parker, Cunningham, Clark, & McMillan, 2004).
San Francisco; Christine Miaskowski, RN, PhD, FAAN, is a profes-
They occur before treatment has begun (Ancoli-Israel, Moore,
sor in the Department of Physiological Nursing at the University of
& Jones, 2001) and often seem to be directly related to the
California, San Francisco; Victoria Mock, DNSc, FAAN, is a profes-
cancer diagnosis (Lee, Cho, Miaskowski, & Dodd, 2004;
sor in and the director of the Center for Nursing Research at Johns
Savard, Simard, Blanchet, Ivers, & Morin, 2001).
Hopkins University in Baltimore, MD; Judith K. Payne, PhD, RN,
The sleep of caregivers, who often are family members, also
AOCN®, is an assistant professor in the School of Nursing at Duke
is disrupted. Problems with insomnia, other nocturnal sleep
University in Durham, NC; and Martica Hall, PhD, is an assistant
disturbances, and daytime fatigue are common among caregiv-
professor of psychiatry and psychology in the School of Medicine at
ers of people with cancer and other chronic illnesses (Carter,
the University of Pittsburgh in Pennsylvania. The conference on which
2003; Carter & Chang, 2000; Hinds et al., 1999; Jepson, Mc-
this article is based was supported by the National Cancer Institute
(R13 CA108758-01); the ONS Foundation's Center for Leadership,
Corkle, Adler, Nuamah, & Lusk, 1999; Kozachik et al., 2001;
Information and Research and the Oncology Nursing Society, both
McGrath, Paton, & Huff, 2004; Nijboer et al., 2000; Nijboer,
in Pittsburgh; Ambulatory Monitoring, Inc., in Ardsley, NY; and Mini
Triemstra, Tempelaar, Sanderman, & van den Bos, 1999). Re-
Mitter Co., Inc., in Bend, OR. The views of Young-McCaughan are
active depression is very prevalent and can interfere with daily
her own and do not purport to reflect the position of the Army Medical
function, QOL, and ability to continue to provide care (Jepson
Department, Department of the Army, or the Department of Defense.
et al.; Kozachik et al.; Nijboer et al., 1999, 2000). Only recently
(Submitted April 2005. Accepted for publication August 5, 2005.)
has chronic sleep loss related to stress and the 24-hour-per-day
demands of caregiving been described as a significant problem
(Carter & Chang). In addition, caregivers may be at risk for
sleep problems after the loss of a loved one. A recent study
espite the evidence suggesting that sleep/wake dis-
demonstrated that the risk for long-term psychological morbid-
turbances, particularly insomnia, are among the most
ity, including sleep/wake disturbances, in a surviving partner
common complaints of people with cancer, sleep
significantly increased if a patient's symptoms were unrelieved
problems have received little attention from healthcare provid-
during the last three months of life (Valdimarsdottir, Helgason,
ers (Savard & Morin, 2001). Data suggest that patients with
Furst, Adolfsson, & Steineck, 2004).
cancer have twice the prevalence of sleep problems as that
Many nurses are aware that people with cancer have problems
reported in the general population (Savard, Laroche, Simard,
getting a good night's sleep but do not realize the effects that
Ivers, & Morin, 2003), with the majority of people reporting
poor sleep can have on daytime wakefulness, functional ability,
maintenance insomnia with several awakenings during the
and QOL (Vena et al., 2004). Furthermore, the assessment of
night (Davidson, MacLean, Brundage, & Schulze, 2002; Lee,
sleep/wake disturbances and treatment of problems have not
2003). Insomnia is associated with increased rates of medical
been integrated into routine clinical practice (Savard & Morin,
and psychiatric illnesses and decreased quality of life (QOL)
2001; Sherman et al., 2004). Barriers to integration of current
in the general population (Sateia & Pigeon, 2004). The distur-
knowledge about sleep are similar to those that commonly
bances compound the many challenges faced by people with
occur when working to translate knowledge of pain. The bar-
cancer. The much-needed research on sleep/wake disturbances
riers occur at the level of patients, families, care providers, and
in people with cancer needs to build on the extensive body
healthcare systems. Nonetheless, as part of a comprehensive
of knowledge that exists regarding sleep/wake disturbances
care plan, nurses should assess the sleep needs of patients and
in the general population and specific sleep disorders such as
caregivers regularly and tailor interventions to meet their needs
sleep apnea, periodic leg movement syndrome, restless leg
(Steele & Fitch, 1996). Sleep/wake disturbances increasingly
syndrome, narcolepsy, and insomnia.
are recognized as significant side effects of cancer treatment that
This article summarizes participants' contributions to the
affect physiologic as well as psychological function. In fact, the
Oncology Nursing Society (ONS) State-of-the-Science Con-
U.S. Department of Health and Human Services, the National
ference on Sleep/Wake Disturbances in People With Cancer
Institutes of Health, and the National Heart, Lung, and Blood In-
and Their Caregivers, which was held at ONS headquarters
stitute (2003); the Institute of Medicine (2005); and ONS (2003)
in Pittsburgh, PA, July 15­17, 2004. This article presents
have identified sleep disturbances as a priority research area.
the state of the science on sleep/wake disturbances in people
with cancer and their caregivers. It is divided into four main
sections: sleep/wake disturbances in people with cancer, mea-
Sleep/Wake-Related Terminology
surement of sleep/wake disturbances, nonpharmacologic and
Recently, an increasing number of descriptive studies
pharmacologic interventions, and implications for practice,
have included one or more sleep variables or terms indicat-
education, and research. This article lays the foundation for
ing sleep/wake disturbances in some context in people with
directing increased attention to sleep/wake disturbances in
cancer. Unfortunately, the lack of consistency in terminol-
clinical, educational, and research environments. As a result,
ogy makes it difficult to compare and contrast studies and
the authors anticipate that patients with cancer and their care-
optimally evaluate their quality. Therefore, the authors
givers will experience improved assessment and treatment.
recommend that, when possible, the terminology employed
Sleep/Wake Disturbances
by the American Academy of Sleep Medicine (AASM) in
the International Classification of Sleep Disorders (AASM,
in People With Cancer
2005) be used to provide consistency (Clark et al., 2004).
For example, a primary sleep disorder is a specific diagnostic
Adults and children report that sleep/wake disturbances
entity that includes a wide array of problems characterized by
occur during all phases of cancer care (Clark, Cunningham,
the symptoms of insomnia, excessive daytime sleepiness, or
McMillan, Vena, & Parker, 2004; Gibson, Garnett, Richard-
abnormal movements, behaviors, or sensations during sleep.
intrinsic circadian rhythm (Lazuna & Farr, 2003). The role of
Eight groups of sleep disorders are described by AASM.
homeostatic factors and circadian rhythm in sleep regulation
· Insomnias: disorders that produce repeated difficulty with
is best understood when sleep is examined as a physiologic
sleep initiation, duration, consolidation, or quality that oc-
as well as a behavioral process controlled by a system based
curs despite adequate time and opportunity for sleep and
in the brain and central nervous system. The sleep-regulation
results in some form of daytime impairment
system provides an engine that drives sleep and wakefulness.
· Sleep-related breathing disorders: characterized by dis-
One model commonly used to help understand sleep is
ordered respiration during sleep
the Two-Process Model of Sleep Regulation (see Figure 1)
· Hypersomnias of central origin: characterized by the
(Achermann & Borbely, 2003; Borbely, 1982). According to
primary complaint of daytime sleepiness not related to
the Two-Process Model, sleep begins (sleep onset) when the
circadian rhythm sleep disorders, sleep-related breathing
homeostatic (somnostat) component or need to sleep (Process
disorders, or other causes of disturbed nocturnal sleep
S) is high. The need to sleep (Process S) increases as the time
· Circadian rhythm sleep disorders: recurrent or chronic
of prior wakefulness increases. The process of sleep onset
patterns of sleep disturbance resulting from alterations of
also is modulated by the sympathetic nervous system (Lich-
stein, Wilson, Noe, Aguillard, & Bellur, 1994; Lushington,
the circadian timing system or misalignment between an
Dawson, & Lack, 2000; Vgontzas et al., 1998) and by the
individual's rhythm and the 24-hour social and physical
hypothalamic-pituitary-adrenal (HPA) axis (Vgontzas et al.).
· Parasomnias: undesirable physical events or experiences
Increased activation of the sympathetic nervous system and
that occur during entry into sleep, within sleep, or during
the HPA axis can elevate arousal levels and delay the onset of
arousal from sleep
sleep. Input from the systems is filtered out by the thalamus
· Sleep-related movement disorders: conditions that are
as sleep occurs, a process that must be filtered continually to
characterized primarily by relatively simple, usually ste-
maintain uninterrupted sleep.
reotyped movements that disturb sleep (e.g., periodic limb
In addition to Process S, there is Process C, the circadian
process. The circadian oscillator resides in the suprachiasmatic
· Isolated symptoms, apparently normal variants, and
nuclei, where rhythms are generated and synchronized with the
unresolved issues: symptoms that either lie at the border-
environment by light and dark cues from the retina of the eye.
line between normal and abnormal sleep or that exist on
Additional timing modifications are provided by interaction
a continuum of normal to abnormal events in sleep (e.g.,
with the paraventricular nuclei and the ventromedial hypothal-
amus. Marker rhythms of the circadian oscillator are the core
· Other sleep disorders: disorders that cannot be classified
temperature rhythm and the rhythmic secretion of the hormone
elsewhere (e.g., other physiologic [organic] sleep disor-
melatonin. At sleep onset, sleep initiation is most likely to oc-
cur during the falling phase of the endogenous component of
In contrast, terms such as sleep/wake disturbances, sleep
the temperature rhythm (Strogatz, Knonaur, & Czeisler, 1986;
problems, alterations in sleep, or impaired sleep are more
Zulley, Weaver, & Aschoff, 1981). This occurs approximately
general terms and often are used to describe complaints,
five to six hours before the daily body temperature minimum
symptoms, or groups of symptoms experienced by individu-
is at its lowest point and one to two hours after the evening
als. They are not diagnostic entities as defined by the AASM
rise in plasma melatonin levels (Duffy, Dijk, Klerman, &
and often are used when a specific diagnosis has not or cannot
Czeisler, 1998). Timing of the circadian component is adjusted
be made. Nonetheless, two of the most common complaints
by melatonin secreted from the pineal gland during the dark
or symptoms experienced by the general public that are de-
and inhibited by light exposure (Claustrate, Brun, & Chazot,
fined in the International Classification of Sleep Disorders
2005). Circadian timing also is adjusted by levels of plasma
are insomnia and excessive daytime sleepiness. Insomnia
and central nervous system neurotransmitters and neuroendo-
crine factors such as gamma-amino-butyric acid; dopamine;
may be a primary sleep disorder or a symptom of one of
many other sleep disorders, such as sleep-related breathing
disorders. Daytime sleepiness is defined as the inability to
stay awake and alert during the major waking episodes of
the day, resulting in unintended lapses into drowsiness or
Psychological factors
sleep; it also may be seen in a wide range of sleep disorders.
Thus, understanding the causes of insomnia and excessive
daytime sleepiness is essential to providing effective inter-
Two-Process Model of Sleep Regulation
ventions. Numerous additional sleep-related terms can help
characterize sleep/wake disturbances, including total sleep
time, sleep latency, awakenings during sleep, and wake after
sleep onset.
Waking Sleep
Time: 23
Understanding Sleep/Wake Disturbances in People
With Cancer and Their Caregivers
Demographic factors
Most organisms, including humans, exhibit a daily rest/ac-
Lifestyle factors
tivity (sleep/wake) pattern. The pattern is a functional property
Figure 1. Factors That Affect Sleep Regulation in Patients
of all living matter, including humans, and is controlled by two
With Cancer
primary factors: the amount, timing, and placement of sleep
across the day (i.e., sleep homeostasis) and the underlying
Note. Based on information from Vena et al., 2004.
growth hormone releasing hormone; prostaglandins D2, E2,
Demographic Factors
and F2a; vasoactive inhibitory peptide; and growth hormone
· Age: Although increased age is associated with poorer quality of sleep,
(Harrington & Mistlberger, 2003; Kunz & Achermann, 2003;
comorbidities may play a more important role.
Lazuna & Farr, 2003; Mignot, Taheri, & Nishino, 2002; Obal
· Gender: Women typically report poorer sleep quality than men.
& Krueger, 2004; Ouyang, Hellman, Abel, & Thomas, 2004;
· Race: Caucasian race has been associated with increased sleep com-
Pace-Schott & Hobson, 2002).
Final awakening typically occurs when body temperature
Lifestyle and Environmental Factors
is rising (approximately one to two hours after the minimum
· Poor sleep hygiene: detrimental lifestyle factors such as irregular sleep/wake
temperature of the endogenous circadian rhythm) and sleep
schedules and prolonged naps
pressure (Process S) has decreased (Czeisler & Dijk, 2001).
· Caffeine intake disrupts sleep and causes arousals from sleep.
Therefore, the result is that sleep is regulated and sleep
· Alcohol intake initially causes sedation but can disrupt sleep later in the
consolidation is achieved by an interaction between the cir-
· Smoking may act as a central nervous system stimulant.
cadian oscillator (Process C) and the homeostatic somnostat
· Environment for sleeping: Excess light, noise, temperature extremes, and
(Process S) (Czeisler & Dijk). The two processes appear to
other stimuli (e.g., phone, pets) disrupt sleep.
interact continuously throughout the day to drive physiologic
and psychophysiologic variables (Borbely, Dijk, Acherman,
Psychological Factors
& Tobler, 2001). Thus, the coordination between the two
· Depression and anxiety are common responses to the diagnosis of or treat-
ment for cancer and are associated with disturbed sleep.
dynamic processes modulates the onset and offset of sleep
· Stress and coping: work and family responsibilities, rigorous therapeutic
as well as the rhythms of sleep propensity, wake propensity,
and the degree of daytime alertness (Czeisler & Dijk; Dijk
· Social support and loneliness: number of people available to provide sup-
& Czeisler, 1994, 1995). In addition to the two processes,
portive assistance to change in self-image
numerous demographic, lifestyle and environmental, psy-
Disease-Related Factors
chological, and disease- and treatment-related factors are
· Symptoms: Pain is experienced by as many as 80% of patients with cancer
likely to contribute to poor sleep and daytime sleepiness
and may significantly interfere with sleep; multiple symptoms occurring in
experienced by people with cancer and their caregivers (see
clusters have been linked to poor sleep.
Figure 2).
· Hormone and cytokine secretion: Factors produced by tumor cells such as
For people with cancer, altered physiology directly related
growth factors and other humoral factors can affect the timing, duration, and
to the disease may play a prominent role in disrupting sleep
quality of sleep. Tumor-related factors, such as tumor growth factor, have
and circadian regulatory processes. For example, abnormalities
been shown to inhibit sleep. Cytokines, such as interleukin (IL)-1, IL-2, IL-6,
in the circadian production of cortisol have been reported in
tumor necrosis factor (TNF), and interferon produced by the immune system
patients with cancer (Mazzoccoli et al., 2003; Raida et al.,
in response to tumors, also interfere with sleep and circadian regulation.
2002). An absent or blunted rhythm of melatonin secretion
· Site and stage of cancer: progression of disease
also has been noted in patients with lung or colorectal cancer
Disease- and Treatment-Related Factors
(Bartsch & Bartsch, 1999; Khoory & Stemme, 1988; Viviani,
· Surgery produces pain and inflammatory responses that can affect sleep;
Bidoli, Spinazze, Rovelli, & Lissoni, 1992). Changes over
anesthetics also may adversely affect sleep and waking for prolonged peri-
time in nighttime melatonin levels in patients with breast or
ovarian cancer also have been reported (Payne, 2002). In addi-
· Chemotherapy: Problems with sleep, particularly night awakenings, have
been described, but the etiology is unknown; hot flashes are common.
tion, cancer cells produce and induce production of cytokines,
· Radiation therapy: Problems with nocturnal sleep and daytime sleepiness
substances that promote sleep (Ardestani, Inserra, Solkoff, &
are documented.
Watson, 1999). More detailed summaries have been published
· Biotherapy: Cytokines (IL-2, TNF-a) are associated with daytime sleepiness,
(Clark et al., 2004; Lee, Dantzer, et al., 2004; Lee, Landis,
disturbed sleep, and depression.
et al., 2004; Vena et al., 2004). Further research is needed to
· Medication use: Many medications such as analgesics, antidepressants,
understand more fully the extent to which the pathophysiology
antiemetics, anxiolytics, and corticosteroids may adversely affect sleep.
of cancer affects sleep and wakefulness.
Figure 2. Factors That Can Interfere With Normal Sleep
Other models derived from or related to the Two-Process
Regulation in Patients With Cancer and Their Caregivers
Model of Sleep Regulation have been proposed and address
sleep/wake disturbances or health outcomes. The Conceptual
Note. Based on information from Vena et al., 2004.
Model of Impaired Sleep (Lee, 2003) describes impaired sleep
as being related primarily to either sleep deprivation or sleep
disruption, both of which can lead to adverse health outcomes
may underlie sleep/wake disturbances in people with cancer
in physiologic, cognitive-behavioral, emotional, and social
and their caregivers. Use of a particular model is dependent
domains. The PPP Model (Spielman & Glovinksy, 2004)
on the focus of a specific research study.
proposes three main categories affecting sleep, particularly
insomnia: predisposing, precipitating, and perpetuating fac-
Symptom Clusters
tors. Predisposing factors are the biologic, genetic, and demo-
Clinical experience suggests that patients with cancer often
graphic traits that increase a person's risk and susceptibility to
present with multiple symptoms, which Dodd, Miaskowski,
insomnia. Precipitating factors are situations and conditions
and Paul (2001) labeled a symptom cluster. Findings from
that, although they may be temporary, actually trigger insom-
several studies suggest that sleep disturbance may be part of
nia. Perpetuating factors are those that reinforce insomnia
a symptom cluster that includes pain, depression, and fatigue
over longer periods of time. The Two-Process Model of Sleep
(Dodd et al.; Gift, Jablonski, Stommel, & Given, 2004; Gift,
Regulation, the Conceptual Model of Impaired Sleep, and the
Stommel, Jablonski, & Given, 2003; Given et al., 2002; Given,
PPP Model offer various approaches to examining factors that
Given, Azzouz, Kozachik, & Stommel, 2001). In most of the
and researchers participating in the ONS State-of-the-Science
studies to date, the symptom cluster of pain, fatigue, and sleep
Conference on Sleep/Wake Disturbances in People With Can-
cer and Their Caregivers, nine parameters of sleep disturbance
disturbance was associated with poorer functional status or
are proposed to provide a common language of sleep distur-
decreased QOL. The question of which comes first remains
bance for measurement of the problem in all studies involving
adults and children with cancer as well as their caregivers. The
Genetics and Sleep
parameters are total sleep time, sleep latency, awakenings,
wake time after sleep onset, napping during the day, exces-
Most sleep disturbances in healthy people result from com-
sive daytime sleepiness, quality of perceived sleep, stability
plex interactions between an individual's genes and his or her
of circadian rhythms, and sleep efficiency (see Table 1). No
environment (Taheri, 2004). Modern techniques in molecular
single parameter is recommended because poor sleep cannot
genetics are being used in insect and animal models to begin
be captured in its totality by any one parameter. Instead, the
to elucidate the genetic basis for circadian rhythms and sleep
nine parameters collectively describe the characteristics of a
disturbances. Studies in animals and humans have shown that
sleep/wake disturbance.
the sleep/wake and circadian clock systems are linked closely
Table 1 includes clinical tools and research instruments to as-
with each other on the molecular and systems levels (Turek,
sess the nine sleep parameters. In addition to the few sleep tools
2004). An excellent review on the genetics of sleep disorders
or instruments that have been validated in patients with cancer
(Taheri) described studies in animals that are beginning to
(e.g., the Pittsburgh Sleep Quality Index [PSQI]), clinical tools
elucidate the molecular mechanisms that underlie narcolepsy,
and research instruments that have been used with other patient
obstructive sleep apnea, and restless leg syndrome. At least 10
populations also have been included. Clinical tools were cho-
circadian clock core genes have been identified in mammals to
sen based on their ease of use in busy clinical settings and the
date, including findings related to diurnal preferences (morn-
ability to modify therapy quickly based on findings. Research
ingness and eveningness) (Archer et al., 2003). Undoubtedly,
instruments were chosen based on what is known about reliabil-
future studies will help to determine the molecular basis for
ity (precision), validity (accuracy), and sensitivity of the instru-
sleep disturbances that occur in the context of other medical
ments in research settings (Beck, Schwartz, Towsley, Dudley, &
conditions such as cancer.
Barsevick, 2004; Carpenter & Andrykowski, 1998).
Of note, the Clinical Sleep Assessment for Adults and the
Clinical Sleep Assessment for Children were developed as
of Sleep/Wake Disturbances
brief assessments for sleep problems in clinical settings (see
Figure 3) (Lee & Ward, 2005). The tools were derived from an
Key Sleep Parameters
extensive clinical assessment protocol called BEARS, an acro-
One of the major challenges facing sleep research is measure-
nym that stands for bedtime issues, excessive daytime sleepi-
ment. "Difficulty sleeping," a term commonly used by patients,
ness, night awakenings, regularity and duration of sleep, and
caregivers, and practitioners, can have very different meanings
snoring (Owens & Dalzell, 2005). Both need extensive testing
and causes. Symptoms of sleep/wake disturbances may include
in clinical practice but are administered easily, can be scored
difficulty getting to sleep; difficulty staying asleep; restless
for research purposes, and have excellent face validity. They
sleep or the feeling that sleep is not refreshing; or sleepiness,
are useful for obtaining a brief sleep history in inpatient and
the inability to stay awake when desired. In otherwise healthy
ambulatory care settings, as well as for identifying people in
populations, relatively little is known about the various mani-
need of specific intervention strategies such as improvements
festations of sleep disturbances and how they interact to result
in sleep hygiene, changes in lifestyle, or referrals to sleep dis-
in a patient's report of "difficulty sleeping." In patients with
order specialists. When time is limited, a subset of four of the
cancer, particularly children, even less is known. Between one-
seven questions can be used (see notes in Figure 3). Using the
third and three-quarters of adult patients with cancer experience
tools can facilitate a focused approach to understanding sleep
difficulty sleeping (Andrykowski et al., 1997; Davidson et al.,
disturbances, improving sleep, and maximizing QOL for all
2002; Engstrom, Strohl, Rose, Lewandowski, & Stefanek,
people with cancer during and after treatment.
1999; Fortner, Stepanski, Wang, Kasprowicz, & Durrence,
As with any tool or instrument, each of the clinical tools
2002; Miaskowski & Lee, 1999). Children with cancer and
and research instruments listed in Table 1 has limitations.
their caregivers or parents also report sleep disturbances, but
Clinicians and researchers are urged to read more about a par-
the actual incidence of disturbed sleep in children is unknown
ticular tool or instrument before using it in practice or research.
(Ferrell, Rhiner, Shapiro, & Dierkes, 1994; Gedaly-Duff et al.,
Validation of clinical tools and research instruments in patients
2002; Hockenberry-Eaton et al., 1998).
with cancer and their caregivers is highly encouraged.
In each of the descriptive studies described in the earlier
Considerations for Assessment of Sleep
paragraph, different parameters were used to define difficulty
sleeping. Daytime sleepiness that may be a result of nocturnal
One consideration in the choice of a sleep measurement is
sleep difficulties has not been included in the studies yet, but
the temporal nature of the sleep/wake cycle. The assessment
it is an important component of poor sleep, as well as an in-
of a biologic process that normally fluctuates in a circadian
dicator of poor QOL. As researchers and clinicians attempt to
pattern is inherently challenging. Instruments that are easy to
explore sleep quality and its importance to health, their efforts
administer and interpret are needed to more fully appreciate
would be advanced significantly if consensus were achieved
the nature of sleep and sleep disturbance.
regarding which sleep parameters to measure and which mea-
Although sleep/wake disturbances can be assessed subjectively
surement instruments to use in a standard fashion. Based on a
and objectively, the subjective and objective measurements do
review of the literature and consensus of the expert clinicians
not necessarily correlate (Berger & Johnson, 2004; Young-
Table 1. Recommended Sleep Parameter Tools for Clinical and Research Settings
Sleep Parameter
Tools in Clinical Settings
Tools in Research Settings
Adults: Clinical Sleep Assessment
Adults: actigraph (Littner, Kushida, et
Total sleep time while in bed: number
Adults normally attempt to sleep
(Adult) (Lee & Ward, 2005), par-
al., 2003; Sadeh et al., 1995), Pitts-
of minutes of sleep while in bed
7­9 hours (420­540 minutes) in
24 hours. Newborns normally sleep
ent/caregiver interview, and Sleep
burgh Sleep Quality Indexa (Buysse
16­20 hours in a 24-hour period.
Diary (Ellis et al., 1981; Kryger,
et al., 1989), polysomnography,
Infants normally sleep 13­16 hours
2004; Morin & Espie, 2003, Na-
and time-lapse video recordings
in a 24-hour period. Children 2­5
tional Sleep Foundation, 1999;
(Anders & Sostek, 1976)
years old normally sleep 11­13 hours
Rogers et al., 1993)
Children and adolescents: actigraph
in a 24-hour period. Children 6­12
Children and adolescents: Clinical
(Littner, Hirshkowitz, et al., 2003;
years old normally sleep 10 hours in
Sleep Assessment (Child) (Lee &
Sadeh et al., 1995), Children's
a 24-hour period. Adolescents 13­18
Ward, 2005) and parent/caregiver
Sleep Habits Questionnaire (Owens
years old normally sleep 8.5­9.25
& Dalzell, 2005), and polysom-
hours in a 24-hour period (Grigg-
Damberger, 2004; Hoban, 2004).
Latency: number of minutes between
Adult latency normally is less than
Adults: BEARS (B­bedtime) (Owens
Adults: actigraph (Littner, Kushida, et
when someone lays down to bed and
20 minutes. Children and adolescent
& Dalzell, 2005), Clinical Sleep
al., 2003), Pittsburgh Sleep Quality
actually goes to sleep
latency normally is 30­60 minutes
Assessment (Adult) (Lee & Ward,
Indexa (Buysse et al., 1989), and
(Glaze, 2004). Preadolescent chil-
2005), parent/caregiver interview,
dren's latency averages 19 minutes
and sleep diary (Ellis et al., 1981;
Children and adolescents: actigraph
+ 1.6 minutes (Hoban & Chervin,
Kryger, 2004; Morin, 1993; Morin
(Littner, Kushida, et al., 2003; Sa-
& Espie, 2003; National Sleep Foun-
deh et al., 1995), Children's Sleep
dation, 1999; Rogers et al., 1993)
Habits Questionnaire (Owens &
Children and adolescents: Clinical
Dalzell, 2005), and polysomnog-
Sleep Assessment (Child) (Lee &
Ward, 2005) and parent/caregiver
Adults normally awaken two to six
Adults: BEARS (A­night awaken-
Adults: actigraph (Littner, Kushida, et
Awakenings during sleep period:
times during a typical night's sleep
ings, S­snoring) (Owens & Dalzell,
al., 2003), Pittsburgh Sleep Quality
the number of awakenings during a
sleep period
of 420 minutes. Children normally
2005), Clinical Sleep Assessment
Indexa (Buysse et al., 1989), and
awaken one to five times (an aver-
(Adult) (Lee & Ward, 2005), par-
age of three times) during a typical
ent/caregiver interview, and sleep
Children and adolescents: actigraph
night's sleep (Riter & Wills, 2004).
diary (Ellis et al., 1981; Kryger,
(Littner, Kushida, et al., 2003; Sa-
Night awakenings gradually decrease
2004; Morin & Espie, 2003; Na-
deh et al., 1995), Children's Sleep
in children two to five years of age
tional Sleep Foundation, 1999;
Habits Questionnaire (Owens &
(Glaze, 2004).
Rogers et al., 1993)
Dalzell, 2005), Pediatric Sleep
Children and adolescents: Clinical
Questionnaire (Chervin et al.,
Sleep Assessment (Child) (Lee
2000), and polysomnography
& Ward, 2005), parent/caregiver
interview, and sleep diary (parent
reports) (D'Andrea, 2004)
Wake after sleep onset (WASO):
Adult WASO time normally is less
Adults: No tools assessing WASO in
Adults: actigraph (Littner, Kushida, et
number of minutes awake or percent-
than 10% of the total sleep minutes,
the clinical setting were identified.
al., 2003), Pittsburgh Sleep Quality
age of time awake after sleep onset
or 42 minutes if the person sleeps
Children and adolescents: No tools
Indexa (Buysse et al., 1989), and
during the sleep period
420 minutes (seven hours) during
assessing WASO in the clinical
the night. Children's WASO time
setting were identified.
Children and adolescents: actigraph
normally is less than 20 minutes for
(Littner, Kushida, et al., 2003; Sadeh
each awakening.
et al., 1995), Children's Sleep Habits
Questionnaire (Owens & Dalzell,
2005), and polysomnography
Napping during the day: total num-
Adult napping normally can vary
Adults: parent/caregiver interview,
Adults: actigraph (Littner, Kushi-
ber of minutes of sleep during the
from five minutes to two hours. Nap-
Sleep Diary (Ellis et al., 1981;
da, et al., 2003) and Pittsburgh
daytime; can be intentional or unin-
ping for children tends to conclude at
Kryger, 2004; Morin & Espie, 2003;
Sleep Quality Indexa (Buysse et
tentional sleep
about age six (Weissbluth, 1995).
National Sleep Foundation, 1999;
al., 1989)
Rogers et al., 1993)
Children and adolescents: actigraph
Children and adolescents: parent/
(Littner, Kushida, et al., 2003; Sa-
caregiver interview, sleep diary,
deh et al., 1995) and Children's
parent report, or sleep logs
Sleep Habits Questionnaire (Owens
& Dalzell, 2005)
(Continued on next page)
These tools are highly recommended.
Although commonly reported, sleep efficiency was not believed to be the most helpful factor to assess because it is difficult to determine without polysom-
Table 1. Recommended Sleep Parameter Tools for Clinical and Research Settings (Continued)
Sleep Parameter
Tools in Clinical Settings
Tools in Research Settings
Excessive daytime sleepiness: epi-
Adults: actigraph (Littner, Kushida,
Adults normally have a minimal
Adults: BEARS (E­excessive daytime
et al., 2003), Epworth Sleepiness
sodes of lapses into sleep of short
chance of dozing while engaged in
sleepiness, S­snoring) (Owens
Scalea (Johns, 1992), Maldanado
duration, usually in situations in
routine activities. 1.7% of children
& Dalzell, 2005), Clinical Sleep
which the person is inactive for even
4­12 years old experience excessive
Assessment (Adult) (Lee & Ward,
Sleepiness Scale With Cartoon
brief periods; excessive daytime
daytime sleepiness.
2005), parent/caregiver interview,
Faces (Maldonado et al., 2004),
sleepiness can result from acute or
and sleep diary (Ellis et al., 1981;
Multiple Sleep Latency Test as
chronic sleep deprivation or loss or
Kryger, 2004; Morin & Espie, 2003;
measured with polysomnography,
other pathophysiologic causes.
National Sleep Foundation, 1999;
Pittsburgh Sleep Quality Indexa
Rogers et al., 1993)
(Buysse et al., 1989), and Stan-
Children and adolescents: Clinical
ford Sleepiness Scale (Hoddes et
Sleep Assessment (Child) (Lee
al., 1972)
& Ward, 2005), parent/caregiver
Children and adolescents: actigraph
interview, and visual analog scale
(Littner, Kushida, et al., 2003; Sa-
(Fallone et al., 2002)
deh et al., 1995), Children's Sleep
Habits Questionnaire (Owens &
Dalzell, 2005), and Multiple Sleep
Latency Test as measured with
Quality of perceived sleep: multidi-
Adults normally feel satisfied or
Adults: Clinical Sleep Assessment
Adults: Pittsburgh Sleep Quality
(Adult) (Lee & Ward, 2005), par-
Indexa (Buysse et al., 1989)
mensional perceptions of length and
very satisfied with their usual sleep
ent/caregiver interview, and sleep
Children and adolescents: Children's
depth of sleep and feelings of being
patterns and believe that their sleep
diary (Ellis et al., 1981; Kryger,
Sleep Habits Questionnaire (Owens
rested upon awakening; subjective
enhances their daily functioning. This
2004; Morin & Espie, 2003; Na-
& Dalzell, 2005)
assessment of sufficiency of sleep
has not been defined for children.
for daytime functioning
tional Sleep Foundation, 1999;
Rogers et al., 1993)
Children and adolescents: Clinical
Sleep Assessment (Child) (Lee &
Ward, 2005) and parent/caregiver
Circadian rhythm: biobehavioral
Adults: BEARS (R­regularity and
Adults: actigraph (Littner, Kushida,
Adults, children, and adolescents:
phenomenon associated with fluc-
duration of sleep) (Owens & Dal-
et al., 2003), dim light melatonin
Circadian rhythm peaks and troughs
tuations in light, hormones, eating,
within a 24-hour period.
zell, 2005) and parent/caregiver
onset, Morningness/Evening-
or socializing that repeats approxi-
ness Scalea (Taillard et al., 2004),
mately every 24 hours
Children and adolescents: parent/
Pittsburgh Sleep Quality Indexa
caregiver interview
(Buysse et al., 1989), social
rhythm metric, and 24-hour core
Children and adolescents: actigraph
(Littner, Kushida, et al., 2003; Sa-
deh et al., 1995) and Children's
Sleep Habits Questionnaire (Owens
& Dalzell, 2005)
Sleep efficiencyb: the number of
In adults, 95% sleep efficiency indi-
Adults: No tools assessing sleep ef-
Adults: actigraph (Littner, Kushida, et
minutes of sleep divided by the total
cates a good night's sleep; less than
ficiency in the clinical setting were
al., 2003) and polysomnography
number of minutes in bed, multiplied
80% indicates a bad night's sleep; in a
Children and adolescents: actigraph
by 100
night's sleep of 420 minutes (7 hours),
Children and adolescents: No tools
(Littner, Kushida, et al., 2003; Sadeh
this would be equivalent to 20 minutes
assessing sleep efficiency in the
et al., 1995), Children's Sleep Habits
to fall asleep and three awakenings of
clinical setting were identified.
Questionnaire (Owens & Dalzell,
10 minutes each. No references were
2005), and Pediatric Sleep Ques-
found for children.
tionnaire (Chervin et al., 2000)
These tools are highly recommended.
Although commonly reported, sleep efficiency was not believed to be the most helpful factor to assess because it is difficult to determine without polysom-
McCaughan et al., 2003). Certain research instruments, such as
section of the PSQI includes specific questions to be answered
sleep diaries, assess individuals' perceptions of problems falling
by bed partners, and the Clinical Sleep Assessment (Child) is
asleep and sleep quality, whereas others, such as wrist actigraphy,
completed by parents of young or otherwise nonverbal children.
assess actual physical movements during sleep-activity periods.
Understandably, sleep information obtained from bed partners
Valuable sleep information can be obtained from bed part-
or caregivers is affected by their own sleep patterns. Skilled
ners, parents of children, and caregivers. For example, the final
clinicians are able to synthesize the data as treatments plans are
Clinical Sleep Assessment (Adult)
Clinical Sleep Assessment (Child)
1. Does your child attend
1. What is your employment status?
___ Day care (part-time)?
___ Retired or unemployed
___ Day care (full time)?
___ Self-employed or homemaker
___ School?
___ Employed
School starts at ___ am/pm
___ Day shift
School ends at ___ am/pm
___ Night shift
2. In the past month, has your child taken a medication or alcohol to sleep?
___ Rotating shifts
___ No ___ Yes
Work hours: ___ am/pm to ___ am/pm
If yes, list name ____________________________
2. In the past month, have you taken medication or alcohol to help you
Was it taken
___ Fewer than one time per week?
___ No ___ Yes
___ One or two nights per week?
If yes, list name. ____________________________
___ Three or more nights per week?
Did you take it
3. Considering the past month, how would you rate your child's sleep quality?
___ Fewer than one time per week?
___ Very good
___ One or two nights per week?
___ Fairly good
___ Three or more nights per week?
___ Fairly bad
3. Considering the past month, how would you rate your sleep quality?
___ Very bad
___ Very good
4. In the past week, what time did your child typically go to bed?
___ Fairly good
___ am/pm weeknights/school days
___ Fairly bad
___ am/pm on weekends
___ Very bad
How difficult was it for your child to settle and fall asleep after bedtime rituals?
4. In the past week, what time did you typically turn out the light to go to
___ Not at all difficult
___ A little difficult
___ Somewhat of a struggle
___ am/pm on weeknights/workdays
___ A constant struggle
___ am/pm on weekends/days off
5. In the past week, what time was your child's typical final awakening?
5. In the past week, what time was your typical final awakening?
___ am/pm on weeknights/school days
___ am/pm on weeknights/workdays
___ am/pm on weekends
Total hours of sleep on weeknights ___
How difficult was it for your child to get up in the morning?
___ am/pm on weekends/days off
___ Not at all difficult
Total hours of sleep on weekends/days off ___
___ A little difficult
6. In the past week, how many times did you typically wake up during the
___ Somewhat of a struggle
___ A constant struggle
___ Never or once
6. In the past week, how many times did your child wake up during the night?
___ One or two times, depending on the night
___ Never or once
Total minutes ___
___ One or two times, depending on the night
___ Two or three times, depending on the night
___ Two to three times, depending on the night
Total minutes ___
___ Three to four times, depending on the night
___ Three to four times, depending on the night
___ Always three or more times
Total minutes ___
What is the main reason for waking up?
___ Always three or more times
___ Thirst
Total minutes ___
___ Bladder
What is the main reason for waking up?
___ Other (Please note snoring, choking, kicking, hitting, head banging,
___ Bladder
teeth grinding, bed wetting, sleep walking or talking, nightmares, night
___ Other (Please note snoring, choking, suffocating, kicking, hitting, etc.)
terrors, etc.)
7. In the past week, how sleepy did you feel during the day?
7. In the past week, how sleepy was your child during the day?
___ Not at all sleepy
___ Not at all sleepy
___ A little sleepy
___ A little sleepy
___ Very sleepy
___ Very sleepy
___ I would fall asleep when I did not really want to or plan to fall asleep.
___ Takes a nap most days
___ Falls asleep during class
___ Falls asleep on the way home from school
Note. A short form of this screen should include items 2, 3, 6, and 7. For
the long or short form, a referral to a sleep specialist is indicated for any of
Note. A short form of this screen should include items 2, 3, 6, and 7. For the
the following responses: #2, medication taken three or more nights/week,
long or short form, referral to a sleep specialist is indicated for any of the fol-
#3, sleep quality is very bad; #6, wake up three or four times depending
lowing responses: #2, medication taken three or more nights/week; #3, sleep
on night; or wake up three or more times/night; #7, fall asleep when not
quality is very bad; #6, wake up three or four times depending on night or wake
up three or more times/night; #7, fall asleep on the way home from school.
Note. For content validity, questions were reviewed by a panel of sleep experts
in research and clinical practice. Neither this seven-item assessment tool nor
Note. For content validity, questions were reviewed by a panel of sleep experts
the essential four items for rapid screening have been rigorously tested in
in research and clinical practice. Neither this seven-item assessment tool nor
clinical practice.
the essential four items for rapid screening have been rigorously tested in
clinical practice.
Note. From "Critical Components of a Sleep Assessment for Clinical Practice
Settings," by K.A. Lee and T.M. Ward, 2005, Issues in Mental Health Nurs-
Note. From "Critical Components of a Sleep Assessment for Clinical Practice
ing, 26, p. 742. Reprinted with permission. Copyright 2005 by Taylor and
Settings," by K.A. Lee and T.M. Ward, 2005, Issues in Mental Health Nursing,
Francis. Reprinted with permission.
26, p. 743. Copyright 2005 by Taylor and Francis. Reprinted with permission.
Figure 3. Clinical Sleep Assessment for Adults and Children
tions or strategies suggested for implementation. However,
developed in clinical settings; however, quantitative methods to
definitions and rationales for the use of interventions were
handle the data should be developed for research purposes.
consistent across sources.
Time frame is an important consideration in the choice of
Meta-analyses of nonpharmacologic interventions for
a sleep measurement. For example, the two most frequently
disturbances in sleep: To determine possible therapeutic
used measurements of daytime sleepiness are the Stanford
modalities to improve sleep in people with cancer and their
Sleepiness Scale (Hoddes, Zarcone, & Dement, 1972) and
caregivers, a search was conducted for meta-analyses of
the Epworth Sleepiness Scale (Johns, 1992), both self-report
nonpharmacologic cognitive and behavioral interventions
measurements. The Stanford Sleepiness Scale was devel-
oped to assess sleepiness several times throughout a day
for insomnia in noncancer populations. Searches of MED-
LINE, CINAHL, PsycINFO, and the Cochrane Library from
in conjunction with the multiple sleep onset latency test (a
1994­2004 were conducted using the key term sleep inter-
polysomnography study of daytime sleepiness) and can be
ventions and limit of meta-analysis. Five analyses were found
performed on an hourly basis to examine change in sleepi-
ness during the day as well as time of day when sleepiness
that provided the state-of-the-science information on the
is greatest. The Epworth Sleepiness Scale, on the other hand,
effectiveness of nonpharmacologic interventions to improve
asks about how likely it would be for a person to doze off in
sleep in presumably healthy noncancer populations (see Table
various activities that typically are performed over the course
3). None of the reviews addressed insomnia interventions in
of a day (e.g., watching TV), week (e.g., as a passenger in a
populations with cancer.
car), or month (e.g., attending the theater). Prospective mea-
The Cochrane Database has published highly controlled,
surements, such as a sleep diary or the Stanford Sleepiness
systematic reviews on bright light therapy, physical exercise,
Scale, and retrospective measurements, such as the Epworth
and cognitive and behavioral interventions to improve sleep.
Sleepiness Scale, have strengths and weaknesses related to
Each systematic review includes information about the search
reliability that clinicians and researchers need to consider
strategy used. Cognitive and behavioral interventions are de-
fined as treatments that aim to improve sleep by changing poor
when matching a research instrument to a specific research
question or clinical concern.
sleep habits and challenging negative thoughts, attitudes, and
beliefs about sleep. The interventions include a broad range of
treatments, from educational packages to those that employ
behavioral strategies. Only cognitive and behavioral treatments
provided level 1 (strongest) evidence of short-term effects in re-
Since 1985, psychological and behavioral factors increas-
ducing sleep onset latency and wake after sleep onset; however,
ingly have been recognized as contributing to the development
the treatment effects were not durable over time (Montgomery
of insomnia. More than a dozen nonpharmacologic, cognitive,
& Dennis, 2003). Two earlier meta-analyses (Morin, Culbert, &
and behavioral interventions have been developed; however,
Schwartz, 1994; Murtagh & Greenwood, 1995) provided stron-
ger evidence that nonpharmacologic interventions produced
the populations in which they have been tested have been
reliable and durable clinical benefits in reducing sleep onset la-
otherwise healthy people with insomnia. Only discussion
tency and wake after sleep onset. Differences in the conclusions
of interventions for insomnia in adults is possible because
of the two earlier meta-analyses compared to a more recent
information is limited in regard to children. To identify com-
monly used nonpharmacologic interventions to promote
report by Montgomery and Dennis (2003) may be attributed,
sleep or prevent insomnia, a two-step search of the literature
in part, to more rigorous inclusion criteria that resulted in fewer
indexed was conducted using the major literature search en-
studies being included in the recent analysis. Larger effects
gines (MEDLINE®, Cumulative Index to Nursing and Allied
were found for people who were referred by clinicians and not
Health Literature®, and PsycINFO). Key words used for the
regular users of hypnotics (Murtagh & Greenwood). Current
primary search included sleep, sleep disturbance, intervention,
reviews are limited by the lack of randomized, controlled trials
and insomnia. From the findings, primary research, research
of sleep in people with cancer and their caregivers. As evidence
summary, and meta-analysis articles were selected.
grows regarding the effects of interventions for insomnia and
The secondary search of the literature was conducted using
other sleep/wake disturbances in cancer and noncancer popula-
tions, additional meta-analyses are recommended.
the names of the interventions used in the articles identified
Nonpharmacologic intervention studies of sleep/wake
in the primary search (e.g., sleep restriction, sleep hygiene,
disturbances in patients with cancer: Clark et al. (2004)
relaxation therapy, massage, yoga). When possible, classic
examined research published from 1980­2003 on sleep/wake
sources were identified and retrieved for each intervention and
disturbances in people with cancer. Fifty-two pieces of evi-
compared with recent sources to provide definitions, rationale,
dence, including descriptive and intervention studies, were used
and recommendations or strategies for each intervention. The
in the evidence tables and rated from level 1 (strongest) to level
findings are summarized in Table 2. Although the authors
3 (weakest) of evidence. Readers are encouraged to review the
recognize the limitations of the generalizability of the stud-
ies, the body of knowledge provides a foundation from which
evidence tables to gain perspective on the state of the science.
to generate knowledge of interventions that are effective in
From 2003­2004, an increasing volume of descriptive
people with cancer and their caregivers.
studies was published that included the variable of sleep in
Nonpharmacologic interventions to promote sleep and
some context in patients with cancer. A search of MEDLINE,
prevent or treat insomnia often were used as combination
CINAHL, and PsycINFO was conducted to identify all
therapies (cognitive and behavioral therapies) depending on
nonpharmacologic intervention studies that examined sleep
the needs or restrictions of the study populations. The use of
disturbance or sleep quality outcomes in adults with cancer;
different combinations of interventions may explain some of
the search revealed 20 studies (as of March 2005) that met
the differences across studies in terms of the recommenda-
the criteria (see Table 4).
Table 2. Cognitive and Behavioral Interventions Used to Promote Sleep
Definition and Rationale
Recommendations and Strategies
Stimulus control (Bootzin et al., 1991)
Definition: reassociating the bed and bedroom with
· Go to bed only when sleepy.
rapid sleep onset by limiting sleep-incompatible
· Use the bedroom only for sleep and sex.
activities that cue wakefulness and by enforcing
· Maintain a regular arising time in the morning
a consistent sleep-wake schedule
regardless of sleep duration the previous night.
Rationale: based on the theory that insomnia is
· Get out of bed and go into another room when-
an altered conditioned response to time and
ever unable to fall asleep or return to sleep within
environmental cues that usually are associated
15­20 minutes, and return to bed only when
with sleep
sleepy again.
· Avoid daytime napping.
Sleep restriction (Morin et al., 1999; Petit et al.,
Definition: limiting the amount of time spent in
· Log sleep efficiency for one week.
2003; Spielman et al., 1987)
bed to closer approximate the amount of time
· Prescribe the amount of time in bed to match the
time slept on the log.
Rationale: creates a mild state of sleep deprivation
· Evaluate sleep efficiency weekly.
and promotes more rapid sleep onset, more ef-
· If > 90%, increase time in bed by 15 minutes.
ficient sleep, and less internight variability
· If < 80%, decrease time in bed by 15 minutes.
· If 80%­89%, keep the same sleep time.
· To avoid excessive daytime sleepiness, do not
prescribe less than five hours in bed per night.
Paradoxical intention (Morin et al., 1999)
Definition: encouragement of patients to engage in
· Instruct patient to attempt to stay awake without
the use of stimulants.
their most feared activity, staying awake
Rationale: based on the theory that performance
anxiety inhibits sleep onset. Thus, if a patient
stops trying to sleep and instead genuinely at-
tempts to stay awake, performance anxiety will
be lessened and sleep may come more easily.
Definition: health practices and environmental
· Avoid caffeine and nicotine a minimum of four to
Sleep hygiene (healthy sleep practices) (Hauri,
1991; Lacks & Rotert, 1986)
factors that may be either barriers to or facilita-
six hours before bed.
tors for sleep
· Avoid strenuous exercise five to six hours before
Rationale: If patients are able to identify factors (in-
ternal and external) that are affecting their sleep,
· Avoid use of alcohol as a sleep aid.
they will be able to choose to remove barriers
· Reduce noise, light, and temperature in bedroom.
and increase facilitators of their sleep.
· Keep a regular sleep schedule.
· Participate in relaxing activities prior to bed.
Cognitive therapy for insomnia (cognitive restruc-
Definition: use of restructuring techniques such
· Target current perceptions of sleep to be ad-
turing) (Morin, 1993)
as decatastrophizing, hypothesis testing, reap-
dressed by cognitive therapy
praisal, and attention shifting to alter false beliefs
· Avoid unrealistic sleep expectations (e.g., "I must
and attitudes about sleep
get eight hours of sleep every night.").
Rationale: short-circuits the vicious cycle of insom-
· Avoid misconceptions about the causes of in-
nia, emotional distress, dysfunctional cognitions,
somnia (e.g., "My insomnia is entirely due to a
and further sleep disturbances by identifying
chemical imbalance.").
patient-specific dysfunctional sleep thoughts,
· Avoid amplifications of the consequences of
challenging their validity, and replacing them with
insomnia (e.g., "I cannot accomplish anything
more adaptive substitutes
after a poor night's sleep.").
· Avoid performance anxiety resulting from exces-
sive attempts at controlling the sleep process
(e.g., "I just need to try harder to sleep.").
Relaxation therapy (Benson & Stark, 1996; Lich-
Definition: activities that reduce somatic or cognitive
· Selection of specific techniques varies depend-
stein, 1988; Seaward, 2002)
arousal. Somatic: Therapies focus on reducing
ing on whether somatic or cognitive arousal
physiologic arousal (e.g., muscle tension, hyper-
is targeted for treatment. Duration, frequency,
tension), including but not limited to progressive
and intensity of treatments are dependent on
muscle relaxation, massage, biofeedback, autogen-
particular therapies selected.
ic training, and self-hypnosis. Cognitive: Therapies
focus on reducing cognitive arousal (e.g., intrusive
thoughts, racing mind), including but not limited
to imagery and visualization training, thought stop-
ping, and mindfulness meditation.
Rationale: Patients with insomnia often have high
levels of somatic and cognitive arousal, both at
night and during the daytime. Relaxation meth-
ods are used to deactivate the arousal system
(Continued on next page)
and promote sleep.
Table 2. Cognitive and Behavioral Interventions Used to Promote Sleep (Continued)
Definition and Rationale
Recommendations and Strategies
Phototherapy, bright light therapy (Smith et al.,
Definition: Phototherapy is the use of bright light
· Phase delay insomnia (go to bed late and wake
to influence the circadian rhythm. Phototherapy
up late)--Bright light exposure in the morning
often is accomplished via a light box that gener-
for a minimum of 30 minutes is recommended.
ates blue spectrum light at 4,000­10,000 lux.
· Phase advance insomnia (go to bed early and
The dose is adjusted by altering the distance and
wake up early)--Bright light exposure in the late
duration of light exposure.
evening/early night for a minimum of 30 minutes
Rationale: Bright light in the morning advances the
is recommended.
timing of the circadian phase, whereas exposure
in the evening delays the circadian phase. The
sleep-promoting effects of bright light may occur
via several mechanisms, including shifting the
circadian system, enhancing the amplitude of
the circadian pacemaker, promoting wakefulness
during the day and sleep at night, or indirectly,
via its antidepressant effects.
Exercise (Buxton et al., 2003; Li et al., 2004; Na-
Definition: bodily exertion at regular intervals for
· Exercise that incorporates stretching, concentra-
tional Center for Complementary and Alternative
sustained periods of time
tion, and strength may improve mood as well as
Medicine, 2004; Neubauer, 1999)
Rationale: Regular exercise, such as brisk walking,
provide circadian influences. Recommendations
may help to improve sleep quality and shorten
are for 60-minute sessions at least three times
per week.
sleep latency. Other types of exercise have
· Timing of exercise can affect circadian phase.
been linked to improved overall sleep quality,
­ Significant phase delays were reported for
decreased sleep latency, and decreased daytime
nocturnal exercise.
sleepiness. Exercise also has been shown to be
­ Significant phase advances were reported for
effective as an intervention for circadian phase
evening exercise.
· Duration and intensity recommendations vary--
typically, study exercise activities last for 60
minutes and are of the intensity of 60%­75%
Definition: Yoga is a comprehensive discipline that
· Daily practice is encouraged. Can be done at the
Yoga (Cohen et al., 2004; National Center for
includes physical exercises, postures, breathing
beginning or end of each day to release stress
Complementary and Alternative Medicine, 2004;
Seaward, 2002)
techniques, and meditation for the purpose of
and frustration.
improving health and well-being.
Rationale: Yoga is used to decrease cognitive and
physiologic arousal, thus improving sleep onset
latency times.
Acupuncture (National Center for Complementary
Definition: Acupuncture is a family of procedures
Recommended method, duration, frequency, and
and Alternative Medicine, 2004; Seaward, 2002;
involving stimulation of anatomical points on
total number of treatments vary widely.
Sok et al., 2003; Spence et al., 2004)
the body for the purpose of regulating energy
All treatments need to be administered by an in-
(Qi) flow.
dividual trained in traditional Chinese medicine
Rationale: Insomnia is a result of stagnated Qi.
Acupuncture regulates Qi, resulting in improved
duration, latency, and efficiency measures.
All of the interventions employed combinations of cognitive
eight sessions; exercise interventions (Mock et al.; Young-Mc-
or behavioral techniques. The most commonly used interven-
Caughan et al.) and mindfulness stress reduction (Carlson et
tion technique was training in relaxation strategies (Allison et
al., 2003, 2004) included practice two to seven times per week.
al., 2004; Berger et al., 2002, 2003; Cannici, Malcolm, & Peek,
The individualized sleep-promotion plan (Berger et al., 2002,
1983; Carlson, Speca, Patel, & Goodey, 2003, 2004; Davidson,
2003) was reinforced and revised every 21­30 days.
Waisberg, Brundage, & MacLean, 2001; Simeit, Deck, &
Notwithstanding the positive findings of most of the cited
Conta-Marx, 2004). All of the studies reported improvement
studies, the quality of the evidence in support of cognitive and
in or stability with some sleep parameters. Four studies tested
behavioral interventions for sleep was generally consistent but
the effect of mindfulness meditation techniques (Carlson
lacking in randomized multienter trials with at least 100 par-
ticipants (Hadorn, Baker, Hodges, & Hicks, 1996). Only six of
et al., 2003, 2004; Cohen, Warneke, Fouladi, Rodriguez, &
the 20 studies used randomized clinical trial designs (Cannici
Chaoul-Reich, 2004; Shapiro, Bootzin, Figueredo, Lopez, &
et al., 1983; Coleman et al., 2003; Dalton, Keefe, Carlson, &
Schwartz, 2003) and reported similar improvements in some
Youngblood, 2004; de Moor et al., 2002; Kim, Roscoe, & Mor-
sleep parameters. Exercise was tested in three studies (Cole-
man et al., 2003; Mock et al., 2001; Young-McCaughan et al.,
row, 2002; Shapiro et al., 2003); four used quasiexperimental
2003). Most of the interventions were conducted in three to
designs (Cohen et al., 2004; Mock et al., 2001; Simeit et al.,
Table 3. Meta-Analyses of Nonpharmacologic Interventions for Sleep Disturbances in Noncancer Populations
Study Information
Conclusions and Implications
Search strategy: MEDLINE (1966­January 2001); EMBASE
No trials met the inclusion criteria for this review by the
Montgomery & Dennis, 2002a
(1980­January 2001); Cumulative Index to Nursing and
Cochrane Collaboration.
Allied Health Literature (CINAHL) (1982­January 2001);
PsycINFO (1970­2001); the Cochrane Library Issue 1,
(2001); National Research Register (NRR [2001]). Bibli-
ographies of existing reviews in the area, as well as of all
trial reports obtained, were searched. Experts in the field
were consulted.
Sample: no studies of bright light therapy for older adults
(age > 60) with primary insomnia
Montgomery & Dennis, 2002b
Search strategy: MEDLINE (1966­January 2001); EMBASE
Only one study met the criteria for review by the Cochrane
(1980­January 2002); CINAHL (1982­January 2002);
PsycINFO (1987­2002); the Cochrane Library Issue 1,
Insufficient evidence exists to determine the effectiveness
(2002); NRR (2002). Bibliographies of existing reviews
of physical exercise for the treatment of sleep problems in
in the area, as well as of all trial reports obtained, were
healthy older adults.
searched. Experts in the field were consulted.
Sample: one study of physical exercise for older adults (> 60
years) with primary insomnia
Outcomes: sleep onset latency, total sleep time, and sleep
Treatment evaluated: 16 weeks of moderate-intensity, com-
munity-based exercise (4 x 40 minutes of endurance
training per week)
Montgomery & Dennis, 2003
Search strategy: MEDLINE (1966­October 2001); EMBASE
Cognitive and behavioral treatments for sleep problems are
(1980­January 2002); CINAHL (1982­January 2002);
mildly effective for some aspects of sleep in the short term;
PsycINFO (1970­2002); the Cochrane Library Issue 1,
the treatments are not always durable.
(2002); NRR (2002). Bibliographies of existing reviews
The interventions had the following effects.
in the area, as well as of all trial reports obtained, were
· Sleep onset latency (three studies) for the treated group
searched. Experts in the field were consulted.
improved slightly after treatment, three minutes less than
Sample: six trials of cognitive and behavioral interventions for
the control group; after one year, it was 11.5 minutes better
older adults (> 60 years) with primary insomnia
than the control group.
Outcomes: sleep onset latency, total sleep time, sleep ef-
· Wake after sleep onset (four studies) improved modestly
ficiency, wake after sleep onset, total wake time, early
immediately after treatment, 21.9 minutes less for the treat-
morning awakening, and sleep quality
ed group; at three months post-treatment (two studies),
Treatments evaluated: sleep hygiene, stimulus control, muscle
wake time was 33 minutes less for the treated group; after
relaxation, sleep restriction, and cognitive therapy
one year (one study) wake time was only 13 minutes less
for the treated group.
· Total wake time (one study) lessened for the treated group
immediately after treatment, one hour less by sleep diary,
and 38 minutes less by polysomnography (PSG).
· Sleep duration (four studies) improved mildly, 14.6 minutes
more sleep by sleep diary; however, by PSG, it decreased by
19 minutes; at three months (one study), treated people had
14.8 minutes less sleep; after one year (one study), treated
people had 32 minutes more sleep; however, by PSG, the
control group was sleeping almost seven minutes more per
· Early morning awakening (one study) improved modestly
after treatment, with the treated group waking 17 minutes
later than the control group by sleep diary; by PSG, waken-
ing was 14.9 minutes later.
· Sleep efficiency (three studies) improved modestly post-
treatment with 7.5% improvement; at three months (one
study), the effect was stronger, 9.6%; after one year (one
study), the effect was weaker, 4.4%.
· Sleep quality (one study) measured by the Pittsburgh Sleep
Quality Index was modestly improved, score = 7.8 for the
treated group compared with 10.6 for the control group; at
three months, score for the treated group was 6.2 compared
with 10.2 for the control group; however, all scores were
higher than the 5.0 cut-off indicating sleep disturbance.
(Continued on next page)
Table 3. Meta-Analyses of Nonpharmacologic Interventions for Sleep Disturbances in Noncancer Populations (Continued)
Study Information
Conclusions and Implications
Nonpharmacologic interventions provide reliable and durable
Morin et al., 1994
Search strategy: computer search of studies that met the
clinical benefits. The average treated person had
following criteria
· Shorter sleep onset latency than 81% of untreated people.
· The primary target problem was sleep onset, maintenance,
· Less time awake after sleep onset latency than 74% of
or mixed insomnia.
· The treatment was nonpharmacologic.
untreated people.
· The study used a group design.
· Fewer number of awakenings than 70% of untreated
· The outcome measures consisted of one or more of the
following: sleep onset latency, time awake after sleep onset,
· Longer total sleep time than 66% of untreated people.
number of awakenings, and total sleep time.
Clinical gains at the end of treatment were maintained at fol-
Case reports or studies based on a single-subject design
low-ups averaging six months in duration.
and studies evaluating pharmacologic treatments were
Comparison of treatments determined that stimulus control
was the most effective single therapy for sleep onset or
Sample: 59 nonpharmacologic treatment outcomes studies of
maintenance difficulties.
people with sleep onset, maintenance, or mixed insomnia
Outcomes: sleep onset latency, total sleep time, number of
awakenings, and time awake after sleep onset
Improvement rates evaluated from pre- to post-treatment and
from post-treatment to follow-up
Treatments evaluated: nonpharmacologic treatment, including
stimulus control, sleep restriction, relaxation, paradoxical
intention, sleep hygiene education, and other therapies
Murtagh & Greenwood, 1995
Search strategy: Key word "insomnia"; MEDLINE and Psy-
Psychological treatments of insomnia enhanced sleep pat-
chLIT (1973­1993) and the reference lists of relevant review
terns and subjective sleep experience.
articles and books. Unpublished studies were identified from
Average treatment effects for the hypothetical average case
listings of dissertations and theses over the same period.
of insomnia may be expected to
Sample: 66 treatment outcome studies of healthy people
· Reduce sleep onset latency from 61 to 37 minutes.
with insomnia
· Increase total sleep time from 5.65 to 6.18 hours.
Outcomes: sleep onset latency, total sleep time, nightly
· Decrease number of nightly awakenings from 1.63 to 0.44.
awakenings, and sleep quality ratings
The majority of active treatment effect sizes was greater
Treatments evaluated: relaxation-based therapies such as
than corresponding placebo effect sizes but did not differ
progressive muscle relaxation, meditation, desensitization,
greatly in efficacy.
imagery relaxation, hypnosis, and autogenic training
Greater gains were found for participants who were clinically
Nonrelaxation therapies, such as stimulus control, paradoxical
referred and who were not regular users of hypnotics.
intention, sleep restriction, and combination treatments
2004; Smith, Kemp, Hemphill, & Vojir, 2002). Only three
half-lives. Those with longer half-lives can cause daytime
studies had sample sizes greater than 100 participants (Dalton
sleepiness and impair waking cognitive and motor function,
et al., 2004; Kim et al.; Simeit et al.), a criterion for the stron-
and those with shorter half-lives may wear off before desired
gest level of evidence (Hadorn et al.; Ropka & Spencer-Cisek,
or scheduled wake-up times.
2001). Because combinations of intervention strategies were
Antidepressants including serotonin reuptake inhibitors
used, specifying the essential elements that accounted for the
(SSRIs) and tricyclic antidepressants also may aid sleep,
success of a given intervention is difficult. Future studies need
depending on dosages used. They have demonstrated some
to disentangle such issues by statistical methods before conclu-
benefit for treatment of concomitant depression, neuropathic
sions can be drawn about the most effective interventions to
pain, hot flashes, and night sweats. SSRIs have varying effects
improve sleep in patients with cancer. The studies also were
on anxiety and daytime activity. The drugs have a narrow
limited by a lack of clear definition of terms regarding sleep
therapeutic index, and signs of toxicity should be monitored
or use of measurements that may contain only a single-item
rating of sleep.
To evaluate the efficacy of current pharmacologic approach-
es, a search was conducted to identify meta-analyses. The
search strategy included MEDLINE, CINAHL, PsycINFO,
Numerous pharmacologic approaches can be used to en-
and the Cochrane Library covering 1994­2004, using the
hance sleep. Few of the approaches have been studied in popu-
key terms insomnia, sleep disturbance, hypnotics, pharmaco-
lations with cancer, although the drugs are used commonly
therapy, and meta-analysis. Stimulants were not included in
clinically in people with cancer. Table 5 summarizes specific
the search; however, they are being tested to improve daytime
medications by class or category, hypnotic dose, and onset and
sleepiness and fatigue in people with cancer. The efficacy
duration of action. Specific features of drugs also are noted.
of hypnotic drugs has been evaluated in five meta-analyses
Hypnotics are the most commonly prescribed medications
published from 1997­2004. None of them was specific to the
and include benzodiazepines and nonbenzodiazepine hypnot-
use of the drugs in patients with cancer or their caregivers.
ics, sometimes referred to as Z drugs. The drugs vary in their
Specific details are summarized in Table 6. The reports were
Table 4. Nonpharmacologic Intervention Studies in Patients With Cancer
Intervention Tested
Sample, Setting, and
Author and Year
or Compared
Treatment Phase
Allison et al., 2004
NuCare Coping Strat-
Pilot feasibility study
66 patients with head
Statistically significant
Priority Symptom Man-
egies Program:
Prospective nonran-
and neck cancer en-
improvement in
a g e m e n t (PRISM)
self-study book and
domized design; no
rolled; 59 completed;
· Sleep disturbance
level of evidence
audiocassette de-
control group; one
50 gave outcome
· Global quality of life
(Ropka & Spencer-
s i g n e d to enhance
· Fatigue
Cisek, 2001) = II (6)
personal control and
Groups: choice of small
· Depression
· No placebo control
teach emotional and
group (n = 3), one
instrumental coping
on one (n = 33), or
· Small sample size
home format without
· Feasibility study not
Training in
therapist (n = 23)
designed to test ef-
· Problem solving
Setting: Canada
· Relaxation
Treatment phase: not
· No objective sleep
· Coping skills
· Goal setting
Compliance: 59 of 66
· Communication
completed the study.
· Social support
· Lifestyle factors
Berger et al., 2002
Individualized sleep
Pilot feasibility study
25 patients with breast
Sleep latency, efficiency,
P R I S M level of evi-
promotion plan
One group; pre- and
total rest, and rating
dence (Ropka &
· Sleep hygiene
post-test design
Setting: midwestern
of awakening were
Spencer-Cisek, 2001)
· Relaxation therapy
United States, oncol-
consistently within
= II (6)
· Stimulus control
ogy clinics and pa-
the desired range.
· No placebo control
· Sleep restriction
tient homes
Time awake after sleep
One two-hour session;
Treatment phase: active
onset and nighttime
· Small sample size
four 15-minute prob-
awakenings exceeded
· Feasibility study not
lem-solving sessions;
Compliance: 22 of 25
desired levels.
designed to test ef-
three 30-minute revi-
completed the study;
No statistically signifi-
sion sessions
moderate to high ad-
c a n t changes over
herence (46%­80%)
time in any variables
with components of
Conclusion: interven-
sleep plan
tion feasible, adher-
ence improved over
time, most sleep/
wake patterns con-
sistent with normal
Berger et al., 2003
Individualized sleep-
One group; pre- and
21 patients with breast
Self-reported and ac-
P R I S M level of evi-
promotion plan
post-test design
tigraph readings of
dence (Ropka &
· Sleep hygiene
Setting: midwestern
sleep latency dem-
Spencer-Cisek, 2001)
· Relaxation therapy
United States, patient
onstrated that it was
= II (6)
· Stimulus control
stable and within de-
· No placebo control
· Sleep restriction
Treatment phase: long-
sired range.
One two-hour session;
term follow-up
Actigraph readings dem-
· Small sample size
four 15-minute prob-
Adherence to compo-
onstrated that 74%­
· Feasibility study not
lem-solving sessions;
n e n t s of interven-
88% were awake for
designed to test ef-
three 30-minute revi-
more than the desired
sion sessions
· 77%­88% for sleep
30 minutes after sleep
Inter vention may be
hygiene counseling
onset. Awakenings
beneficial in reducing
· 83%­88% for relax-
were more frequent
nighttime awakenings
ation therapy
than six per night.
and daytime naps.
· 36%­56% for stimu-
Number of nighttime
lus control
awakenings and length
· 83%­88% for sleep
of daytime naps de-
creased over time.
Cannici et al., 1983
Muscle relaxation train-
Randomized clinical
30 patients with can-
Sleep onset latency re-
P R I S M level of evi-
duced in relaxation
dence (Ropka &
Three sessions
Groups: relaxation, n
group compared with
Spencer-Cisek, 2001)
= 15; usual care, n
usual care group.
= II (6)
= 15
(Continued on next page)
Table 4. Nonpharmacologic Intervention Studies in Patients With Cancer (Continued)
Intervention Tested
Sample, Setting, and
Author and Year
or Compared
Treatment Phase
Setting: southeastern
After three months, dif-
· Small sample size
United States, hospi-
· No objective sleep
ferences in sleep laten-
tal or patient's home
cy were maintained.
Treatment phase: active
No differences in hours
Inter vention may be
treatment or long-
of sleep, number of
effective in reducing
term follow-up
night awakenings,
sleep onset latency.
Compliance: 26 of 30
sleep satisfaction, or
completed the study.
feeling rested
One group; pre- and
Carlson et al., 2003
Pretest: n = 59 patients
Data were presented for
P R I S M level of evi-
post-test design
with breast and pros-
42 participants who
dence (Ropka &
meditation program
tate cancer
completed the study.
Spencer-Cisek, 2001)
· Relaxation
Setting: Canada, outpa-
Significant improve-
= II (6)
· Meditation
tient department
ments in
· No placebo control
· Gentle yoga
Treatment phase: long-
· Overall quality of life
Nine sessions, daily
term follow-up
· Symptom of appetite
· Small sample size
· Findings were not
· 42 of 59 completed
· Stress symptoms
correlated with pro-
the study.
No significant change
gram attendance or
· 52 of 59 completed
in sleep quality
home practice.
five or more interven-
Immune cells:
· Compliance prob-
tion sessions.
· No change occurred
· Median sessions at-
in overall number of
· No objective sleep
tended = eight
· Average practice time
· T-cell production
per day = 24 minutes
of interleukin-4 in-
(meditation) and 13
minutes (yoga)
· Interferon-gamma
· Nonkiller cell produc-
tion of interleukin-10
Shift in immune profile
was consistent with
shift in depressive
symptoms to a more
normal profile.
Carlson et al., 2004
One group; pre- and
59 patients with breast
Postintervention data
P R I S M level of evi-
post-test design
cancer and 10 pa-
available for 31 of 42
dence (Ropka &
meditation program
tients with prostate
Spencer-Cisek, 2001)
· Relaxation
Significant improve-
= II (6)
· Meditation
Setting: Canada, outpa-
ments from pre- and
· No placebo control
· Gentle yoga
tient department
post-test interven-
Nine sessions, daily
Treatment phase: long-
tion in
· Small sample size
term follow-up
· Overall quality of life
· Findings were not
· Symptoms of stress
correlated with pro-
· 42 of 59 completed
· Sleep quality
gram attendance or
the study.
Improvements not sig-
home practice.
· 52 of 59 completed
nificantly correlated
· Compliance prob-
the intervention.
with the degree of
· Median sessions at-
program attendance
· No objective sleep
tended = eight
or minutes of home
· Average practice time
Inter vention may be
per day = 24 minutes
Average daily cortisol
effective in improving
(meditation) and 13
level did not change
sleep quality.
minutes (yoga)
from pre- to post-test
Cohen et al., 2004
Tibetan yoga (TY) ver-
Quasiexperimental two-
39 patients with lym-
TY group reported low-
P R I S M level of evi-
sus wait list control
er sleep disturbance
dence (Ropka &
group design with se-
quential assignment
G r o u p s : TY, n = 20;
(Continued on next page)
Table 4. Nonpharmacologic Intervention Studies in Patients With Cancer (Continued)
Intervention Tested
Sample, Setting, and
Author and Year
or Compared
Treatment Phase
control, n = 19 (wait
scores than control
Spencer-Cisek, 2001)
· Controlled breathing
= II (6)
· Visualization
Setting: southern Unit-
· Better subjective
· No randomization
· Mindfulness tech-
ed States, outpatient
sleep quality
· Small sample size
· Faster sleep latency
· No objective sleep
· L o w - i m p a c t p o s-
Treatment phase: active
· Longer sleep dura-
treatment or long-
· Compliance prob-
Seven sessions
term follow-up (15
· Less use of sleep
of 39)
Intervention may be ef-
No differences in
fective for subjective
· 38 of 39 completed
· Intrusion
sleep quality, latency,
the study.
· Avoidance
duration, and less use
· 32% attended seven
· State anxiety
of sleep drugs.
· Depression
· 26% attended five or
· Fatigue
six sessions.
· 32% attended two or
three sessions.
· 10% attended one
Coleman et al., 2003
Randomized clinical
24 patients with multiple
Home-based exercise
Compared with control
P R I S M level of evi-
versus control group
myeloma receiving
dence (Ropka &
group, exercise group
high-dose chemother-
Spencer-Cisek, 2001)
had lower weight loss
· Warm-up walking
apy and autologous
= II (6)
than control group (p
· Stretching
peripheral blood stem
= 0.01).
· Small sample size
· Endurance walking
cell transplant
No patient injuries
· Compliance prob-
· Strength resistance
Groups: exercise = 14;
Individualized exercise
control = 10 (usual
program during ag-
· Analysis poorly de-
· Cool down
gressive treatment is
Frequency and length
Setting: midwestern
of program not de-
United States, outpa-
tient setting
Treatment phase: active
Compliance: 58% com-
pleted study; no in-
formation on compli-
ance with exercise
Dalton et al., 2004
S t a n d a r d cognitive
Three-group, experi-
131 patients who had
Compared with stan-
P R I S M level of evi-
behavioral therapy
mental, repeated
elevated scores on
dard CBT, profile-tai-
dence (Ropka &
(CBT), profile-tailored
measures, random-
the BPP and who
lored CBT led to sub-
Spencer-Cisek, 2001)
CBT, or usual care
ized clinical trial de-
were receiving treat-
stantial improvement
= II (6)
Standard CBT
ment for cancer
from baseline to pos-
Significant attrition
Profile-tailored CBT:
pain at three cancer
tintervention in
Inter vention may be
matched to modules
treatment centers
· Worst pain
beneficial for less
based on score on
and one hospital in
· Least pain
interference of pain
Biobehavioral Pain
three southeastern
· Less interference of
with sleep.
Profile (BPP) (Dalton
pain with sleep
et al., 1994)
Groups: Standard,
· Less confusion.
Usual care: five one-
CBT = 47; profile-
From baseline to one
hour sessions
tailored, CBT = 50;
month postinterven-
usual care = 34
tion, profile-tailored
Setting: southeastern
patients saw greater
United States, inpa-
improvement in
tient and outpatient
· Less interference of
pain with activities
Treatment phase: Ac-
· Walking
t i v e treatment for
· Relationships
cancer pain
· Sleep.
(Continued on next page)
Table 4. Nonpharmacologic Intervention Studies in Patients With Cancer (Continued)
Intervention Tested
Sample, Setting, and
Author and Year
or Compared
Treatment Phase
· 121 of 131 completed
· 53 of 131 completed
immediate follow-up.
· 43 of 131 completed
one-month follow-up.
· 28 of 131 completed
six-month follow-up.
Davidson et al., 2001
Sleep therapy program
One-group; pre- and
14 patients; eight with
Significant improve-
P R I S M level of evi-
· Stimulus control
post-test design
breast cancer, one with
ment over baseline in
dence (Ropka &
cervical cancer, two
· Number of awaken-
Spencer-Cisek, 2001)
· Relaxation training
with lymphoma, and
= II (6)
· Strategies to consoli-
one with melanoma
· Ti m e a w a k e a f t e r
· No placebo control
date sleep
Setting: central Canada,
sleep onset
· Strategies to reduce
outpatient clinics
· Total sleep time
· Small sample size
Treatment phase: long-
· Sleep efficiency
· No objective sleep
Six sessions
term follow-up
· Rating of feeling
Compliance: 12 of 14
Intervention may be ef-
completed the inter-
· Rating of sleep qual-
fective for improving
number of awaken-
ings, time awake af-
ter sleep onset, total
sleep time, sleep ef-
ficiency, and ratings
of feeling rested and
sleep quality.
Expressive writing
Randomized clinical
de Moor et al., 2002
42 patients with meta-
Compared with the NW
P R I S M level of evi-
(EW) versus neutral
static renal cell carci-
group, the EW group
dence (Ropka &
writing (NW)
noma receiving vac-
Spencer-Cisek, 2001)
Four sessions
cine therapy
· Less sleep distur-
= II (6)
Groups: EW, n = 21;
Small sample size
NW, n = 21
· Better sleep quality
Inter vention may be
Setting: southwestern
and duration
effective in improv-
United States, outpa-
· Less daytime dys-
ing sleep quality and
tient clinic
duration and feel-
Treatment phase: active
· Increased vigor.
ings of vigor; it may
No differences between
reduce sleep distur-
groups in
bance and daytime
· 35 of 42 completed
· Distress
four writing assign-
· Perceived stress
· Mood disturbance
· 42 of 42 completed
at least one writing
Fobair et al., 2002
One group; pre- and
20 lesbians diagnosed
Improved sleep, less
P R I S M level of evi-
group therapy dis-
post-test design
with early-stage breast
dence (Ropka &
cancer within the pre-
Decreased emotional
Spencer-Cisek, 2001)
· Problem of new diag-
vious 12 months
distress, intrusive-
= II (6)
Setting: western United
ness, and avoidance
· No placebo control
· Coping with illness,
States, outpatient
Improved coping
treatment, and mood
Reduced social support
· Small sample size
Treatment phase: active
Tr e n d toward more
· No objective sleep
· Self-efficacy
treatment or long-
family cohesiveness
· Improving relation-
term follow-up
and expressiveness
Inter vention may be
No change in body im-
effective in improving
· Managing pain, sleep,
· 20 of 20 completed
age, sexuality, or at-
body image, and sex-
baseline, three- and
titude toward health-
six-month measures,
care providers
and participated in
(Continued on next page)
Table 4. Nonpharmacologic Intervention Studies in Patients With Cancer (Continued)
Intervention Tested
Sample, Setting, and
Author and Year
or Compared
Treatment Phase
group intervention
· 17 of 20 completed
12-month measure.
Kim et al., 2002
I n f o r m a t i o n a l inter-
184 patients with pros-
Information group had
Randomized clinical
P R I S M level of evi-
vention on severity
tate cancer receiving
· Fewer sleep prob-
dence (Ropka &
of side effects of ra-
Spencer-Cisek, 2001)
diation therapy for
Groups: information,
· Less fatigue.
= II (6)
prostate cancer
n = 77; control, n =
No difference in nega-
· No baseline measure-
Two audiocassettes, 8
75 (general informa-
tive attitude
and 12 minutes
· No objective sleep
Setting: northeastern
United States, outpa-
Intervention may be
tient setting
effective in reducing
Treatment phase: active
sleep problems.
Compliance: 152 of 184
were fully evaluable.
Moderate, self-paced,
Two-group; pre- and
46 patients with breast
Exercise group report-
P R I S M level of evi-
Mock et al., 1997
progressive, home-
post-test experimen-
cancer receiving
ed positive findings
dence (Ropka &
based exercise pro-
outpatient radiation
tal design
as compared to the
Spencer-Cisek, 2001)
control group.
= II (6)
Walking exercise ver-
Groups: exercise, n =
· Improved physical
· Small sample
sus usual care
22; usual care, n = 24
· Possible diffusion
Individualized walking
Setting: eastern United
· Reduced symptom
between groups
program based on age,
intensity (fatigue,
· Lack of control over
level of fitness, and
Treatment phase: ac-
anxiety, and difficulty
intervention in home
history of exercise
· 20­30 minute brisk
Compliance: 152 of 184
Issues with adherence
were fully evaluable;
to five-day-per-week
· Five-minute slow
issues with adher-
walk cool down
ence to five-day-per-
· Four to five sessions
week regimen
a week for six weeks
Quesnel et al., 2003
Multimodal cognitive-
Prospective, nonran-
10 patients with chron-
Daily variability in sleep
P R I S M level of evi-
behavioral therapy
d o m i z e d , repeated
ic insomnia and non-
decreased after initia-
dence (Ropka &
for insomnia (eight
measure design (21
m e t a s t a t i c breast
tion of the interven-
Spencer-Cisek, 2001)
data points; as long
cancer (stages I­III)
= II (6)
· Stimulus control
as six months fol-
who had completed
Significant improve-
· Small sample
· Sleep restriction
low-up); single case
chemotherapy and
ment in sleep effi-
· Inconsistent comple-
· Coping strategies for
research comparing
radiation therapy
ciency and total wake
tion of sleep diaries
each individual over
2­175 months prior;
· Reframing maladap-
this time
all had a diagnosis of
Significant improve-
tive cognitions
chronic insomnia per
ments in mood, gen-
· Sleep hygiene
Diagnostic and Sta-
eral and physical fa-
tistical Manual IV.
tigue, and global and
Setting: Canada
cognitive dimensions
Treatment phase: long-
of quality of life
term follow-up
Compliance: 8 of 10
completed the study.
Shapiro et al., 2003
Randomized clinical
63 women with a his-
Both MBSR and free-
P R I S M level of evi-
stress reduction
tor y of breast can-
choice groups had
dence (Ropka &
significant improve-
Spencer-Cisek, 2001)
· Sitting meditation
Groups: MBSR, n = 31;
ment in sleep quality.
= II (6)
· Body scan
free-choice control,
Neither group had im-
· Small sample
· Hatha yoga (stretches
n = 32
provement in sleep
· Inconsistent com-
and postures)
Setting: western United
pliance with MBSR
(Continued on next page)
Table 4. Nonpharmacologic Intervention Studies in Patients With Cancer (Continued)
Intervention Tested
Sample, Setting, and
Author and Year
or Compared
Treatment Phase
MBSR group members
Six sessions and a one-
Treatment phase: long-
who practiced more
day silent retreat
term follow-up
· Wide range of relax-
improved more on
ation techniques ac-
sleep measure asso-
ciated with distress.
Quasiexperimental de-
Simeit et al., 2004
P r o g r e s s i v e muscle
239 patients with
C o m p a r e d with the
P R I S M level of evi-
relaxation (PMR)
sign; sequential re-
breast, kidney, or
control group, both
dence (Ropka &
versus autogenic
cruitment of groups;
prostate cancer stay-
intervention groups
Spencer-Cisek, 2001)
training (AT)
patient choice for
ing for three to four
had improvement in
= II (6)
I n f o r m a t i o n : myths,
weeks in oncology
· Sleep latency
· Compliance with
relaxation technique
rehabilitation center
· Duration
MBSR techniques
(PMR or AT), sleep
Groups: PMR, n = 80;
· Efficiency
not described
hygiene, stimulus
AT, n = 71; control,
· Quality.
· Covariates like hor-
control; three one-
n = 78
Quality-of-life scales
mone therapy not de-
hour sessions
Setting: Germany
showed improvement
scribed or controlled
Practice session: re-
Treatment phase: long-
over time.
· Sleep tool translated;
lapse prevention and
term follow-up
no information about
Compliance: not de-
validity and reliabil-
Smith, Kemp, et al.,
Therapeutic massage:
Quasiexperimental de-
41 patients admitted
Mean scores for pain,
P R I S M level of evi-
15­30 minutes of light
sign; sequential re-
to inpatient oncol-
sleep quality, symp-
dence (Ropka &
Swedish massage,
cruitment of groups
ogy unit for chemo-
t o m distress, and
Spencer-Cisek, 2001)
three sessions per
therapy or radiation
anxiety improved
= II (6)
from baseline for the
· Small sample
Groups: massage, n =
massage group.
· Lack of randomiza-
20; control, n = 21
Only anxiety improved
tion; cohorts treated
(nurse interaction)
for control group.
Setting: midwestern
Sleep deteriorated for
· High withdrawal rate
United States, veter-
the control group.
ans' hospital
Interactions were found
Treatment phase: ac-
for pain, symptom
distress, and sleep
C o m p l i a n c e : 18 pa-
tients were with-
drawn or lost from
the study because of
their discharge be-
fore the end of the
one-week period or
because of death (8
from massage group
and 10 from control
Weze et al., 2004
Healing touch, a non-
One group; pre- and
35 patients with vari-
Statistically significant
P R I S M level of evi-
invasive, noncondi-
post-test design
ety of cancers, males
improvement pre- to
dence (Ropka &
tion-specific method
and females, half had
post-test on sleep
Spencer-Cisek, 2001)
involving placing of
less than one-year
disturbances per Wil-
= II (6)
hands on various
duration of cancer,
coxon Signed Rank
· <