Original Research

Medical Journal of the Islamic Republic of Iran.Vol. 23, No. 4, February, 2010. pp. 184-188

The impact of anxiety on sleep quality

Azizeh Afkham-Ebrahimi, MSc1, Maryam Rasoulian, MD.2, ZahraTaherifar3, Maryam Zare4

Tehran Psychiatric Institute, Rasoul Akram Hospital, Iran University of Medical Science, Tehran, Iran.

Abstract
Background: A significant relationship between psychiatric disorders principally
depression and anxiety and insomnia is well-known in general population. A high per-
cent of insomnia sufferers report anxiety symptoms. Anxiety is also frequently seen in
medical patients whom complaints of sleep problems are often prominent.
Method: 250 outpatients with various medical complaints participated in the study
and completed Pittsburgh Sleep Quality Index(PSQI) and Beck Anxiety inventory
(BAI).
Results: The patients reported moderate anxiety on BAI. Significant correlations
were found between anxiety and four components of sleep; sleep quality, sleep latency,
sleep duration and habitual sleep efficiency.
Conclusion: The correlation of sleep complaints with anxiety symptoms indicated a
high interrelatedness between anxiety and sleep complaints. Insomnia is a disorder of
hyperarousal. Nonetheless inqury into mechanisms of arousal regulation could further
explain the anxiety and sleep disorders as well.

Keywords: Anxiety, sleep quality, Beck Anxiety Inventory, Pittsburgh Sleep Quali-
ty Index

ta from the large-scale Epidemiologic Catch-

Introduction ment Area (ECA) project demonstrates the rela-
Sleep is an active state, critical for our physi- tively high percentage of individuals in the gen-
cal, mental and emotional well-being and im- eral population who suffer from significant in-
portant for optimal cognitive and overall func- somnia symptoms and meet the criteria for
tioning. However sleep problems are very com- mood and anxiety disorders. 40% of these in-
mon in the general population, affecting 10- somnia sufferers met the criteria for at least 1
20% of adults [1]. Sleep disorders coexist with psychiatric disorder. Anxiety disorder was di-
a number of physical and psychiatric condi- agnosed in 24% of insomniacs [5-7]. There is a
tions, including psychoses, anxiety and mood high interrelatedness between sleep and anxiety
disorders [2-4]. In general population studies, a . Anxiety complaint such as worrying and rumi-
significant relationship between psychiatric nation are named to be important in sleep prob-
disorders principally depression and anxiety lems [8,9] . Sleep disorders are also a common
and insomnia is well-known. An analysis of da- but under-recognized problem in general hospi-
1. Corresponding author, MSc in Clinical Psychology. Academic Member of Department of Clinical Psychology, Tehran. Psychiatric Insti-
tute. Satarkhan, Niayesh, Next to Shahid Mansouri St. Tel: +989121597189. Email: [email protected]
2. Psychiatrist. Assistant Professor of Psychiatry Department. Rasoul Akram hospital .Tehran-Iran
3&4 . MSc in Clinical Psychology. Tehran Psychiatric Institute. Tehran, Iran
A. Afkham-Ebrahimi, et al.

Fig. 1. The correlation between the PSQI and anxiety

tals and approximately 35% of the patients month . The exclusion criteria was current or a
complain of insomnia [10]. From an epidemio- history of psychiatric disorders All of these in-
logic perspective anxiety symptoms often her- formations were obtained through a brief inter-
ald the onset of sleep problems. Anxiety is also view with the patients.
frequently seen in medical patient whom com- The assessment was conducted in two stages.
plaints of sleep problems are often prominent In the first stage, participants completed the
[11]. The aim of this study was to investigate PSQI and then those who reported frequent
the impact of anxiety on components of sleep anxiety on 5th component ( sleep disturbances)
quality (e.g sleep duration, sleep latency, sleep of the PSQI, were entered the second stage and
efficiency, sleep disturbances and daytime dys- the BAI was administered. 96 ( 55 women and
function) in a sample of medical outpatients. 41 men) out of 234 patients qualified for entry
to the second stage and the remaining 138 pa-
Method tients were excluded.
Participants and procedure
A total of 250 ( 138 women and 112 men ) Measures
outpatients with various medical complaints at- 1- The PSQI is a self-rated questionnaire that
tending different internal/surgical clinics ( ex- provides an index of sleep quality for a 1 month
pect for psychiatry clinic) of a training general interval and comprises 19 questions and con-
hospital and agreed to participate in the study, sidered to be an instrument with established re-
were selected . The age range of the participants liability and validity [12]. These 19 questions
was between 18-74 ( Mean= 42.56; SD=17.18). were grouped into 7 component scores, each
The inclusion criteria were 18 years and older, weighted equally on a 0-3 scale. The 7 compo-
high school education , intact cognition being nent were subjective sleep quality (C1), sleep
communicative and cooperative and no regular latency (C2), sleep duration(C3), habitual sleep
sedative-hypnotics consumer, over the past efficiency (C4), sleep disturbances (C5), use of

MJIRI.Vol. 23, No.4, February, 2010. pp. 178-188 185

 The impact of anxiety on sleep quality
7.67( SD= 3.86) and higher than the anchor
point. The mean of anxiety score was 18.11
(SD= 9.75) on BAI which indicated moderate
anxiety in our medical patients.
The associations of PSQI and anxiety scores
are shown in Fig. 1. The scatter of anxiety and
global sleep scores showed a significant linear
Table 1. The relationship of sleep quality components relationship between the two variables.
and anxiety.
Table 1 shows the correlations of anxiety scores
with sleep components. There were significant
sleep medication (C6) and daytime dysfunction
relationships between anxiety , sleep quality,
(C7). The sum of scores for these seven compo-
sleep latency, sleep duration and habitual sleep
nents yielded a global score between 0 and 21
efficiency.
with higher scores indicating worse sleep quali-
ty. The PSQI has been shown to be valid and re-
Discussion
liable in Iranian population [Ebrahimi et al in
There is now compelling evidence for link-
press]. We used the scores >6 on PSQI as an in-
ages between the regulation of sleep, and emo-
dicator of poor sleep quality.
tions. Spielman and colleagues (1996) pro-
2- The BAI is a 21 items, self-reporting ques-
posed a behavioral model of insomnia that in-
tionnaire that measures the severity of anxiety
corporates predisposing or constitutional fac-
in adults and adolescents [13]. The anxiety
tors, precipitating factors that serve as "trig-
symptoms were rated on a 4-point scale from "
gers" and perpetuating factors that sustain the
not at all" (0 points) to "severe" (3 points). The
sleep disruption [14]. Anxiety (which is often
total score had ranges from 0 to 63. Scores from
brought on by a stressful event) is probably the
0-7 points reflected a minimal level of anxiety,
most common cause of difficulty in falling
8-15 mild anxiety, 16-25 moderate anxiety and
asleep. This is called " stress related insomnia".
26-63 severe anxiety. The scale had good psy-
Clinicians have long noted associations be-
chometric properties and was shown to be valid
tween sleep disruption and anxiety symptoms.
and reliable in Iranian subjects (unpublished
Our patients with sleep difficulties reported to
dissertations).
have moderate anxiety on BAI. The severity of
reported anxiety was correlated with poorer
Data analysis
sleep quality, longer sleep latency, shorter sleep
Statistical analysis was performed using
duration and poor perceived sleep efficiency
SPSS 15 for windows. Descriptive statistics
which comprised the criteria of insomnia. The
and scatter-plot graph were used for demon-
study of lamberg [15] show that bedtime state
strating the distribution and correlation of anxi-
anxiety has been positively related to perceived
ety and PSQI scores. The bivariate relations be-
sleep latency and negative emotions were
tween continuous variables ( anxiety , and com-
found to influence particular types of insomnia;
ponents of PSQI ) were calculated with Pearson
sleep onset versus maintenance which is con-
product-moment correlations.
sistent with our results. We electronically
searched MEDLINE and PSYCHLINC data-
Results
base for further relevant references on anxiety
41% of the patients were reported to have
and components of insomnia. . The majority of
anxiety on PSQI-C5 in the initial screening and
the articles were focused on relationship of in-
the mean for PSQI in the screened patients was
somnia with an anxiety disorder or the impact
186 MJIRI.Vol. 23, No.4, February, 2010. pp. 178-188
A. Afkham-Ebrahimi, et al.
of anxiety on a certain sleep disorder. Although Furthermore, systematic work has not yielded a
we did ruled out the recent or history of any psy- practical, valid, and objective method of assess-
chiatric disorder in our patients, but the severity ing a person's arousal level although brain im-
of anxiety symptoms in some of our patients aging along with elevations in cortisol and
suggested unaddressed anxiety symptoms and changes in circadian rhythms of cortisol secre-
disorder in this population. The high correlation tion shows evidence of hyperarousal in insom-
of anxiety and four components of insomnia in niacs. Insights into mechanisms of arousal reg-
our patients also showed the comorbidity of ulation would have been applicable to both anx-
these two clinical entities. iety and sleep disorders. It is also important to
In fact, there are theoretical links between note that the association between cognitive er-
anxiety and insomnia. Sleep is a necessary and rors and sleep problems was mediated by the
restorative state of diminished cortical arousal. presence of both anxiety and depressive symp-
A global reduction in metabolism as well as rel- toms [20]. Thus cognitive biases seemed to be
ative deactivation of a broad range of other one of several potential mechanisms involved
physiologic processes are hallmarks of the tran- in the co-occurrence of these problems in our
sition of sleep. Anxiety and fear states manifest studied population.
with heightened cortical and peripheral arousal.
The studies had shown that insomnia is a disor- Conclusion
der of hyperarousal [16]. The hyperarousal has We found a high interrelatedness between
been evidenced among insomniacs with in- anxiety and sleep components in our study. Al-
creased brain metabolism, increases in auto- though the nature of these association does not
nomic nervous system activity, EEG spectral allow us to make directionality, the relation-
analysis, and norepinephrine levels. It is there- ships between sleep and affective functioning
fore not surprising that heightened physiologic seemed to be bidirectional and remains a chal-
arousal, indexed by such changes as increased lenge for the clinical evaluation.
body temperature, heart rate and muscle ten- Whereas sleep has been extensively studied
sion induced by anxiety is incompatible with in depressive disorders but the PSG study of
sleep [17]. These measures of increased arousal anxiety disorders was less developed. The PSG
are not only present when patients are lying in studies of anxiety disorders provide informa-
their beds awake trying to sleep, but also when tion regarding measurable objective distur-
subjects are asleep and more importantly dur- bances in sleep initiation, maintenance and oth-
ing the day when subjects are not even trying to er components of sleep quality as well as the
sleep. [1] Presleep cognitive arousal is also re- presence of a primary sleep pathologic process.
ported to be the primary predictor of sleep qual- This study had the following Limitations:
ity in medical patients [18]. This was reported The same interviewers assessed the insomnia
also by our patients in the related subscale of complaints and psychiatric diagnosis and lack
PSQI. of blinding can lead to experimenter bias. The
Because increases in nocturnal secretion of absence of information related to medical con-
cortisol also has been found among adults with ditions and drugs the patients were taking pre-
insomnia, alterations in the hypothalamic-pitu- cluded a deep analysis of the relationship be-
itary-adrenal axis seem to have implication tween these conditions and insomnia. Our data
both for the development of psychopathology were self-reported, whereas sleep disturbances
as well as the sleep-wake cycle [19]. Patients are measured most accurately with the use of
may be unaware of their hyperarousal, however polysomnography. Nonetheless subjective and
which poses a problem for the diagnostician. objective measures of sleep quality did not cor-
MJIRI.Vol. 23, No.4, February, 2010. pp. 178-188 187
 Association between sdLDL ...

relate well with each other. Finally, our findings with and without insomnia. Biological Research For
were cross-sectional in nature, and the direction Nursing 2004; 6: 46-58.
15. Lamberg L. Insomnia shows strong link to psychi-
of causality was not determined definitively. atric disorders. Psychiatric News, 2005 ; 40: 21-3.
16. Drake CL, Schwartz JR.L, Roth T. The evolu-
tion of insomnia in relation to comorbidity. Psychiatric
Annals 2008; 38: 621-6.
References 17. Bonnet MH, Arand DI. 24 hour metabolic rate in
1. Roth T. Insomnia and sleep-related disorders. Psy- insomniacs and matched normal sleepers. Sleep 1995;
chiatric Annals 2008; 38: 575-77. 19: 581-88.
2. Walsh JK. Clinical and socioeconomic correlates of 18. Smith MT, Perlis ML, Smith MS, Giles DE, Car-
insomnia. Journal of Clinical psychiatry 2004; 659 suppl mody TP. Sleep quality and presleep arousal in chronic
8) : 13-19. pain. Journal of behavior Medicine 2000; 23: 1-13.
3. Riemann D, Voderholzer U. Primary insomnia: A 19. Alfano Candice A, Zakem Alan H, Costa Natalie
risk factor to develop depression? Journal of affective M, Taylor Leslie K, Weems Carl F. Sleep problems and
disorders 2003; 76: 255-59. their relations to cognitive factors, anxiety, and depres-
4. Verhulst FC, van der Ende J, Ferdinand RF, Kasius sive symptoms in children and adolescents. Depression
MC. The prevalence of DSM-III-R diagnosis in a nation- and Anxiety 2008; 0: 1-10.
al sample of dutch adolescents. Archives of General Psy- 20. Morin CM, Blais F, Savard J. Are changes in be-
chiatry 1997; 54: 329-36. liefs and attitudes about sleep related to sleep improve-
5. Ohayon MM, Caulet M, Lemoine P. Comorbidity of ments in the treatment of insomnia? Behavior Research
mental and insomnia disorders in the general population. and Therapy 2002; 40: 741-52.
Comprehensive Psychiatry 1998; 39: 185-97.
6. Weyerer S, Dilling H. prevalence and treatment of
insomnia in the community: results from the Upper
Bavarian Field Study. Sleep 1991; 14: 392-98.
7. Ford DE, Kamerow DB. Epidemiologic study of
sleep disturbances and psychiatric disorders. An oppor-
tunity for prevention? JAMA 1989; 262: 1479-84.
8. Spoormaker Victor I, Bout Jan van den. Depression
and anxiety complaints; relations with sleep distur-
bances. European Psychiatry 2005; 20: 243-45.
9. Harvey A.G, Jones C, Shmidt DA. Sleep and post-
traumatic stress disorder: a review. Clinical Psychology
review 2003; 23: 377-407.
10. Rocha Fabio Lopes, Hara Claudia, Vaz Rodrigues
Cristina, Antunes costa Milena, Castro e Costa Erico,
fuzikawa Cintia, Guimaraes Santos valeria. Is insomnia a
marker for psychiatric disorders in general hospitals?
Sleep Medicine 2005; 6: 549-53.
11. Stein MB. Attending to anxiety disorders in pri-
mary care. Journal of Clinical psychiatry 2003; 64(suppl
15) : 35-9.
12. Buysse DJ, Reynolds CF, Berman SR, Kupfer DJ.
The Pittsburgh Sleep Quality Index : a new instrument
for psychiatric practice and research. Psychiatry Re-
search 1989; 28: 193-213.
13. Beck AT, Epstein N, Brown G, Steer RA. An inven-
tory for measuring clinical anxiety psychometric proper-
ties. Journal of consulting and clinical Psychology 1988;
56: 893-97.
14. Cheek Rita E, Shaver Joan L, Lentz Martha J.
Lifestyle practices and nocturnal sleep in midlife women

188 MJIRI.Vol. 23, No.4, February, 2010. pp. 178-188