Nonrestorative sleep: Symptom or unique diagnostic entity?
Kate Wilkinsona,⇑, Colin Shapiroa,b,c
aDepartment of Cell and Systems Biology, University of Toronto, 25 Harbord Street, Toronto, ON M5S 3G5, Canada
bSleep and Alertness Clinic, Toronto Western Hospital, 750 Dundas Street West, Suite 221, Toronto, ON M6J 1H1, Canada
cYouthdale Child and Adolescent Sleep Centre, 227 Victoria Street, Lower Level 2, Toronto, ON M5B 1T8, Canada
a r t i c l e i n f o
Available online 5 May 2012
Restorative sleep theory
Chronic fatigue syndrome
a b s t r a c t
Nonrestorative sleep (NRS) refers to the subjective experience of sleep as insufficiently refreshing, often
despite the appearance of normal sleep according to traditionally assessed objective parameters. This has
led researchers to pursue alternative physiological markers of nonrestorative or unrefreshing sleep,
though much of this research remains controversial and inconclusive. This review summarizes the recent
findings on NRS in the literature and discusses some of the issues inherent in current efforts to define and
measure NRS. We offer a summary of recommended clinical approaches to NRS and discuss a new poten-
tial paradigm for the assessment of NRS—an approach modelled on current diagnosis of insomnia.
? 2012 Elsevier B.V. All rights reserved.
Conditions like sleep apnea, narcolepsy, and insomnia are
increasingly recognized for their detrimental effects on both phys-
ical and mental health, and this recognition has led to the develop-
ment ofgold-standard approaches
treatment. However, issues with sleep often defy easy categoriza-
tion and, increasingly, sleep physicians have begun to recognize
that objective sleep data obtained through polysomnography
(PSG), actigraphy, or otherwise may not be sufficient to explain
poor subjective quality of sleep. Nonrestorative sleep (NRS)—or
the subjective feeling that sleep has been insufficiently refresh-
ing—is a complex sleep problem that remains contested in the lit-
erature and difficult to define using objective physiological data.
Still, the feeling that sleep has not been restorative has a direct
bearing on patient quality of life, making the recognition and treat-
ment of NRS a priority for physicians in the field of sleep medicine.
In the ongoing effort to understand NRS, two research questions
are of paramount importance. First, what distinguishes NRS from
sleep that is comparatively refreshing? This is not simply an issue
of epidemiology or of clinical care, but one that strikes to the core
of sleep medicine, addressing the functions of sleep itself. More
specifically, can we pinpoint a neurophysiological or endocrinolog-
ical mechanism for the subjective feeling of refreshing sleep? How-
ever, in order to address this question, the task of defining NRS
across a variety of disorders and conditions becomes vital. Thus,
is there a way to differentiate between ‘‘pure’’ NRS and sleep that
is unrefreshing due to another primary sleep disorder? In other
words, should NRS be considered a primary diagnosis as well as
In an effort to address these questions, this review will evaluate
existing literature on nonrestorative sleep. We will begin by
attempting to define what is meant by nonrestorative sleep and
by discussing both clinical and research approaches to its assess-
ment. Next, we will discuss clinical approaches to the diagnosis
and treatment of NRS and issues involved in its assessment. Finally,
we propose a new paradigm for the assessment of NRS, one that al-
lows for better standardization across disciplines.
2. Defining and assessing nonrestorative sleep
A general definition of NRS is sleep that is subjectively experi-
enced as unrefreshing, light, restless, or of poor quality [1,2]. While
such a definition may be valuable from a theoretical perspective,
its utility becomes questionable in practical situations. With its
lack of specificity, almost any individual could be labelled as hav-
ing NRS, making it very difficult to distinguish occasional poor
sleep quality from clinically significant NRS. There are two per-
spectives from which to approach the evaluation of NRS: as a re-
searcher intending to screen for NRS, and as a clinician interested
2.1. Defining NRS in research
As evidenced by the widely varying definitions and vocabularies
applied to the experience of unrefreshing sleep by the major clas-
sification manuals, a consensus on its role in clinic diagnosis has
yet to be reached. Similarly, efforts to investigate the epidemiology
and presentation of NRS have been hindered by the lack of a
1389-9457/$ - see front matter ? 2012 Elsevier B.V. All rights reserved.
E-mail address: email@example.com (K. Wilkinson).
Sleep Medicine 13 (2012) 561–569
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/sleep
standardized definition. In a recent review of the literature on NRS
, Stone and colleagues found that, of 10 studies attempting to
determine the prevalence rate of NRS, each used a different meth-
od for screening.
Approaches to measuring NRS vary, but typically they consist of
a short screening tool ranging from one to four items in length. For
example, in a study of insomnia in adolescents, Roberts and col-
leagues asked participants to report the frequency with which they
were ‘‘Not feeling really rested’’ over the past month, with re-
sponses of ‘‘often’’ or ‘‘almost every day’’ indicating NRS . Other
researchers have chosen pre-existing questionnaires for the pur-
pose of assessment, focusing on items that query feelings of
refreshment or fatigue following sleep in order to screen for NRS
. This is not only true in the field of sleep research, but else-
where as well, and this lack of standardization has been cited as
an issue in literature regarding fibromyalgia and chronic fatigue
syndrome as well .
A number of studies conducted by Ohayon and Roth have begun
to address the insufficiency of these single-item assessment tech-
niques . In their most recent research efforts, they have tended
to approach NRS as the report that sleep feels unrefreshing at least
three nights per week despite sufficient duration of sleep. Stipula-
tions regarding the length of the sleep problem are also often
As part of this effort, Roth and colleagues developed a four-item
scale that encompasses most of the criteria listed above . Scale
items query the frequency with which respondents experience
the following issues: having trouble getting up in the morning,
waking up not feeling rested, feeling as though they had not slept
long enough despite having enough time in bed, and not feeling re-
freshed after sleep. In their study, participants could select ‘‘often,’’
‘‘sometimes,’’ ‘‘rarely,’’ and ‘‘never’’ in response. While the scale
does not specifically address the frequency of these problems dur-
ing the week, it does incorporate some criteria for frequency that
could be used to satisfy this requirement. The scale has been sub-
jected to factor analysis and found to have a single-factor structure.
However, it has not been systematically analyzed for reliability and
validity and no questionnaire to date has been designed and sub-
jected to psychometric analysis for NRS . Thus, measurement
of NRS has been left to the discretion of individual researchers
and has yet to be standardized, making it difficult to draw compar-
isons between studies.
In answer to this lack of standardization, Stone and colleagues
have proposed several general principles for evaluating NRS .
These are: (1) the term NRS should be applied to individuals who
feel unrested or unrefreshed upon awakening despite having re-
ceived a sufficient duration of sleep, (2) the problem should be
present for a minimum of one month, and (3) the complaint should
be experienced at least three times per week. The researchers have
also suggested that, in order for the label NRS to be applied, indi-
viduals should not present with difficulties initiating or maintain-
ing sleep and should not be diagnosed with an organic sleep
disorder. However, these guidelines represent a fundamental ques-
tion regarding NRS which remains unanswered: is NRS its own
condition or does it occur primarily as a symptom of other sleep
and medical disorders? This question can be answered only
through an approach that integrates findings on NRS from a variety
of different disorders.
2.2. Clinical diagnosis
From a clinical point-of-view, NRS is most typically treated as a
symptom of another primary sleep or medical disorder. In particu-
lar, it is often regarded as a symptom of insomnia. NRS receives
mention in almost every diagnostic classification manual, though
the way it is defined and the degree of attention it receives varies.
2.2.1. Current diagnostic manuals
In the ICSD-2, NRS—or sleep that is experienced as ‘‘unrefresh-
ing’’—is listed as a symptom of disorders as varied as obstructive
sleep apnea, periodic limb movement disorder, shift work sleep dis-
as a common complaint of patients with insomnia due to medical
condition. Ineachcasethe manualoffersnoguidelinesfordetermin-
ing how frequency of symptoms may factor into diagnosis, though it
suggests that symptoms should persist for at least one month.
In its general description of insomnia, the ICD-10 Classification of
Mental and Behavioural Disorders suggests that quantity of sleep
alone should not be the primary criteria for diagnosis, as even those
who experience a sufficient amount of sleep both objectively and
subjectively may still ‘‘suffer immensely from the poor quality of
their sleep’’ . A complaint of ‘‘poor quality of sleep’’ is listed as
ficulties falling asleep and maintaining sleep. The manual suggests
for a period of one month before the diagnosis of insomnia is given.
Of all three diagnostic manuals, the DSM-IV-TR is the only one
to specifically define what is meant by ‘‘nonrestorative sleep’’ .
Here the label is used to describe sleep that is ‘‘restless, light, or
of poor quality’’ and is listed as one of the predominant complaints
associated with primary insomnia. The manual suggests that NRS
should be present for at least one month in order to merit a diag-
nosis of insomnia, though no frequency criteria are given. NRS is
also listed as a feature of primary hypersomnia, and ‘‘unrefreshing’’
sleep is offered as one of the less frequent symptoms of breathing-
related sleep disorder. Finally, the complaint of NRS associated
with fibromyalgia is supplied as an example of a sleep disorder
due to a general medical condition.
With the upcoming release of the DSM-5 in 2013, certain pro-
posed changes to the manual may help to clarify the label of NRS
and its use in the diagnosis of a number of sleep disorders. One
of the changes suggested by the Sleep-Wake Disorders Work Group
is the addition of a frequency criterion for the diagnosis of insom-
nia, requiring nonrestorative sleep (or other insomnia-related
sleep problems) to be present at least three times per week .
In conjunction with this, the work group has recommended a
greater emphasis on the importance of dissatisfaction with sleep
in insomnia, moving away from a single-minded focus on difficul-
ties initiating sleep and maintaining sleep.
Another proposed change includes a greater role for NRS in the
diagnosis of primary hypersomnia/narcolepsy without cataplexy.
Sleep periods that are nonrestorative or unrefreshing (despite
being of sufficient or greater-than-average length) could poten-
tially become one symptom listed in this category, though hyper-
somnia and excessive sleepiness would remain the predominant
criteria. While these proposed revisions are still under develop-
ment, many of these suggested changes appear to reflect a greater
emphasis on the importance of subjective sleep quality. In this
way, the DSM-5 is poised to grant greater clinical significance to
NRS—a sleep problem that, without many satisfactory objective
markers, is currently subjective in nature. Most importantly, a fo-
cus on sleep quality may allow for more sensitive diagnosis and
screening as individuals who report sleep dissatisfaction have
higher prevalence rates of mental disorders and present with com-
plaints of poor quality sleep and daytime sleepiness more often
than those with insomnia symptoms alone .
2.2.3. Clinical management of NRS
There are several challenges inherent in attempting to address
NRS in a clinical setting. First, despite an ongoing effort to identify
the biological variables potentially implicated in the experience of
K. Wilkinson, C. Shapiro/Sleep Medicine 13 (2012) 561–569
NRS and its associated conditions, these findings have yet to be syn-
thesized and reinterpreted for use in a clinical setting. A number of
these potential physiological markers are summarized in a review
by Stone and colleagues  and they include variables such as al-
pha-wave intrusion in slow wave sleep, cyclic alternating pattern,
and sleep macroarchitectural changes. Unfortunately, much of the
research investigating these possible biomarkers has been incon-
clusive, and NRS remains a symptom that can be detected exclu-
sively by self-report. To complicate matters further, sleep
researchers have only recently begun to develop consensus regard-
ing the definition and measurement NRS. Without agreement on
these issues,ithas beenverydifficultto drawcomparisonsbetween
research investigations of NRS conducted by different sleep groups,
let alone to extend their findings to the clinical setting. Even now,
there is still no definitively accepted self-report tool for the mea-
surement of NRS in insomnia , and many of the tools used to as-
sess its presence in other disorders are not yet widely employed.
A review by Vernon and colleagues evaluated 59 different
sleep-related questionnaires for content relating to NRS . Of
the 26 questionnaires found to query unrefreshing sleep and its
daytime consequences, only one featured a domain score specifi-
cally devoted to NRS—the Sleep Assessment Questionnaire (SAQ).
The SAQ has been found to be sensitive to Fibromyalgia Syndrome
(FS) and Chronic Fatigue Syndrome (CFS) compared to healthy
controls . However, some have argued that the scale includes
only one item that technically queries NRS—a question asking
respondents to indicate if they awaken feeling unrefreshed or
A number of other questionnaires were also identified for their
ability to assess NRS. The Pittsburgh Insomnia Rating Scale (PIRS)
contains four NRS-related questions relating to quality of sleep
and feelings of being refreshed from sleep and asking about both
daytime and night time consequences of NRS . The PIRS does
ask respondents to specify the frequency with which they awaken
feeling unrefreshed, but no durationcriteria are specified. Similarly,
the Insomnia Symptom Questionnaire includes one item relating to
unrefreshing sleep, though it asks for both the number of times per
week that the symptom is experienced and for the duration of the
problem, making it potentially useful in identifying chronic NRS
. Clearly, there are widely varying approaches to the measure-
ment of NRS and there is still no standardized tool approved for
the purpose. A scale validated for use at the clinical level—across
all manner of sleep and medical conditions—would provide an
invaluable resource for physicians as they screen for NRS and at-
tempt to evaluate its progress over the course of treatment.
The second issue associated with the clinical treatment of NRS
regards its presence in a number of heterogeneous conditions.
NRS is a rheumatology issue, a problem in the sleep clinic, and a
complaint one might discuss with a family doctor. At this time,
the vast majority of research studies focus on NRS as it relates spe-
cifically to insomnia, chronic fatigue, or fibromyalgia. Very little is
understood about NRS as a distinct entity or NRS as a symptom
across multiple conditions. Thus, when unrefreshing sleep persists
despite the treatment of its presumed antecedent, clinicians are
left with little guidance regarding potential next steps. With no
obvious mechanism of restorative sleep to pinpoint and address,
the task of treatment becomes quite challenging.
Several groups have proposed guidelines for addressing NRS as
a symptom of insomnia. Riemann and colleagues offer a clinical
algorithm for its step-by-step diagnosis and treatment . Pro-
gressing through the algorithm, clinicians are instructed to rule
out potential causes of NRS, including improper sleep hygiene, cir-
cadian rhythm sleep disorder, and psychiatric disorder. The algo-
rithm culminates in a referral to a sleep disorder clinic for PSG
where, if a diagnosis of insomnia is obtained, a number of different
behavioural and pharmacological interventions may be pursued.
The guidelines offered in a review by Stone and colleagues are
similarly preliminary, advising clinicians to screen for psychiatric
conditions, sleep disorders, CFS, and FS and, once those issues have
been treated, to reassess the sleep complaint . The researchers
also recommend that unrefreshing sleep should be present at least
three times per week and for a minimum duration of one month in
order to be labelled as NRS. However, there is little evidence to
support the selection of these criteria. For the most part, these par-
ticular guidelines are based on the criteria used to define other
sleep complaints like insomnia. Thus, the distinction between
NRS that occurs two versus three times per week, for example, re-
mains arbitrary. Further, the assumption inherent in both of these
approaches is that a physiological antecedent for NRS will eventu-
ally be located over the course of the diagnostic process. Neither
offers guidelines for managing patients for whom PSG and medical
diagnostic tests are less clear.
In our own sleep clinic, we tend to take a ‘‘three strikes’’ ap-
proach to diagnosis and treatment of unspecified sleep dissatisfac-
tion. This method emphasizes that, in the absence of large
deviations from normal sleep architecture marking obvious sleep
pathology, the combination of a multiplicity of relatively small is-
sues with sleep architecture may lead to the perception of sleep as
nonrestorative. Using a sleep index we developed, we have made
preliminary attempts to investigate the possibilities of this method
. For each sleep parameter we set cut-off points that indicate
deviation from normal—the greater the deviation, the higher the
score the patient receives on a scale from 1 to 10. The current lim-
itation of this approach is that the large data sets of PSG recordings
needed to properly model the interactions between sleep parame-
ters are not available. However, by building on this idea gradually
we hope to begin to clarify some of the underlying physiological
mechanisms implicated in the experience of unrefreshing sleep.
3. A new paradigm for the assessment of NRS
With a label as vague and potentially all-encompassing as NRS,
efforts to reach a consensus regarding its definition are of para-
mount importance. Without specific guidelines adopted across dif-
communication will remain difficult.
Even with potential revisions for the DSM-V poised to place
greater emphasis on complaints of subjectively poor sleep, the task
of then treating NRS remains challenging. Unrefreshing sleep may
tentative evidence for various biomarkers, it remains unclear how
the reports of poor sleep quality associated with these many condi-
tions may be associated. It is for this reason that we recommend a
new paradigm for the assessment of NRS—one that more closely
cross-discipline collaboration on the topic of NRS, enhancing efforts
to isolate biomarkers for subjectively poor sleep quality.
Thus, three broad categories of NRS might then be created: NRS
due to medical illness (incorporating diagnoses such as chronic fati-
gue syndrome and fibromyalgia, as well as other chronic conditions
ness (which would, for example, help to incorporate issues of attri-
bution and patient perception into the definition of NRS), and
isolated or ‘‘pure’’ NRS appearing to occur without antecedent.
3.1. NRS due to medical illness
Unrefreshing sleep is likely to manifest as a result of any num-
ber of medical conditions. It can occur as both the direct result of a
K. Wilkinson, C. Shapiro/Sleep Medicine 13 (2012) 561–569
medical disorder (e.g., the hormonal dysregulation caused by
hypothyroidism) and as the indirect result, as in the case of envi-
ronmental sleep disruption associated with a prolonged hospital
visit. However, while both antecedents are likely to result in clini-
cally relevant effects on sleep quality, the label of NRS resulting
from medical illness would be most beneficially applied to unre-
freshing sleep that results directly from the illness itself. In this
way, disorders that share NRS as one of their primary symptoms
might be examined under the same label, facilitating a search for
potential physiological commonalities between these conditions.
The two illnesses most frequently discussed for their relation-
ship with NRS are (FS) and (CFS). FS is a heterogeneous illness char-
acterized by complaints of widespread pain, unrefreshing sleep,
fatigue that significantly impairs daytime function, and increased
sensitivity to stimuli. Similarly, CFS is defined by persistent fatigue
unexplained by other medical disorders and a variety of other
symptoms including joint and muscle pain, depression, and unre-
freshing sleep. For many, these supposedly separate conditions
represent a single entity due to their high degree of symptom over-
lap (see  for a discussion of this). Regardless of their etiologies,
with unrefreshing sleep and persistent fatigue as two of the defin-
ing features of these disorders, FS and CFS represent two of the
most significant candidates for inclusion under the label of NRS
due to medical illness.
Researchers administered the Medical Outcomes Study Sleep
Scale to more than 3000 FS patients in a recent survey study and
found that only about 30% reported ‘‘optimal sleep’’ . Partici-
pants reported a mean score of 32.2 on a scale of 0–100 to an item
evaluating ‘‘sleep adequacy.’’ In another study of 43 women with
FS, 72% of those surveyed reported ‘‘nonrestorative’’ sleep—a sig-
nificantly greater proportion than that found in the healthy con-
trols . And in one internet survey study of FS, respondents
labelled ‘‘nonrestorative sleep’’ as one of their most severe symp-
toms (ranked only slightly below morning stiffness and fatigue)
. In terms of CFS, approximately 85–95% of patients report
unrefreshing or ‘‘nonrestorative’’ sleep [22,23]. Indeed, one study
found that individuals with CFS were 28 times as likely to score
above threshold on a factor evaluating NRS than individuals with
no fatigue . Gender appears to be a significant factor in terms
of the demographic makeup of FS and CFS patient groups. Preva-
lence estimates suggest that more than 90% of all cases of FS occur
in women , while studies of CFS suggest a gender ratio that is
less skewed but still indicative of greater rates in females .
Unfortunately, in the majority of research studies examining FS
sleep are far less stringent than those currently employed in insom-
getting up’’ and ‘‘difficulties staying awake after getting up,’’ with no
guidelines regarding the frequency or duration of problems and no
attempts made to take into account sleep quantity . Other inves-
defined ‘‘nonrestorative’’ or unrefreshing sleep, likely using only a
single item . Table 1, which is modeled after a table created by
stone and colleagues to summarize definitions of NRS used across
sleep research, offers a summary of medical illness and psychiatric
studies in which an explicit definition of NRS is offered.
Beyond FS and CFS, a number of other medical conditions are
associated with complaints of unrefreshing sleep. First, chronic
pain patients may be considered for inclusion under this label. In
a sample of individuals referred for assessment at one pain clinic,
70% were found to report poor sleep quality . However, this sit-
uation may be more akin to the sleep complaints seen in patients
with sleep apnea, where arousals from deep sleep are caused by
restlessness, difficulties finding a comfortable sleeping position,
and the intrusion of pain sensation [27–29].
Another, perhaps more promising, application of the label may
be among patients with autoimmune disorders. The relationship
between sleep and immune function has been established for dec-
ades, but evidence implicating immunological variables in NRS has
recently begun to build. Both CFS and FS have been linked to im-
mune system function. The onset of CFS is often preceded by some
form of acute inflammatory insult, and individuals with CFS appear
to have abnormalities in their immune cell function [30,31]. Simi-
larly, autoimmune disease and acute infections have been corre-
lated with the onset of fibromyalgia . Even in healthy
individuals, experimentally induced sleep loss has been shown to
interfere with the proper functioning of the immune system .
A burgeoning body of experimental evidence suggests a bi-
directional relationship between changes in sleep regulation and
pro-inflammatory cytokines such as interleukin-1 (IL-1), interleu-
kin-6 (IL-6), and tumor necrotic factor. Studies have found that
increasing circulating levels of IL-6 correlate with decreases in
SWS  and increases in REM percentage and density [35,36].
Subjective sleep quality also appears to be affected, with studies
showing a negative correlation between circulating levels of IL-6
and subjective ratings of sleep (assessed using questions regarding
‘‘feeling refreshed in the morning’’) . Perhaps most strikingly,
the stimulation of evening IL-6 production in healthy controls is
also associated with decreased SWS and increased REM sleep, as
well as increased fatigue reported during the following day .
Though a causal relationship cannot be assumed from these re-
sults, they are highly suggestive of an influence on sleep architec-
ture by circulating levels of cytokines and they provide preliminary
support for the role of immune dysregulation in NRS.
Naturally, the high degree of interconnectedness between im-
mune function, stress, and sleep makes it very difficult to deter-
mine the direction of any potential cause and effect relationships.
Regardless of the underlying mechanisms of the relationship,
symptoms of fatigue and sleepiness are commonly reported in
these disorders, making them potential candidates for this classifi-
cation . For example, infection with Human Immunodeficiency
Virus (HIV) causes a pathological immune response that leads to a
number of sleep-related changes, including an increase in slow-
wave sleep and the dysregulation of growth hormone secretion
[39,40]. These changes are accompanied by increased subjective
reports of fatigue and poor sleep quality . Similarly, with rheu-
matoid arthritis, a disease characterized by the chronic inflamma-
tion of the joints, increases in disease activity have been tentatively
linked with greater amounts of slow-wave sleep . In this case a
bi-directional relationship between pain and subjective levels of
fatigue appears to exist .
Unfortunately, phrases like ‘‘nonrestorative sleep’’ and ‘‘unre-
freshing sleep’’ emerged primarily within the context of sleep re-
search and, as such, have yet to be employed widely in research
outside this domain. For most of the above-mentioned chronic
pain, inflammatory, and immune disorders, sleep pathology has
only recently become a recognized issue and, thus, the language
used to assess subjective sleep quality is markedly different. The
introduction of a label like NRS due to medical illness may help
to unite these disparate disciplines in an effort to improve our
understanding of the potential physiological correlates of unre-
This goal of standardization is the issue most relevant to the
creation of a new diagnostic paradigm for NRS. If NRS is already
a well-recognized diagnostic feature of disorders like FS and CFS,
one may wonder what the clinical utility of a label like NRS due
to medical illness would be. Is it not merely a restating of a disor-
der’s primary symptoms? This is potentially the case in clinical sit-
uations where the medical disorder at root has already been
identified. However, when a diagnosis has yet to be made, such a
descriptive category may be helpful in priming physicians to
K. Wilkinson, C. Shapiro/Sleep Medicine 13 (2012) 561–569
consider particular medical screening measures (e.g., a thyroid pa-
nel to investigate hormonal dysregulation).
Beyond diagnostic concerns, as a body of standardized research
begins to grow, such a label may provide a foundation for cross-
disciplinary communication regarding issues like symptom man-
agement and outcome monitoring. As can be seen in Table 1, a
number of disparate approaches to the definition of NRS confound
the literature on chronic fatigue and fibromyalgia. Many of these
studies treat different descriptors of sleep quality as interchange-
able. ‘‘Tired and worn out’’ , ‘‘not rested’’ , and ‘‘restoring
sleep’’  are just a few of the phrases employed, and many more
studies do not specify at all how they screened for these issues.
NRS is becoming increasingly recognized for its relevancy at the
clinical level. It is clear that sleep quality has significance in terms
of ongoing presentation of the illness and its clinical outcomes.
Complaints of poor sleep quality were found to be significantly cor-
related with the total number of pain sites, as well as reports of
pain throughout the body in patients with fibromyalgia . Sim-
ilarly, deprivation of deep sleep induced in healthy controls has
been shown to cause temporary musculoskeletal complaints that
mimic those seen in FS patients . In patients with HIV, subjec-
tive sleep quality as measured by the Pittsburgh Sleep Quality
Index has been tentatively correlated with symptom severity
. Placing additional clinical emphasis on the subjective quality
of sleep is one way to improve patient quality of life and treatment
outcomes. Such a diagnostic paradigm may represent one of the
first steps in doing so.
3.2. NRS due to psychiatric condition
According to an analysis of eight epidemiological studies, the
presence of insomnia symptoms is significantly predictive of an in-
creased risk for developing depression up to three years after initial
assessment . In one survey of primary care patients with
insomnia, 22% of participants were being treated for depression
and 32% reported being anxious . Given that NRS is frequently
categorized under the umbrella of insomnia, it is unsurprising that
it also appears to be linked to a number of psychiatric issues. In one
prevalence study by Ohayon, individuals with depressive and bipo-
lar disorders were roughly four times as likely to report experienc-
ing frequent NRS (occurring three or more times per week) as other
individuals in the general population—rates of 43% in those with
depression, 41% in those with bipolar disorder, and roughly 10%
in those without any mood disorder . Of those with anxiety,
Summary of definitions for NRS across multiple disciplines and patient populations.
Study Population Definition of NRSFindings
et al. 
HIV positive patients (N = 53) Subjects rated whether they ‘‘wake refreshed’’ on a four-
Almost half of all patients reported NRS. NRS
was associated with past or present
Greater disease severity in rheumatoid
arthritis associated with ‘‘superficial and non-
Unrefreshing sleep was present in 50% of
et al. 
Outpatients with rheumatoid arthritis
(N = 35)
Used the Spiegel Sleep Questionnaire to define NRS. Precise
Resta et al.
Severely obese patients without
obstructive sleep apnea (N = 78).
Healthy controls (N = 40)
Consecutive oncology patients admitted
to a pain relief clinic (N = 123)
Representative cross-section of adults
(N = 118, 336)
Patients with transformed migraine
(N = 147)
Community sample of adults with
complaints of insomnia and mild to
moderate sleep-disordered breathing
(N = 91)
National probability sample of full-time
employees (N = 1965)
‘‘When you wake up in the morning, do you feel that you
have not got enough rest during the night? (unrefreshing
One unspecified questionnaire item addressing ‘‘restoring
Subjects asked how frequently they found sleep refreshing.
Responses of ‘‘never’’ were considered unrefreshing sleep
NRS assess by asking patients to describe their state upon
awakening as ‘‘refreshed’’ or ‘‘tired’’
Subjects asked to rate on a scale from 0 to 10 their sleep on
five indices: refreshing sleep, sleep quality, depth of sleep,
ability to sustain sleep, and ability to return to sleep if
‘‘In the past month, on how many days did you have
difficulty waking up in the morning to make it to work on
time?’’ Yes or no response
A questionnaire asking subjects to indicate the ‘‘frequency
and intensity of unrefreshing sleep’’
et al. 
Power et al.
et al. 
et al. 
Anxiety and depression linked with less
Prevalence of unrefreshing sleep in individuals
with arthritis was 11.9%
All patients surveyed reported feeling ‘‘tired’’
Treatment group demonstrated improved
ratings for NRS
et al. 
NRS was associated with job autonomy and
role conflict in the workplace
Majer et al.
Patients with chronic fatigue syndrome
(N = 35). Healthy controls (N = 40)
Subjects with chronic fatigue reported
significantly more unrefreshing sleep than
60% of the sample reported NRS.
et al. 
Cross-sectional study of patients with
fibromyalgia (N = 2580)
Using the National Fibromyalgia Association
Questionnaire, intensity of symptoms rated on a scale from
0 to 10
‘‘During the past month did you wake up after your normal
amount of sleep feeling tired and worn out?’’ Severity of
NRS assessed using a five-point frequency scale
A complaint of non-refreshing sleep was considered
present if subjects responded ‘‘Yes’’ to this statement: ‘‘I do
not feel refreshed when I get up in the morning.’’ The
severity of the issue was also ranked on a scale from 1 to 10
One item relating to ‘‘a feeling of non-refreshing sleep
upon awakening,’’ scored on a six-point scale ranging from
‘‘never’’ to ‘‘always’’
‘‘Do you find your sleep is not restoring?’’ Presence of NRS
determined by a frequency of >3 nights per week and
reported negative impact
Unspecified. Subjects placed in one of two groups based on
presence or absence of ‘‘NRS complaint’’
Davies et al.
Sample of patients registered at three
general practice clinics (N = 1061)
Restorative sleep significantly correlated with
resolution of chronic widespread pain
Creti et al.
Subjects with chronic fatigue syndrome
(N = 49)
47 of 49 subjects reported non-refreshing
sleep. Assessments of non-refreshing sleep
were not reflected in objective measures like
Non-refreshing sleep was the most common
complaint with 23% reporting it
et al. 
Consecutive myocardial infarction
patients admitted to a coronary care
clinic (N = 204)
A sample of individuals with insomnia
(N = 5293)
Léger et al.
Prevalence of NRS-alone (no other insomnia
symptoms) of 1.4%, all symptoms of insomnia
NRS linked to shorter sleep duration and
poorer scores on measures of fatigue, anxiety,
NRS negatively associated with insomnia, and
daytime physical, cognitive, and emotional
Neu et al.
Community sample (N = 150)
et al. 
Comparing subjects with NRS (N = 541)
to those who reported never
experiencing NRS symptoms (N = 717)
‘‘In the past 4 weeks, how often have you had a full night’s
sleep but woken up feeling unrefreshed or not rested?’’
NRS present if three or more times per week
K. Wilkinson, C. Shapiro/Sleep Medicine 13 (2012) 561–569
13% reported NRS, which was a rate only slightly higher than in
Similarly, in a study of insomnia patients, Sarsour and col-
leagues found depression to be significantly correlated with fre-
quent NRS, even after controlling for the severity of insomnia
symptoms and patients’ total sleep duration . However,
researchers in this study used only two items to screen for depres-
sion: (1) ‘‘During the past month have you often been bothered by
feeling down, depressed, or hopeless?’’ and (2) ‘‘During the past
month, have you been bothered by little interest or pleasure in
doing things?’’ Though these screening items have been validated
for their utility in detecting the presence or absence of depression,
they do not allow for an in-depth exploration of the connection be-
tween specific depressive symptoms and NRS.
In a recent study by Roth and colleagues, slightly more specific
diagnostic techniques were employed. Researchers used data col-
lected as part of the National Comorbidity Survey to evaluate the
connection between DSM-IV psychiatric diagnoses and individual
symptoms of insomnia (including NRS) . Participants were diag-
nosed following a structured clinical interview and insomnia
symptoms were considered present if they occurred three or more
times per week for a minimum of one month. Of those participants
found to have NRS according to these criteria, 54% also had at least
one DSM-IV diagnosis (e.g., mood disorder, anxiety disorder, sub-
stance use disorder) with an odds ratio of 4.4. Additionally, 17%
of those individuals with NRS were found to have three or more
diagnoses, indicating a high degree of psychiatric comorbidity in
the NRS population.
In the case of NRS due to psychiatric illness, a cause and effect
relationship is almost impossible to determine. In much the same
way that NRS remains a self-reported symptom with no adequate
biomarkers, the majority of psychiatric illnesses are also diagnosed
on the basis of subjective assessment. Thus, patient perception
may contribute to both. In their recent study, Léger and colleagues
observed that the strongest factor associated with severe self-re-
ported daytime impairment was the perception that sleep quality
was poor . Indeed, daytime functioning has been shown to be
strongly linked with perceived quality of sleep, regardless of its
objective quality as determined by biological measures . Thus,
the impression that sleep has not been restorative may, in itself,
contribute to outcomes resembling NRS.
The relationship between psychiatric illness and subjective
sleep quality has been investigated extensively. Depressed patients
have been shown to report significantly lower satisfaction with
sleep than healthy controls as assessed by a number of subjective
variables, including feeling ‘‘rested upon waking’’ . Indeed,
there appears to be a discrepancy between objective and subjective
sleep quality in many of these patients . One study found that
sleep satisfaction improved in remitted depression patients despite
a lack of objective improvement recorded through PSG . The
combined weight of this evidence has led some to propose that
the negative cognitions inherent in depression may bias patient
reporting of sleep .
one, perhaps further pathologizing individuals who are already
struggling with these issues. It is possible that patients will be
encouraged to report problems where none had previously existed.
Or, perhaps, by providing further diagnostic weight to these issues,
we may be creating a vicious cycle that mirrors the kind seen in pa-
tients with insomnia. Namely, anxiety regarding not obtaining suf-
ficient sleep may interfere further with the sleep cycle, which then
However, the benefits of such a label are likely to outweigh the
potential downsides. First, it places a greater clinical emphasis on
the patients’ perceptions, which are likely to be highly influential
in terms of management and outcomes—regardless of what objec-
tive sleep measures may indicate. Roth and colleagues have ob-
served that clinician-rated assessments of symptom severity for
patients with NRS tend to be lower than those reported for patients
with difficulties initiating or maintaining sleep . Despite this,
Sarsour and colleagues have found that close to 50% of patients
who reported NRS three or more times per week rated their NRS
symptoms as severe or very severe .
It is possible that these lower clinician ratings may result from
the physicians’ perception of NRS as a vague complaint—one with-
out reliable biological markers or validated treatment approaches
that is better treated in conjunction with other issues than ad-
dressed separately. A diagnostic category that recognizes the
importance of NRS may help to ensure that sleep issues will be
treated with increased relevancy at the time of clinical interview,
resulting in better outcomes for these patients and a greater
impression of support from the clinician.
3.3. NRS alone
Historically, the label of ‘‘nonrestorative sleep’’ has been con-
ceptualized as the third—and often overlooked – symptom of
insomnia in the field of sleep research. This is likely related to its
treatment in the DSM-IV, where it receives a cursory mention
alongside the better-recognized symptoms of difficulties initiating
and maintaining sleep. However, with the changes proposed for
the DSM-V, subjective sleep quality is likely to become a greater
component of sleep disorder diagnosis. If this occurs, NRS may be-
gin to be considered independently outside the framework of
insomnia, a research direction that has been pursued minimally
The relative lack of attention placed on NRS may be understand-
able given its high degree of comorbidity with other diagnoses.
Much of the literature on NRS indicates that it frequently occurs
in the presence of other symptoms of insomnia. A recent study
examining a sample of insomnia patients in a primary care setting
used logistic regression to demonstrate that all symptoms of
insomniaaresignificantlyrelatedto oneanother.The research-
ers found that the most frequently reported type of insomnia was a
combination of all symptoms (difficulties initiating and maintain-
ing sleep, early morning awakenings, and NRS), with 38.6% of the
population reporting some issues with each of them. Similarly, a
study of insomnia patients conducted by Sarsour and colleagues
demonstrated that, of those reporting NRS symptoms three times
per week, more than 80% also reported at least one other symptom
of insomnia . The frequency of reported NRS symptoms was
also found to be significantly correlated with difficulties initiating
and maintaining sleep and early morning awakenings.
The high degree of interconnectedness between symptoms pre-
sents something of a problem when attempting to isolate and de-
fine NRS. Many issues associated with insomnia—difficulties
initiating and maintaining sleep, early morning awakenings—have
the potential to disrupt sleep continuity and duration and to inter-
fere with individuals’ attempts to sleep at times that are optimal in
terms of their circadian cycles. In this way, it becomes unclear
whether NRS is its own symptom/diagnosis or simply a product
of the disturbances in sleep caused by other symptoms of insomnia.
In an attempt to clarify the issue and to isolate NRS as a distinct
entity, Roth and colleagues conducted an analysis of 579 subjects
reporting unrefreshing or unrestful sleep . After ruling out
any sleep disorders or medical conditions that could be causing
the subjective experience of unrefreshing sleep, they identified a
cohort of individuals experiencing ‘‘pure NRS’’ (meaning unrefresh-
ing sleep occurring three times per week or more in the absence of
other insomnia symptoms).
K. Wilkinson, C. Shapiro/Sleep Medicine 13 (2012) 561–569
Using PSG, researchers confirmed that individuals in the NRS-
only group did indeed have normal mean sleep latency, a normal
amount of wakefulness following sleep onset, and a normal sleep
duration when compared to healthy controls. Despite having re-
ceived an adequate duration of sleep, the levels of daytime impair-
ment, depression, and anxiety in this NRS-only group were
comparable to levels found in individuals with other insomnia
symptoms (difficulties initiating or maintaining sleep). These re-
sults suggest not only that NRS exists as a distinct entity, but that
it causes equivalent issues with functioning to other better recog-
nized sleep disorder symptoms.
out NRS, Sarsour and colleagues found that almost 15% of those
belonging to the NRS group were classified as having no insomnia
using the Insomnia Severity Index . Further, 12% were found to
have no insomnia and no depression, allowing for the elimination
ciated with an increase in total sleep duration. Though all sleep vari-
ables were subjectively reported and therefore not verified using
objective data, these findings suggest that NRS does indeed exist
independent of other insomnia symptoms and psychiatric diagnoses
despite the presence of a reasonable duration of sleep.
Sarsour and colleagues also investigated daytime function, ask-
ing respondents how frequently they experienced the following
three issues due to their sleep pattern: (1) feeling sleepy, tired,
or low energy during the day; (2) having poor concentration or
memory, or needing more effort to get things done; and (3) feeling
irritable, stressed, or being in a bad mood. Researchers found that
NRS was associated with decreased physical, cognitive, and emo-
tional function. The relationship remained even after controlling
for insomnia severity, sleep duration, and depression.
Another study released only in the past month has also at-
tempted to isolate NRS from other insomnia and psychiatric diag-
noses . Researchers found that, in a cross-sectional sample of
the population, 5.8% had NRS occurring three times per week for
a minimum of one month in the absence of other sleep and psychi-
atric disorders. Within this ‘‘pure’’ NRS population, 56% of individ-
uals reported global dissatisfaction with their sleep while 67%
reported daytime consequences in areas that included cognitive
functioning, affective symptoms, fatigue, and sleepiness. According
to these preliminary findings, NRS appears to be a distinct entity
with serious implications for quality of life.
Unfortunately, with so little currently understood regarding this
‘‘pure’’ NRS, the clinical value of such a diagnosis is still unclear. In-
deed, according to the criteria of NRS proposed by some research-
ers, many of the parameters recorded by objective measures
should appear very similar for both NRS patients and those with
normal sleep, making it difficult to determine what treatment
course to take when dealing with these patients. Only one study
to date has analyzed polysomnographic data in individuals with
NRS alone (occurring in the absence of other sleep or medical dis-
orders). The researchers confirmed that many factors evaluated by
PSG are similar between these NRS-only patients and normal con-
trols, including total sleep time, sleep latency, and wakefulness
after sleep onset . In terms of the differences found, the NRS-
only group spent somewhat less time in stages 3 and 4 sleep and
less time in REM sleep, though the distinctions were minimal.
Still, with a categorization like NRS alone available, an increased
directive to investigate the antecedents of such a symptom may re-
sult. In the very least, such a diagnostic category will provide indi-
viduals who experience unexplained subjective sleep complaints
with additional clinical support, increasing the attention paid to
subjective sleep satisfaction and its implications for quality of life.
Sleep is a multi-factorial process, one that is not easily captured
by the diagnostic tools sleep specialists currently have at their dis-
posal. Polysomnographic testing acts as a window onto a number
of the complex physiological processes involved in sleep, but
parameters outside the scope of that window remain obscured.
Electroencephalography, for example, primarily records surface
electrical potential, while deep brain structures contribute only
minimally to the signal. Similarly, the impracticality of monitoring
endocrine variables during routine PSG dictates that an integral
element of the sleep quality picture must be neglected during diag-
nosis. A number of promising physiological markers of NRS have
been identified. Alpha EEG sleep, CAP, and various endocrine and
immunological variables offer researchers new perspectives on
sleep and its functions. But until the complex interrelationships
between these factors can be teased apart, cause and effect rela-
tionships are difficult to determine. And with no satisfactory con-
firmatory markers currently established to detect NRS, the issue
may go unrecognized or be untreated in the clinical setting and pa-
tient quality of life will continue to suffer.
In response to this, we recommend a new approach to the def-
inition of NRS. This approach would not necessarily be mutually
exclusive with the one recently suggested by Stone and colleagues
where specific frequency and duration criteria are concerned ,
though it would clearly be in conflict with the recommendation
that the label of NRS should only be applied when it appears in
the absence of other diagnoses. With this approach we allow for
the possibility that such a thing as ‘‘pure’’ NRS may not exist. Per-
haps more involved screening of those patients who report such a
phenomenon would reveal an underlying issue to be addressed—
an unrecognized issue with growth hormone release or an uniden-
tified psychiatric illness. Still, their complaint of NRS is no less valid
and the impact on daytime symptoms no less significant.
There is the chance that categorizing all conditions related to
NRS together may be a useless exercise. It is possible, for example,
that NRS seen in fibromyalgia is entirely distinct from NRS resulting
from depression. However, with the current disconnect between
how NRS is defined across different disorders, this potentiality re-
mains impossible to test. Our new suggested paradigm for defining
NRS allows for the possibility that the antecedents of NRS in differ-
ent conditions may be completely unrelated—much as the diagno-
sis of insomnia allows for many different sources. By uniting NRS
under a single umbrella definition, we move towards a kind of stan-
dardization that makes this analysis possible.
It is our view that, in a way that is analogous to the evolution of
the understanding of insomnia, NRS will eventually be given status
as a separate and clearly distinct condition. Two decades ago we
raised the issue of insomnia as a separate entity and advocated that
secondary insomnia receive more focus in the literature. It took
many years for this approach to gain traction, but the concept of
comorbid insomnia is now well established. Similarly, the status
of NRS must evolve and, based on the putative notion of a different
pathophysiological etiology, gain an identity that is distinct from
insomnia. We can easily accept different causes of insomnia—med-
ical disorders, psychological stress, and drug misuse, to name a
few. So too should we begin to consider NRS as a symptom with
many causes, and perhaps even a condition in its own right.
Conflict of Interest
The ICMJE Uniform Disclosure Form for Potential Conflicts of
Interest associated with this article can be viewed by clicking on
the following link: doi:10.1016/j.sleep.2012.02.002.
K. Wilkinson, C. Shapiro/Sleep Medicine 13 (2012) 561–569
 Stone K, Taylor D, McCrae C, Kalsekar A, Lichstein K. Nonrestorative sleep.
Sleep Med 2008;12:275–88.
 American Psychiatric Association. Diagnostic and statistical manual of mental
disorders. 4th ed. Washington: American Psychiatric Association; 2000.
 Roberts R, Ramsay Roberts C, Ger Chen I. Impact of insomnia on future
functioning of adolescents. J Psychosom Res 2002;53:561–9.
 Phillips B, Mannino D. Correlates of sleep complaints in adults: the ARIC study.
J Clin Sleep Med 2005;1:277–83.
 Spaeth M, Rizzi M, Sarzi-Puttini P. Fibromyalgia and sleep. Best Pract Res Clin
 Ohayon M, Roth T. What are the contributing factors for insomnia in the
general population? J Psychosom Res 2001;51:745–55.
 Roth T, Jaeger S, Jin R, Kalsekar S, Stang P, Kessler R. Sleep problems, comorbid
mental disorders, and role functioning in the National Comorbidity Replication
Survey. Biol Psychiatry 2006;60:1364–71.
 Vernon M, Dugar A, Revicki D, Treglia M, Buysse D. Measurement of non-
restorative sleep in insomnia: a review of the literature. Sleep Med Rev
 American Sleep Disorders Association. International classification of sleep
disorders: diagnostic and coding manual, ICSD-2. 2nd ed. Westchester:
American Academy of Sleep Medicine; 2005.
 AmericanPsychiatric Association.
Pages/Sleep-WakeDisorders.aspx> [accessed on June 16, 2010].
 Ohayon M, Caulet M, Priest R, Guilleminault C. DSM-IV and ICSD-90 insomnia
symptoms and sleep dissatisfaction. Br J Psychiatry 1997;171:382–8.
 Cesta A, Moldofsky H, Sammut C. The sensitivity and specificity of the Sleep
Assessment Questionnaire (SAQ?) as a measure of nonrestorative sleep. Sleep
 Moul D, Pilkonis P, Miewald J, Carey T, Buysse D. Preliminary study of the test-
retest reliability and concurrent validities of the Pittsburgh Insomnia Rating
Scale (PIRS). Sleep 2002;2002(25):A246–7.
 Okun M, Kravitz H, Sowers F, Moul D, Buysse D, Hall M. Psychometric
evaluation of the insomnia symptom questionnaire: a self-report measure to
identify chronic insomnia. J Clin Sleep Med 2009;5:41–52.
 Riemann D, Fischer J, Mayer G, Peter H. The guidelines for ‘non-restorative
sleep’: relevance for the diagnosis and therapy of insomnia. Somnologie
 Hossain J. Sleep, fatigue and sleepiness in shift-workers and sleep disordered
individuals. Thesis (Ph.D.)—University of Toronto; 2004.
 Wessely S, Hotopf M. Is fibromyalgia a distinct clinical entity? Historical and
epidemiological evidence. Best Pract Res Clin Rheumatol 1999;13:427–36.
 Rehm S, Koroschetz J, Gockel U, Brosz M, Freynhagen R, Tölle T, et al. A cross-
sectional survey of 3035 patients with fibromyalgia: subgroups of patients
with typical comorbidities and sensory symptom profiles. Rheumatology
 Drewes A, Gade J, Nielsen K, Bjerregard K, Taagholt S, Svendsen L. Clustering of
syndrome. Rheumatology 1995;1995(34):1151–6.
 Bennett R, Jones J, Turk D, Russell I, Matallana L. An internet survey of 2,596
people with fibromyalgia. BMC Musculoskelet Disord 2007;8.
 Jason L, Richman J, Rademaker A, Jordon K, Plioplys A, Taylor R, et al. A
community-based study of chronic fatigue syndrome. Arch Intern Med
 Nisembaum R, Reyes M, Unger E, Reeves W. Factor analysis of symptoms
among subjects with unexplained chronic fatigue: what can we learn about
chronic fatigue syndrome? J Psychosom Res 2004;56:171–8.
 Unger E, Nisenbaum R, Moldofsky H, Cesta A, Sammut C, Reyes M, et al. Sleep
assessment in a population-based study of chronic fatigue syndrome. BMC
 Wolfe F. Epidemiology of fibromyalgia. J Musculoskelet Pain 1993;1:137–48.
 Pilowsky I, Crettenden I, Townley M. Sleep disturbance in pain clinic patients.
 Drewes A, Nielsen K, Arendt-Nielsen L, Birket-Smith L, Hansen L. The effect of
cutaneous and deep pain on the electroencephalogram during sleep—an
experimental study. Sleep 1997;20:632–40.
 Moldofsky H, Lue F, Smythe H. Alpha EEG sleep and morning symptoms in
rheumatoid arthritis. J Rheumatol 1983;10:373–9.
 Devins G, Edworthy S, Leendert C, Mandin H, Seland P, Klein G, et al. Restless
sleep, illness intrusiveness, and depressive symptoms in three chronic illness
conditions: rheumatoid arthritis, end-stage renal disease, and multiple
sclerosis. J Psychosom Res 1993;37:163–70.
 Hickie I, Davenport T, Wakefield D, Volmer-Conna U, Cameron B, Vernon S,
et al. Post-infective and chronic fatigue syndromes precipitated by viral and
non-viral pathogens: prospective cohort study. BMJ 2006;2006(333):575.
 Lorusso L, Mikhaylova S, Capelli E, Ferrari D, Ngonga G, Ricevuti G.
Immunological aspects of chronic fatigue syndrome. Autoimmun Rev
 Di Franco M, Iannucelli C, Valesini G. Neuroendocrine immunology of
fibromyalgia. Ann NY Acad Sci 2010;1193:84–90.
in patientswith fibromyalgia
 Irwin M. Effects of sleep and sleep loss on immunity and cytokines. Brain
Behav Immun 2002;16:503–12.
 Burgos I, Richter L, Klein T, Fiebich B, Feige B, Lieb K, et al. Increased nocturnal
interleukin-6 excretion in patients with primary insomnia: a pilot study. Brain
Behav Immun 2006;20:246–53.
 Irwin M, Rinetti G, Redwine L, Motivala S, Dang J, Ehlers C. Nocturnal
proinflammatory cytokine-associated sleep disturbances in abstinent African
American alcoholics. Brain Behav Immun 2004;18:349–60.
 Motivala S, Sarfotti A, Olmos L, Irwin M. Inflammatory markers and sleep
disturbance in major depression. Psychosom Med 2005;2005(67):187–94.
 Thomas K, Motivala S, Olmstead R, Irwin M. Sleep depth and fatigue: role of
cellular inflammatory activation. Brian Behav Immun 2011;25:53–8.
 Lashley F. A review of sleep in selected immune and autoimmune disorders.
Holistic Nurs Pract 2003;17:65.
 White J, Darko D, Brown S, Miller J, Hayduk R, Kelly T. Early central nervous
system response to HIV infection: sleep distortion and cognitive-motor
decrements. AIDS 1995;9:1043–50.
 Darko D, Mitler M, Miller J. Growth hormone, fatigue, poor sleep, and disability
in HIV infection. Neuroendocrinol 1998;67:317–24.
 Rubinstein M, Selwyn P. High prevalence of insomnia in an out-patient
population with HIV infection. J Acquir Immune Defic Syndr Hum Retrovirol
 Drewes A, Nielsen K, Hansen B, Taagholt S, Bjerregârd K. A longitudinal study
of clinicalsymptomsand sleep
 Stone A, Broderick J, Porter L, Kaell A. The experience of rheumatoid arthritis
pain and fatigue: examining momentary reports and correlates over one week.
Arthritis Care Res 1997;10:185–93.
 Yunus M, Ahles T, Aldag J, Masi A. Relationship of clinical features with
psychological status in primary fibromyalgia. Arthritis Rheum 1991;34:15–21.
 Moldofsky H, Scarisbrick P, England R, Smythe H. Musculoskeletal symptoms
and non-REM sleep disturbance in patients with ‘‘fibrositis syndrome’’ and
healthy subjects. Psychosom Med 1975;37:341–51.
 Nokes K, Kendrew J. Sleep quality in people with HIV disease. J Assoc Nurses
AIDS Care 1996;7:43–50.
 Riemann D, Voderholzer U. Primary insomnia: a risk factor to develop
depression? J Affect Disord 2003;76:255–9.
 Léger D, Partinen M, Hirshkowitz M, Chokroverty S, Hedner J. Characteristics of
insomnia in a primary care setting: EQUINOX survey of 5293 insomniacs from
10 countries. Sleep Med 2010;11:987–98.
 Ohayon M. Prevalence and correlates of nonrestorative sleep complaints. Arch
Intern Med 2005;165:35–41.
 Sarsour K, Van Brunt D, Johnston J, Foley K, Morin C, Walsh J. Associations with
nonrestorative sleep with insomnia, depression, and daytime function. Sleep
 Semler C, Harvey A. Misperception of sleep can adversely affect daytime
functioning in insomnia. Behav Res Ther 2005;43:843–56.
 Mayers A, Van Hooff J, Baldwin D. Quantifying subjective assessment of sleep
Psychopharmacol Clin Exp 2003;18:21–7.
 Argyropoulos S, Hicks J, Nash J, Bell C, Rich A, Nutt D, et al. Correlation of
subjective and objective sleep measurements at different stages of the
treatment of depression. Psychiatry Res 2003;120:179–90.
 Lee J, Reynolds C, Hoch C, Buysse D, Mazumdar S, George C, et al.
 Andrew G, Mayers A, Baldwin D. The relationship between sleep disturbance
and depression. Int J Psychiatry Clin Pract 2006;10:2–16.
 Roth T, Zammit G, Lankford A, Mayleben D, Stern T, Pitman V, et al.
 Ohayon M, Riemann D, Morin C, Reynolds C. Hierarchy of insomnia criteria
based on daytime consequences. Sleep Med 2012;13:52–7.
 Gardner G, Petrin J, Collier A, Paauw D. Myalgias, non-restorative sleep, and
fatigue in patients infected with the human immunodeficiency virus. J
Musculoskelet Pain 1997;5:29–39.
 Resta O, Barbaro M, Bonfitto P, Giliberti T, Depalo A, Pannacciulli N, et al. Low
sleep quality and daytime sleepiness in obese patients without obstructive
sleep apnoea syndrome. J Intern Med 2003;253:536–43.
 Mercadante S, Girelli D, Casuccio A. Sleep disorders in advanced cancer
 Power D, Peruccio A, Badley E. Pain as a mediator of sleep problems in arthritis
and other chronic conditions. Arthritis Rheum 2005;53:911–9.
 Calhoun A, Ford S, Finkel A, Kahn K, Mann J. The prevalence and spectrum of
 Krakow B, Meledrez D, Sisley B, Warner T, Krakow J, Leahigh L, et al. Nasal
dilator strip therapy for chronic sleep-maintenance insomnia and symptoms
of sleep-disordered breathing: a randomized controlled trial. Sleep Breath
 Knudsen H, Ducharme L, Roman P. Job stress and poor sleep quality: data from
an American sample of full-time workers. Soc Sci Med 2007;64:1997–2007.
 Majer M, Jones J, Unger E, Youngblood L, Decker M, Gurbaxani B, et al.
Perception versus polysomnographic assessment of sleep in CFS and non-
recently remitted,elderly depressed
component ofinsomnia. Sleep
associated. SupportiveCare Cancer
K. Wilkinson, C. Shapiro/Sleep Medicine 13 (2012) 561–569
fatigued control subjects: results from a population-based study. BMC Neurol
 Rutledge D, Jones K, Jones C. Predicting high physical function in people with
fibromyalgia. J Nurs Scholarsh 2007;39:319–24.
 Davies K, Macfarlane G, Nicholl B, Dickens C, Morriss R, Ray D, et al. Restorative
sleep predicts the resolution of chronic widespread pain: results from the
EPIFUND study. Rheumatology 2008;47:1809–13.
 Creti L, Libman E, Baltzan M, Rizzo D, Bailes S, Fichten C. Impaired sleep in
chronic fatigue syndrome: how is it best measured? J Health Psychol
 Johannson I, Karlson B, Grankvist G, Brink E. Disturbed sleep, fatigue, anxiety
and depression in myocardial infarction patients. Eur J Cardiovasc Nurs
 Neu D, Mairesse O, Hoffmann G, Valsamis J, Verbanck P, Linkowski P, et al. Do
‘sleepy’ and ‘tired’ go together? Rasch analysis of the relationships between
sleepiness, fatigue and nonrestorative sleep complaints in a nonclinical
population sample. Neuroepidemiol 2010;35:1–11.
K. Wilkinson, C. Shapiro/Sleep Medicine 13 (2012) 561–569