<%BANNER%>

Sleep Disturbance in Chronic Pain Patients

Permanent Link: http://ufdc.ufl.edu/UFE0021908/00001

Material Information

Title: Sleep Disturbance in Chronic Pain Patients
Physical Description: 1 online resource (90 p.)
Language: english
Creator: O'Brien, Erin
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: actigraphy, anxiety, back, chronic, depression, diary, facial, fibromyalgia, insomnia, mediation, mood, pain, psychophysical, sem, sleep
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Chronic pain is a prevalent problem and it is associated with a number of negative consequences. Sleep disturbances are a common complaint reported by chronic pain patients, with 50-70% of patients endorsing significant sleep disturbance. The presence of concomitant sleep problems can significantly complicate both the course and the management of chronic pain. Further, there is evidence to suggest that the relationship between sleep disturbance and pain might be reciprocal, such that pain can disrupt sleep and poor or disrupted sleep may lead to increased pain. Additionally, associations among pain, negative mood, and sleep disturbance among chronic pain patients have been inconsistent. Specifically, some investigators have reported greater negative mood (as well as higher pain intensity) among self-reported 'poor sleepers,' whereas other researchers have reported higher pain ratings but no differences on measures of negative mood among good and poor sleepers. More clearly defining the relationship between pain and sleep disturbance, as well as the roll of negative mood, may further clarify the shared pathophysiology of sleep and pain. This study examined sleep, pain, and negative mood in 292 adults, 18 to 65 years of age, with chronic back pain, facial pain, or fibromyalgia. Additionally, a subset of 22 participants completed two weeks of daily sleep diaries and actigraphy monitoring and participated in psychophysical pain testing procedures. A generalized pattern of sleep disturbance (difficulties with sleep onset, sleep maintenance, and poor sleep quality) was reported by all groups of chronic pain patients, with facial pain patients reporting relatively less disturbed sleep overall. Results also indicated a direct relationship between poor sleep and increased pain, and further revealed that negative mood mediated the relationship between poor sleep and increased pain when it was included in the model. No significant results emerged from analyses examining pain response to psychophysical testing among good and poor sleepers, although moderate to large effect sizes were found. Findings suggest multiple pathways between sleep disturbance and individuals' pain experience, such that poor sleep may lead to increased pain but higher levels of negative mood may also lead to decreased sleep, resulting in more pain.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Erin O'Brien.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Robinson, Michael E.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021908:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021908/00001

Material Information

Title: Sleep Disturbance in Chronic Pain Patients
Physical Description: 1 online resource (90 p.)
Language: english
Creator: O'Brien, Erin
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: actigraphy, anxiety, back, chronic, depression, diary, facial, fibromyalgia, insomnia, mediation, mood, pain, psychophysical, sem, sleep
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Chronic pain is a prevalent problem and it is associated with a number of negative consequences. Sleep disturbances are a common complaint reported by chronic pain patients, with 50-70% of patients endorsing significant sleep disturbance. The presence of concomitant sleep problems can significantly complicate both the course and the management of chronic pain. Further, there is evidence to suggest that the relationship between sleep disturbance and pain might be reciprocal, such that pain can disrupt sleep and poor or disrupted sleep may lead to increased pain. Additionally, associations among pain, negative mood, and sleep disturbance among chronic pain patients have been inconsistent. Specifically, some investigators have reported greater negative mood (as well as higher pain intensity) among self-reported 'poor sleepers,' whereas other researchers have reported higher pain ratings but no differences on measures of negative mood among good and poor sleepers. More clearly defining the relationship between pain and sleep disturbance, as well as the roll of negative mood, may further clarify the shared pathophysiology of sleep and pain. This study examined sleep, pain, and negative mood in 292 adults, 18 to 65 years of age, with chronic back pain, facial pain, or fibromyalgia. Additionally, a subset of 22 participants completed two weeks of daily sleep diaries and actigraphy monitoring and participated in psychophysical pain testing procedures. A generalized pattern of sleep disturbance (difficulties with sleep onset, sleep maintenance, and poor sleep quality) was reported by all groups of chronic pain patients, with facial pain patients reporting relatively less disturbed sleep overall. Results also indicated a direct relationship between poor sleep and increased pain, and further revealed that negative mood mediated the relationship between poor sleep and increased pain when it was included in the model. No significant results emerged from analyses examining pain response to psychophysical testing among good and poor sleepers, although moderate to large effect sizes were found. Findings suggest multiple pathways between sleep disturbance and individuals' pain experience, such that poor sleep may lead to increased pain but higher levels of negative mood may also lead to decreased sleep, resulting in more pain.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Erin O'Brien.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Robinson, Michael E.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021908:00001


This item has the following downloads:


Full Text





SLEEP DISTURBANCE IN CHRONIC PAIN PATIENTS


By

ERIN M. O'BRIEN




















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008






























2008 Erin M. O'Brien


























To my parents ...
You've cheered me on in all that I do, supported me when I needed you to stand with me, given
me the courage to the believe in myself and to stand on my own, and always told me that there
was nothing I could not do.

You are my inspiration now and always.









ACKNOWLEDGEMENTS

I thank my supervisory committee for their support and mentoring. I especially thank my

supervisory committee chair, Dr. Michael Robinson, for his continual mentoring and feedback. I

would also like to thank my participants for their dedicated participation, and the

MiniMitter/Respironics Corporation for the provision of equipment for this research. Finally, I

want to thank my family for their constant support and encouragement, without which this

accomplishment would not have been possible.









TABLE OF CONTENTS



A C K N O W L E D G E M E N T S ............................................................................................................4

L IST O F T A B L E S ......................................................................................................... ........ .. 7

LIST OF FIGURES ............................................. .. .......... ............ ...............8

LIST OF ABBREVIATIONS .............................................. ...............................................9.......

A B S T R A C T .......................................................................................................... ..................... 1 1

CHAPTER

1 INTRODUCTION ..................................... .. ........... ............................... 13

Prevalence and Impact of Chronic Pain............................................................................ 13
Subtypes of P ain D isorders................................................... ............................................ 15
B ack P ain ....................................................................................................... ....... .. 15
F racial P ain ............... ...................................................................................... ........ .. 16
Fibromyalgia and Rheumatic Conditions.................................................... 17
B asic Sleep Inform ation ................................................................... ............... 18
Assessment of Sleep Patterns and Sleep Problems............................................................ 19
Experimentally Induced Sleep Disturbances in Healthy Participants ............................... 22
Sleep D isturbances in Pain P atients......................................... ....................... ................ 23
B ack P ain ....................................................................................................... ....... .. 2 4
F racial P ain ............... ...... ............................................................................. ....... .. 2 5
Fibromyalgia and Rheumatic Conditions................................................................... 27
Neurobiology Findings on the Sleep-Pain Relationship.................................................... 29
Sleep Disturbances and Experimental Pain Testing ............... ...................................31
Negative Mood, Sleep, and Pain ..................................................................................... 32
C u rrent Stu dy ......................................................................................................... ........ .. 3 3
S p e cific A im s ................................................................................................................. 3 7
H y p o th e se s ......................................................................................................................3 7

2 M E T H O D S ................................................................................................... ..................... 3 9

P a rtic ip a n ts .............................................................................................................................3 9
P ro c e d u re ..................................................................................................... ..................... 4 0
M e a su re s ................................................................................................................................. 4 2
Graded Thermal Stimulation or RAMP and HOLD (RH) ...........................................46
T em poral Sum m ation (W ind-up) .................................................................................... 47
Sleep D iary and A ctigraphy ....................................................................... ...............47
S statistic al A n aly se s .................................................................................................................4 9










3 R E S U L T S ............................................................................................................................... 5 1

Differences in Subjective Sleep Across the Three Pain Groups.......................................51
Role of Negative Mood in the Sleep-Pain Relationship....................................................52
Psychophysical Testing-Temporal Summation/"Wind up" and Ramp and Hold............... 54
A ctigraphy D ata and Sleep D iary D ata ..................................................................................55

4 D IS C U S S IO N .........................................................................................................................6 3

Comparison of Study Results to Previous Findings ............... ....................................68
Role for Negative Mood in the Sleep-Pain Relationship...................................................70
Implications for Conceptualization and Treatment of Chronic Pain.................................71
L im station s ........................................................................................................... ....... .. 74
F u tu re D ire ctio n s .................................................................................................................. .. 7 5

L IST O F R EFER EN CE S ............................................................................................. 77

B IO G R A PH IC A L SK E T C H .............. ...................................................................... 90









LIST OF TABLES


Table page

3-1 Demographic information for the 3 chronic pain patient groups..................................59

3-2 Comparison of diary- and actigraphy-measured sleep variables across 3 pain groups .....60

3-3 Multivariate mixed model MANOVA results for pain ratings in temporal summation
and ram p and hold procedures .......................................... ......................... ................ 6 1

3-4 Correlation between actigraphy- and diary-measured sleep variables for good and
p o o r sle e p e rs ................................................................................................................... ... 6 2









LIST OF FIGURES


Figure page

3-1 Self-reported sleep (as reported by PSQI) for 3 chronic pain groups...............................57

3-2 Structural equation model for the relationship among sleep, pain, and negative mood. ...58











BDI

BMI

BP

CBT

CBT-I

CNS

CPH

CSQ

CSQ-R

DSM-IV-TR

EEG

FMS

FP

HPA

ICSD

IL-1

MANOVA

MCV

MPQ

NK cell

NREM

PASS

PDI


LIST OF ABBREVIATIONS

Beck Depression Inventory

Body mass index

Back pain

Cognitive behavior therapy

Cognitive behavior therapy for insomnia

Central nervous system

Chronic paroxysmal hemicrania

Coping Strategies Questionnaire

Coping Strategies Questionnaire Revised

Diagnostic and Statistical Manual of Mental Disorders, 4th edition, text revision

Electroencephalogram/electroencephalographic

Fibromyalgia

Facial pain

Hypothalamic-pituitary-adrenal

International Classification of Sleep Disorders

Interleukin-1

Multivariate analysis of variance

Medical College of Virginia Pain Questionnaire

McGill Pain Questionnaire

Natural killer cell

Non-rapid eye movement

Pain Anxiety Symptom Scale

Pain Disability Index










PILL Pennebaker Inventory of Limbic Languidness

PSG Polysomnography

PSQI Pittsburgh Sleep Quality Index

REM Rapid eye movement

RM Raphe magnus

RMSEA Root mean square error of approximation

SE Sleep efficiency

SEM Structural equation modeling

SES Socio-economic status

SF-36 Medical Outcomes Survey Short Form-36

SOL Sleep onset latency

STAXI State-Trait Anger Expression Inventory

TENS Transcutaneous electrical nerve stimulation

TMD Temporomandibular disorder

TST Total sleep time

VAS Visual analogue scale

WASO Wake after sleep onset









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

SLEEP DISTURBANCE IN CHRONIC PAIN PATIENTS

By

Erin M. O'Brien

August 2008

Chair: Michael E. Robinson
Major: Psychology

Chronic pain is a prevalent problem and it is associated with a number of negative

consequences. Sleep disturbances are a common complaint reported by chronic pain patients,

with 50-70% of patients endorsing significant sleep disturbance. The presence of concomitant

sleep problems can significantly complicate both the course and the management of chronic pain.

Further, there is evidence to suggest that the relationship between sleep disturbance and pain

might be reciprocal, such that pain can disrupt sleep and poor or disrupted sleep may lead to

increased pain.

Additionally, associations among pain, negative mood, and sleep disturbance among

chronic pain patients have been inconsistent. Specifically, some investigators have reported

greater negative mood (as well as higher pain intensity) among self-reported "poor sleepers,"

whereas other researchers have reported higher pain ratings but no differences on measures of

negative mood among good and poor sleepers. More clearly defining the relationship between

pain and sleep disturbance, as well as the roll of negative mood, may further clarify the shared

pathophysiology of sleep and pain.

This study examined sleep, pain, and negative mood in 292 adults, 18 to 65 years of age,

with chronic back pain, facial pain, or fibromyalgia. Additionally, a subset of 22 participants









completed two weeks of daily sleep diaries and actigraphy monitoring and participated in

psychophysical pain testing procedures.

A generalized pattern of sleep disturbance (difficulties with sleep onset, sleep

maintenance, and poor sleep quality) was reported by all groups of chronic pain patients, with

facial pain patients reporting relatively less disturbed sleep overall. Results also indicated a

direct relationship between poor sleep and increased pain, and further revealed that negative

mood mediated the relationship between poor sleep and increased pain when it was included in

the model. No significant results emerged from analyses examining pain response to

psychophysical testing among good and poor sleepers, although moderate to large effect sizes

were found. Findings suggest multiple pathways between sleep disturbance and individuals'

pain experience, such that poor sleep may lead to increased pain but higher levels of negative

mood may also lead to decreased sleep, resulting in more pain.









CHAPTER 1
INTRODUCTION

Chronic pain conditions constitute a major challenge facing the health care system.

These conditions are often associated with substantial disability and distress for the individual

and also result in significant burdens on the health care system, as well as considerable economic

and social costs. Similarly, sleep disturbances are prevalent among the general population and

can also substantially impact individuals' physical, emotional, social, and occupational

functioning, particularly when these difficulties evolve into chronic problems. Recent attention

has begun to focus on the relationship between sleep disturbances and chronic pain conditions.

When these occur in concert, the resultant impact on the individual is even greater, leading to

higher levels of physical and emotional disability and greater functional impairments. Research

needs to clarify the relationship between sleep disturbances and chronic pain, in order to better

understand the potential reciprocal influences of these conditions on one another, as well as to

develop ways to intervene more effectively. As pain and sleep can be assessed in various ways,

it is important to consider both subjective and objective measures of sleep, and to consider

subjective pain reports as well as response to experimentally controlled painful stimuli. The

following review will examine the literature on these topics to date, and highlight the rationale

and proposed contribution of the current study.

Prevalence and Impact of Chronic Pain

Chronic nonmalignant pain is a prevalent problem among the adult population and is

associated with a number of negative physical, emotional, and social consequences. A review of

epidemiological studies conducted by Verhaak and colleagues (1998) found that the median

prevalence rate of chronic pain for adults across all studies was 15% (Verhaak, Kerssens,

Dekker, Sorbi, & Bensing, 1998). Subsequent epidemiological studies have reported similar









prevalence rates (Breivik, Collett, Ventafridda, Cohen, & Gallacher, 2005; Moulin, Clark,

Speechley, & Morley-Forster, 2002). Several studies have noted higher rates among females and

among older age groups (Moulin et al., 2002; B. H. Smith et al., 2001; Verhaak et al., 1998).

Andersson and colleagues (1999) examined health care and medication use among

Swedish adult reporting chronic pain (H. I. Andersson, Ejlertsson, Leden, & Schersten, 1999).

Results indicated that individuals with chronic pain were more likely to consult a physician or

physiotherapist, which is consistent with findings from other studies (Breivik et al., 2005). In

particular, perception of pain intensity was noted to be the most important factor motivating

health-seeking behaviors, although ethnicity, SES, age, and depressive symptoms were also

found to be important. Another commonly report among chronic pain samples is that, despite

increased health care usage, many patients report that their pain is inadequately managed

(Breivik et al., 2005).

Increased levels of chronic pain have been associated with increasingly negative

associations with employment, interference with daily activities, and general health (B. H. Smith

et al., 2001). Similarly, individuals with chronic pain also report decreased ability to participate

in social and occupational activities due to pain (Breivik et al., 2005; Moulin et al., 2002).

When it was examined, positive associations were also noted between chronic pain and

psychological symptoms (Verhaak et al., 1998). Psychological disorders are common correlates

of chronic pain conditions (Breivik et al., 2005; Duquesnoy, Allaert, & Verdoncq, 1998), and

when assessed, negative effects of pain on sexual activity are also frequently noted (Duquesnoy

et al., 1998).









Subtypes of Pain Disorders


Back Pain

Research has suggested that as many as 80% of adults will experience significant back

pain over the course of their lives (Lanes et al., 1995), and recurrence rates for low back pain

have been reported to be as high as 85% (Binkley, Finch, Hall, Black, & Gowland, 1993). In an

epidemiologic review of several cross-sectional studies (evidence level 1) by Anderson (1999),

the annual incidence of back pain of at least moderate intensity and duration was estimated to be

10-15% among the adult population (G. B. Andersson, 1999). Further, this review also

estimated the point prevalence of back pain of at least moderate intensity and duration to be 15-

30% among adults.

Low back pain is a commonly experienced pain condition, which involves substantial

social and economic costs (Malanga & Nadler, 1999). Grabois (2005) reported that, in the

United States, back pain accounts for expenditures of $14 billion per year, 19 million physician

visits, and half of all workers' compensation cases (Grabois, 2005). In addition, approximately

10 million Americans are disabled by chronic low back pain, and 250 million workdays are lost

per year due to this condition (Kriegler & Ashenberg, 1987). While 90% of patients with back

pain recover over a 3 month period, the remaining 10% of patients have a slow recovery process,

which involves large resource-intensive demands being placed on the healthcare system (G. B.

Andersson, 1999). According to Gallagher and Verma (1999), major depression commonly

accompanies chronic pain and can increase patients' level of impairment and disability

(Gallagher & Verma, 1999). Thus, a review by Verma and Gallagher (2002) concluded that, if

present, depression and anxiety need to be addressed in order to obtain a good functional

outcome from treatment (Verma & Gallagher, 2002).









Farasyn and Meeusen (2005) examined pressure pain thresholds in a group of 87 patients

with sub-acute low back pain, compared to 64 healthy control subjects. Results indicated that

the low back pain group had significantly lower pressure pain thresholds than the healthy

controls at several body sites (Farasyn & Meeusen, 2005). Flor and colleagues (2004) found

support for the idea of enhanced central and peripheral reactivity among chronic back pain

patients, due to lower pressure pain threshold and tolerance levels in these patients compared to

patients with tension headaches or healthy controls (Flor, Diers, & Birbaumer, 2004). Clauw

and colleagues (1999) found that, even after accounting for structural, demographic, and

psychosocial variables, pain sensitivity (threshold and tolerance) accounted for a significant

amount of variance in chronic low back pain patients' pain scores (from SF-36 subscale) and

functional status (Clauw et al., 1999).

Facial Pain

In a large random sample of adults, Macfarlane and colleagues (2002a) reported an

overall prevalence of 26% for orofacial pain, but found that only 46% of these individuals

reported seeking professional advice regarding their pain (Macfarlane, Blinkhorn, Davies,

Kincey, & Worthington, 2002). Prevalence of orofacial pain symptoms was noted to be higher

among women, and among men and women in the 18-25 year old age group. A review of 23

epidemiological studies by Pau and colleagues (2003) reported similar estimates for the

prevalence of "oral and facial pain" (40-44%) (Pau, Croucher, & Marcenes, 2003). While

gender was not associated with dental pain in this review, younger subjects and subjects from

lower SES groups were noted to be more likely to report pain. The impact of orofacial pain

includes, an inability to engage in normal activities, having to take time off from work due to

pain, and higher levels of psychologic distress (Macfarlane, Blinkhorn et al., 2002; Macfarlane,

Kincey, & Worthington, 2002).









Experimental pain testing in facial pain patient samples has produced conflicting

findings. A study by Bragdon and colleagues (2002) reported that women with TMD and

women without pain did not exhibit significant differences in pain threshold or tolerance

measures to either heat or ischemic pain (Bragdon et al., 2002). Similarly, Curran and

colleagues (1996) found no differences in pain or any psychological variables between TMD

patients and matched controls undergoing a pressure-pain task (Curran, Carlson, & Okeson,

1996). However, Maixner and colleagues (1995) found significantly lower thermal pain

thresholds (and a trend for lower thermal pain tolerance), as well as significantly lower ischemic

pain threshold and tolerance levels in a group of TMD patients compared to matched controls

(Maixner, Fillingim, Booker, & Sigurdsson, 1995). Moreover, findings from a study conducted

by Widerstrom-Noga and colleagues (1998) suggested that psychological variables, such as

stress or anxiety, can attenuated the degree of analgesia obtained from various methods

(acupuncture, TENS) in patients with tooth pain (Widerstrom-Noga et al., 1998).

Fibromyalgia and Rheumatic Conditions

According to a multi-site prospective study conducted by Wolfe and colleagues (1997),

the average yearly cost per patient (in 1996 dollars) was $2,274, although this number was

affected by those patients who utilized services at a higher level (Wolfe et al., 1997). Overall,

fibromyalgia patients reported attending an average of 10 outpatient medical visits per year, with

this number increasing even further when non-traditional treatments were included. Additionally

across the entire sample of fibromyalgia patients, an average of 2.7 disease-related drugs were

used in each 6-month study period, and hospitalization frequency averaged one stay every 3

years, with over half of the hospitalizations stemming from fibromyalgia-related symptoms. The

number of comorbid conditions reported by fibromyalgia patients has been associated with the

total costs and health care usage (Wolfe et al., 1997). Further, total cost and utilization level was









associated with both functional disability ratings, and global disease-severity in this sample of

fibromyalgia patients.

Laursen and colleagues (2005) found significantly lower pressure pain thresholds among

chronic pain patients of diverse etiologies compared to healthy female controls, and noted

significantly higher VAS ratings of habitual pain among female fibromyalgia/whiplash patients

compared to other female chronic pain patients (Laursen, Bajaj, Olesen, Delmar, & Arendt-

Nielsen, 2005). Carli and colleagues (2002) examined pain thresholds, using different modalities

of noxious stimuli (pressure, heat, cold, and ischemic), in fibromyalgia patients and control

participants (Carli, Suman, Biasi, & Marcolongo, 2002). The fibromyalgia patients were further

divided into 5 groups, based on tender point evaluations and how diffuse their pain was noted to

be. Fibromyalgia subjects had significantly lower heat pain threshold and cold pressure pain

threshold, and lower ischemic pain tolerance, compared to healthy subjects. Additionally, 4 of

the 5 fibromyalgia groups also demonstrated lower pressure pain threshold levels compared to

healthy subjects. Similarly, in a study by Hurtig and colleagues (2001) differences were noted

between fibromyalgia patients and healthy controls in cold and heat pain threshold levels, but not

in perception levels for warmth and cold (Hurtig, Raak, Kendall, Gerdle, & Wahren, 2001).

Sleep quality was not examined in relationship to pain threshold measures in this study.

Basic Sleep Information

Insomnia symptoms are among the most common sleep complaints, with prevalence

estimates commonly reported among 20-40% of adults (Bailey, 1997; Ohayon, 2002). Rates of

insomnia symptoms are also generally noted to be higher among women, and among elderly

individuals (Ohayon, 2002). The Diagnostic and Statistical Manual of Mental Disorders, 4th

edition, text revision's (DSM-IV-TR; (American Psychiatric Association, 1994) description of

insomnia includes difficulties with either the initiation or maintenance or sleep, as well as









complaints of nonrestorative sleep. In terms of duration, insomnia can be intermittent, last for a

short duration, or become a chronic problem (Bailey, 1997).

Sleep bruxism is a movement disorder whose primary symptom is clenching or grinding

the teeth during sleep (Bailey, 1997), and prevalence estimates have indicated that 5-20% of the

population is affected by this disorder (Glaros, 1981). Patients with sleep bruxism present with

almost daily complaints, which may include musculoskeletal pain or temporomandibular

disorders (Bailey, 1997), and as the severity and chronicity of this condition increases, there is

decreased likelihood of experiencing restorative sleep.

Assessment of Sleep Patterns and Sleep Problems

Sleep patterns and sleep problems can be assessed in a number of different ways, each

involving certain strengths and limitations. The three main methods of sleep assessment are

subjective measures (e.g., questionnaires, sleep diaries), behavioral measures (e.g., actigraphy),

and physiological measures (e.g., polysomnography; PSG), and the recent research has

demonstrated the utility of employing a multi-dimensional approach to the study of sleep,

particularly in the evaluation of insomnia complaints (see (Morin, 2003) for a review).

Questionnaires involve low burden on participants, and so are widely used measures for

the assessment of sleep, as well as factors related to sleep (e.g., emotional status, daytime

sleepiness, functioning, quality of life). Daily sleep diaries are widely used measures of sleep

patterns, especially in the evaluation of insomnia complaints, and can be especially useful tools

for assessing individuals' sleep patterns prior to, during, and following the implementation of a

treatment for a sleep problem. Although daily diary reports may evidence significant

discrepancies with information obtained via PSG, they permit evaluation of individuals'

perception of their sleep, and allow for prospective monitoring over longer periods of time.









Actigraphy is a behavioral measure of sleep, and can also be useful for assessing an

individual's response to treatment for a sleep problem. Research has demonstrated that

actigraphy provides accurate estimates of several global sleep parameters (such as total sleep

time, time in bed, total wake time) relative to PSG, while remaining unobtrusive as a

measurement device (Hauri & Wisbey, 1992; Sadeh, Hauri, Kripke, & Lavie, 1995; Vallieres &

Morin, 2003). Ancoli-Israel and colleagues (2003) reviewed research on the use of actigraphy

for sleep research, and reported correlations between actigraphy and polysomnography ranging

from 0.81 to 0.97 for total sleep time; 0.61 to 0.78 for percent of sleep; 0.53 to 0.94 for sleep

onset latency (Ancoli-Israel et al., 2003). It is important to note that correlations between

actigraphy and PSG, and between actigraphy and sleep diaries, are often moderate, especially

when estimating more specific sleep parameters (e.g., sleep latency, 0.12 to 0.69; time awake

after sleep onset; 0.22 to 0.37) (de Souza et al., 2003; Lockley, Skene, & Arendt, 1999).

However, for the assessment of certain sleep problems, particularly the night-to-night variability

prominent among insomnia patients, actigraphy may actually be a more appropriate diagnostic

tool than traditional PSG (Ancoli-Israel et al., 2003). Additionally, a recent validation study

conducted by Lichstein and colleagues (2006) concluded that actigraphy is a satisfactory

objective measure of number of awakenings, wake time after sleep onset (WASO), total sleep

time (TST), and sleep efficiency in insomnia patients, relative to PSG recordings (Lichstein et

al., 2006).

Kushida and colleagues (2001) reported that measurement of total sleep time and sleep

efficiency did not differ significantly between PSG data and combined data from actigraphy and

questionnaire reports in a group of 100 consecutive sleep-disordered patients (Kushida et al.,

2001). Further, when actigraphy parameters included a high-threshold (low-wake-sensitivity)









algorithm, the number of awakenings recorded was similar to those measured using PSG. These

researchers concluded that both subjective data and actigraphy data should be used to estimate

sleep parameters in sleep-disordered patients.

Studies have also compared the use of sleep logs and actigraphy in the assessment of

sleep patterns (Lockley et al., 1999; Wilson, Watson, & Currie, 1998). Lockley and colleagues

(1999) demonstrated good correlations between sleep logs and actigraphy for timing of sleep (r =

0.77 for sleep onset and r = 0.88 for sleep offset) and sleep duration (r = 0.57), as well as good

agreement in measurement of changes in sleep patterns over time. However, these

methodologies were not as highly correlated in measuring transitional variables during sleep,

such as sleep latency and number and duration of awakenings. Wilson and colleagues (1998)

also compared sleep reports using a diary measure and actigraphy in a group of patients with

chronic musculoskeletal pain (82.5% back pain). Results indicated that the two measures

provided similar estimates of TST, WASO, sleep efficiency, but differed in estimation of sleep

onset latency and number of awakenings. Pain severity was noted to be the variable that

evidenced the strongest association with sleep disturbance overall (as measured by sleep diary).

PSG is beneficial in the screening for sleep disorders such as obstructive sleep apnea or periodic

limb movement disorder, but is still not routinely used in the clinical evaluation of insomnia

complaints (Reite, Buysse, Reynolds, & Mendelson, 1995; Sateia, Doghramji, Hauri, & Morin,

2000). Despite its usefulness for screening purposes, PSG involves significant burden on

participants, which increases with the number of recording nights and if the recording is being

conducted in a lab versus the individual's home. Furthermore, while PSG provides a wealth of

valuable physiological data, it does not adequately address the subjective experience that is often

a central component to insomnia complaints.









An additional benefit provided by PSG evaluation of sleep is the ability to examine not

only the macrostructure of sleep architecture (stage REM, NREM stages 1-4), but also the ability

to assess sleep microstructure using advances in technology of EEG analysis. Sleep

microstructure analysis permits examination of the proportion of different frequencies of brain

waves (alpha, beta, gamma, delta, theta) throughout an individual's sleep. Different frequencies

of brain waves are associated with different stages of sleep (for example, lower frequency delta

waves are most often associated with stage 3 or 4 NREM "slow wave" sleep), and abnormalities

in the brain wave patterns during sleep have been noted in individuals who complain of certain

sleep disturbances.

Experimentally Induced Sleep Disturbances in Healthy Participants

Experimental studies have found that selective disruption of stage 4 NREM sleep (slow

wave sleep) in healthy participants led to musculoskeletal tenderness the next day in these

participants; furthermore, these symptoms mimicked the symptoms of fibromyalgia (Moldofsky

& Scarisbrick, 1976). A more recent, and very carefully conducted, study by Onen and

colleagues (2001a) reported that healthy males (PSQI and BMI in normal range) who had

undergone 40 hours of sleep deprivation showed hyperalgesia to mechanical stimuli, but not to

thermal stimuli (Onen, Alloui, Gross, Eschallier, & Dubray, 2001). In addition, these

participants also showed a robust analgesic effect following selective slow wave recovery sleep

after undergoing slow wave sleep interruption. Additional studies have supported these findings

of hyperalgesia to painful stimuli following sleep deprivation in healthy subjects, with no

changes noted in somatosensory threshold detection levels (Kundermann, Spernal, Huber, Krieg,

& Lautenbacher, 2004).

Experimental studies examining the effects of painful stimuli on sleep have consistently

shown that pain causes microarousal, which was measured by increased high frequency EEG









activity (in the alpha and beta ranges) at the expense of slow frequency EEG activity (delta

range) (Drewes, Nielsen, Arendt-Nielsen, Birket-Smith, & Hansen, 1997). These results have

led researchers to conclude that pain can cause changes to the sleep architecture of normal

controls in ways that lighten sleep and diminish the reputed restorative effects of slow wave

sleep. However, results have not been consistent and no significant effects on either EMG

activity (measuring bruxism) or on pressure pain thresholds were noted in a group of 10 healthy

males undergoing slow-wave sleep deprivation (Arima et al., 2001).

Sleep Disturbances in Pain Patients

Sleep disturbances are among the most common complaints reported by patients

experiencing chronically painful conditions (M. T. Smith & Haythornthwaite, 2004). "Sleep

disturbance" is a term used to denote subjective reports of problems in either sleep quality or

quantity (Cohen, Menefee, Doghramji, Anderson, & Frank, 2000). Studies commonly report that

at least 50% of patients with diverse chronic pain conditions complain of significant sleep

disturbance, with many studies reporting an even higher prevalence of sleep disturbance in

chronic pain patients, along the order of 70% (Atkinson, Ancoli-Israel, Slater, Garfin, & Gillin,

1988; Pilowsky, Crettenden, & Townley, 1985). Sleep efficiency (total sleep time/time in bed *

100) has also been noted to be significantly affected in chronic pain patients (Wittig, Zorick,

Blumer, Heilbronn, & Roth, 1982). The presence of concomitant sleep problems can

significantly complicate both the course and the management of chronic pain patients (Cohen et

al., 2000). Experimental data gathered from studies of healthy participants and cross-sectional

research in clinical populations suggests that the relationship between sleep disturbance and pain

might be reciprocal. In some cases, disturbed or poor sleep appears to contribute to the pain

problem, whereas in others ongoing pain diminishes individuals' ability to sleep and leads to a

cycle of increasing pain and continued degradation of sleep quality (Bailey, 1997). In other









words, pain may disrupt sleep continuity and/or sleep quality, and poor sleep may contribute to

the exacerbation of pain in these patients.

Additional support for the hypothesis that sleep disturbance contributes to chronic pain

comes from clinical studies, which have demonstrated that chronic pain patients often

demonstrate reduced delta (i.e. slow wave) sleep and/or increased alpha sleep (Harding, 1998;

Moldofsky, Lue, & Smythe, 1983). The abnormalities found in these studies appear to be similar

to the experimental studies in healthy participants, where sleep disruption appears to alter pain

sensitivity. Similarly, correlational studies involving clinical samples have consistently found a

positive association between sleep disturbance and pain severity (Pilowsky et al., 1985).

Back Pain

Studies in samples of chronic back pain patients have demonstrated high rates of sleep

problems, including difficulties with both the initiation of sleep and difficulties with sleep

maintenance (Lobbezoo, Visscher, & Naeije, 2004; Widerstrom-Noga, Felipe-Cuervo, &

Yezierski, 2001). Interpretation of these results is made difficult by the fact that there was no

indication of whether these groups were composed of different individuals, or whether the

groups contained the same participants presenting with both types of sleep problems. The

patients who reported frequent interference (3 or more times per week) in falling asleep due to

pain also indicated higher average pain intensity ratings, and used more descriptors when

describing their pain. Other studies involving patients with chronic back pain have combined

difficulties falling asleep and difficulties maintaining sleep into an overarching description

termed "insomnia", which prevents examination of patients' specific sleep problems (Lobbezoo

et al., 2004). Atkinson and colleagues (1988) also examined potential influential factors

involved in sleep disturbances among a sample of chronic low back pain patients. Results

indicated that sleep dissatisfaction was most strongly associated with greater depressed mood









and shorter duration of pain. Furthermore, patients who reported high pain intensity also

reported shorter TST, longer sleep latency, and more frequent awakenings, compared to the low

pain intensity patients. Wilson and colleagues (1998) also noted that patients with chronic

musculoskeletal pain (82.5% back pain) who had high pain severity reported greater sleep

impairment than those patients with low pain severity.

Facial Pain

Studies have documented sleep disturbances in patients with facial pain (Riley et al.,

2001; Yatani, Studts, Cordova, Carlson, & Okeson, 2002), although the exact nature of these

disturbances has not been well-defined. Riley and colleagues (2001) conducted both cross-

sectional and longitudinal analyses to examine the relationships between pain, sleep disturbance,

and depression in a sample of orofacial pain patients. Results indicated that reduced amount of

sleep (sleep quantity) was associated with depression and pain, and reduced sleep quality was

associated with negative affect. Additionally, when longitudinal analyses were conducted, initial

depression and pain predicted sleep disturbance at follow-up, and initial pain also predicted

negative affect at follow-up. However, in this study, sleep at time one did not predict pain at

follow-up.

As reported by Bailey (1997), a variety of headache disorders have been found to be

related to sleep disorders in different ways, which can be useful in both the diagnosis and

treatment of these disorders. In fact, the International Classification of Sleep Disorders (1990)

(ICSD; (Diagnostic Classification Steering Committee, 1990) includes a broad category of

classification entitled Sleep-Related Headaches under neurologic disorders, and this group of

headaches is defined as occurring during sleep with their onset most often during REM sleep.

The relationship between sleep disturbance and headache conditions is complex and difficult to

assess, with symptoms of both conditions potentially having causal relations or having mutual









reinforcements in an individual (Paiva, Batista, Martins, & Martins, 1995); see (Sahota &

Dexter, 1990) for a detailed review). The onset of certain types of headache conditions, such as

chronic paroxysmal hemicrania (CPH), cluster headache, and migraine headache, are known to

be associated with specific sleep stages (Sahota & Dexter, 1990). There are also well-established

effects of sleep disruptions (Blau, 1990; Headache Classification Committee of the International

Headache Society, 1988), as well as sleep disturbances (such as somnambulism, see (Paiva,

Martins, Batista, Esperanca, & Martins, 1994), on headache conditions. In fact, the presence of

early awakening or morning headaches has been conceptualized as possibly suggesting the

presence of a sleep disturbance in some patients (Paiva et al., 1995). Additionally, patients with

chronic headache complaints also indicate long-standing experiences of subjective sleep

problems (Paiva, Esperanca, Martins, Batista, & Martins, 1992). However, complicating the

picture is the fact that, for some individuals, sleep can also serve as an effective treatment for

some headaches, such as migraine (Blau, 1982).

A review by Bailey (1997) identified the nature and prevalence of sleep disorders that are

most often associated with orofacial pain conditions, and indicated that the presence of a

concurrent sleep disorder in patients with a pain condition necessitates the treatment of both

conditions. This review also reported that alpha intrusion on the EEG sleep recordings is one of

the most consistent findings displayed by chronic pain patients. This alpha rhythm is seen in a

condition of relaxed wakefulness, but normally disappears when an individual moves into stage 1

NREM sleep (Bailey, 1997). Alpha waves are also reportedly seen during arousals, which are

frequently found in individuals with pain conditions, and a condition termed alpha-delta sleep,

which is associated with subjectively nonrestorative sleep and feelings of fatigue, is also

reportedly found among pain patients (Bailey, 1997). While fibromyalgia has been strongly









linked with this phenomenon of alpha intrusion, and is reportedly the most common pain

condition associated with subjectively nonrestorative sleep, many researchers are noting

similarities between fibromyalgia and myofascial pain. Additionally, bruxism is commonly

found in patients with TMD, and because of the wide range of symptoms that can be produced

by TMD, sleep disturbances are often found in these patients (Bailey, 1997).

In a study by Paiva and colleagues (1995), 13 of the 25 patients from an outpatient

headache clinic who reported morning or nocturnal headaches received a change in their

diagnosis after PSG data were obtained. In 10 patients, the onset (or period of worsening) of

their headaches coincided with the onset of the sleep disturbances. However, the majority of

cases in this study reflected a complex association between patients' sleep disturbances and

headache conditions. A subsequent study by Paiva and colleagues (1997) examined the sleep of

patients with headache complaints who identified the onset of their headaches as occurring

during the night or early morning at least 75% of time (Paiva, Farinha, Martins, Batista, &

Guilleminault, 1997). In this sample, 53% of patients were identified as having a primary sleep

disorder, and all of these patients reported fragmented sleep. There were no significant

differences in total sleep time or percentage of REM sleep, as measured by polysomnography,

found between the headache group with a sleep disorder and the headache group without sleep

disorders. However, on a self-report sleep questionnaire, the patients who were identified as

having a sleep disorder reported a greater number of sleep complaints compared to the group of

headache patients who did not evidence any objective evidence of a sleep disorder.

Fibromyalgia and Rheumatic Conditions

Subjective sleep complaints, such as nonrestorative sleep and insomnia complaints,

among fibromyalgia patients have been reported by numerous studies (Campbell, Clark, Tindall,

Forehand, & Bennett, 1983; Drewes et al., 1995; Schaefer, 1995). Further studies have reported









links between sleep difficulties and pain in fibromyalgia, such that reports of restful sleep have

been associated with less reported discomfort and fatigue (Moldofsky, 1989) and nonrestorative

sleep is associated with exacerbation of pain in fibromyalgia patients (Affleck, Urrows, Tennen,

Higgins, & Abeles, 1996). Polysomnographic studies have also found specific differences

between the sleep architecture of fibromyalgia patients and that of healthy controls, including

polysomnographic evidence of increased sleep onset latency (Home & Shackell, 1991),

increased amounts of stage 1 sleep (Home & Shackell, 1991; Shaver et al., 1997), reduced

amounts of stage 3 and 4 sleep (Branco, Atalaia, & Paiva, 1994; Home & Shackell, 1991;

Shapiro, Devins, & Hussain, 1993), and increased number of arousals (Branco et al., 1994;

Jennum, Drewes, Andreasen, & Nielsen, 1993; Shapiro et al., 1993) in fibromyalgia patients.

Shapiro and colleagues (1993) also reported that fibromyalgia patients evidenced lower amounts

of age-corrected REM sleep, and total sleep time, and long awakenings (>10 minutes), and an

EEG pattern of intrusive alpha frequency waves, compared to healthy controls. As has been

demonstrated in previous studies, patients with fibromyalgia frequently exhibit an alpha-delta

sleep rhythm, which is also produced during stage 4 sleep deprivation, and by deep pain induced

during sleep, in normal control subjects (Harding, 1998). The alpha wave anomaly has long

been hypothesized to be involved in the pathophysiology of fibromyalgia (Moldofsky,

Scarisbrick, England, & Smythe, 1975), and more recent data has supported the hypothesis that

alpha intrusion is an inherent characteristic of NREM sleep in fibromyalgia patients (Branco et

al., 1994; Smythe, 1995). This alpha intrusion sleep anomaly is associated with indications of

vigilance during sleep and reports of nonrestorative sleep (Anch, Lue, MacLean, & Moldofsky,

1991), as well as pain, energy, and mood in fibromyalgia patients (Moldofsky & Lue, 1980).

The amount of alpha frequency that occurred during sleep has also been shown to correlate with









increases in overnight pain measures (Moldofsky & Lue, 1980). Furthermore, Harding (1998)

concluded that the compilation of evidence suggests this alpha intrusion is found in the majority

of patients with fibromyalgia, the amount of alpha intrusion correlates with objective

measurements of pain, and decreasing the amount of alpha intrusion results in improvements in

pain.

Landis and colleagues (2003) found that the women with fibromyalgia reported poorer

sleep quality and more fatigue than controls, although actigraphy sleep indicators were not

different between groups (Landis et al., 2003). In the women with fibromyalgia only, self-

reported sleep quality was directly related to actigraphy indicators of TST, and was inversely

related to sleep fragmentation. Additionally, fatigue in the women with fibromyalgia was

directly related to the actigraphy indicators of wake after sleep onset, and inversely related to

sleep efficiency.

According to Wolfe and colleagues (1990), nonrestorative sleep is a prevalent complaint

among patients with fibromyalgia (Wolfe et al., 1990). Harding (1998) reviewed the literature

regarding sleep in fibromyalgia, and further stated that, although sleep disturbance is a prominent

aspect in the clinical picture of fibromyalgia and that pain in fibromyalgia may increase due to a

lack of sleep, it is still unclear whether sleep disturbance plays a causal role in fibromyalgia, or is

simply an outcome resulting from the disorder.

Neurobiology Findings on the Sleep-Pain Relationship

Studies have implicated a number of areas in the central nervous system, as well as

several different chemical substances, in both the control/disturbance of the sleep/wake cycle and

in the experience of chronic pain. The mesencephalic periaqueductal gray area, the thalamus,

and the reticular nucleus of the thalamus have been implicated in the generation and maintenance

of sleep, and also in pain modulation. The mesencephalic periaqueductal gray area has been









shown to modulate both sleep states (Sastre, Buda, Kitahama, & Jouvet, 1996) and nociception

(Demarco, Baghdoyan, & Lydic, 2003). The thalamus is involved in both arousal and in the

processing of nociceptive stimuli in the cortex (Casey, Morrow, Lorenz, & Minoshima, 2001),

and the reticular nucleus of the thalamus is thought to actively regulate the synchronization of

the cortex during delta sleep (Steriade & Llinas, 1988). A study by Mountz and colleagues

(1995) demonstrated that fibromyalgia patients had reduced regional blood flow to the thalamus

and caudate nucleus (Mountz et al., 1995), which may be involved in abnormalities of growth

hormone secretion that have been observed in patients with fibromyalgia (Culebras & Miller,

1984). Bennett (1993) reported low levels of somatomedin C, which is a growth hormone

responsible for muscle regeneration and homeostasis, in fibromyalgia patients, and further

indicated that growth hormone is primarily secreted during stage 4 NREM (Bennett, 1993). Foo

and Mason (2003) have argued that persistent pain, unlike acute pain, is associated with

functional changes in the raphe magnus (RM) cells, which modulate both pain and arousal (Foo

& Mason, 2003). Paulson and colleagues (2002) have also hypothesized that persistent pain may

lead to lasting functional changes in the neural systems that regulate both sleep and pain

(Paulson, Casey, & Morrow, 2002). Specifically, they present data that suggest that persistent

pain might promote changes in the ascending arousal system, which could ultimately lead to

disturbance of sleep continuity.

Moldofsky theorizes that the diffuse myalgia, fatigue, and psychological distress

experienced by patients with fibromyalgia are not only related to a disorder of their sleep-wake

system, but also to circadian alterations of associated biologic systems of the body. Specifically,

studies have shown that these systems include, neurotransmitters (e.g. serotonin, substance P),

neuroimmune and neuroendocrine (e.g. IL-1, NK cell activity, HPA and thyroid axes), and the









autonomic nervous systems that are altered in patients with fibromyalgia (McAlpine, 1987;

Moldofsky, 1994; Paulson et al., 2002; Pillemer, Bradley, Crofford, Moldofsky, & Chrousos,

1997). Several studies have reported that patients with fibromyalgia have decreased levels of

serotonin in their cerebrospinal fluid and blood (Russell et al., 1992), and there is evidence for an

inverse relationship between pain levels and serotonergic activity in the brain (Bailey, 1997).

Furthermore, Moldofsky (1989) reported strong evidence from animal studies demonstrating a

relationship between CNS metabolism of serotonin and its role in regulating both pain and

NREM sleep. Additionally, Carette and colleagues (1986) reported that low doses of tricyclic

antidepressants, which influence the metabolism of serotonin in the central nervous system, have

been found to be beneficial for sleep in patients with fibromyalgia (Carette, McCain, Bell, &

Fam, 1986). Taken together, findings suggest that low levels of serotonin in fibromyalgia

patients' central nervous systems may play a role in their decreased delta (slow wave) sleep, and

may predispose these patients to developing the alpha intrusion phenomenon. Unstable

serotonin levels have also been proposed as a common factor in both migraine and

somnambulism (Barabas, Ferrari, & Matthews, 1983), while Vaeroy and colleagues (1988)

reported finding elevated cerebrospinal fluid levels of substance P in patients with fibromyalgia

(Vaeroy, Helle, Forre, Kass, & Terenius, 1988). Moreover, sleep and pain are both associated

with activation of a number of regions in the central nervous system. Studies have suggested

that the supraspinal regions, thalamocortical pathways, or the anterior cingulated cortex may be

involved in the interaction between sleep and pain (Chase & Morales, 1994; Jones, 1994; Seigel,

1994).

Sleep Disturbances and Experimental Pain Testing

Onen and colleagues (2001b) reported that REM sleep deprivation in rats led to increased

behavioral responses to noxious thermal, mechanical, and electrical stimuli, but not to a noxious









chemical stimulus (Onen, Alloui, Jourdan, Eschalier, & Dubray, 2001). These authors

hypothesize that the normal duration of REM sleep may be important for anti-nociceptive

activity of endogenous and exogenous opioids. They further suggest that the nociceptive process

may be affected by REM sleep deprivation, producing a relative hypersensitivity to noxious

electrical stimuli. It is known that both REM sleep and nociception are modulated by the

cholinergic system. Additionally, as serotonin is involved in both REM sleep and in pain

mechanisms, these authors hypothesize that a possible serotonin depletion, due to increased

serotonin metabolism, may partly induce a hyperalgesia in REM sleep deprived animals.

A number of studies have demonstrated that fibromyalgia patients exhibit hyperalgesia in

response to various pain stimuli (see (Hurtig et al., 2001), Table 1), compared to healthy

controls. In addition, Agargun, et al. (1999) demonstrated a significant negative correlation

between pressure pain threshold and an overall measure of sleep quality, as well as measures of

subjective sleep quality, sleep efficiency, and sleep disturbance in 16 fibromyalgia patients

(Agargun et al., 1999).

Negative Mood, Sleep, and Pain

The associations between pain, depression, and sleep disturbance have been examined in

several chronic pain patient samples (Affleck et al., 1996; Atkinson et al., 1988; Morin, Gibson,

& Wade, 1998; Nicassio & Wallston, 1992). Some studies have reported that "poor" sleepers

reported greater pain, but do not differ from "good" sleepers on measures of depression or

anxiety (Moffitt, Kalucy, Kalucy, Baum, & Cooke, 1991; Morin et al., 1998). However, other

studies have reported that "poor sleepers" have higher scores on measures of depression and

anxiety, in addition to higher pain intensity and more physical disability, compared to "good

sleepers" (Atkinson et al., 1988; Pilowsky et al., 1985; Sayar, Arikan, & Yontem, 2002).









Longitudinal analyses have supported pain predicting sleep (Atkinson et al., 1988;

Nicassio & Wallston, 1992), although results for sleep predicting pain over time have been less

consistent (Nicassio & Wallston, 1992). Additionally, sleep problems and pain have been shown

to be predictive of depression over time in chronic pain populations (Nicassio & Wallston,

1992). Conversely, measures of depression have also been found to be predictive of sleep in

chronic pain samples (Sayar et al., 2002).

Chiu and colleagues (2005) examined the relationship between psychological factors,

sleep disturbances, and pressure pain threshold in a population study, stratifying the sample by

presence and extent of current pain (Chiu et al., 2005). Results indicated that those participants

who reported the greatest sleep disturbance and highest levels of depression had a 2-fold

increased chance of being in the lowest tertile for pain threshold. Importantly, these two

variables were found to be independently associated with lower pain threshold, and this

relationship remained even after adjusting for participants' initial pain status. Wittig and

colleagues (1982) examined polysomnographically-measured sleep patterns in a group of pain

patients, and compared these to findings in a group of patients with insomnia secondary to a

psychiatric disorder and a group of patients with subjective insomnia complaints, but no

objective findings of sleep disturbance. The pain patients were found to have more difficulty

initiating and maintaining sleep compared to the group with subjective insomnia complaints;

however, the group with insomnia secondary to a psychiatric disorder evidenced poorer sleep

efficiency and greater early morning awake time than the pain patients. Additionally, 8 of the 26

pain patients demonstrated alpha rhythm intrusion into NREM stages of sleep.

Current Study

Defining the relationships between particular pain conditions and sleep disturbance may

further clarify the shared pathophysiology of sleep and pain. There have not yet been any









comparative studies simultaneously carried out in patients with different types of chronic pain

conditions to determine whether there are physiological sleep differences among such patients.

Previous investigations have examined specific groups of chronic pain patients, or unspecified
"mixed" groups of patients, without examining whether any differences exist among patients

with different pain conditions. Although the mechanism is not well studied, pain sensations

could interrupt sleep via direct or indirect pathways. If specific relationships between chronic

pain conditions and sleep disturbance can be identified, this could clarify the shared

physiological underpinnings of sleep and pain, and treatments that target the specific sleep

disruptions in a particular chronic pain condition could be developed.

Furthermore, although several investigations have examined sleep patterns in chronic

pain patients, these investigations have used a variety of measures which limits the comparison

of results across various studies. Similarly, several studies have used measures that were

idiosyncratically generated and/or not validated, reducing confidence in the robustness of their

findings. Sleep problems appear to be a ubiquitous finding among samples of chronic pain

patients. In order to accurately and reliably delineate the nature of this relationship, as well as

the influence of negative mood on this relationship, it is essential to use valid measurement

instruments when assessing these patients. Previous studies have also indicated varying levels of

agreement between different measures of sleep, usually related to the variable being considered

(e.g., total sleep time vs. number of awakenings). This suggests that further information is

needed to determine the utility and adequacy of different measures of individuals' sleep patterns

and sleep problems.

Finally, previous investigations have indicated that individuals' subjective pain reports

and their response to experimentally-induced painful stimuli are not perfectly correlated. Both of









these variables are important to examine, as each represents an important target for intervention

in chronic pain populations. Additionally, differences among types of experimental pain

procedures have been noted. Thermal pain appears to be an appropriate means of measuring

pain sensitivity among chronic pain patients.

The present study addressed some of the limitations noted in previous research examining

the relationship between sleep disturbance and chronic pain. First, different groups of chronic

pain patients were recruited and completed the same procedures. This allows for comparison of

results across different pain groups. Second, the self-report instruments used to measure pain,

mood, and sleep are all valid and reliable instruments, which permits increased confidence in the

findings of this study. Furthermore, sleep was measured using various methodologies

(questionnaire, sleep diary, actigraphy) in a portion of the participants, which allows for

comparisons across these methods to determine their level of agreement, as well as their utility

for measuring different sleep variables within a chronic pain population. Similarly, pain was

measured using various methodologies, including self-report (VAS), questionnaires, and

psychophysical testing. This again permits examination of the relationships among these various

measures and provides a comprehensive picture of participants' pain experience.

It was hypothesized that the prevalence of various types of sleep disturbances would

differ between the different groups of chronic pain patients under study, underscoring the

importance of implementing targeted treatments for specific sleep problems in these populations

instead of assuming that a one-size-fits-all approach is appropriate. The current multimodal

approach to nonpharmacological treatment of sleep disturbances has been shown to be effective;

however, identifying the active treatment components in these multimodal packages will enable

more efficient and cost-effective treatment delivery, which can ultimately target the specific









sleep problem being experienced by patients. There has been some evidence to suggest the

differential effectiveness of specific treatment components for particular sleep problems, such as

difficulties with sleep onset versus problems with sleep maintenance (Espie, Brooks, & Lindsay,

1989; Harvey, 2000; Waters et al., 2003).

Further, examination of the reported sleep disturbances among these chronic pain

populations was undertaken to investigate interactions between the pain condition and the

specific sleep disturbance that is being reported, such that the pain condition may be maintaining

the sleep problem, the sleep problem may be exacerbating the pain conditions, or some

reciprocal influence may be occurring between the sleep problem and the pain condition. It was

also deemed important to note whether all 3 groups differed in the sleep disturbances that they

presented, or if there was one group (such as fibromyalgia patients) that was distinguishable

from the other two groups. If present, the emergence of this pattern of results might have lent

support for a greater role of the central nervous system in mediating both the painful symptom

presentation, as well as the particular type of sleep problem that is reported, by such a group of

patients. Alternatively, if the pattern of results suggested that all 3 groups were distinct from one

another, future studies should aim to identify what mechanisms may be underlying both the pain

condition and the associated sleep problems in each group of patients.

Along these lines, it was hypothesized that differences in pain sensitivity and sleep

patterns would be found among these groups of chronic pain patients, supporting the hypothesis

of different mechanisms underlying these painful conditions. For instance, conditions with

greater central nervous system involvement may involve both greater pain hypersensitivity and a

more pervasive type of sleep disturbance, such as non-restorative sleep. As experimental studies

have demonstrated hyperalgesia in healthy participants following sleep deprivation (Moldofsky









& Scarisbrick, 1976; Onen, Alloui, Gross et al., 2001), it was hypothesized that greater

hypersensitivity to experimental stimuli in pain patients in the present study would lend further

support to the hypothesis of a shared mechanism underlying both the processing of painful

stimuli as well as regulation of individuals' sleep.

As described above, the presence of sleep disturbance in chronic pain patients warrants

both evaluation and treatment in order to produce the most successful treatment outcome.

Improved understanding of the underlying mechanisms of various chronic pain conditions, as

well as their involvement in the sleep disturbances presented in these patients, may lead to

improved pain management and functional outcome among chronic pain populations.

Specific Aims

Specific aim 1: To measure subjective sleep quality in different populations of chronic
pain patients using validated measures, and to identify the types of sleep disturbances
present in each population.

Specific aim 2: To examine the relationship between sleep, pain, and negative affect in
order to better understand the relationships among these variables.

Specific aim 3: To determine if chronic pain patients who reported concurrent sleep
disturbances had greater sensitivity to painful stimuli compared to chronic pain patients
who did not report concurrent sleep disturbances.

Hypotheses

Hypothesis 1: Although it was expected that sleep disturbances would be reported by
many of the chronic pain patients in this study, it was hypothesized that there would be
an unequal distribution of specific types of sleep disturbances across groups.
Specifically, it was hypothesized that facial pain patients would report more difficulty
initiating sleep (greater sleep latency), back pain patients would report more difficulty
maintaining sleep (greater sleep disturbances), and fibromyalgia patients would report a
greater prevalence of non-restorative sleep (lower sleep quality), compared to the other
groups of pain patients.

Hypothesis 2: It was hypothesized that the relationship between sleep and pain in this
sample of chronic pain patients would be partially mediated, when the influence of
negative affect was included in the model.









* Hypothesis 3: It was predicted that those patients who reported poor sleep (as measured
by the PSQI) would show hypersensitivity (higher ratings of pain intensity) during the
ramp and hold pain testing procedures, and greater temporal summation compared to
those chronic pain patients who reported good sleep (as measured by the PSQI).

* Hypothesis 4: It was predicted that participants who reported sleep disturbances would
evidence longer sleep latency, more wake time after sleep onset, and lower sleep
efficiency, as measured by both actigraphy and sleep diary measures, compared to
participants who reported no sleep disturbances.

* Hypothesis 5: It was hypothesized that participants who did not report sleep disturbances
would demonstrate higher correlations between actigraphy and sleep diary variables than
participants who reported sleep disturbances.









CHAPTER
METHODS

Participants

Participants in this study included 292 individuals with a current chronic pain condition;

116 patients with chronic facial pain, 55 patients with chronic back pain, and 121 patients with

fibromyalgia. All participants were between 18 and 65 years of age (M=46.67, SD=12.05). The

sample included 51 males and 241 females; and of those participants who indicated ethnicity,

88.2% were Caucasian. Additionally, a subgroup of participants was recruited that included 22

females (10 good sleepers, 12 poor sleepers) with a current chronic pain condition: 8 with

chronic facial pain, 8 with chronic back pain, and 6 with fibromyalgia. Subgroup participants

were designated as good or poor sleepers based on their total score from the Pittsburgh Sleep

Quality Index (PSQI), using the modified cutoff score suggested by Carpenter and Andrykowski

(1998). The average age of the subgroup participants was 43.77 years (SD=14.13 years), and

81.8% of subgroup participants identified themselves as Caucasian.

Participants were recruited from three groups of chronic pain patients at the University of

Florida. Patients with chronic back pain were recruited from the Spine Care Center, patients

with chronic facial pain were recruited from the Facial Pain Clinic, and patients with

fibromyalgia were recruited from the Fibromyalgia Clinic. Participants who completed the

subgroup procedures were recruited directly from the pain clinics described above, as well as

through printed advertisements posted on the University of Florida campus.

Power analyses were conducted to determine the number of participants needed to detect

an effect based on findings from previous studies using the PSQI with similar populations of

pain patients. Specifically, comparisons were made based on PSQI global scores reported by

Agargun and colleagues (1999) for fibromyalgia patients, Yatani and colleagues (2002) for









patients with temporomandibular disorder, and Menefee and colleagues (2000) for patients with

back pain (Agargun et al., 1999; Menefee et al., 2000; Yatani et al., 2002). Based on these

analyses, effect sizes of 0.53 and 0.62, were obtained in comparisons of PSQI global scores

between these groups of chronic pain patients. Therefore, with power set at 0.80, and an alpha

value of 0.05, it was determined that approximately 40 participants would be needed in each

group of chronic pain patients in order to detect similar effects. Therefore, recruitment efforts

attempted to secure at least 40 participants from each of the chronic pain clinics to ensure

adequate power in this study.

Procedure

Patients provided demographic information and information related to their health and

pain condition. This included the participant's age, sex, ethnicity, duration of pain, education,

current medications, and pain ratings. In addition, participants completed a standard

questionnaire packet as part of their clinical assessment, including measures of pain, negative

affect, coping strategies, somatic focus, and disability. These questionnaires are described

below. The first pain measure was either the McGill Pain Questionnaire (MPQ; (Melzack,

1975), a self-report questionnaire that assesses the sensory, affective, and evaluative dimensions

of the pain experience, or the Medical College of Virginia Pain Questionnaire (MCV; (Price &

Bushnell, 1994), which asks patients to provide visual analogue scale (VAS) ratings on pain

intensity, pain unpleasantness, and function dimensions. Participants also completed the Coping

Strategies Questionnaire Revised (CSQ-R; (Riley & Robinson, 1997), a self-report instrument

measuring pain coping strategies. The third pain measure that was completed by participants

was the Pain Disability Index (PDI; (Pollard, 1984), a brief self-report measure of the degree to

which pain interferes in seven life areas. The affective measures that participants completed

included the Beck Depression Inventory (BDI; (Beck, Ward, Mendelson, Mock, & Erbaugh,









1961), which assesses the experience of cognitive and affective, and neurovegetative symptoms

of depression during the past week; the State-Trait Anger Expression Inventory (STAXI;

(Spielberger, 1988), which is used to assess both state anger symptoms and more general

constitutional anger symptoms; the Pain-Anxiety Symptom Scale (PASS; (McCracken, Zayfert,

& Gross, 1992), which provides an assessment of pain-related anxiety; the Medical College of

Virginia Pain Questionnaire (MCV; (Price & Bushnell, 1994), which asks patients to provide

visual analogue scale (VAS) ratings on mood, in addition to pain and function dimensions; and

the Pennebaker Inventory of Limbic Languidness (PILL; (Pennebaker, 1982), a self-report

symptom frequency checklist that assesses nonspecific common physical complaints. In addition

to this standard questionnaire packet, participants also completed the Pittsburgh Sleep Quality

Index (PSQI; (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989), a self-report measure of

sleep quality and disturbances over a one-month interval.

A subset of 22 participants (12 reporting sleep disturbances and 10 reporting no sleep

disturbances) also underwent psychophysical testing to assess their pain sensitivity. Subgroup

participants were Subjects underwent quantitative sensory testing using a contact thermode

applied to the volar surface of the forearm. The protocol that was used enabled assessment of

both first pain (primarily A-delta function) and second pain (primarily C-fiber input). All thermal

stimuli were delivered using a computer-controlled Medoc Thermal Sensory Analyzer (TSA-

2001, Ramat Yishai, Israel), which is a peltier-element-based stimulator. Temperature levels

were monitored by a contactor-contained thermistor, and returned to a preset baseline of 32 deg

C by active cooling at a rate of 10 deg C/Sec. All stimuli were delivered to the ventral left

forearm. Alternating stimulation sites were used to prevent carryover effects due to local

sensitization.









Additionally, this subset of 22 participants also completed daily sleep diaries and

actigraphy monitoring for two weeks. The actigraphy monitoring was used to provide an

objective measure of these individuals' sleep patterns. The sleep diaries were completed twice

per day (before participants go to bed, and again in the morning), and provided an additional

measure of participants' self-reported sleep patterns. Comparison of these participants' reports

provided an indication of the degree to which these measures converge.

Measures

Demographic/patient characteristics information: Patients provided information

pertaining to their sex, age, ethnicity, duration of pain, education, current medications, and pain

ratings.

Pittsburgh Sleep Quality Index (PSQI; (Buysse et al., 1989): The PSQI is a self-report

questionnaire that assesses sleep quality and disturbances over a 1-month time interval, and is

designed to be used in clinical populations. This instrument is comprised of 19 items, which

generate 7 "component" scores: subjective sleep quality, sleep latency, sleep duration, habitual

sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction.

Additionally, the sum of the scores for these 7 components yields one global score. The

psychometric properties of this measure have been shown to be sound, suggesting its utility in

both clinical practice and research. This measure has been shown to have acceptable internal

consistency (Cronbach's alpha = .83; (Buysse et al., 1989), and these authors further

demonstrated that this measure has a diagnostic sensitivity of 85.5% and specificity of 86.5% in

distinguishing good and poor sleepers. The psychometrics of the PSQI were subsequently

evaluated by Carpenter and Andrykowski (1998) in 4 different patient populations, including

bone marrow transplant patients, renal transplant patients, women with breast cancer, and

women with benign breast problems. Results indicated that the PSQI had good internal









consistency (Cronbach's alpha coefficients = 0.80, across groups) and construct validity, and

PSQI scores were found to be moderately to highly correlated with measures of sleep quality and

sleep problems in these patients (Carpenter & Andrykowski, 1998).

McGill Pain Questionnaire (MPQ; (Melzack, 1975): The MPQ is a self-report

questionnaire of participants' pain experience. This instrument provides both an overall total

pain score, as well as evaluations of the sensory, affective, and evaluative dimensions of

participants' pain experiences. The McGill Pain Questionnaire (MPQ) has a long history of use

in pain research, and it is the most widely used instrument for evaluating pain (Melzack & Katz,

1992). The reliability of the MPQ was investigated by Love and colleagues (1989), and results

demonstrated very strong test-retest reliability coefficients of this measure in a group of low back

pain patients who were tested on two occasions (Love, Leboeuf, & Crisp, 1989). Additionally,

several studies have replicated the factor structure of the MPQ (Lowe, Walker, & McCallum,

1991; Turk, Rudy, & Salovey, 1985), and results from a study by Pearce and Morley (1989) also

demonstrated the construct validity of this measure (Pearce & Morley, 1989).

Medical College of Virginia Pain Questionnaire (MCV; (Price & Bushnell, 1994):

The MCV questionnaire asks patients to provide visual analogue scale (VAS) ratings on pain,

mood and function dimensions. These dimensions include measures of the pain experience

itself, specifically pain intensity and pain unpleasantness, rated in reference to current levels, as

well as highest, lowest, and usual levels during the preceding week. In addition, negative

feelings associated with the pain experience (i.e., depression, anxiety, frustration, fear, and

anger) are also rated, in reference to the previous week. Ratings were also provided regarding

the extent to which pain prevented an individual from doing what he/she wanted to do, how









difficult an individual found it to endure pain over time, and how concerned an individual is with

his/her health.

Beck Depression Inventory (BDI; (Beck et al., 1961): The BDI is a self-report measure

of depression consisting of 21 items, which assess cognitive and affective, and neurovegetative

symptoms of depression. This measure is designed to determine the extent to which individuals

currently exhibit or experience each of these symptoms. Participants are instructed to indicate the

statement in each item group that is most descriptive of how they have been feeling during the

past week, including the current day. Each item is scored on a scale that ranges from 0 to 3. The

use of the BDI has been evaluated in psychiatric and nonpsychiatric populations, and alpha

coefficients ranging from .73 to .95 have been reported (Beck, Steer, & Garbin, 1988). The BDI

is a well-validated assessment instrument for depression and it is an extensively used measure in

experimental pain research.

Pain Anxiety Symptom Scale (PASS; (McCracken et al., 1992): The PASS is a self-

report questionnaire consisting of 40 items assessing four dimensions of pain-related anxiety-

cognitive anxiety, escape/avoidance, fearful appraisal, and physiological anxiety (McCracken, et

al., 1992). Participants indicate to what extent the items are an accurate description of them on a

6-point scale, which ranges from never (0) to always (5). There are five items that are reverse

scored. Previous studies have examined the reliability and validity of this measure (McCracken

& Dhingra, 2002; Roelofs et al., 2004), and results have shown the PASS to be psychometrically

sound. Cronbach's alpha coefficients of 0.94 have been reported in various samples of chronic

pain patients (fibromyalgia, low back pain) (Roelofs et al., 2004).

State-Trait Anger Expression Inventory (STAXI; (Spielberger, 1988): The STAXI is

used to assess both state anger symptoms and more general trait-like or constitutional anger









symptoms. The factor structure of the STAXI has been supported in analyses with a large

sample of college students (D. G. Forgays, Forgays, & Spielberger, 1997) and a sample of

middle-aged men and women (D. K. Forgays, Spielberger, Ottaway, & Forgays, 1998). The

measurement properties of this assessment instrument have been shown to be acceptable,

including good reliability and adequate validity (Kramer & Conoley, 1992).

Pennebaker Inventory of Limbic Languidness (PILL; (Pennebaker, 1982): The PILL

is a self-report measure of the occurrence and frequency of non-specific common physical

complaints, and is used as a measure of somatic focus. It consists of 54 items, such as upset

stomach, sore throat, headache, and nausea. The following response categories are provided:

"have never or almost never experienced the symptom", "less than 3 or 4 times per year", "every

month or so", "every week or so", and "more than once every week", and responses are indicated

by a five-point Likert scale. This measure evaluates commonly experienced symptoms over an

unspecified time period in the past and assesses a general tendency to experience and report

symptoms instead of the person's specific symptom experience (Gijsbers van Wijk, van Vliet, &

Kolk, 1996). Therefore, the PILL is conceptualized as a trait-like symptom scale that evaluates

somatization or a general propensity to report physical symptoms (Pennebaker, 1982). When

used with healthy subjects, a high score is indicative of somatization. Internal consistency for

this measure is high (Cronbach's alpha = 0.91) (Gijsbers van Wijk et al., 1996). The PILL also

has sufficient test-retest reliability (r = 0.83) and was shown to correlate moderately with similar

symptom scales (Pennebaker, 1982).

Coping Strategies Questionnaire-Revised (CSQ-R; (Riley & Robinson, 1997): The

CSQ-R is a reformulation of the original CSQ (Rosenstiel & Keefe, 1983), which was a

rationally constructed instrument designed to assess pain coping and was formulated to measure









the extent to which patients used six different cognitive coping strategies and two behavioral

coping strategies. The CSQ-R retains 27 of the original 48 items of the CSQ, and proposes a 6-

factor solution. The items that were discarded from the original CSQ did not appear to possess

good factor discrimination across several studies examining the factor structure of this measure

(Riley & Robinson, 1997; Swartzman, Gwadry, Shapiro, & Teasell, 1994; Tuttle, Shutty, &

DeGood, 1991). The six-factor solution for the CSQ-R has been replicated, both in samples of

chronic pain patients (Riley & Robinson, 1997; Robinson et al., 1997) as well as in an ethnically

diverse sample of healthy individuals (Hastie, Riley, & Fillingim, 2004). Additionally, this

measure was found to have acceptable internal consistency (Cronbach's alpha = 0.72 to 0.91)

across ethnic groups (Hastie et al., 2004).

Pain Disability Inventory (PDI; (Pollard, 1984): The PDI is a 7-item measure of the

degree to which chronic pain interferes with patients' functioning in the following areas of life:

family/home responsibilities, recreation, social activity, occupation, sexual behavior, self care,

and life-support activity (Pollard, 1984). An 11-point scale ranging from 0 (no disability) to 10

(total disability) is used to indicate the amount of disability experienced in each of the domains

listed above. The seven ratings are summed to compute a total score (0 70). The PDI has

adequate psychometric properties with an internal consistency coefficient of .86 (Tait, Pollard,

Margolis, Duckro, & Krause, 1987). Additionally, results reported by Tait, Chibnall, and Krause

(1990) demonstrated the construct validity of the PDI in a large sample of chronic pain patients,

and also indicated that this measure had adequate test-retest reliability with a smaller group of

pain patients undergoing inpatient treatment (Tait, Chibnall, & Krause, 1990).

Graded Thermal Stimulation or RAMP and HOLD (RH)

All thermal stimuli were delivered using a computer-controlled Medoc Thermal Sensory

Analyzer (TSA-2001, Ramat Yishai, Israel), which is a peltier-element-based stimulator. The









temperature of the probe was calibrated immediately prior to each session. Visual Analog Scale

(VAS) ratings of 4 graded intensities (45, 47, 49, 510 C) of 3 second temperature stimuli were

obtained in the following fashion. Stimuli were applied in random order to the forearm by a

contact thermode and were 3 seconds in duration. Several sites located on the forearms of both

arms were employed. Stimulus presentation was timed such that no site was stimulated with less

than a 3-minute interval to avoid sensitization of the site. Participants rated 8 stimuli (2 at each

intensity) using a VAS for pain intensity anchored at the right end by "the most intense pain

imaginable." A second random sequence of 8 stimuli (2 at each intensity) was rated by VAS for

pain unpleasantness (anchored at the right end by "the most unpleasant sensation imaginable.")

This method of pain assessment has been shown to yield ratio scale measurement of clinical pain

that is both internally consistent and provides independent sensory intensity and affective

dimensions of experimentally induced pain (Price, Harkins, & Baker, 1987).

Temporal Summation (Wind-up)

Another method of eliciting second pain was employed that mimics natural conditions of

nociceptive thermal stimulation such as when one touches a hot object. Trains of 8 stimuli with

an inter-stimulus interval of 3 seconds were used. The stimuli were pulsed from a baseline

temperature of 45 C to 52 C. When rating sensory magnitude, the participants were instructed

to attend to the peak of late sensations that occur approximately 1.5 to 2 seconds after the probe

leaves the skin on each presentation. This type of stimulus presentation results in a temporal

summation believed to be primarily C-fiber determined.

Sleep Diary and Actigraphy

Although actigraphy does not correlate perfectly with polysomnographic measurement of

sleep, the use of actigraphy provided an objective measure of certain sleep parameters, which

enhanced the methodology of the present research study. As described previously, sleep can be









measured in various ways, and current recommendations involve using a multi-modal assessment

of sleep patterns (meaning a combination of subjective, behavioral, and physiological measures

of sleep). Actigraphy data provided an objective measure of participants' sleep without the

burden inherent in polysomnography, while the sleep diary captured participants' subjective

perception of certain sleep parameters. Both types of information are important and can provide

useful information during both assessment and treatment activities. Furthermore, assessment of

the degree of agreement between these measures can illustrate the parameters for which these

methodologies overlap, and the parameters about which each provided unique, and potentially

important, information.

Actigraphy assessment used a high-sensitivity algorithm. Variables obtained from

actigraphy included: time in bed, total sleep time (TST), wake after sleep onset (WASO),

number of awakenings, sleep efficiency, and sleep onset latency.

For the sleep diary, participants were asked to complete the diary before they went to bed

at night, and then again the following morning. The variables that were included in the morning

sleep diary are listed below:

Time participant got into bed
Approximate time participant fell asleep (SOL)
Wake time
Time participant got out of bed
Number and duration of any awakenings during the night (number of awakenings and
WASO)
Rating of overall sleep quality using a visual analogue scale
Participants were also asked to list any other factors that interfered with their sleep,
such as pain or worries
Visual analogue scale (VAS) for pain intensity and pain unpleasantness upon waking

The portion of the sleep diary that participants completed in the evening (i.e., before bed)

contained a different set of variables, including:

Number and amount of caffeinated beverages consumed during the day









Number and amount of alcoholic beverages consumed during the day
Any medications taken during the day
Amount and type of physical activity
Number and duration of naps taken during the daytime or early evening
Visual analogue scale (VAS) for pain intensity and pain unpleasantness at bedtime

Statistical Analyses

Analysis 1: To determine if there were differences in sleep quality variables across type

of chronic pain group, Kruskal-Wallis nonparametric tests for 3 independent samples were

conducted for the following component scores from the PSQI across each of the three pain

groups: sleep quality, sleep latency, and sleep disturbances. This nonparametric test was used

due to the ordinal scale of the dependent variables in these analyses.

Analysis 2: Structural Equation Modeling (SEM) was used to examine the role of

negative affect in the relationship between sleep and pain. Mediation was indicated if there was

a lower or non-significant path coefficient between the latent factor of sleep and the latent factor

of pain, after the latent factor of negative affect was entered into the model.

Analysis 3: A General Linear Model approach was used to conduct a mixed model

analysis of variance to assess group (good sleeper vs. poor sleeper), heat pulse number, and

group by heat pulse number interactions on pain report in the temporal summation/windup

protocol.

Analysis 4: A General Linear Model approach was used to conduct a mixed model

analysis of variance to assess group (good sleeper vs. poor sleeper), temperature level, and group

by temperature level interactions on pain report in the ramp and hold procedure.

Analysis 5: A General Linear Model approach was used to conduct a mixed model

analysis of variance to assess chronic pain group, heat pulse number, and chronic pain group by

heat pulse number interactions on pain report in the temporal summation/windup protocol.









Analysis 6: A General Linear Model approach was used to conduct a mixed model

analysis of variance to assess chronic pain group, heat pulse number, and chronic pain group by

heat pulse number interactions on pain report in the ramp and hold procedure

Analysis 7: To determine if there were differences in actigraphy and sleep diary variables

for participants with and without complaints of sleep disturbances, 2 multivariate analyses of

variance (MANOVA) were conducted. One MANOVA examined SOL, WASO, sleep

efficiency, as measured by actigraphy, between participants who did and did not report sleep

disturbances. The second MANOVA examined SOL, WASO, sleep efficiency, as measured by

sleep diary, between participants who did and did not report sleep disturbances.

Analysis 8: Pearson correlations were computed for SOL, WASO, TST, and sleep

efficiency, as measured by actigraphy and sleep diary, for participants who did and did not report

sleep disturbances. Differences between groups were then tested using z-tests.









CHAPTER 3
RESULTS

Sample characteristics were examined to determine whether there were any significant

demographic differences among the three groups of chronic pain patients. Significant

differences were found across groups in years of education, (F(2,157)=5.57, p<.01), with post

hoc analyses indicating that facial pain (FP) patients had significantly more years of education

than back pain (BP) patients. Significant differences in sex (2"(2)=36.96, p<.001) and use of

narcotic medications (y2(2)=37.35, p<.001) were also found across the three chronic pain groups.

No other significant differences in sample characteristics were found. Demographic

characteristics of the sample and results of these analyses are provided in Table 3-1.

Differences in Subjective Sleep Across the Three Pain Groups

To examine whether there were differences in subjective sleep quality variables across

the three chronic pain groups, Kruskal-Wallis nonparametric tests for 3 independent samples

were performed. The dependent variables in this analysis were the sleep quality, sleep latency,

and sleep disturbance component scores from the PSQI; the Kruskal-Wallis nonparametric test

was used due to the ordinal scale of these variables. Results revealed significant differences

among the three groups on PSQI sleep quality, X2(2) = 33.74, p<.001; PSQI sleep latency, X2(2)

= 16.63, p<.001; and PSQI sleep disturbances, X2(2) = 44.28,p<.001. For each of the three PSQI

components, post hoc analyses were conducted using Mann-Whitney U tests in order to

determine which groups were significantly different from one another. For sleep quality

component scores, facial pain patients had significantly lower scores (indicating better sleep

quality) compared to back pain patients, z = 3.89, p<.001; and fibromyalgia patients, z = 5.53,

p<.001. For sleep latency component scores, facial pain patients had significantly lower scores

(indicating shorter sleep latency) compared to back pain patients, z = 2.87, p<.01; and









fibromyalgia patients, z = 3.83, p<.001. For sleep disturbances component scores, facial pain

patients had significantly lower scores (indicating fewer sleep disturbances) compared to back

pain patients, z = 3.68, p<.001; and fibromyalgia patients, z = 6.33, p<.001; there was also a

trend for back pain patients to have lower scores compared to fibromyalgia patients, z = 1.73,

p<.10. These results are shown in Figure 3-1. These findings are consistent with results from a

one-way ANOVA comparing PSQI global scores for each of the three pain groups, F(2,262) =

25.86, p<.001, rf = 0.17; post hoc Tukey tests indicated the same pattern of results

(FP
Sleep onset latency (SOL), wake after sleep onset (WASO), and sleep efficiency (SE), as

measured by diary and actigraphy, and average sleep quality VAS ratings from the diaries, were

also compared across subgroup participants from the three pain groups. The omnibus test

revealed no significant differences across group. However, when mean values for each variable

were examined across each group, the pattern of results that emerged was broadly consistent

with the findings revealed by examining PSQI component score across groups. The effect size

of this analysis was large (f = 0.28), suggesting that the sample size of the subgroup could have

played a role in the non-significant findings. These results are presented in Table 3-2.

Role of Negative Mood in the Sleep-Pain Relationship

Structural Equation Modeling (SEM) was used to examine the relationships among sleep,

pain and negative mood in this sample of chronic pain patients. Specifically, the relationships

between the latent factors of sleep and pain, the latent factors of sleep and negative mood, and

the latent factors of negative mood and pain were examined to investigate whether direct causal

paths existed between these factors. Additionally, in the final model including all three latent

factors, the potential mediating influence of the negative mood factor was considered in terms of









its influence on the sleep-pain relationship. Following the procedures of Baron and Kenny

(1986), mediation was indicated if there was a reduced or non-significant path coefficient

between the latent factor of sleep and the latent factor of pain, after the latent factor of negative

mood entered into the model (Baron & Kenny, 1986).

The sleep factor was indicated by three observed variables: PSQI global score, sleep

quantity (raw score) from the PSQI, and the sleep quality component score from the PSQI. The

negative mood factor was indicated by four observed variables, BDI total score, depression score

from the MCV, PASS total score, and anxiety score from the MCV. The pain factor was

indicated by three observed variables, MPQ total score, VAS for average pain intensity, and

usual pain intensity from the MCV.

The relationships between the latent factors were tested sequentially, in accordance with

the procedures outlined by Baron and Kenny (1986), and path coefficients were examined to

determine whether significant causal relationships existed. The first step involved testing the

relationship between the sleep factor and the pain factor. A significant relationship was found,

path coefficient = -0.46, t = -3.72, indicating that poorer sleep predicts greater pain; the model

was also found to provide a good fit for the data, X2(8)=3.85,p=0.87, RMSEA=0.00. Next, the

relationship between the sleep factor and the negative mood factor was tested. A significant

relationship was found, path coefficient = -0.57, t = -7.66, such that poorer sleep also predicts

higher levels of negative affect; the model was also found to provide a good fit for the data,

X2(13)=18.74, p=0.13, RMSEA=0.04. Third, the relationship between the negative mood factor

and the pain factor was tested. A significant relationship was found, path coefficient = 0.69, t =

3.63, indicating that higher levels of negative affect predict greater pain; however, the fit of this

model did not provide an excellent fit for the data, X2(12)=40.59,p=0.00, RMSEA=0.09.









Finally, the full model involving all 3 latent factors was tested. Examination of the path

coefficients in this final model, revealed a non-significant path between the sleep factor and the

pain factor, path coefficient = -0.02, t = -0.18; a significant path between the sleep factor and the

negative mood factor, path coefficient = -0.57, t = -7.72; and significant path between the

negative mood factor and the pain factor, path coefficient = 0.99, t = 4.34. Results indicated that

the full model fit the data well, X2(32)=41.77, p=0.12, RMSEA=0.03. This final model indicates

that negative affect mediates the effect of sleep on pain in this sample of chronic pain patients.

Poorer sleep also continued to predict higher levels of negative affect, and increased levels of

negative affect continued to predict increased pain in the final model.

The final structural equation model is shown in Figure 3-2. As stated above, mediation

of the relationship between sleep and pain by negative mood was indicated by this final model.

Specifically, the path coefficient between the sleep factor and the pain factor was reduced when

negative mood was entered into the model (-0.02), compared to the path coefficient between the

sleep factor and the pain factor without negative mood in the model (-0.46). This indicates that

the direct relationship between sleep and pain was significantly reduced when negative affect

was also included in the analysis, indicating that negative mood mediates the role of sleep on

pain.

Psychophysical Testing-Temporal Summation/"Wind up" and Ramp and Hold

A General Linear Model approach was used to conduct a series of mixed model analyses

of variance. For the temporal summation procedures, these analyses examined group (either:

good sleeper vs. poor sleeper, or chronic pain group), heat pulse number, and group by heat pulse

number interactions on pain report among facial pain, back pain, and fibromyalgia subgroup

participants. For the ramp and hold procedures, these analyses examined group (either: good









sleeper vs. poor sleeper, or chronic pain group), temperature level, and group by temperature

level interactions on pain intensity ratings and pain unpleasantness ratings among facial pain,

back pain, and fibromyalgia subgroup participants.

Omnibus multivariate tests for the temporal summation procedures indicated no

significant differences in pain report across either good vs. poor sleepers, or across chronic pain

groups. Omnibus multivariate tests for the ramp and hold procedures indicated no significant

differences in pain intensity ratings or pain unpleasantness ratings across either good vs. poor

sleepers, or across chronic pain groups. Significantly higher pain intensity ratings and

significantly higher pain unpleasantness ratings were found at higher temperatures, in all ramp

and hold analyses. Results of these mixed model analyses of variance are provided in Table 3-3.

Actigraphy Data and Sleep Diary Data

A MANOVA examined sleep onset latency (SOL), wake after sleep onset (WASO), sleep

efficiency, and total sleep time (TST), as measured by actigraphy, between subgroup participants

who did and did not report sleep disturbances. The overall omnibus test was non-significant,

F(4,17) = 0.79,p>.05, r2 = 0.16. A second MANOVA examined sleep onset latency (SOL),

wake after sleep onset (WASO), sleep efficiency, and total sleep time (TST), as measured by

participants' sleep diaries, between subgroup participants who did and did not report sleep

disturbances. The overall omnibus test for this analysis was also non-significant, F(4,17) = 1.47,

p>.05, l = 0.26.

Finally, separate Pearson correlations were computed between actigraphy and diary

measurement of SOL, WASO, sleep efficiency (SE), and total sleep time (TST) for good and

poor sleepers in the subgroup (Table 3-4). The correlations for each variable (SOL, WASO, SE,

TST) were compared between good and poor sleepers. Results revealed significantly stronger









correlations between diary and actigraphy measurements of WASO for good sleepers.

Correlations between diary and actigraphy measurement of SOL, SE, and TST were also higher

among good sleepers; however, the difference in the magnitude of the correlations between good

and poor sleepers for each of these variables did not reach significance (Table 3-4).











3 H Facial Pain

U Back Pain
2.5 F i bromyalgia
.......... 1:. -. .r. .r .







< 1


0.5
I s





0
sleep quality sleep latency sleep disturbances
PSQI subscales

Figure 3-1. Self-reported sleep (as reported by PSQI) for 3 chronic pain groups. **p<.01;
***p<.001.


















-0.57


-0.02


Figure 3-2. Structural equation model for the relationship among sleep, pain, and negative
mood. x2(32)=41.77, p=0.12, RMSEA=0.03.










Table 3-1. Demographic information for the 3 chronic pain patient groups
Facial pain Back pain Fibromyalgia
N= 116 N= 55 N= 121
M SD M SD M SD F df p

Age 46.28 14.12 46.44 10.28 47.53 10.82 0.35 2,291 0.706
Pain duration 91.32 109.91 128.70 111.57 130.29 70.80 2.37 2,175 0.097
Years of education 14.20 2.76 12.89 2.32 15.86 4.71 5.57 2,157 0.005


Facial pain


Back pain


Fibromyalgia


N (%) N (%) N (%)0 p
Sex
Female 95 (80.5) 34 (61.8) 147 ( 95.5) 36.96 0.000
Male 23 (19.5) 21 (38.2) 7 ( 4.5)
Race
Caucasian 107 (90.7) 46 (83.6) 6( 85.7) 9.30 0.318
Black/African- 4 ( 3.4) 8 (14.5) 1 ( 14.3)
American
Asian 1 ( 0.8) 0 ( 0.0) 0 ( 0.0)
Hispanic 4 ( 3.4) 1 ( 1.8) 0 ( 0.0)
Other 2( 1.7) 0( 0.0) 0( 0.0)
Work status
Full-time 23 (19.5) 8 (15.1) 0( 0.0) 15.32 0.053
Part-time 8 ( 6.8) 2 ( 3.8) 0 ( 0.0)
Working, 25 (21.2) 7 (13.2) 0 ( 0.0)
unspecified
Student 13 (11.0) 1( 1.9) 1 ( 14.3)
Not employed 49 (41.5) 35 (66.0) 6( 85.7)
Narcotic med
use
Yes 29 (26.1) 41(75.9) 4( 57.1) 37.35 0.000
No 82 (73.9) 13 (24.1) 3 ( 42.9)
Antidepressant
med use
Yes 41(36.9) 22 (42.3) 4( 66.7) 2.33 0.313
No 70 (63.1) 30 (57.7) 2( 33.3)
Sleep med use
Yes 47 (41.2) 20 (45.5) 3 (100.0) 4.51 0.341
No 62 (54.4) 23 (52.3) 0 ( 0.0)









Table 3-2. Comparison of diary- and actigraphy-measured sleep variables across 3 pain groups
Facial pain Back pain Fibromyalgia
M SD M SD M SD F p rq
Sleep diary variables df(2,19)
SOL 21.74 13.24 42.42 24.91 32.73 19.08 2.20 0.138 0.19
WASO 13.17 12.35 47.54 79.53 54.40 39.83 1.29 0.298 0.12
SE 91.12 3.89 86.87 7.23 85.22 5.77 2.00 0.163 0.17
VAS sleep quality 5.46 1.22 5.04 1.31 4.49 1.24 1.02 0.380 0.10

M SD M SD M SD F p rf
Actigraphy variables df(2,19)
SOL 16.93 13.37 29.68 23.47 23.46 16.67 0.95 0.404 0.09
WASO 47.26 13.06 57.58 14.77 67.25 27.92 1.99 0.164 0.17
SE 83.66 3.56 78.48 9.26 79.05 6.00 1.37 0.278 0.13

Note. SOL=Sleep onset latency; WASO=Wake after sleep onset; SE=Sleep efficiency; VAS=Visual analogue scale.









Table 3-3. Multivariate mixed model MANOVA results for pain ratings in temporal summation
and ramp and hold procedures
Temporal summation
Source Hypothesis df Error df F p 7r2
Pulse number 3 18 1.24 0.326 0.17
Pulse number x type 3 18 1.24 0.325 0.17
of sleeper
Pulse number 3 17 1.28 0.312 0.19
Pulse number x 6 36 0.90 0.509 0.13
chronic pain group

Ramp and hold -pain intensity
ratings
Source Hypothesis df Error df F p q2
Temperature 3 18 24.70 0.000 0.81
Temperature x type of 3 18 0.26 0.855 0.04
sleeper
Temperature 3 17 24.10 0.000 0.81
Temperature x 6 36 0.99 0.448 0.14
chronic pain group
Ramp and hold -pain
unpleasantness ratings
Source Hypothesis df Error df F p rf
Temperature 3 18 18.82 0.000 0.76
Temperature x type of 3 18 0.26 0.852 0.04
Sleeper
Temperature 3 17 20.14 0.000 0.78
Temperature x 6 36 1.08 0.392 0.15
chronic pain group









Table 3-4. Correlation between actigraphy- and diary-measured sleep variables for good and
poor sleepers


SE WASO SOL TST
actigraphy actigraphy actigraphy actigraphy
Good sleepers
(N=9)
SE
da 0.47
diary
WASO
0.91**
diary
SOL 0.77*
diary
TST 0.92***
diary

Poor sleepers
(N= 13)
SE
SE 0.34
diary
WASO 0.19
diary
SOL 0.47
diary
TST 0.82**
diary

Difference
Difference 0.30 2.59** 0.99 0.84
z-score


Note. SOL=
TST=Total


:Sleep onset latency; WASO=Wake after sleep onset; SE=Sleep efficiency;
sleep time. *p<.05; **p<.01; ***p<.001.









CHAPTER 4
DISCUSSION

Results revealed significant differences in subjective sleep reports across the three

chronic pain populations in the current sample. Most notably, the facial pain patients reported

better sleep quality, shorter latency to sleep onset, and fewer sleep disturbances, compared to the

patients in the back pain and fibromyalgia groups. This suggests that the facial pain patients were

experiencing less disturbed sleep overall, compared to the other two groups. The back pain

patients and the fibromyalgia patients endorsed similar reports regarding sleep quality, latency to

sleep onset, and sleep disturbances overall. They reported higher levels of subjective sleep

problems on a validated self-report measure of sleep quality, compared to the facial pain patients.

Descriptive analyses indicated significant differences in sex, education, and narcotic

medication use across chronic pain groups. No significant differences on any of the three sleep

variables were found across sex. Significant differences were found for sleep quality and sleep

latency across years of education, with poorer sleep being related to fewer years of education.

Facial pain patients also had significantly higher education compared to back pain patients.

Finally, significant differences in narcotic medication use was found across all three sleep

variables, with users having higher scores on each of the three sleep variables (indicating poorer

sleep) compared to non-users.

The highest rates of narcotic use were found among back pain patients, followed by

fibromyalgia patients, with facial pain patients endorsing the lowest rates of narcotic medication

use. As narcotic medications are known to affect sleep continuity and sleep architecture

(reduced sleep quantity, suppression of REM and stage 3-4 sleep) (Kay, Eisenstein, & Jasinski,

1969), it is possible that higher rates of use of these medications among the back pain and

fibromyalgia patients is related to the poorer sleep reported by these patients. However,









fibromyalgia patients reported the most impaired sleep across subscales, and back pain patients

had the greatest rates of narcotic medication use. Further, studies have demonstrated that

individuals typically develop a tolerance to the sleep effects of these medications within a few

weeks after initiating use (Kay et al., 1969). Additionally, average pain levels were compared

across pain groups, and fibromyalgia patients were found to have higher levels of average pain

compared to facial pain patients; back pain patients did not differ from either of the other two

groups. It is possible that higher levels of average pain among fibromyalgia patients were related

to the increased subjective sleep problems reported by these patients; however, back pain

patients also reported similar levels of sleep problems and similar levels of pain, so it is unclear

how important of a role average pain ratings played in the subjective sleep reports of these

patients.

Thus, while some support was found for the hypothesis that there would be differences in

the distribution of sleep problems across the three chronic pain groups, it was not the case that

each group of chronic pain patients exhibited increased levels of specific types of sleep

problems. The differences in sleep that emerged between the three chronic pain groups were in

quantity rather than type of sleep disturbance. Within each of the three chronic pain groups,

multiple types of sleep problems (difficulties with initiation of sleep, frequent disturbances

during sleep, and poor subjective sleep quality) were reported. However, it appears that patients

with back pain and fibromyalgia endorse greater levels of sleep problems when compared to

patients suffering from chronic facial pain. Back pain and fibromyalgia groups reported

significantly greater problems within all three areas: initiation of sleep, frequency of sleep

disturbances, and subjective sleep quality, compared to the group of facial pain patients in this

study.









It may be that the sleep problems of chronic pain patients are similar overall, or it is

possible that the component scores from the Pittsburgh Sleep Quality Index (PSQI) were not

sensitive enough to differentiate each of the specific sleep problems being investigated.

Therefore, the sleep onset latency (SOL), wake after sleep onset (WASO), and sleep quality

VAS ratings from subgroup participants' diaries (or actigraphy and diaries for SOL and WASO)

were also examined to determine whether any differences existed between the three chronic pain

groups on any of these characteristics. No significant differences emerged in these analyses; this

may be due in part to the small sample size of the subgroup as large effect sizes were seen in

these analyses. The overall pattern of results that emerged from analysis of the diary and

actigraphy variables in the subgroup is generally consistent with the findings from the self-report

measures in the larger sample of participants.

The structural equation modeling analysis sought to increase understanding of the

relationship between sleep, pain, and negative mood. In particular, the role of sleep in predicting

pain was of interest, as was the potential mediating effect of negative mood on this relationship.

Results from this analysis supported a direct relationship between sleep and pain, when negative

mood was not considered in the model. This supports and extends previous research by

demonstrating that increased sleep problems predict greater pain, using validated and reliable

measures in a large group of chronic pain patients. Poorer sleep was also shown to be predictive

of higher levels of negative mood, which again supports previous research and highlights the

importance of addressing sleep problems when present in chronic pain patients. Additionally,

when negative mood was included in the model, it was shown to mediate the relationship

between sleep and pain in this sample. Thus, the importance of assessing and treating both sleep

problems and negative mood in chronic pain patients cannot be underestimated.









While the relationship between sleep and pain is complex, these results suggest that there

are multiple pathways by which sleep is related to pain. First, sleep disturbance may lead

directly to increased pain among chronic pain patients. Second, sleep disturbance may lead to

increased pain in chronic pain patients through sleep's influence on negative mood. While

higher levels of negative mood are found among chronic pain patients, not all patients with

chronic pain have increased levels of negative mood. This model demonstrates that the influence

of sleep on pain may take multiple routes, and is also related to whether increased levels of

negative mood are present. This suggests that assessment of both sleep and negative mood

should routinely be undertaken in chronic pain populations. This model also suggests that

treatment efforts can take multiple routes, although research is needed to determine whether

interventions that improve sleep also lead to improvements in mood and/or pain as well.

No differences were found across the three chronic pain groups in their response to the

psychophysical testing. All participants noted higher ratings of pain intensity and pain

unpleasantness at higher temperatures in the ramp and hold procedure, but no other significant

results were found between pain groups in either the ramp and hold or the temporal summation

procedures. Additionally, the results of the psychophysical testing in the subgroup of

participants did not support the hypothesis of greater sensitivity to painful stimuli among those

patients with concurrent sleep disturbance. However, it should be noted that these analyses were

exploratory in nature, and a general pattern of large effect sizes was demonstrated across

analyses.

The results of this study demonstrated the high prevalence of subjective sleep problems

in chronic pain patients. By including different groups of chronic pain patients, and examining

differences in sleep reports across these different groups of patients, this study allows for a









greater understanding of the similarities and differences in the sleep problems reported by

different populations of chronic pain patients. Contrary to the hypothesis that different types of

sleep problems would be more prevalent in specific chronic pain groups, results suggested the

emergence of similar subjective sleep reports across all three chronic pain groups. One

significant difference noted across groups, was that the facial pain patients reported fewer sleep

disturbances overall compared to the other two groups. This pattern of results suggests that a

broad-based sleep treatment, within interventions for chronic pain, may be the most appropriate

approach given the current evidence. Since no specific sleep problems emerged within each of

the chronic pain groups, there is no evidence to suggest targeting sleep treatments for specific

sleep problems within any chronic pain populations.

Examination of the relationships among sleep, pain, and negative mood revealed several

important findings, and underscored the role of both sleep and negative mood in the clinical

picture of chronic pain. First, poor sleep was found to directly predict both negative mood and

pain in this sample of chronic pain patients. This highlights the importance of thoroughly

assessing for and treating sleep problems within chronic pain populations. The direct

relationship between sleep and pain suggests that addressing sleep problems in pain patients will

likely have a beneficial impact on these patients' pain experience. However, since negative

mood mediated the relationship between sleep and pain in this sample of chronic pain patients,

addressing negative mood may have a greater impact on patients' pain experience among those

chronic pain patients who are experiencing both sleep disturbance and negative mood.

Within the subgroup of participants who completed the psychophysical testing, no

significant differences emerged across the three chronic pain groups on either the ramp and hold

or temporal summation protocols. Participants from all three groups responded similarly to the









thermal pain stimuli in each protocol. Additionally, no significant differences were noted

between subgroup participants with and without concurrent sleep disturbances. These results did

not support the hypothesis that chronic pain patients with concurrent sleep problems would

experience greater sensitivity to the painful thermal stimuli, and fail to support previous findings

of increased pain sensitivity among individuals who experienced reduced or disrupted sleep

(Kundermann et al., 2004; Moldofsky & Scarisbrick, 1976; Onen, Alloui, Gross et al., 2001).

Comparison of Study Results to Previous Findings

Similar to previous studies, the current results support the ubiquitous nature of sleep

problems in chronic pain populations. Overall, 63.5% of participants reported subjectively poor

sleep using a modified cut-off score on the PSQI, within this sample of back pain, facial pain,

and fibromyalgia patients. This supports the hypothesis of sleep problems as a pervasive

condition among chronic pain patients, and suggests that sleep problems should be routinely

assessed for and treated in this population. This study adds to the existing knowledge by

examining the relative distribution of specific sleep complaints reported by patients with

different chronic pain conditions, allowing for a comparison of findings across different chronic

pain patient samples. In particular, while similarly disrupted sleep was noted across the entire

sample, results indicated that reported sleep problems were more severe among back pain and

fibromyalgia patients, as compared to the facial pain patients. The back pain and facial pain

patients did not appear to differ from one another in the type or severity of sleep problems

reported. Further study employing polysomnography would be useful in identifying whether

objective sleep differences emerged among different groups of patients, but the present findings

suggest that subjective experience of sleep is similar for back pain and fibromyalgia patients.

While it was hypothesized that the three chronic pain groups would differ in their

subjective sleep reports, results supported the presence of a more generalized pattern of sleep









disturbance across groups. Specifically, all three chronic pain groups reported difficulties with

both sleep onset and frequent sleep disturbances, as well as subjectively poor quality sleep.

Differences among groups emerged in the severity of these sleep problems, with fibromyalgia

patients endorsing significantly greater levels of these problems compared to the facial pain

patients, and back pain patients endorsing sleep problems that fell intermediate of these other

two groups. Thus, it appears that sleep is more broadly disturbed within chronic pain conditions

than initially hypothesized, and no specific differences regarding the sleep problems reported by

various chronic pain populations were evident in this sample. The experience of chronic pain,

either alone or in conjunction with negative mood, may simply serve as a gross disruptive factor

for sleep among these patients.

No significant relationships were found between subjective reports of good vs. poor sleep

and sensitivity to painful thermal stimuli in the current study. Previous studies have reported

increased sensitivity to painful stimuli following conditions of interrupted or restricted sleep

(Kundermann et al., 2004; Moldofsky & Scarisbrick, 1976; Onen, Alloui, Gross et al., 2001).

Thus, the present findings are somewhat inconsistent with these previous reports from the

literature. Interestingly, the effect sizes found in the subgroup analyses for the temporal

summation procedure were large, suggesting that the presence or absence of sleep disturbance

may be an important predictor of experimental pain response in chronic pain populations. Prior

studies have demonstrated that chronic pain patients have an increased likelihood of sensitization

to painful stimuli (Staud et al., 2003), and it may be that this is further compounded by the

experience of disturbed sleep. However, this cannot be determined based on the findings from

the present analyses and needs to be examined in future studies.









Role for Negative Mood in the Sleep-Pain Relationship

Higher levels of negative mood are commonly reported, both among chronic pain

populations and among individuals with sleep problems (Breivik et al., 2005; Breslau, Roth,

Rosenthal, & Andreski, 1996; Buysse et al., 1994; Duquesnoy et al., 1998; Ford & Kamerow,

1989; Ohayon, 1997; Verhaak et al., 1998). However, the interrelationships among sleep, pain,

and negative mood have been inconsistent in the existing literature (Atkinson et al., 1988;

Moffitt et al., 1991; Morin et al., 1998; Pilowsky et al., 1985; Sayar et al., 2002). Some studies

have reported higher levels of depression and/or anxiety among chronic pain patients with sleep

disturbances; other studies have not found these relationships. The current results support a

mediating role for negative mood in the relationship between sleep and pain. In other words,

while significant relationships have been reported between sleep and pain, these relationships

must take into account the presence of negative mood. In particular, among chronic pain patients

endorsing high levels of negative mood, pain is more strongly predicted by negative mood than

by sleep. However, it is not difficult to see how negative mood, sleep disturbances, and pain can

each act to increase the other two, leading to a cycle that perpetuates itself if there is no

intervening action to disrupt it. Addressing negative mood will likely have a beneficial effect on

patients' pain experience or perceptions regarding their ability to cope with their pain, and may

also improve sleep as well. Further, among patients with low levels of negative mood,

intervening to improve sleep disturbances will likely have a positive impact on patients' pain

experience as well. As effective interventions have been developed for both negative mood and

sleep, implementing these treatment will be important as it is likely to improve patients' pain

experience and lead to significant improvements in functioning.









Implications for Conceptualization and Treatment of Chronic Pain

The comorbidity between chronic pain and sleep disturbances has been widely reported

in the literature (Atkinson et al., 1988; Pilowsky et al., 1985; M. T. Smith & Haythornthwaite,

2004). Similarly, numerous studies have reported increased reports of negative mood among

both chronic pain patients (Breivik et al., 2005; Duquesnoy et al., 1998; Robinson & Riley III,

1999; Verhaak et al., 1998), and patients with sleep problems, particularly those reporting

insomnia complaints (Breslau et al., 1996; Chang, Ford, Mead, Cooper-Patrick, & Klag, 1997;

Ford & Kamerow, 1989; Katz & McHorney, 2002). The model investigated in this study

examined the role of sleep and negative mood in predicting pain among chronic pain patients,

and revealed that negative mood mediated the relationship between sleep disturbance and pain.

Thus, previous findings suggesting a causal link between sleep disturbances and pain in chronic

pain patients may be due, in part, to the effects of negative mood. High levels of negative mood

may increase or perpetuate the impact of sleep disturbances on patients' pain experiences,

possibly through the interruption of sleep, which is a common report in the clinical pictures of

both depression and anxiety.

Additionally, certain areas in the central nervous system have been implicated in both

pain perception/regulation, as well as the regulation/facilitation of sleep, such as the

hypothalamus (Montagna, 2006). Abnormalities of the hypothalamic-pituitary-adrenal (HPA)

axis functioning have been noted among 30-70% of individuals suffering from major depression,

and good treatment response has been demonstrated for individuals who evidence normalization

of HPA functioning with use of antidepressant medications (Takahashi, 2002). This lends

further supports to the importance of this pathway in mood regulation, and given the overlap

with functions related to both sleep and pain, this pathway may be a common physiological link

between these oft related conditions. Taken together, these findings support the argument for the









involvement of common pathophysiological pathways in sleep disturbance, mood disorders, and

chronic pain.

Interestingly, no clear relationship was found between sleep disturbance and sensitivity to

painful stimuli in the present study. This is somewhat contradictory to findings from previous

studies, which have reported increased sensitivity to painful stimuli or increased pain reports

among healthy individuals following conditions of sleep restriction or interrupted sleep

(Kundermann et al., 2004; Moldofsky & Scarisbrick, 1976; Onen, Alloui, Gross et al., 2001).

Similarly, studies examining pain sensitivity among chronic pain patients have also suggested an

inverse relationship between pain response to experimental stimuli and sleep (Agargun et al.,

1999). The present results are inconsistent with these findings.

It should be noted that these analyses utilized participants' self-reports about their

subjective sleep to determine good and poor sleepers, rather than experimentally manipulating

participants' sleep. When sleep has been manipulated experimentally among healthy

participants, increased pain sensitivity has been reported (Kundermann et al., 2004; Moldofsky

& Scarisbrick, 1976; Onen, Alloui, Gross et al., 2001). Additionally, a recent study by Smith

and colleagues demonstrated differential effects on pain inhibition for restricting total sleep time

to a shorter duration compared to repeatedly disrupting sleep using forced awakenings to

produce a similar shorter total duration of sleep, within a sample of healthy participants (M. T.

Smith, Edwards, McCann, & Haythornthwaite, 2007). These results suggest that differential

effects on pain processing may be produced by disruptions in sleep continuity as compared to

reduced sleep quantity produced by a shortened schedule. The pattern of more generalized sleep

disturbances reported by the subgroup of chronic pain patients in the present study may be

qualitatively different than the sleep deprivation produced by restricting total sleep time in









healthy participants. Thus, the increased pain sensitivity related to sleep disturbances reported

by previous studies may vary as a function of the type of sleep disturbance, as well as the type of

population (healthy volunteers versus chronic pain patients). In other words, differential effects

on pain sensitivity may be seen following experimentally-produced sleep restriction in healthy

volunteers as compared to the effect of more chronic or cumulative sleep interruptions on pain

response in chronic pain patients. If this hypothesis is true, it could help to explain the lack of

significant findings regarding pain sensitivity in the present analyses.

Current evidence supports cognitive-behavioral approaches to the treatment of chronic

pain as having the most consistent empirical support (Robinson & O'Brien, in press), with

improvements seen in pain, mood, and coping, as well as reductions in interference and

improved functional outcomes (such as fewer absences from work, reduced medication use and

health care visits) (Linton, Boersma, Jansson, Svard, & Botvalde, 2005; Nash, Park, Walker,

Gordon, & Nicholson, 2004; Turner, Mancl, & Aaron, 2006). Within CBT approaches for

chronic pain, negative mood is often addressed either directly or indirectly, and the results of this

study suggest that addressing negative mood within chronic pain treatments is important and

should continue to be a central part of these treatments. However, the assessment and treatment

of sleep is a less standard component in CBT treatments for chronic pain, and most often consists

of education about sleep hygiene when it is included. While this may be useful information,

sleep hygiene alone has not been demonstrated to be effective as a means of implementing

changes to improve sleep (Engle-Friedman, Bootzin, Hazlewood, & Tsao, 1992; Guilleminault et

al., 1995).

Given the high comorbidity of sleep disturbance and chronic pain, it would appear

prudent to include a formal assessment of patients' sleep within chronic pain assessments.









Additionally, incorporating established techniques (stimulus control, sleep restriction) from CBT

for Insomnia (CBT-I) treatments for addressing sleep disturbances into existing CBT treatments

for chronic pain would likely be more effective for addressing sleep problems being experienced

by chronic pain patients. The skill-based, behavioral nature of these techniques allows for their

inclusion within existing cognitive-behavioral treatments for chronic pain. Incorporating these

sleep treatment components with existing techniques for addressing mood and activity, as well as

increasing patients' knowledge regarding chronic pain, will likely lead to better outcomes and

more fully address the full spectrum of impairment experienced by chronic pain patients.

Limitations

Some limitations to the above findings should be noted. Analyses involving the subgroup

of participants completing sleep diaries, actigraphy monitoring, and the psychophysical testing

procedures were intended to be exploratory, in order to identify the presence of potentially

provocative findings that could subsequently be explored in a larger and more representative

sample. As such, the sample size of this subgroup of participants was small and may have made

it difficult to detect significant findings using traditional statistical criteria.

Results from the analyses of participants' response to psychophysical testing revealed

large effect sizes between good and poor sleepers (temporal summation), and across chronic pain

groups (temporal summation and ramp and hold), in these analyses. This suggests that sleep

disturbances may be an important of determinant of chronic pain patients' responses to painful

stimuli and that type of pain condition may also be an important influence on individuals'

response to painful stimuli. However, the present analyses may not have been able to

demonstrate these relationships at traditional statistical significance levels, due to the small

sample size of the subgroup reducing the power of the analyses. There was no pain-free sample









of participants in the present study, so results are not available to examine the role of sleep

disturbance on psychophysical pain response among individuals without chronic pain.

Similarly, when sleep diary and actigraphy measurement of several sleep parameters

across good and poor sleepers were examined, no significant differences emerged. The large

effect sizes of these analyses again suggested that the sample size of the subgroup may have

impacted the ability for significant findings to emerge. Thus, consistent with the general pattern

of correlations found between sleep diary and actigraphy measures found among good and poor

sleepers, it appears that these measures may be more strongly associated in good sleepers

compared to poor sleepers. Examination of these relationships in a larger sample will increase

the power of the analyses and improve the ability to detect such relationships, if they are present.

The large effect sizes found across subgroup analyses suggests avenues for additional studies

using larger numbers of participants, in order to elucidate the nature of the relationships among

sleep disturbance, type of chronic pain, and psychophysical response.

Additionally, the number of female participants was much greater than the number of

male participants in the sample. This is consistent with the literature on chronic pain, where

females tend to be over-represented in most chronic pain populations (Moulin et al., 2002;

Verhaak et al., 1998). While the unequal distribution of males and females makes it difficult to

directly examine sex differences in these findings, the preponderance of data has demonstrated a

lack of sex differences in clinical pain (Robinson, Wise, Riley III, & Atchison, 1998). This

allows for increased confidence in the applicability of the current findings to both male and

female chronic pain populations.

Future Directions

It would be interesting to examine whether incorporating CBT-I techniques into existing

CBT treatments for chronic pain affords any benefit for chronic pain patients. Specifically,









examination of patients' mood, sleep, and pain reports following standard CBT treatment for

pain, CBT for pain plus sleep hygiene information alone, and CBT for pain plus sleep hygiene

information and CBT-I techniques, would provide information about the additive benefits of the

addition of sleep treatment to existing CBT treatments for chronic pain. While it is likely that

the addition of the sleep techniques would confer added utility to chronic pain treatments, it

would also increase the duration and thus, the cost, of these treatments. Thus, it would be

important to provide a justification for the additional time and cost of adding these techniques

into existing treatment paradigms, by demonstrating improved patient outcomes when these

components are part of the treatment package. Additionally, longitudinal evaluations would be

useful to examine any improvement in outcomes that patients experience following treatment.









LIST OF REFERENCES


Affleck, G., Urrows, S., Tennen, H., Higgins, P., & Abeles, M. (1996). Sequential daily relations
of sleep, pain intensity, and attention to pain among women with fibromyalgia. Pain,
68(2-3), 363-368.

Agargun, M. Y., Tekeoglu, I., Gunes, A., Adak, B., Kara, H., & Ercan, M. (1999). Sleep quality
and pain threshold in patients with fibromyalgia. Compr Psychiatry, 40(3), 226-228.

American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental
Disorders (DSM-IV) (4th edition ed.). Washington, DC: American Psychiatric
Association.

Anch, A. M., Lue, F. A., MacLean, A. W., & Moldofsky, H. (1991). Sleep physiology and
psychological aspects of the fibrositis (fibromyalgia) syndrome. Can JPsychol, 45(2),
179-184.

Ancoli-Israel, S., Cole, R., Alessi, C., Chambers, M., Moorcroft, W., & Pollak, C. P. (2003). The
role of actigraphy in the study of sleep and circadian rhythms. Sleep, 26(3), 342-392.

Andersson, G. B. (1999). Epidemiological features of chronic low-back pain. Lancet, 354(9178),
581-585.

Andersson, H. I., Ejlertsson, G., Leden, I., & Schersten, B. (1999). Impact of chronic pain on
health care seeking, self care, and medication. Results from a population-based Swedish
study. JEpidemiol Community Health, 53(8), 503-509.

Arima, T., Svensson, P., Rasmussen, C., Nielsen, K. D., Drewes, A. M., & Arendt-Nielsen, L.
(2001). The relationship between selective sleep deprivation, nocturnal jaw-muscle
activity and pain in healthy men. J Oral Rehabil, 28(2), 140-148.

Atkinson, J. H., Ancoli-Israel, S., Slater, M., Garfin, S. R., & Gillin, J. C. (1988). Subjective
sleep disturbance in chronic pain. Clinical Journal of Pain, 4, 225-232.

Bailey, D. R. (1997). Sleep disorders. Overview and relationship to orofacial pain. Dent Clin
North Am, 41(2), 189-209.

Barabas, G., Ferrari, M., & Matthews, W. S. (1983). Childhood migraine and somnambulism.
Neurology, 33(7), 948-949.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: conceptual, strategic, and statistical considerations. JPers Soc
Psychol, 51(6), 1173-1182.

Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric properties of the Beck
Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review, 8,
77-100.










Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for
measuring depression. Arch Gen Psychiatry, 4, 561-571.

Bennett, R. M. (1993). The origin of myopain: An integrated hypothesis of focal muscle changes
and sleep disturbance in patients with the fibromyalgia syndrome. JMusculoskelPain, 1,
105.

Binkley, J., Finch, E., Hall, J., Black, T., & Gowland, C. (1993). Diagnostic classification of
patients with low back pain: report on a survey of physical therapy experts. Phys Ther,
73(3), 138-150; discussion 150-135.

Blau, J. N. (1982). Resolution of migraine attacks: sleep and the recovery phase. JNeurol
Neurosurg Psychiatry, 45(3), 223-226.

Blau, J. N. (1990). Sleep deprivation headache. Cephalalgia, 10(4), 157-160.

Bragdon, E. E., Light, K. C., Costello, N. L., Sigurdsson, A., Bunting, S., Bhalang, K., et al.
(2002). Group differences in pain modulation: pain-free women compared to pain-free
men and to women with TMD. Pain, 96(3), 227-237.

Branco, J., Atalaia, A., & Paiva, T. (1994). Sleep cycles and alpha-delta sleep in fibromyalgia
syndrome. JRheumatol, 21(6), 1113-1117.

Breivik, H., Collett, B., Ventafridda, V., Cohen, R., & Gallacher, D. (2005). Survey of chronic
pain in Europe: Prevalence, impact on daily life, and treatment. Eur JPain.

Breslau, N., Roth, T., Rosenthal, L., & Andreski, P. (1996). Sleep disturbance and psychiatric
disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry, 39(6),
411-418.

Buysse, D. J., Reynolds, C. F., 3rd, Kupfer, D. J., Thorpy, M. J., Bixler, E., Manfredi, R., et al.
(1994). Clinical diagnoses in 216 insomnia patients using the International Classification
of Sleep Disorders (ICSD), DSM-IV and ICD-10 categories: a report from the
APA/NIMH DSM-IV Field Trial. Sleep, 17(7), 630-637.

Buysse, D. J., Reynolds, C. F., 3rd, Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The
Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research.
Psychiatry Res, 28(2), 193-213.

Campbell, S. M., Clark, S., Tindall, E. A., Forehand, M. E., & Bennett, R. M. (1983). Clinical
characteristics of fibrositis. I. A "blinded," controlled study of symptoms and tender
points. Aiin iti,% Rheum, 26(7), 817-824.









Carette, S., McCain, G. A., Bell, D. A., & Fam, A. G. (1986). Evaluation of amitriptyline in
primary fibrositis. A double-blind, placebo-controlled study. At ith iti, Rheum, 29(5), 655-
659.

Carli, G., Suman, A. L., Biasi, G., & Marcolongo, R. (2002). Reactivity to superficial and deep
stimuli in patients with chronic musculoskeletal pain. Pain, 100(3), 259-269.

Carpenter, J. S., & Andrykowski, M. A. (1998). Psychometric evaluation of the Pittsburgh Sleep
Quality Index. JPsychosom Res, 45(1 Spec No), 5-13.

Casey, K. L., Morrow, T. J., Lorenz, J., & Minoshima, S. (2001). Temporal and spatial dynamics
of human forebrain activity during heat pain: analysis by positron emission tomography.
JNeurophysiol, 85(2), 951-959.

Chang, P. P., Ford, D. E., Mead, L. A., Cooper-Patrick, L., & Klag, M. J. (1997). Insomnia in
young men and subsequent depression. The Johns Hopkins Precursors Study. Am J
Epidemiol, 146(2), 105-114.

Chase, M. A., & Morales, F. R. (1994). The control of motorneurons during sleep. In M. H.
Kryger, T. Roth, D. W. C. & e. al. (Eds.), Principles and Practice of Sleep Medicine (2nd
ed., pp. 163-176). Philadelphia, PA: W. B. Saunders.

Chiu, Y. H., Silman, A. J., Macfarlane, G. J., Ray, D., Gupta, A., Dickens, C., et al. (2005). Poor
sleep and depression are independently associated with a reduced pain threshold. Results
of a population based study. Pain, 115(3), 316-321.

Clauw, D. J., Williams, D., Lauerman, W., Dahlman, M., Aslami, A., Nachemson, A. L., et al.
(1999). Pain sensitivity as a correlate of clinical status in individuals with chronic low
back pain. Spine, 24(19), 2035-2041.

Cohen, M. J. M., Menefee, L. A., Doghramji, K., Anderson, W. R., & Frank, E. D. (2000). Sleep
in chronic pain: Problems and treatments. International Review of Psychiatry, 12, 115-
126.

Culebras, A., & Miller, M. (1984). Dissociated patterns of nocturnal prolactin, cortisol, and
growth hormone secretion after stroke. Neurology, 34(5), 631-636.

Curran, S. L., Carlson, C. R., & Okeson, J. P. (1996). Emotional and physiologic responses to
laboratory challenges: patients with temporomandibular disorders versus matched control
subjects. J Orofac Pain, 10(2), 141-150.

Demarco, G. J., Baghdoyan, H. A., & Lydic, R. (2003). Differential cholinergic activation of G
proteins in rat and mouse brainstem: relevance for sleep and nociception. J Comp Neurol,
457(2), 175-184.









de Souza, L., Benedito-Silva, A. A., Pires, M. L., Poyares, D., Tufik, S., & Calil, H. M. (2003).
Further validation of actigraphy for sleep studies. Sleep, 26(1), 81-85.

Diagnostic Classification Steering Committee, T. M., Chairman. (1990). International
Classification of Sleep Disorders: Diagnostic and Coding Manual (ICSD). Rochester,
MN: American Sleep Disorders Association.

Drewes, A. M., Nielsen, K. D., Arendt-Nielsen, L., Birket-Smith, L., & Hansen, L. M. (1997).
The effect of cutaneous and deep pain on the electroencephalogram during sleep--an
experimental study. Sleep, 20(8), 632-640.

Drewes, A. M., Nielsen, K. D., Taagholt, S. J., Bjerregard, K., Svendsen, L., & Gade, J. (1995).
Sleep intensity in fibromyalgia: focus on the microstructure of the sleep process. Br J
Rheumatol, 34(7), 629-635.

Duquesnoy, B., Allaert, F. A., & Verdoncq, B. (1998). Psychosocial and occupational impact of
chronic low back pain. Rev Rhum Engl Ed, 65(1), 33-40.

Engle-Friedman, M., Bootzin, R. R., Hazlewood, L., & Tsao, C. (1992). An evaluation of
behavioral treatments for insomnia in the older adult. J Clin Psychol, 48(1), 77-90.

Espie, C. A., Brooks, D. N., & Lindsay, W. R. (1989). An evaluation of tailored psychological
treatment of insomnia. JBehav Ther Exp Psychiatry, 20(2), 143-153.

Farasyn, A., & Meeusen, R. (2005). The influence of non-specific low back pain on pressure
pain thresholds and disability. Eur JPain, 9(4), 375-381.

Flor, H., Diers, M., & Birbaumer, N. (2004). Peripheral and electrocortical responses to painful
and non-painful stimulation in chronic pain patients, tension headache patients and
healthy controls. Neurosci Lett, 361(1-3), 147-150.

Foo, H., & Mason, P. (2003). Brainstem modulation of pain during sleep and waking. Sleep Med
Rev, 7(2), 145-154.

Ford, D. E., & Kamerow, D. B. (1989). Epidemiologic study of sleep disturbances and
psychiatric disorders. An opportunity for prevention? Jama, 262(11), 1479-1484.

Forgays, D. G., Forgays, D. K., & Spielberger, C. D. (1997). Factor structure of the State-Trait
Anger Expression Inventory. JPers Assess, 69(3), 497-507.

Forgays, D. K., Spielberger, C. D., Ottaway, S. A., & Forgays, D. G. (1998). Factor structure of
the State-Trait Anger Expression Inventory for middle-aged men and women.
Assessment, 5(2), 141-155.

Gallagher, R. M., & Verma, S. (1999). Managing pain and comorbid depression: A public health
challenge. Semin Clin Neuropsychiatry, 4(3), 203-220.









Gijsbers van Wijk, C. M., van Vliet, K. P., & Kolk, A. M. (1996). Gender perspectives and
quality of care: towards appropriate and adequate health care for women. Soc Sci Med,
43(5), 707-720.

Glaros, A. G. (1981). Incidence of diurnal and nocturnal bruxism. JProsthet Dent, 45(5), 545-
549.

Grabois, M. (2005). Management of chronic low back pain. Am JPhys MedRehabil, 84(3
Suppl), S29-41.

Guilleminault, C., Clerk, A., Black, J., Labanowski, M., Pelayo, R., & Claman, D. (1995).
Nondrug treatment trials in psychophysiologic insomnia. Arch Intern Med, 155(8), 838-
844.

Harding, S. M. (1998). Sleep in fibromyalgia patients: subjective and objective findings. Am J
Med Sci, 315(6), 367-376.

Harvey, A. G. (2000). Sleep hygiene and sleep-onset insomnia. JNervMentDis, 188(1), 53-55.

Hastie, B. A., Riley, J. L., 3rd, & Fillingim, R. B. (2004). Ethnic differences in pain coping:
factor structure of the coping strategies questionnaire and coping strategies questionnaire-
revised. JPain, 5(6), 304-316.

Hauri, P. J., & Wisbey, J. (1992). Wrist actigraphy in insomnia. Sleep, 15(4), 293-301.

Headache Classification Committee of the International Headache Society. (1988). Classification
and diagnostic criteria for headache disorders, cranial neuralgias and facial pain.
Cephalalgia, 8(suppl 7), 10-73.

Horne, J. A., & Shackell, B. S. (1991). Alpha-like EEG activity in non-REM sleep and the
fibromyalgia (fibrositis) syndrome. Electroencephalogr Clin Neurophysiol, 79(4), 271-
276.

Hurtig, I. M., Raak, R. I., Kendall, S. A., Gerdle, B., & Wahren, L. K. (2001). Quantitative
sensory testing in fibromyalgia patients and in healthy subjects: identification of
subgroups. Clin JPain, 17(4), 316-322.

Jennum, P., Drewes, A. M., Andreasen, A., & Nielsen, K. D. (1993). Sleep and other symptoms
in primary fibromyalgia and in healthy controls. JRheumatol, 20(10), 1756-1759.

Jones, B. E. (1994). Basic mechanisms of sleep-wake states. In M. H. Kryger, T. Roth, D. W. C.
& e. al. (Eds.), Principles and Practice of Sleep Medicine (2nd ed., pp. 145-162).
Philadelphia, PA: W. B. Saunders.

Katz, D. A., & McHorney, C. A. (2002). The relationship between insomnia and health-related
quality of life in patients with chronic illness. JFam Pract, 51(3), 229-235.









Kay, D. C., Eisenstein, R. B., & Jasinski, D. R. (1969). Morphine effects on human REM state,
waking state and NREM sleep. Psychopharmacologia, 14(5), 404-416.

Kramer, J. J., & Conoley, J. C. (Eds.). (1992). The eleventh mental measurements yearbook.
Lincoln, NE: Buros Institute of Mental Measurements.

Kriegler, J. S., & Ashenberg, Z. S. (1987). Management of chronic low back pain: a
comprehensive approach. Semin Neurol, 7(4), 303-312.

Kundermann, B., Spernal, J., Huber, M. T., Krieg, J. C., & Lautenbacher, S. (2004). Sleep
deprivation affects thermal pain thresholds but not somatosensory thresholds in healthy
volunteers. Psychosom Med, 66(6), 932-937.

Kushida, C. A., Chang, A., Gadkary, C., Guilleminault, C., Carrillo, 0., & Dement, W. C.
(2001). Comparison of actigraphic, polysomnographic, and subjective assessment of
sleep parameters in sleep-disordered patients. Sleep Med, 2(5), 389-396.

Landis, C. A., Frey, C. A., Lentz, M. J., Rothermel, J., Buchwald, D., & Shaver, J. L. (2003).
Self-reported sleep quality and fatigue correlates with actigraphy in midlife women with
fibromyalgia. Nurs Res, 52(3), 140-147.

Lanes, T. C., Gauron, E. F., Spratt, K. F., Wernimont, T. J., Found, E. M., & Weinstein, J. N.
(1995). Long-term follow-up of patients with chronic back pain treated in a
multidisciplinary rehabilitation program. Spine, 20(7), 801-806.

Laursen, B. S., Bajaj, P., Olesen, A. S., Delmar, C., & Arendt-Nielsen, L. (2005). Health related
quality of life and quantitative pain measurement in females with chronic non-malignant
pain. Eur JPain, 9(3), 267-275.

Lichstein, K. L., Stone, K. C., Donaldson, J., Nau, S. D., Soeffing, J. P., Murray, D., et al.
(2006). Actigraphy validation with insomnia. Sleep, 29(2), 232-239.

Linton, S. J., Boersma, K., Jansson, M., Svard, L., & Botvalde, M. (2005). The effects of
cognitive-behavioral and physical therapy preventive interventions on pain-related sick
leave: a randomized controlled trial. Clin JPain, 21(2), 109-119.

Lobbezoo, F., Visscher, C. M., & Naeije, M. (2004). Impaired health status, sleep disorders, and
pain in the craniomandibular and cervical spinal regions. Eur JPain, 8(1), 23-30.

Lockley, S. W., Skene, D. J., & Arendt, J. (1999). Comparison between subjective and
actigraphic measurement of sleep and sleep rhythms. JSleep Res, 8(3), 175-183.

Love, A., Leboeuf, C., & Crisp, T. C. (1989). Chiropractic chronic low back pain sufferers and
self-report assessment methods. Part I. A reliability study of the Visual Analogue Scale,
the Pain Drawing and the McGill Pain Questionnaire. JManipulative Physiol Ther,
12(1), 21-25.









Lowe, N. K., Walker, S. N., & McCallum, R. C. (1991). Confirming the theoretical structure of
the McGill Pain Questionnaire in acute clinical pain. Pain, 46, 53-60.

Macfarlane, T. V., Blinkhorn, A. S., Davies, R. M., Kincey, J., & Worthington, H. V. (2002).
Oro-facial pain in the community: prevalence and associated impact. Community Dent
Oral Epidemiol, 30(1), 52-60.

Macfarlane, T. V., Kincey, J., & Worthington, H. V. (2002). The association between
psychological factors and oro-facial pain: a community-based study. Eur JPain, 6(6),
427-434.

Maixner, W., Fillingim, R., Booker, D., & Sigurdsson, A. (1995). Sensitivity of patients with
painful temporomandibular disorders to experimentally evoked pain. Pain, 63(3), 341-
351.

Malanga, G. A., & Nadler, S. F. (1999). Nonoperative treatment of low back pain. Mayo Clin
Proc, 74(11), 1135-1148.

McAlpine, T. H. (1987). Sleep, Divine and Human in the Old Testament. Sheffield: Academic
Press.

McCracken, L. M., & Dhingra, L. (2002). A short version of the Pain Anxiety Symptoms Scale
(PASS-20): preliminary development and validity. Pain Res 3/,/ui. 7(1), 45-50.

McCracken, L. M., Zayfert, C., & Gross, R. T. (1992). The Pain Anxiety Symptoms Scale:
development and validation of a scale to measure fear of pain. Pain, 50(1), 67-73.

Melzack, R. (1975). The McGill Pain Questionnaire: major properties and scoring methods.
Pain, 1(3), 277-299.

Melzack, R., & Katz, J. (1992). The McGill Pain Questionnaire: Appraisal and current status. In
D. C. Turk & R. Melzack (Eds.), Handbook ofpain assessment (pp. 152-168). New York,
NY: Guilford Press.

Menefee, L. A., Frank, E. D., Doghramji, K., Picarello, K., Park, J. J., Jalali, S., et al. (2000).
Self-reported sleep quality and quality of life for individuals with chronic pain
conditions. Clin JPain, 16(4), 290-297.

Moffitt, P. F., Kalucy, E. C., Kalucy, R. S., Baum, F. E., & Cooke, R. D. (1991). Sleep
difficulties, pain and other correlates. Jlntern Med, 230(3), 245-249.

Moldofsky, H. (1989). Sleep and fibrositis syndrome. Rheum Dis Clin North Am, 15(1), 91-103.

Moldofsky, H. (1994). Central nervous system and peripheral immune functions and the sleep-
wake system. JPsychiatry Neurosci, 19(5), 368-374.









Moldofsky, H., & Lue, F. A. (1980). The relationship of alpha and delta EEG frequencies to pain
and mood in 'fibrositis' patients treated with chlorpromazine and L-tryptophan.
Electroencephalogr Clin Neurophysiol, 50(1-2), 71-80.

Moldofsky, H., Lue, F. A., & Smythe, H. A. (1983). Alpha EEG sleep and morning symptoms in
rheumatoid arthritis. JRheumatol, 10(3), 373-379.

Moldofsky, H., & Scarisbrick, P. (1976). Induction of neurasthenic musculoskeletal pain
syndrome by selective sleep stage deprivation. Psychosom Med, 38(1), 35-44.

Moldofsky, H., Scarisbrick, P., England, R., & Smythe, H. (1975). Musculosketal symptoms and
non-REM sleep disturbance in patients with "fibrositis syndrome" and healthy subjects.
Psychosom Med, 37(4), 341-351.

Montagna, P. (2006). Hypothalamus, sleep and headaches. Neurol Sci, 27 Suppl 2, S138-143.

Morin, C. M. (2003). Measuring outcomes in randomized clinical trials of insomnia treatments.
Sleep Med Rev, 7(3), 263-279.

Morin, C. M., Gibson, D., & Wade, J. (1998). Self-reported sleep and mood disturbance in
chronic pain patients. Clin JPain, 14(4), 311-314.

Moulin, D. E., Clark, A. J., Speechley, M., & Morley-Forster, P. K. (2002). Chronic pain in
Canada--prevalence, treatment, impact and the role of opioid analgesia. Pain Res fa ,/Ig.
7(4), 179-184.

Mountz, J. M., Bradley, L. A., Modell, J. G., Alexander, R. W., Triana-Alexander, M., Aaron, L.
A., et al. (1995). Fibromyalgia in women. Abnormalities of regional cerebral blood flow
in the thalamus and the caudate nucleus are associated with low pain threshold levels.
Ai /th iti, Rheum, 38(7), 926-938.

Nash, J. M., Park, E. R., Walker, B. B., Gordon, N., & Nicholson, R. A. (2004). Cognitive-
behavioral group treatment for disabling headache. Pain Med, 5(2), 178-186.

Nicassio, P. M., & Wallston, K. A. (1992). Longitudinal relationships among pain, sleep
problems, and depression in rheumatoid arthritis. JAbnorm Psychol, 101(3), 514-520.

Ohayon, M. M. (1997). Prevalence of DSM-IV diagnostic criteria of insomnia: distinguishing
insomnia related to mental disorders from sleep disorders. JPsychiatr Res, 31(3), 333-
346.

Ohayon, M. M. (2002). Epidemiology of insomnia: what we know and what we still need to
learn. Sleep MedRev, 6(2), 97-111.









Onen, S. H., Alloui, A., Gross, A., Eschallier, A., & Dubray, C. (2001). The effects of total sleep
deprivation, selective sleep interruption and sleep recovery on pain tolerance thresholds
in healthy subjects. JSleep Res, 10(1), 35-42.

Onen, S. H., Alloui, A., Jourdan, D., Eschalier, A., & Dubray, C. (2001). Effects of rapid eye
movement (REM) sleep deprivation on pain sensitivity in the rat. Brain Research, 900,
261-267.

Paiva, T., Batista, A., Martins, P., & Martins, A. (1995). The relationship between headaches and
sleep disturbances. Headache, 35(10), 590-596.

Paiva, T., Esperanca, P., Martins, I., Batista, A., & Martins, P. (1992). Sleep disorders in
headache patients. Headache Quarterly, 3, 438-442.

Paiva, T., Farinha, A., Martins, A., Batista, A., & Guilleminault, C. (1997). Chronic headaches
and sleep disorders. Arch Intern Med, 157(15), 1701-1705.

Paiva, T., Martins, P., Batista, A., Esperanca, P., & Martins, I. (1994). Sleep disturbances in
chronic headache patients: A comparison with healthy controls. Headache Quarterly, 5,
135-141.

Pau, A. K., Croucher, R., & Marcenes, W. (2003). Prevalence estimates and associated factors
for dental pain: a review. Oral Health Prev Dent, 1(3), 209-220.

Paulson, P. E., Casey, K. L., & Morrow, T. J. (2002). Long-term changes in behavior and
regional cerebral blood flow associated with painful peripheral mononeuropathy in the
rat. Pain, 95(1-2), 31-40.

Pearce, J., & Morley, S. (1989). An experimental investigation of the construct validity of the
McGill Pain Questionnaire. Pain, 39(1), 115-121.

Pennebaker, J. W. (1982). The psychology ofphysical symptoms. New York: Springer Verlag.

Pillemer, S. R., Bradley, L. A., Crofford, L. J., Moldofsky, H., & Chrousos, G. P. (1997). The
neuroscience and endocrinology of fibromyalgia. A it /i/i/ Rheum, 40(11), 1928-1939.

Pilowsky, I., Crettenden, I., & Townley, M. (1985). Sleep disturbance in pain clinic patients.
Pain, 23(1), 27-33.

Pollard, C. A. (1984). Preliminary validity study of the pain disability index. Percept Mot ,kill%,
59(3), 974.

Price, D. D., & Bushnell, M. C. (1994). Overview of pain dimensions and their psychological
modulation. In D. D. Price & M. C. Bushnell (Eds.), Psychological methods ofpain
control. Basic science and clinical perspectives (pp. 3-17). Seattle: IASP Press.









Price, D. D., Harkins, S. W., & Baker, C. (1987). Sensory-affective relationships among different
types of clinical and experimental pain. Pain, 28(3), 297-307.

Reite, M., Buysse, D., Reynolds, C., & Mendelson, W. (1995). The use of polysomnography in
the evaluation of insomnia. Sleep, 18(1), 58-70.

Riley, J. L., 3rd, Benson, M. B., Gremillion, H. A., Myers, C. D., Robinson, M. E., Smith, C. L.,
Jr., et al. (2001). Sleep disturbance in orofacial pain patients: pain-related or emotional
distress? Cranio, 19(2), 106-113.

Riley, J. L., 3rd, & Robinson, M. E. (1997). CSQ: five factors or fiction? Clin JPain, 13(2), 156-
162.

Robinson, M. E., & O'Brien, E. M. (in press). Chronic pain. In Handbook of rehabilitation
psychology (2nd edition ed.).

Robinson, M. E., & Riley III, J. L. (1999). The role of emotion in pain. In R. J. Gatchel & D. C.
Turk (Eds.), Psychosocialfactors in pain (pp. 74-88). New York: Guilford Press.

Robinson, M. E., Riley, J. L., 3rd, Myers, C. D., Sadler, I. J., Kvaal, S. A., Geisser, M. E., et al.
(1997). The Coping Strategies Questionnaire: a large sample, item level factor analysis.
Clin JPain, 13(1), 43-49.

Robinson, M. E., Wise, E. A., Riley III, J. L., & Atchison, J. W. (1998). Sex differences in
clinical pain: A multisample study. Journal of Clinical Psychology in Medical Settings,
5(4), 413-424.

Roelofs, J., McCracken, L., Peters, M. L., Crombez, G., van Breukelen, G., & Vlaeyen, J. W.
(2004). Psychometric evaluation of the Pain Anxiety Symptoms Scale (PASS) in chronic
pain patients. JBehav Med, 27(2), 167-183.

Rosenstiel, A. K., & Keefe, F. J. (1983). The use of coping strategies in chronic low back pain
patients: relationship to patient characteristics and current adjustment. Pain, 17(1), 33-44.

Russell, I. J., Michalek, J. E., Vipraio, G. A., Fletcher, E. M., Javors, M. A., & Bowden, C. A.
(1992). Platelet 3H-imipramine uptake receptor density and serum serotonin levels in
patients with fibromyalgia/fibrositis syndrome. JRheumatol, 19(1), 104-109.

Sadeh, A., Hauri, P. J., Kripke, D. F., & Lavie, P. (1995). The role of actigraphy in the
evaluation of sleep disorders. Sleep, 18(4), 288-302.

Sahota, P. K., & Dexter, J. D. (1990). Sleep and headache syndromes: a clinical review.
Headache, 30(2), 80-84.









Sastre, J. P., Buda, C., Kitahama, K., & Jouvet, M. (1996). Importance of the ventrolateral region
of the periaqueductal gray and adjacent tegmentum in the control of paradoxical sleep as
studied by muscimol microinjections in the cat. Neuroscience, 74(2), 415-426.

Sateia, M. J., Doghramji, K., Hauri, P. J., & Morin, C. M. (2000). Evaluation of chronic
insomnia. An American Academy of Sleep Medicine review. Sleep, 23(2), 243-308.

Sayar, K., Arikan, M., & Yontem, T. (2002). Sleep quality in chronic pain patients. Can J
Psychiatry, 47(9), 844-848.

Schaefer, K. M. (1995). Sleep disturbances and fatigue in women with fibromyalgia and chronic
fatigue syndrome. J Obstet Gynecol Neonatal Nurs, 24(3), 229-233.

Seigel, J. M. (1994). Brainstem mechanism generating REM sleep. In M. H. Kryger, T. Roth, D.
W. C. & e. al. (Eds.), Principles and Practice of Sleep Medicine (2nd ed., pp. 122-144).
Philadelphia, PA: W. B. Saunders.

Shapiro, C. M., Devins, G. M., & Hussain, M. R. (1993). ABC of sleep disorders. Sleep
problems in patients with medical illness. Bmj, 306(6891), 1532-1535.

Shaver, J. L., Lentz, M., Landis, C. A., Heitkemper, M. M., Buchwald, D. S., & Woods, N. F.
(1997). Sleep, psychological distress, and stress arousal in women with fibromyalgia. Res
Nurs Health, 20(3), 247-257.

Smith, B. H., Elliott, A. M., Chambers, W. A., Smith, W. C., Hannaford, P. C., & Penny, K.
(2001). The impact of chronic pain in the community. Fam Pract, 18(3), 292-299.

Smith, M. T., Edwards, R. R., McCann, U. D., & Haythornthwaite, J. A. (2007). The effects of
sleep deprivation on pain inhibition and spontaneous pain in women. Sleep, 30(4), 494-
505.

Smith, M. T., & Haythornthwaite, J. A. (2004). How do sleep disturbance and chronic pain inter-
relate? Insights from the longitudinal and cognitive-behavioral clinical trials literature.
Sleep MedRev, 8(2), 119-132.

Smythe, H. A. (1995). Studies of sleep in fibromyalgia; techniques, clinical significance, and
future directions. Br JRheumatol, 34(10), 897-899.

Spielberger, C. (1988). State-Trait Anger Expression Inventory, research edition. Professional
manual. Odessa, FL: Psychological Assessment Resources.

Staud, R., Cannon, R. C., Mauderli, A. P., Robinson, M. E., Price, D. D., & Vierck, C. J., Jr.
(2003). Temporal summation of pain from mechanical stimulation of muscle tissue in
normal controls and subjects with fibromyalgia syndrome. Pain, 102(1-2), 87-95.









Steriade, M., & Llinas, R. R. (1988). The functional states of the thalamus and the associated
neuronal interplay. Physiol Rev, 68(3), 649-742.

Swartzman, L. C., Gwadry, F. G., Shapiro, A. P., & Teasell, R. W. (1994). The factor structure
of the Coping Strategies Questionnaire. Pain, 57(3), 311-316.

Tait, R. C., Chibnall, J. T., & Krause, S. (1990). The Pain Disability Index: psychometric
properties. Pain, 40(2), 171-182.

Tait, R. C., Pollard, C. A., Margolis, R. B., Duckro, P. N., & Krause, S. J. (1987). The Pain
Disability Index: psychometric and validity data. Arch Phys Med Rehabil, 68(7), 438-
441.

Takahashi, L. K. (2002). Neurobiology of schizophrenia, mood disorders, and anxiety disorders.
In K. L. McCance & S. E. Huether (Eds.), Pathophysiology: The biologic basis for
disease in adults and children (4th edition ed., pp. 550-565). St. Louis: Mosby.

Turk, D. C., Rudy, T. E., & Salovey, P. (1985). The McGill Pain Questionnaire reconsidered:
confirming the factor structure and examining appropriate uses. Pain, 21(4), 385-397.

Turner, J. A., Mancl, L., & Aaron, L. A. (2006). Short- and long-term efficacy of brief cognitive-
behavioral therapy for patients with chronic temporomandibular disorder pain: a
randomized, controlled trial. Pain, 121(3), 181-194.

Tuttle, D. H., Shutty, M. S., & DeGood, D. E. (1991). Empirical dimensions of coping pain
patients: A factorial analysis. Rehabilitation Psychology, 36(3), 179-188.

Vaeroy, H., Helle, R., Forre, 0., Kass, E., & Terenius, L. (1988). Cerebrospinal fluid levels of
beta-endorphin in patients with fibromyalgia (fibrositis syndrome). JRheumatol, 15(12),
1804-1806.

Vallieres, A., & Morin, C. M. (2003). Actigraphy in the assessment of insomnia. Sleep, 26(7),
902-906.

Verhaak, P. F., Kerssens, J. J., Dekker, J., Sorbi, M. J., & Bensing, J. M. (1998). Prevalence of
chronic benign pain disorder among adults: a review of the literature. Pain, 77(3), 231-
239.

Verma, S., & Gallagher, R. M. (2002). The psychopharmacologic treatment of depression and
anxiety in the context of chronic pain. Curr Pain Headache Rep, 6(1), 30-39.

Waters, W. F., Hurry, M. J., Binks, P. G., Carney, C. E., Lajos, L. E., Fuller, K. H., et al. (2003).
Behavioral and hypnotic treatments for insomnia subtypes. Behav Sleep Med, 1(2), 81-
101.









Widerstrom-Noga, E., Dyrehag, L. E., Borglum-Jensen, L., Aslund, P. G., Wenneberg, B., &
Andersson, S. A. (1998). Pain threshold responses to two different modes of sensory
stimulation in patients with orofacial muscular pain: psychologic considerations. J
Orofac Pain, 12(1), 27-34.

Widerstrom-Noga, E. G., Felipe-Cuervo, E., & Yezierski, R. P. (2001). Chronic pain after spinal
injury: interference with sleep and daily activities. Arch PhysMedRehabil, 82(11), 1571-
1577.

Wilson, K. G., Watson, S. T., & Currie, S. R. (1998). Daily diary and ambulatory activity
monitoring of sleep in patients with insomnia associated with chronic musculoskeletal
pain. Pain, 75(1), 75-84.

Wittig, R. M., Zorick, F. J., Blumer, D., Heilbronn, M., & Roth, T. (1982). Disturbed sleep in
patients complaining of chronic pain. JNerv Ment Dis, 170(7), 429-431.

Wolfe, F., Anderson, J., Harkness, D., Bennett, R. M., Caro, X. J., Goldenberg, D. L., et al.
(1997). A prospective, longitudinal, multicenter study of service utilization and costs in
fibromyalgia. Arthritis Rheum, 40(9), 1560-1570.

Wolfe, F., Smythe, H. A., Yunus, M. B., Bennett, R. M., Bombardier, C., Goldenberg, D. L., et
al. (1990). The American College of Rheumatology 1990 Criteria for the Classification of
Fibromyalgia. Report of the Multicenter Criteria Committee. A/ ih ti, Rheum, 33(2), 160-
172.

Yatani, H., Studts, J., Cordova, M., Carlson, C. R., & Okeson, J. P. (2002). Comparison of sleep
quality and clinical and psychologic characteristics in patients with temporomandibular
disorders. J Orofac Pain, 16(3), 221-228.









BIOGRAPHICAL SKETCH

Erin Maureen O'Brien was born in 1978 in Piscataway, New Jersey. The oldest of three

daughters, she grew up in Marlton, New Jersey and graduated from Cherokee High School in

1996. She attended Saint Joseph's University in Philadelphia, Pennsylvania, and was accepted

into a 5-year B.S./M.S. program, where she earned a B.S. in psychology and an M.S. in

experimental psychology in 2000 and 2001, respectively. Her master's thesis was titled, "Sleep

and Risk-Taking Behavior in Adolescents" and this work was published in 2005.

Upon graduating in 2001 with her M.S. in experimental psychology, Erin obtained a

position as a clinical research coordinator in the Center for Sleep and Respiratory Neurobiology

at the University of Pennsylvania. Here she coordinated the progression of several large NIH-

funded clinical research protocols under the direction of Dr. Allan Pack. This experience

solidified Erin's interest in pursuing additional study in the field of psychology and resulted in

her application and acceptance into the doctoral program at the University of Florida so that she

could earn her Ph.D. in clinical psychology. Here she pursued her dual interests in the study of

sleep disturbances and chronic pain, which culminated in her dissertation research.

As part of her doctoral program, Erin completed a year-long clinical internship at the

Warren Alpert Medical School at Brown University where she pursued additional training in the

area of behavioral medicine. Upon completion of her Ph.D. program, Erin will be continuing her

research and clinical work as a post-doctoral fellow in the Methods to Improve Diagnostic

Assessment and Services (MIDAS) clinical-research program at Brown University, with a focus

on working with patients presenting with insomnia and other sleep disorders.





PAGE 1

1 SLEEP DISTURBANCE IN CHRONIC PAIN PATIENTS By ERIN M. OBRIEN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Erin M. OBrien

PAGE 3

3 To my parents Youve cheered me on in all that I do, supported me when I need ed you to stand with me, given me the courage to the believe in myself and to stand on my own, and always told me that there was nothing I could not do. You are my inspiration now and always.

PAGE 4

4 ACKNOWLEDGEMENTS I thank m y supervisory committee for their suppo rt and mentoring. I especially thank my supervisory committee chair, Dr. Michael Robinson, for his continual mentoring and feedback. I would also like to thank my participants for their dedicated participation, and the MiniMitter/Respironics Corporati on for the provision of equipment for this research. Finally, I want to thank my family for their constant support and encouragement, without which this accomplishment would not have been possible.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS .............................................................................................................4LIST OF TABLES ...........................................................................................................................7LIST OF FIGURES .........................................................................................................................8LIST OF ABBREVIATIONS ......................................................................................................... .9ABSTRACT ...................................................................................................................... .............11 CHAP TER 1 INTRODUCTION .................................................................................................................. 13Prevalence and Impact of Chronic Pain .................................................................................. 13Subtypes of Pain Disorders .....................................................................................................15Back Pain .........................................................................................................................15Facial Pain .......................................................................................................................16Fibromyalgia and Rheumatic Conditions ........................................................................ 17Basic Sleep Information ....................................................................................................... ..18Assessment of Sleep Patterns and Sleep Problems ................................................................. 19Experimentally Induced Sleep Distur bances in Healthy Participants .................................... 22Sleep Disturbances in Pain Patients ........................................................................................23Back Pain .........................................................................................................................24Facial Pain .......................................................................................................................25Fibromyalgia and Rheumatic Conditions ........................................................................ 27Neurobiology Findings on the Sleep-Pain Relationship ......................................................... 29Sleep Disturbances and Experimental Pain Testing ............................................................... 31Negative Mood, Sleep, and Pain ............................................................................................. 32Current Study ..........................................................................................................................33Specific Aims ..................................................................................................................37Hypotheses .................................................................................................................... ..372 METHODS ....................................................................................................................... ......39Participants .................................................................................................................. ...........39Procedure ..................................................................................................................... ...........40Measures ...................................................................................................................... ...........42Graded Thermal Stimulation or RAMP and HOLD (RH) .............................................. 46Temporal Summation (Wind-up) ....................................................................................47Sleep Diary and Actigraphy ............................................................................................ 47Statistical Analyses .......................................................................................................... .......49

PAGE 6

6 3 RESULTS ....................................................................................................................... ........51Differences in Subjective Sleep Across the Three Pain Groups ............................................. 51Role of Negative Mood in the Sleep-Pain Relationship ......................................................... 52Psychophysical TestingTemporal Summa tion/Wind up and Ramp and Hold .................. 54Actigraphy Data and Sleep Diary Data .................................................................................. 554 DISCUSSION .................................................................................................................... .....63Comparison of Study Results to Previous Findings ...............................................................68Role for Negative Mood in the Sleep-Pain Relationship ........................................................ 70Implications for Conceptualization and Treatment of Chronic Pain ...................................... 71Limitations ................................................................................................................... ...........74Future Directions ....................................................................................................................75LIST OF REFERENCES ...............................................................................................................77BIOGRAPHICAL SKETCH .........................................................................................................90

PAGE 7

7 LIST OF TABLES Table page 3-1 Demographic information for the 3 chronic pain patient groups .......................................59 3-2 Comparison of diaryand actigraphy-measur ed sleep variables across 3 pain groups .....60 3-3 Multivariate mixed model MANOVA results for pain ratings in tem poral summation and ramp and hold procedures ........................................................................................... 61 3-4 Correlation between actigraphyand di ary-m easured sleep variables for good and poor sleepers ......................................................................................................................62

PAGE 8

8 LIST OF FIGURES Figure page 3-1 Self-reported sleep (as reported by P SQI) for 3 chronic pain groups.. .............................. 57 3-2 Structural equation model for the rela tionship among sleep, pain, and negative m ood. ... 58

PAGE 9

9 LIST OF ABBREVIATIONS BDI Beck Depression Inventory BMI Body mass index BP Back pain CBT Cognitive behavior therapy CBT-I Cognitive behavior therapy for insomnia CNS Central nervous system CPH Chronic paroxysmal hemicrania CSQ Coping Strategies Questionnaire CSQ-R Coping Strategies Questionnaire Revised DSM-IV-TR Diagnostic and Statistica l Manual of Mental Disorders, 4th edition, text revision EEG Electroencephalogram/electroencephalographic FMS Fibromyalgia FP Facial pain HPA Hypothalamic-pituitary-adrenal ICSD International Classi fication of Sleep Disorders IL-1 Interleukin-1 MANOVA Multivariate analysis of variance MCV Medical College of Vi rginia Pain Questionnaire MPQ McGill Pain Questionnaire NK cell Natural killer cell NREM Non-rapid eye movement PASS Pain Anxiety Symptom Scale PDI Pain Disability Index

PAGE 10

10 PILL Pennebaker Inventory of Limbic Languidness PSG Polysomnography PSQI Pittsburgh Sleep Quality Index REM Rapid eye movement RM Raphe magnus RMSEA Root mean square error of approximation SE Sleep efficiency SEM Structural equation modeling SES Socio-economic status SF-36 Medical Outcomes Survey Short Form-36 SOL Sleep onset latency STAXI State-Trait Anger Expression Inventory TENS Transcutaneous elec trical nerve stimulation TMD Temporomandibular disorder TST Total sleep time VAS Visual analogue scale WASO Wake after sleep onset

PAGE 11

11 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SLEEP DISTURBANCE IN CHRONIC PAIN PATIENTS By Erin M. OBrien August 2008 Chair: Michael E. Robinson Major: Psychology Chronic pain is a prevalent problem and it is associated with a number of negative consequences. Sleep disturbances are a common complaint reported by chronic pain patients, with 50-70% of patients endorsing significant sleep disturbance. The presence of concomitant sleep problems can significantly complicate both the course and the management of chronic pain. Further, there is evidence to suggest that th e relationship between sleep disturbance and pain might be reciprocal, such that pain can disrupt sleep and poor or disrup ted sleep may lead to increased pain. Additionally, associations among pain, ne gative mood, and sleep disturbance among chronic pain patients have been inconsistent. Specifically, some inve stigators have reported greater negative mood (as well as higher pain in tensity) among self-report ed poor sleepers, whereas other researchers have reported higher pain ratings but no differences on measures of negative mood among good and poor sleepers. More clearly defining the relationship between pain and sleep disturbance, as well as the roll of negative mood, may furt her clarify the shared pathophysiology of sleep and pain. This study examined sleep, pain, and negative mood in 292 adults, 18 to 65 years of age, with chronic back pain, facial pain, or fibromya lgia. Additionally, a subset of 22 participants

PAGE 12

12 completed two weeks of daily sleep diaries and actigraphy monitoring and participated in psychophysical pain testing procedures. A generalized pattern of sleep disturban ce (difficulties with sleep onset, sleep maintenance, and poor sleep quality) was reported by all groups of chronic pain patients, with facial pain patients reporting relatively less distur bed sleep overall. Results also indicated a direct relationship between poor sleep and incr eased pain, and further revealed that negative mood mediated the relationship between poor slee p and increased pain when it was included in the model. No significant results emerged from analyses examining pain response to psychophysical testing among good and poor sleepers, although moderate to large effect sizes were found. Findings suggest multiple pathways between sleep disturbance and individuals pain experience, such that poor sleep may lead to increased pain but higher levels of negative mood may also lead to decreased sl eep, resulting in more pain.

PAGE 13

13 CHAPTER 1 INTRODUCTION Chronic pain conditions constitute a majo r challenge facing the health care system. These conditions are often associated with subs tantial disability and di stress for the individual and also result in significant burdens on the health care system, as well as considerable economic and social costs. Similarly, sleep disturbanc es are prevalent among the general population and can also substantially impact individuals physical, emotional, social, and occupational functioning, particularly when thes e difficulties evolve into chroni c problems. Recent attention has begun to focus on the relationship between sleep disturbances and chronic pain conditions. When these occur in concert, the resultant imp act on the individual is ev en greater, leading to higher levels of physical and emo tional disability and greater func tional impairments. Research needs to clarify the relationship between sleep di sturbances and chronic pain, in order to better understand the potential reciprocal influences of these conditions on one another, as well as to develop ways to intervene more effectively. As pain and sleep can be assessed in various ways, it is important to consider both subjective and objective measures of sleep, and to consider subjective pain reports as well as response to experimentally c ontrolled painful stimuli. The following review will examine the literature on thes e topics to date, and highlight the rationale and proposed contribution of the current study. Prevalence and Impact of Chronic Pain Chronic nonm alignant pain is a prevalen t problem among the adult population and is associated with a number of negative physical, emoti onal, and social consequences. A review of epidemiological studies conducted by Verhaak and colleagues (1998) f ound that the median prevalence rate of chronic pain for adults across all studies was 15% (Verhaak, Kerssens, Dekker, Sorbi, & Bensing, 1998). Subsequent ep idemiological studies have reported similar

PAGE 14

14 prevalence rates (Breivik, Collett, Ventafri dda, Cohen, & Gallacher, 2005; Moulin, Clark, Speechley, & Morley-Forster, 2002). Several studi es have noted higher rates among females and among older age groups (Moulin et al., 2002; B. H. Smith et al., 2001; Verhaak et al., 1998). Andersson and colleagues (1999) examined health care and medication use among Swedish adult reporting chronic pain (H. I. A ndersson, Ejlertsson, Leden, & Schersten, 1999). Results indicated that individuals with chronic pa in were more likely to consult a physician or physiotherapist, which is consistent with findings from other studi es (Breivik et al., 2005). In particular, perception of pain intensity was noted to be the most important factor motivating health-seeking behaviors, although ethnicity, SES, age, and depr essive symptoms were also found to be important. Another commonly report among chronic pain samp les is that, despite increased health care usage, many patients repo rt that their pain is inadequately managed (Breivik et al., 2005). Increased levels of chronic pain have b een associated with increasingly negative associations with employment, in terference with daily activities, and general health (B. H. Smith et al., 2001). Similarly, individuals with chronic pain also report decreased ability to participate in social and occupational acitivities due to pain (Breivik et al., 2005; Moulin et al., 2002). When it was examined, positive associations were also noted between chronic pain and psychological symptoms (Verhaak et al., 1998). Psychological disorders are common correlates of chronic pain conditions (B reivik et al., 2005; Duquesnoy, A llaert, & Verdoncq, 1998), and when assessed, negative effects of pain on se xual activity are also fr equently noted (Duquesnoy et al., 1998).

PAGE 15

15 Subtypes of Pain Disorders Back Pain Research has suggested that as m any as 80% of adults will experience significant back pain over the course of their liv es (Lanes et al., 1995), and recu rrence rates for low back pain have been reported to be as high as 85% (Bi nkley, Finch, Hall, Black, & Gowland, 1993). In an epidemiologic review of severa l cross-sectional studi es (evidence level 1) by Anderson (1999), the annual incidence of back pain of at least m oderate intensity and duration was estimated to be 10-15% among the adult population (G. B. Ander sson, 1999). Further, this review also estimated the point prevalence of back pain of at least moderate intensity and duration to be 1530% among adults. Low back pain is a commonly experienced pain condition, which i nvolves substantial social and economic costs (Malanga & Nadler, 1999). Grabois (2005) re ported that, in the United States, back pain accounts for expenditu res of $14 billion per year, 19 million physician visits, and half of all workers compensation cases (Grabois, 2005). In addition, approximately 10 million Americans are disabled by chronic low back pain, and 250 million workdays are lost per year due to this condition (Kriegler & Ashenberg, 1987). While 90% of patients with back pain recover over a 3 month period, the remaining 10% of patients have a slow recovery process, which involves large resource-intensive demands being placed on the healthcare system (G. B. Andersson, 1999). According to Gallagher and Verm a (1999), major depression commonly accompanies chronic pain and can increase patients level of impairment and disability (Gallagher & Verma, 1999). Thus, a review by Verma and Gallagher (2002) concluded that, if present, depression and anxiety need to be addressed in order to obtain a good functional outcome from treatment (Verma & Gallagher, 2002).

PAGE 16

16 Farasyn and Meeusen (2005) examined pressure pain thresholds in a group of 87 patients with sub-acute low back pain, compared to 64 healthy control subjects. Results indicated that the low back pain group had significantly lower pressure pain thresholds than the healthy controls at several body sites (Farasyn & M eeusen, 2005). Flor and colleagues (2004) found support for the idea of enhanced central and pe ripheral reactivity among chronic back pain patients, due to lower pressure pa in threshold and tolerance levels in these patients compared to partients with tension headaches or healthy cont rols (Flor, Diers, & Birbaumer, 2004). Clauw and colleagues (1999) found that, even after accounting for structural demographic, and psychosocial variables, pain se nsitivity (threshold a nd tolerance) accounted for a significant amount of variance in chronic lo w back pain patients pain sc ores (from SF-36 subscale) and functional status (Clauw et al., 1999). Facial Pain In a large random sample of adults, Macf arlane and colleagues (2002a) reported an overall prevalence of 26% for or ofacial pain, but found that on ly 46% of these individuals reported seeking professional ad vice regarding their pain (M acfarlane, Blinkhorn, Davies, Kincey, & Worthington, 2002). Prev alence of orofacial pain symp toms was noted to be higher among women, and among men and women in the 18-25 year old age group. A review of 23 epidemiological studies by Pau and colleague s (2003) reported similar estimates for the prevalence of oral and facial pain (40-44 %) (Pau, Croucher, & Marcenes, 2003). While gender was not associated with dental pain in this review, younger subjects and subjects from lower SES groups were noted to be more likely to report pain. The impact of orofacial pain includes, an inability to engage in normal activit ies, having to take time off from work due to pain, and higher levels of psychologic distress (Macfarlane, Blinkhorn et al., 2002; Macfarlane, Kincey, & Worthington, 2002).

PAGE 17

17 Experimental pain testing in facial pa in patient samples has produced conflicting findings. A study by Bragdon and colleagues (2 002) reported that women with TMD and women without pain did not exhi bit significant differences in pain threshold or tolerance measures to either heat or ischemic pain (Bragdon et al., 2002). Similarly, Curran and colleagues (1996) found no differences in pain or any psychological variables between TMD patients and matched controls undergoing a pressure-pain task (Curran, Carlson, & Okeson, 1996). However, Maixner and colleagues ( 1995) found significantly lower thermal pain thresholds (and a trend for lower thermal pain tolerance), as well as significantly lower ischemic pain threshold and tolerance levels in a group of TMD patients compared to matched controls (Maixner, Fillingim, Booker, & Sigurdsson, 1995). Moreover, findings from a study conducted by Widerstrom-Noga and colleagues (1998) sugges ted that psychological variables, such as stress or anxiety, can attenuated the degree of analgesia obtained from various methods (acupuncture, TENS) in patients with tooth pain (Widerstrm-Noga et al., 1998). Fibromyalgia and Rheumatic Conditions According to a m ulti-site prospective study conducted by Wolfe and colleagues (1997), the average yearly cost per patient (in 1996 dollars) was $2,274, although this number was affected by those patients who utilized services at a higher level (Wolfe et al., 1997). Overall, fibromyalgia patients reported attending an averag e of 10 outpatient medical visits per year, with this number increasing even further when non-tradit ional treatments were included. Additionally across the entire sample of fibromyalgia patients, an average of 2.7 dis ease-related drugs were used in each 6-month study period, and hospitali zation frequency averaged one stay every 3 years, with over half of the hos pitalizations stemming from fibromyalgia-related symptoms. The number of comorbid conditions re ported by fibromyalgia patients has been associated with the total costs and health care usage (Wolfe et al., 1997). Further, total cost and utilization level was

PAGE 18

18 associated with both functional disability ratings, and global disease-severity in this sample of fibromyalgia patients. Laursen and colleagues (2005) found significantl y lower pressure pain thresholds among chronic pain patients of divers e etiologies compared to hea lthy female controls, and noted significantly higher VAS ratings of habitual pa in among female fibromya lgia/whiplash patients compared to other female chronic pain patien ts (Laursen, Bajaj, Olesen, Delmar, & ArendtNielsen, 2005). Carli and colleagues (2002) examin ed pain thresholds, using different modalities of noxious stimuli (pressure, heat, cold, and isch emic), in fibromyalgia patients and control participants (Carli, Suman, Bias i, & Marcolongo, 2002). The fibrom yalgia patients were further divided into 5 groups, based on te nder point evaluations and how di ffuse their pain was noted to be. Fibromyalgia subjects had significantly lower heat pain thre shold and cold pressure pain threshold, and lower ischemic pain tolerance, comp ared to healthy subjects Additionally, 4 of the 5 fibromyalgia groups also demonstrated lower pressure pain threshold levels compared to healthy subjects. Similarly, in a study by Hurtig and colleague s (2001) differences were noted between fibromyalgia patients and he althy controls in cold and heat pain threshold levels, but not in perception levels for warmth and cold (H urtig, Raak, Kendall, Gerdle, & Wahren, 2001). Sleep quality was not examined in relationship to pain threshold measures in this study. Basic Sleep Information Insom nia symptoms are among the most co mmon sleep complaints, with prevalence estimates commonly reported among 20-40% of a dults (Bailey, 1997; Ohayon, 2002). Rates of insomnia symptoms are also generally noted to be higher among wo men, and among elderly individuals (Ohayon, 2002). The Diagnostic and Statistical Manua l of Mental Disorders, 4th edition, text revisions (DSM-IV -TR; (American Psychiatric A ssociation, 1994) description of insomnia includes difficulties with either the in itiation or maintenance or sleep, as well as

PAGE 19

19 complaints of nonrestorative sleep. In terms of duration, insomnia can be intermittent, last for a short duration, or become a chro nic problem (Bailey, 1997). Sleep bruxism is a movement disorder whos e primary symptom is clenching or grinding the teeth during sleep (Bailey, 1997), and prevalen ce estimates have indicated that 5-20% of the population is affected by this disorder (Glaros, 19 81). Patients with sleep bruxism present with almost daily complaints, which may include musculoskeletal pain or temporomandibular disorders (Bailey, 1997), and as the severity and ch ronicity of this condition increases, there is decreased likelihood of experien cing restorative sleep. Assessment of Sleep Patterns and Sleep Problems Sleep patterns and sleep problem s can be asse ssed in a number of different ways, each involving certain strengths and limitations. The three main methods of sleep assessment are subjective measures (e.g., questi onnaires, sleep diaries), behavi oral measures (e.g., actigraphy), and physiological measures (e.g., polysomnography; PSG), and the recent research has demonstrated the utility of employing a multidimensional approach to the study of sleep, particularly in the evaluation of insomnia complaints (see (M orin, 2003) for a review). Questionnaires involve low burden on participan ts, and so are widely used measures for the assessment of sleep, as well as factors related to sleep (e .g., emotional status, daytime sleepiness, functioning, quality of life). Daily sleep diaries are widely used measures of sleep patterns, especially in the evaluation of insomnia complaints, and can be especially useful tools for assessing individuals sleep patterns prior to, during, and following the implementation of a treatment for a sleep problem. Although da ily diary reports may evidence significant discrepancies with information obtained via PSG, they permit evaluation of individuals perception of their sleep, and allow for prospectiv e monitoring over longer periods of time.

PAGE 20

20 Actigraphy is a behavioral measure of sleep, and can also be useful for assessing an individuals response to treatme nt for a sleep problem. Research has demonstrated that actigraphy provides accurate estima tes of several global sleep para meters (such as total sleep time, time in bed, total wake time) relativ e to PSG, while remaining unobtrusive as a measurement device (Hauri & Wisbey, 1992; Sade h, Hauri, Kripke, & Lavie, 1995; Vallieres & Morin, 2003). Ancoli-Israel and colleagues (2003) reviewed rese arch on the use of actigraphy for sleep research, and reported correlati ons between actigraphy and polysomnography ranging from 0.81 to 0.97 for total sleep time; 0.61 to 0.7 8 for percent of sleep; 0.53 to 0.94 for sleep onset latency (Ancoli-Israel et al., 2003). It is important to note that correlations between actigraphy and PSG, and between ac tigraphy and sleep diaries, are often moderate, especially when estimating more specific sleep parameters (e.g., sleep latency, 0.12 to 0.69; time awake after sleep onset; 0.22 to 0.37) (de Souza et al ., 2003; Lockley, Skene, & Arendt, 1999). However, for the assessment of certain sleep problems, particularly the night-to-night variability prominent among insomnia patients, actigraphy may actually be a more appropriate diagnostic tool than traditional PSG (Ancoli-Israel et al., 2003). Additionally, a recent validation study conducted by Lichstein and colleagues (2006) concluded that actigraphy is a satisfactory objective measure of number of awakenings, wa ke time after sleep onset (WASO), total sleep time (TST), and sleep efficiency in insomnia patie nts, relative to PSG r ecordings (Lichstein et al., 2006). Kushida and colleagues (2001) reported that m easurement of total sleep time and sleep efficiency did not differ significantly between PSG data and combined data from actigraphy and questionnaire reports in a group of 100 consecutiv e sleep-disordered patients (Kushida et al., 2001). Further, when actigraphy parameters in cluded a high-threshold (low-wake-sensitivity)

PAGE 21

21 algorithm, the number of awakenings recorded was similar to those measured using PSG. These researchers concluded that both su bjective data and actigraphy data should be used to estimate sleep parameters in slee p-disordered patients. Studies have also compared the use of sleep logs and actigraphy in the assessment of sleep patterns (Lockley et al ., 1999; Wilson, Watson, & Currie, 1998). Lockley and colleagues (1999) demonstrated good correlations between sl eep logs and actigraphy for timing of sleep ( r = 0.77 for sleep onset and r = 0.88 for sleep offset) and sleep duration ( r = 0.57), as well as good agreement in measurement of changes in sleep patterns over time. However, these methodologies were not as highly correlated in measuring transitional variables during sleep, such as sleep latency and number and durati on of awakenings. Wils on and colleagues (1998) also compared sleep reports using a diary m easure and actigraphy in a group of patients with chronic musculoskeletal pain ( 82.5% back pain). Results indi cated that the two measures provided similar estimates of TST, WASO, sleep efficiency, but differed in estimation of sleep onset latency and number of awakenings. Pain severity was noted to be the variable that evidenced the strongest a ssociation with sleep disturbance ove rall (as measured by sleep diary). PSG is beneficial in the screening for sleep diso rders such as obstructive sleep apnea or periodic limb movement disorder, but is still not routinely used in the clinical evaluation of insomnia complaints (Reite, Buysse, Reynolds, & Mendelson, 1995; Sateia, Doghramji, Hauri, & Morin, 2000). Despite its usefulness for screening purposes, PSG involves significant burden on participants, which increases with the number of recording nights and if the recording is being conducted in a lab versus the individuals home. Furthermore, while PSG provides a wealth of valuable physiological data, it does not adequately address the subjective experience that is often a central component to insomnia complaints.

PAGE 22

22 An additional benefit provided by PSG evaluati on of sleep is the ability to examine not only the macrostructure of sleep architecture (stage REM, NREM st ages 1-4), but also the ability to assess sleep microstructure using advances in technology of EEG analysis. Sleep microstructure analysis permits examination of the proportion of different frequencies of brain waves (alpha, beta, gamma, delta, theta) throughout an individuals sleep. Different frequencies of brain waves are associated with different stages of sleep (for example, lower frequency delta waves are most often associated with stage 3 or 4 NREM slow wave sleep), and abnormalities in the brain wave patterns during sleep have been noted in indivi duals who complain of certain sleep disturbances. Experimentally Induced Sleep Distur bances in Healthy Participants Experim ental studies have found that selec tive disruption of stage 4 NREM sleep (slow wave sleep) in healthy particip ants led to musculoskeletal te nderness the next day in these participants; furthermore, these symptoms mimi cked the symptoms of fibromyalgia (Moldofsky & Scarisbrick, 1976). A more recent, and ve ry carefully conducted, study by Onen and colleagues (2001a) reported th at healthy males (PSQI and BMI in normal range) who had undergone 40 hours of sleep deprivation showed hyperalgesia to mechanical stimuli, but not to thermal stimuli (Onen, Alloui, Gross, Eschal lier, & Dubray, 2001). In addition, these participants also showed a robus t analgesic effect following select ive slow wave recovery sleep after undergoing slow wave sleep interruption. Additional studies have supported these findings of hyperalgesia to painful stimuli following sl eep deprivation in hea lthy subjects, with no changes noted in somatosensory threshold detec tion levels (Kundermann, Spernal, Huber, Krieg, & Lautenbacher, 2004). Experimental studies examining the effects of painful stimuli on sleep have consistently shown that pain causes microarousal, which was measured by increased high frequency EEG

PAGE 23

23 activity (in the alpha and beta ranges) at the expense of slow frequency EEG activity (delta range) (Drewes, Nielsen, Arendt-Nielsen, Birk et-Smith, & Hansen, 1997). These results have led researchers to conclude that pain can cause changes to the sleep architecture of normal controls in ways that lighten sleep and diminish the reputed restorative effects of slow wave sleep. However, results have not been consistent and no significant effects on either EMG activity (measuring bruxism) or on pressure pain thresholds were noted in a group of 10 healthy males undergoing slow-wave sleep deprivation (Arima et al., 2001). Sleep Disturbances in Pain Patients Sleep disturbances are among the most comm on complaints reported by patients experiencing chronically painful conditions (M. T. Smith & Haythornthwaite, 2004). Sleep disturbance is a term used to denote subjective reports of problems in either sleep quality or quantity (Cohen, Menefee, Doghramji, Anderson, & Frank, 2000). Studies commonly report that at least 50% of patients with diverse chronic pain conditions complain of significant sleep disturbance, with many studies reporting an ev en higher prevalence of sleep disturbance in chronic pain patients, along the order of 70% (Atkinson, Ancoli-Is rael, Slater, Garfin, & Gillin, 1988; Pilowsky, Crettenden, & Townley, 1985). Sleep efficiency (total sleep time/time in bed 100) has also been noted to be significantly a ffected in chronic pain patients (Wittig, Zorick, Blumer, Heilbronn, & Roth, 1982). The presen ce of concomitant sleep problems can significantly complicate both the course and the ma nagement of chronic pain patients (Cohen et al., 2000). Experimental data gathered from studi es of healthy participants and cross-sectional research in clinical populations suggests that the relationship be tween sleep disturbance and pain might be reciprocal. In some cases, disturbed or poor sleep ap pears to contribute to the pain problem, whereas in others ongoing pain diminishes individuals ab ility to sleep and leads to a cycle of increasing pain and continued degradat ion of sleep quality (Bailey, 1997). In other

PAGE 24

24 words, pain may disrupt sleep continuity and/or sleep quality, and poor sleep may contribute to the exacerbation of pain in these patients. Additional support for the hypothesis that sleep disturbance contribute s to chronic pain comes from clinical studies, which have demonstrated that chronic pain patients often demonstrate reduced delta (i.e. slow wave) sleep and/or increa sed alpha sleep (Harding, 1998; Moldofsky, Lue, & Smythe, 1983). The abnormalities found in these studies appear to be similar to the experimental studies in healthy participan ts, where sleep disruption appears to alter pain sensitivity. Similarly, correlati onal studies involving clinical sa mples have consistently found a positive association between sleep disturbance a nd pain severity (Pilowsky et al., 1985). Back Pain Studies in samples of chronic back pain pati ents have dem onstrated high rates of sleep problems, including difficulties with both the in itiation of sleep and difficulties with sleep maintenance (Lobbezoo, Visscher, & Naeije 2004; Widerstrom-Noga, Felipe-Cuervo, & Yezierski, 2001). Interp retation of these results is made di fficult by the fact that there was no indication of whether these groups were compos ed of different indivi duals, or whether the groups contained the same participants presenting with both types of sleep problems. The patients who reported frequent inte rference (3 or more times per week) in falling asleep due to pain also indicated higher average pain intens ity ratings, and used more descriptors when describing their pain. Other studi es involving patients with chr onic back pain have combined difficulties falling asleep and difficulties maintaining sleep into an overarching description termed insomnia, which prevents examinati on of patients specific sleep problems (Lobbezoo et al., 2004). Atkinson and co lleagues (1988) also examined potential influential factors involved in sleep disturbances among a sample of chronic low back pain patients. Results indicated that sleep dissatisfac tion was most strongly associated with greater depressed mood

PAGE 25

25 and shorter duration of pain. Furthermore, pa tients who reported high pain intensity also reported shorter TST, longer sleep latency, and mo re frequent awakenings, compared to the low pain intensity patients. Wilson and colleagues (1998) also noted that patients with chronic musculoskeletal pain (82.5% back pain) who ha d high pain severity re ported greater sleep impairment than those patients with low pain severity. Facial Pain Studies have docum ented sleep disturbances in patients with facial pain (Riley et al., 2001; Yatani, Studts, Cordova, Carlson, & Okes on, 2002), although the exact nature of these disturbances has not been well-defined. Rile y and colleagues (2001) conducted both crosssectional and longitudinal analyses to examine th e relationships between pa in, sleep disturbance, and depression in a sample of orofacial pain pati ents. Results indicated that reduced amount of sleep (sleep quantity) was asso ciated with depression and pain and reduced sleep quality was associated with negative affect. Additionally, wh en longitudinal analyses were conducted, initial depression and pain predicted sl eep disturbance at follow-up, and initial pain also predicted negative affect at follow-up. However, in this study, sleep at time one did not predict pain at follow-up. As reported by Bailey (1997), a variety of h eadache disorders have been found to be related to sleep disorders in different ways, which can be us eful in both the diagnosis and treatment of these disorders. In fact, the International Classi fication of Sleep Disorders (1990) (ICSD; (Diagnostic Classification Steering Comm ittee, 1990) includes a broad category of classification entitled Sleep-Relat ed Headaches under neurologic disorders, and this group of headaches is defined as occurring during sleep with their onset most often during REM sleep. The relationship between sleep disturbance and headache conditions is complex and difficult to assess, with symptoms of both conditions potentially having causal relations or having mutual

PAGE 26

26 reinforcements in an individual (Paiva, Batis ta, Martins, & Martins, 1995); see (Sahota & Dexter, 1990) for a detailed review ). The onset of certain types of headache conditions, such as chronic paroxysmal hemicrania (CPH), cluster headache, and migraine headache, are known to be associated with specific sleep stages (Sahota & Dexter, 1990). There ar e also well-established effects of sleep disruptions (B lau, 1990; Headache Classification Committee of the International Headache Society, 1988), as well as sleep distur bances (such as somnambulism, see (Paiva, Martins, Batista, Esperanca, & Martins, 1994), on headache conditions. In fact, the presence of early awakening or morning headaches has been conceptualized as possibly suggesting the presence of a sleep disturbance in some patients (Paiva et al., 1995). Additionally, patients with chronic headache complaints also indicate long-standing experiences of subjective sleep problems (Paiva, Esperanca, Martins, Batista, & Martins, 1992). However, complicating the picture is the fact that, for some individuals, sleep can also serv e as an effective treatment for some headaches, such as migraine (Blau, 1982). A review by Bailey (1997) identif ied the nature and prevalence of sleep disorders that are most often associated with orofacial pain c onditions, and indicated th at the presence of a concurrent sleep disorder in patients with a pain condition necessitates the treatment of both conditions. This review also reported that alpha intrusion on the EEG sleep recordings is one of the most consistent findings displayed by chronic pa in patients. This alpha rhythm is seen in a condition of relaxed wakefulness, but normally di sappears when an individual moves into stage 1 NREM sleep (Bailey, 1997). Alpha waves are also reportedly seen durin g arousals, which are frequently found in individuals with pain condi tions, and a condition termed alpha-delta sleep, which is associated with subjectively nonrestora tive sleep and feelings of fatigue, is also reportedly found among pain patients (Bailey, 19 97). While fibromyalgia has been strongly

PAGE 27

27 linked with this phenomenon of alpha intrusi on, and is reportedly the most common pain condition associated with subjectively nonres torative sleep, many re searchers are noting similarities between fibromyalg ia and myofascial pain. A dditionally, bruxism is commonly found in patients with TMD, and because of the wide range of symptoms that can be produced by TMD, sleep disturbances are often found in these patients (Bailey, 1997). In a study by Paiva and colleagues (1995), 13 of the 25 patients from an outpatient headache clinic who reported morning or noctu rnal headaches received a change in their diagnosis after PSG data were obt ained. In 10 patients, the onset (or period of worsening) of their headaches coincided with the onset of the sleep disturbances. However, the majority of cases in this study reflected a complex associa tion between patients sl eep disturbances and headache conditions. A subsequent study by Paiva and colleagues (1997) examined the sleep of patients with headache complaints who identifie d the onset of their headaches as occurring during the night or early morni ng at least 75% of time (Paiva Farinha, Martin s, Batista, & Guilleminault, 1997). In this sample, 53% of pati ents were identified as having a primary sleep disorder, and all of these pati ents reported fragmented slee p. There were no significant differences in total sleep time or percentage of REM sleep, as measured by polysomnography, found between the headache group with a sleep disorder and the headache group without sleep disorders. However, on a self-report sleep ques tionnaire, the patients who were identified as having a sleep disorder reported a greater number of sleep compla ints compared to the group of headache patients who did not evidence any objec tive evidence of a sleep disorder. Fibromyalgia and Rheumatic Conditions Subjective sleep com plaints, such as nonres torative sleep and insomnia complaints, among fibromyalgia patients have been reported by numerous studies (Campbell, Clark, Tindall, Forehand, & Bennett, 1983; Drewes et al., 1995; Schaefer, 1995). Further studies have reported

PAGE 28

28 links between sleep difficulties and pain in fibrom yalgia, such that reports of restful sleep have been associated with less reported discomfo rt and fatigue (Moldofsky, 1989) and nonrestorative sleep is associated with exacerbation of pain in fibromyalgia patients (Affleck, Urrows, Tennen, Higgins, & Abeles, 1996). Polysomnographic st udies have also found specific differences between the sleep architecture of fibromyalgia patients and that of h ealthy controls, including polysomnographic evidence of increased sleep onset latency (Horne & Shackell, 1991), increased amounts of stage 1 sleep (Horne & Shackell, 1991; Shaver et al., 1997), reduced amounts of stage 3 and 4 sleep (Branco, Atalai a, & Paiva, 1994; Horne & Shackell, 1991; Shapiro, Devins, & Hussain, 1993), and increased number of arousals (Branco et al., 1994; Jennum, Drewes, Andreasen, & Nielsen, 1993; Shapir o et al., 1993) in fibromyalgia patients. Shapiro and colleagues (1993) also reported that fibromyalgia pa tients evidenced lower amounts of age-corrected REM sleep, and total sleep tim e, and long awakenings (>10 minutes), and an EEG pattern of intrusive alpha fr equency waves, compared to healthy controls. As has been demonstrated in previous studies, patients with fibromyalgia frequently exhibit an alpha-delta sleep rhythm, which is also produced during stage 4 sleep deprivation, and by deep pain induced during sleep, in normal control subjects (Har ding, 1998). The alpha wave anomaly has long been hypothesized to be involved in the pa thophysiology of fibromyalgia (Moldofsky, Scarisbrick, England, & Smythe, 1975), and more recent data has supported the hypothesis that alpha intrusion is an inherent ch aracteristic of NREM sleep in fi bromyalgia patients (Branco et al., 1994; Smythe, 1995). This alpha intrusion sleep anomaly is associated with indications of vigilance during sleep and reports of nonresto rative sleep (Anch, Lue, MacLean, & Moldofsky, 1991), as well as pain, energy, and mood in fibr omyalgia patients (Moldofsky & Lue, 1980). The amount of alpha frequency that occurred duri ng sleep has also been shown to correlate with

PAGE 29

29 increases in overnight pain measures (Moldofs ky & Lue, 1980). Furthermore, Harding (1998) concluded that the compilation of evidence suggests this alpha intrusion is found in the majority of patients with fibromyalgia, the amount of alpha intrusion correlates with objective measurements of pain, and decreasing the amount of alpha intrusion results in improvements in pain. Landis and colleagues (2003) found that the women with fibromyalgia reported poorer sleep quality and more fatigue than controls, al though actigraphy sleep in dicators were not different between groups (Landis et al., 2003). In the women with fibr omyalgia only, selfreported sleep quality was directly related to ac tigraphy indicators of TST, and was inversely related to sleep fragmentation. Additionally, fatigue in the women with fibromyalgia was directly related to the actigraphy indicators of wake after sleep onset, and inversely related to sleep efficiency. According to Wolfe and colleagues (1990), nonr estorative sleep is a prevalent complaint among patients with fibromyalgia (Wolfe et al., 1990). Harding (1998) re viewed the literature regarding sleep in fibromyalgia, and further stated th at, although sleep disturbance is a prominent aspect in the clinical picture of fibromyalgia and that pain in fibromyalgia may increase due to a lack of sleep, it is still unclear whether sleep disturbance plays a causal role in fibromyalgia, or is simply an outcome resulting from the disorder. Neurobiology Findings on the Sleep-Pain Relationship Studies have im plicated a number of areas in the central nervous system, as well as several different chemical substa nces, in both the contro l/disturbance of the sleep/wake cycle and in the experience of chronic pain. The mesen cephalic periaqueductal gray area, the thalamus, and the reticular nucleus of the thalamus have b een implicated in the generation and maintenance of sleep, and also in pain m odulation. The mesencephalic periaqueductal gray area has been

PAGE 30

30 shown to modulate both sleep st ates (Sastre, Buda, Kitahama, & Jouvet, 1996) and nociception (Demarco, Baghdoyan, & Lydic, 2003). The thalamus is involved in both arousal and in the processing of nociceptive stimuli in the cortex (Casey, Morrow, Lorenz, & Minoshima, 2001), and the reticular nucleus of the thalamus is t hought to actively regulat e the synchronization of the cortex during delta sleep (Steriade & Llinas, 1988). A study by Mountz and colleagues (1995) demonstrated that fibromya lgia patients had reduced regiona l blood flow to the thalamus and caudate nucleus (Mountz et al., 1995), which may be involved in abnormalities of growth hormone secretion that have been observed in patients with fibromyalgia (Culebras & Miller, 1984). Bennett (1993) reported low levels of somatomedin C, which is a growth hormone responsible for muscle regeneration and homeost asis, in fibromyalgia patients, and further indicated that growth hormone is primarily s ecreted during stage 4 NREM (Bennett, 1993). Foo and Mason (2003) have argued that persistent pain, unlike acut e pain, is associated with functional changes in the raphe magnus (RM) ce lls, which modulate both pain and arousal (Foo & Mason, 2003). Paulson and colleagues (2002) have also hypothesized that persistent pain may lead to lasting functional cha nges in the neural systems that regulate both sleep and pain (Paulson, Casey, & Morrow, 2002). Specifically, they present data that sugge st that persistent pain might promote changes in the ascending arousal system, which could ultimately lead to disturbance of sleep continuity. Moldofsky theorizes that the diffuse my algia, fatigue, and psychological distress experienced by patients with fibrom yalgia are not only related to a disorder of their sleep-wake system, but also to circadian alterations of asso ciated biologic systems of the body. Specifically, studies have shown that these systems include, neurotransmitters (e.g. serotonin, substance P), neuroimmune and neuroendocrine (e.g. IL-1, NK cell activity, HP A and thyroid axes), and the

PAGE 31

31 autonomic nervous systems that are altered in patients with fibromyalgia (McAlpine, 1987; Moldofsky, 1994; Paulson et al., 2002; Pillemer, Bradley, Crofford, Moldofsky, & Chrousos, 1997). Several studies have report ed that patients with fibromyalg ia have decreased levels of serotonin in their cerebrospinal fluid and blood (R ussell et al., 1992), and there is evidence for an inverse relationship between pain levels and se rotonergic activity in the brain (Bailey, 1997). Furthermore, Moldofsky (1989) reported strong evidence from animal studies demonstrating a relationship between CNS metabo lism of serotonin and its role in regulating both pain and NREM sleep. Additionally, Carette and colleague s (1986) reported that lo w doses of tricyclic antidepressants, which influence the metabolism of serotonin in the centra l nervous system, have been found to be beneficial for sleep in patien ts with fibromyalgia (Carette, McCain, Bell, & Fam, 1986). Taken together, findings suggest that low levels of serotonin in fibromyalgia patients central nervous systems may play a role in their decreas ed delta (slow wave) sleep, and may predispose these patients to developi ng the alpha intrusion phenomenon. Unstable serotonin levels have also been proposed as a common factor in both migraine and somnambulism (Barabas, Ferrari, & Matthews, 1983), while Vaeroy and colleagues (1988) reported finding elevated cerebrosp inal fluid levels of substance P in patients with fibromyalgia (Vaeroy, Helle, Forre, Kass, & Terenius, 1988). Moreover, sleep and pain are both associated with activation of a number of regions in the central nervous system. Studies have suggested that the supraspinal regions, thal amocortical pathways, or the ante rior cingulated cortex may be involved in the interactio n between sleep and pain (Chase & Mo rales, 1994; Jones, 1994; Seigel, 1994). Sleep Disturbances and Experimental Pain Testing Onen and colleagues (2001b) reported that REM sleep deprivation in ra ts led to increased behavioral responses to noxious therm al, mechani cal, and electrical stimu li, but not to a noxious

PAGE 32

32 chemical stimulus (Onen, Alloui, Jourdan, Eschalier, & Dubray, 2001). These authors hypothesize that the normal duration of REM sl eep may be important for anti-nociceptive activity of endogenous and exogenous opioids. They further suggest that the nociceptive process may be affected by REM sleep deprivation, pr oducing a relative hypersensitivity to noxious electrical stimuli. It is known that both REM sleep and nociception are modulated by the cholinergic system. Additionally, as serotoni n is involved in both REM sleep and in pain mechanisms, these authors hypothesize that a possi ble serotonin depletion, due to increased serotonin metabolism, may partly induce a hyp eralgesia in REM sleep deprived animals. A number of studies have demonstrated that fibromyalgia patients e xhibit hyperalgesia in response to various pain stimu li (see (Hurtig et al., 2001), Table 1), compared to healthy controls. In addition, Agargun, et al. (1999) demonstrated a si gnificant negative correlation between pressure pain threshold and an overall measure of sleep qua lity, as well as measures of subjective sleep quality, sleep efficiency, and sleep disturbance in 16 fibromyalgia patients (Agargun et al., 1999). Negative Mood, Sleep, and Pain The associations between pain, depression, and sleep disturbance have been exam ined in several chronic pain patient samples (Affleck et al., 1996; Atkinson et al., 1988; Morin, Gibson, & Wade, 1998; Nicassio & Wallston, 1992). Some studies have reported that poor sleepers reported greater pain, but do not differ from g ood sleepers on measur es of depression or anxiety (Moffitt, Kalucy, Kalucy, Baum, & Cooke 1991; Morin et al., 1998). However, other studies have reported that poor sleepers have higher scores on measures of depression and anxiety, in addition to higher pain intensity a nd more physical disabili ty, compared to good sleepers (Atkinson et al., 1988; Pilowsky et al., 1985; Sayar, Arikan, & Yontem, 2002).

PAGE 33

33 Longitudinal analyses have supported pain predicting sleep (A tkinson et al., 1988; Nicassio & Wallston, 1992), althoug h results for sleep predicting pain over time have been less consistent (Nicassio & Wallston, 1992). Additionally, sleep problem s and pain have been shown to be predictive of depressi on over time in chronic pain populations (Nicassio & Wallston, 1992). Conversely, measures of depression have also been found to be predictive of sleep in chronic pain samples (Sayar et al., 2002). Chiu and colleagues (2005) examined the re lationship between psychological factors, sleep disturbances, and pressure pain threshold in a populatio n study, stratifying the sample by presence and extent of current pain (Chiu et al ., 2005). Results indicated that those participants who reported the greatest sleep disturbance and highest levels of depr ession had a 2-fold increased chance of being in the lowest tertile for pain threshold. Importantly, these two variables were found to be independently associated with lo wer pain threshold, and this relationship remained even after adjusting for participants initial pain status. Wittig and colleagues (1982) examined polysomnographically-meas ured sleep patterns in a group of pain patients, and compared these to findings in a gr oup of patients with in somnia secondary to a psychiatric disorder and a gr oup of patients with subjective insomnia complaints, but no objective findings of sleep disturbance. The pa in patients were found to have more difficulty initiating and maintainin g sleep compared to the group with subjective insomnia complaints; however, the group with insomnia secondary to a psychiatric disorder evidenced poorer sleep efficiency and greater early morning awake time th an the pain patients. Additionally, 8 of the 26 pain patients demonstrated alpha rhythm intrusion into NREM stages of sleep. Current Study Defining the relationships betw een particular pain conditions and sleep disturbance may further clarify the shared pathophysiology of sleep and pain. There have not yet been any

PAGE 34

34 comparative studies simultaneously carried out in patients with di fferent types of chronic pain conditions to determine whether there are physiol ogical sleep differences among such patients. Previous investigations have examined specific groups of chronic pain patients, or unspecified mixed groups of patients, w ithout examining whether any di fferences exist among patients with different pain conditions. Although the me chanism is not well studied, pain sensations could interrupt sleep via direct or indirect pathways. If specific relationships between chronic pain conditions and sleep disturbance can be identified, this coul d clarify the shared physiological underpinnings of sleep and pain, an d treatments that target the specific sleep disruptions in a particular chronic pain condition could be developed. Furthermore, although several investigations have examined sleep patterns in chronic pain patients, these investigati ons have used a variety of meas ures which limits the comparison of results across various studies. Similarly, se veral studies have used measures that were idiosyncratically generated and/or not validated, reducing confidence in the robustness of their findings. Sleep problems appear to be a ubi quitous finding among samples of chronic pain patients. In order to accurately and reliably delineate the nature of this relationship, as well as the influence of negative mood on this relations hip, it is essential to use valid measurement instruments when assessing these pa tients. Previous studies have al so indicated vary ing levels of agreement between different measures of sleep, us ually related to the variable being considered (e.g., total sleep time vs. number of awakenings). This suggests that further information is needed to determine the utility a nd adequacy of different measures of individuals sleep patterns and sleep problems. Finally, previous investigations have indicat ed that individuals s ubjective pain reports and their response to experimentally-induced painfu l stimuli are not perfectly correlated. Both of

PAGE 35

35 these variables are important to ex amine, as each represents an important target for intervention in chronic pain populations. Additionally, differences among types of experimental pain procedures have been noted. Thermal pain appears to be an appropriate means of measuring pain sensitivity among chroni c pain patients. The present study addressed some of the limitations noted in previous research examining the relationship between sleep dist urbance and chronic pain. Firs t, different groups of chronic pain patients were recruited and completed the sa me procedures. This allows for comparison of results across different pain groups. Second, the self-report instruments used to measure pain, mood, and sleep are all valid and re liable instruments, which permits increased confidence in the findings of this study. Furthermore, sleep was measured using various methodologies (questionnaire, sleep diary, actigraphy) in a portion of the participan ts, which allows for comparisons across these methods to determine thei r level of agreement, as well as their utility for measuring different sleep va riables within a chr onic pain population. Similarly, pain was measured using various methodologies, includi ng self-report (VAS), questionnaires, and psychophysical testing. This again permits examin ation of the relationships among these various measures and provides a comprehensive pict ure of participants pain experience. It was hypothesized that the pr evalence of various types of sleep disturbances would differ between the different groups of chroni c pain patients under study, underscoring the importance of implementing targeted treatments for specific sleep problems in these populations instead of assuming that a one-size-fits-all appr oach is appropriate. The current multimodal approach to nonpharmacological treatm ent of sleep disturbances has been shown to be effective; however, identifying the active treatment componen ts in these multimodal packages will enable more efficient and cost-effective treatment delivery, which can ultimately target the specific

PAGE 36

36 sleep problem being experienced by patients. Th ere has been some evidence to suggest the differential effectiveness of speci fic treatment components for partic ular sleep problems, such as difficulties with sleep onset versus problems w ith sleep maintenance (Espie, Brooks, & Lindsay, 1989; Harvey, 2000; Waters et al., 2003). Further, examination of the reported sleep disturbances among these chronic pain populations was undertaken to investigate inte ractions between the pain condition and the specific sleep disturbance that is being reported, such that the pa in condition may be maintaining the sleep problem, the sleep problem may be exacerbating the pain conditions, or some reciprocal influence may be occurring between th e sleep problem and the pain condition. It was also deemed important to note whether all 3 groups differed in the sleep di sturbances that they presented, or if there was one group (such as fi bromyalgia patients) th at was distinguishable from the other two groups. If present, the emerge nce of this pattern of results might have lent support for a greater role of the central nervous system in mediating both the painful symptom presentation, as well as the part icular type of sleep problem th at is reported, by such a group of patients. Alternatively, if the pa ttern of results suggested that a ll 3 groups were distinct from one another, future studies should aim to identify wh at mechanisms may be underlying both the pain condition and the associated sleep problems in each group of patients. Along these lines, it was hypothesized that diffe rences in pain sensitivity and sleep patterns would be found among these groups of ch ronic pain patients, supporting the hypothesis of different mechanisms underlying these painfu l conditions. For inst ance, conditions with greater central nervous system involvement may involve both greater pain hypersensitivity and a more pervasive type of sleep disturbance, such as non-restorative sleep. As experimental studies have demonstrated hyperalgesia in healthy part icipants following sleep deprivation (Moldofsky

PAGE 37

37 & Scarisbrick, 1976; Onen, Alloui, Gross et al., 2001), it was hypothesized that greater hypersensitivity to experimental stimuli in pain patients in the present st udy would lend further support to the hypothesis of a shared mechanis m underlying both the processing of painful stimuli as well as regulation of individuals sleep. As described above, the presence of sleep dist urbance in chronic pain patients warrants both evaluation and treatment in order to produce the most successful treatment outcome. Improved understanding of the underlying mechanis ms of various chronic pain conditions, as well as their involvement in the sleep disturbances pr esented in these patients, may lead to improved pain management and functiona l outcome among chronic pain populations. Specific Aims Specific aim 1: To m easure subjective sleep quality in different populations of chronic pain patients using validated measures, and to identify the types of sleep disturbances present in each population. Specific aim 2: To examine the relationship between sleep, pain, and negative affect in order to better understand the relationships among these variables. Specific aim 3: To determine if chronic pain pa tients who reported concurrent sleep disturbances had greater sensitivity to pain ful stimuli compared to chronic pain patients who did not report concurre nt sleep di sturbances. Hypotheses Hypothesis 1: Although it was expected that sleep disturbances would be reported by m any of the chronic pain patients in this study, it was hypothesized th at there would be an unequal distribution of specific types of sleep disturbances across groups. Specifically, it was hypothesized that facial pain patients would report more difficulty initiating sleep (greater sleep latency), back pain patients would report more difficulty maintaining sleep (greater sleep disturbances), and fibromya lgia patients would report a greater prevalence of non-restorative sleep (l ower sleep quality), compared to the other groups of pain patients. Hypothesis 2: It was hypothesized that the relations hip between sleep and pain in this sample of chronic pain patients would be partially mediated, when the influence of negative affect was in cluded in the model.

PAGE 38

38 Hypothesis 3: It was predicted that those patients who reported poor sleep (as measured by the PSQI) would show hypersensitivity (highe r ratings of pain intensity) during the ramp and hold pain testing procedures, a nd greater temporal summation compared to those chronic pain patients who reporte d good sleep (as measured by the PSQI). Hypothesis 4: It was predicted that participants who reported sleep disturbances would evidence longer sleep latency, more wake time after sleep onset, and lower sleep efficiency, as measured by both actigraphy and sleep diary measures, compared to participants who reporte d no sleep disturbances. Hypothesis 5: It was hypothesized that participants who did not report sleep disturbances would demonstrate higher correlations betw een actigraphy and sleep diary variables than participants who reporte d sleep disturbances.

PAGE 39

39 CHAPTER2 METHODS Participants Participants in this study in cluded 292 individuals with a cu rrent chronic pain condition; 116 patients with chronic facial pain, 55 patients with chronic ba ck pain, and 121 patients with f ibromyalgia. All participants we re between 18 and 65 years of age ( M =46.67, SD =12.05). The sample included 51 males and 241 females; and of those participants who indicated ethnicity, 88.2% were Caucasian. Additionally, a subgroup of participants was recruited that included 22 females (10 good sleepers, 12 poor sleepers) with a current chronic pa in condition: 8 with chronic facial pain, 8 with chronic back pain, and 6 with fibromyalgia Subgroup participants were designated as good or poor sleepers based on their total score from the Pittsburgh Sleep Quality Index (PSQI), using the modified cuto ff score suggested by Carpenter and Andrykowski (1998). The average age of the s ubgroup participants was 43.77 years ( SD =14.13 years), and 81.8% of subgroup participants identifi ed themselves as Caucasian. Participants were recruited from three groups of chronic pain patients at the University of Florida. Patients with chronic back pain were recruited from the Spin e Care Center, patients with chronic facial pain were recruited from the Facial Pain Clinic, and patients with fibromyalgia were recruited from the Fibromya lgia Clinic. Participants who completed the subgroup procedures were recruite d directly from the pain clin ics described above, as well as through printed advertisements posted on the University of Florida campus. Power analyses were conducted to determine th e number of participan ts needed to detect an effect based on findings from previous stud ies using the PSQI with similar populations of pain patients. Specifically, comparisons were made based on PSQI global scores reported by Agargun and colleagues (1999) for fibromyalgia patients, Yatani and colleagues (2002) for

PAGE 40

40 patients with temporomandibular disorder, and Menefee and colleagues (2000) for patients with back pain (Agargun et al., 1999; Menefee et al., 2000; Yatani et al., 2002). Based on these analyses, effect sizes of 0.53 and 0.62, were obt ained in comparisons of PSQI global scores between these groups of chronic pain patients. Therefore, with power set at 0.80, and an alpha value of 0.05, it was determined that approximate ly 40 participants would be needed in each group of chronic pain patients in order to detect similar effects. Therefore, recruitment efforts attempted to secure at least 40 participants from each of the chronic pain clinics to ensure adequate power in this study. Procedure Patients provided dem ographic information a nd information related to their health and pain condition. This included the participants age, sex, ethnicity, duration of pain, education, current medications, and pain ratings. In addition, pa rticipants completed a standard questionnaire packet as part of their clinical assessment, incl uding measures of pain, negative affect, coping strategies, somatic focus, and disability. These questionnaires are described below. The first pain measure was either the McGill Pain Questionnaire (MPQ; (Melzack, 1975), a self-report questionnaire that assesses the se nsory, affective, and evaluative dimensions of the pain experience, or the Medical College of Virginia Pain Questionnaire (MCV; (Price & Bushnell, 1994), which asks patients to provide visual analogue scale (VAS) ratings on pain intensity, pain unpleasantness, and function dimensions. Participants also completed the Coping Strategies Questionnaire Revised (CSQ-R; (R iley & Robinson, 1997), a se lf-report instrument measuring pain coping strategies. The third pa in measure that was completed by participants was the Pain Disability Index (PDI; (Pollard, 198 4), a brief self-report me asure of the degree to which pain interferes in seven life areas. The affective measures that participants completed included the Beck Depression Inventory (B DI; (Beck, Ward, Mendelson, Mock, & Erbaugh,

PAGE 41

41 1961), which assesses the experience of cognitive and affective, and neurovegetative symptoms of depression during the past week; the State-Trait Anger Expre ssion Inventory (STAXI; (Spielberger, 1988), which is used to assess both stat e anger symptoms and more general constitutional anger symptoms; the Pain-Anxiety Symptom Scale (PASS; (McCracken, Zayfert, & Gross, 1992), which provides an assessment of pain-related anxiety; the Medical College of Virginia Pain Questionnaire (M CV; (Price & Bushnell, 1994), whic h asks patients to provide visual analogue scale (VAS) ratings on mood, in addition to pain and function dimensions; and the Pennebaker Inventory of Limbic Languidness (PILL; (Pennebaker, 1982), a self-report symptom frequency checklist that assesses nonsp ecific common physical complaints. In addition to this standard questionnaire packet, particip ants also completed the Pittsburgh Sleep Quality Index (PSQI; (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989), a self-report measure of sleep quality and disturbances over a one-month interval. A subset of 22 participants (12 reporting sleep disturbances and 10 reporting no sleep disturbances) also underwent psychophysical testing to assess their pain sensitivity. Subgroup participants were Subjects underwent quantitative se nsory testing using a contact thermode applied to the volar surf ace of the forearm. The protocol th at was used enabled assessment of both first pain (primarily A-delta function) and second pain (prima rily C-fiber input). All thermal stimuli were delivered using a computer-contro lled Medoc Thermal Sensory Analyzer (TSA2001, Ramat Yishai, Israel), which is a peltier-element-based stim ulator. Temperature levels were monitored by a contactor-contained thermistor, and returned to a preset baseline of 32 deg C by active cooling at a rate of 10 deg C/Sec. All stimuli were delivered to the ventral left forearm. Alternating stimulation sites were used to prevent carryover effects due to local sensitization.

PAGE 42

42 Additionally, this subset of 22 participants also completed daily sleep diaries and actigraphy monitoring for two w eeks. The actigraphy monitori ng was used to provide an objective measure of these individuals sleep patt erns. The sleep diaries were completed twice per day (before participants go to bed, and again in the morning), and provided an additional measure of participants self-repo rted sleep patterns. Comparison of these participants reports provided an indication of the degree to which these measures converge. Measures Demographic/patient characteristics information : Patients provided infor mation pertaining to their sex, age, ethnicity, duration of pain, education, cu rrent medications, and pain ratings. Pittsburgh Sleep Quality Index (PSQI; (Buysse et al., 1989) : The PSQI is a self-report questionnaire that assesses sleep quality and disturbances over a 1-month time interval, and is designed to be used in clinical populations. Th is instrument is comprised of 19 items, which generate 7 component scores: s ubjective sleep quality, sleep late ncy, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and daytime dysfunction. Additionally, the sum of the scores for these 7 components yields one global score. The psychometric properties of this measure have b een shown to be sound, suggesting its utility in both clinical practice and research. This measure has been shown to have acceptable internal consistency (Cronbachs alpha = .83; (Buysse et al., 1989), and these authors further demonstrated that this measure has a diagnostic sensitivity of 85.5% and specificity of 86.5% in distinguishing good and poor sleepers. The psyc hometrics of the PSQI were subsequently evaluated by Carpenter and A ndrykowski (1998) in 4 different patient populations, including bone marrow transplant patients, renal transpla nt patients, women with breast cancer, and women with benign breast problems. Results indicated that the PSQI had good internal

PAGE 43

43 consistency (Cronbachs alpha coefficients = 0.80, across groups) and construct validity, and PSQI scores were found to be modera tely to highly correl ated with measures of sleep quality and sleep problems in these patients (Carpenter & Andrykowski, 1998). McGill Pain Questionnaire (MPQ; (Melzack, 1975): Th e MPQ is a self-report questionnaire of participants pa in experience. This instrument provides both an overall total pain score, as well as evaluations of the se nsory, affective, and ev aluative dimensions of participants pain experiences. The McGill Pain Questionnaire (M PQ) has a long history of use in pain research, and it is the most widely used instrument for evaluati ng pain (Melzack & Katz, 1992). The reliability of the MPQ was investigated by Love and colleagues (1989), and results demonstrated very strong test-rete st reliability coefficients of this measure in a group of low back pain patients who were tested on two occasi ons (Love, Leboeuf, & Crisp, 1989). Additionally, several studies have replicated the factor structure of the MPQ (Lowe, Walker, & McCallum, 1991; Turk, Rudy, & Salovey, 1985), and results from a study by P earce and Morley (1989) also demonstrated the construct validity of this measure (Pearce & Morley, 1989). Medical College of Virginia Pain Questionnaire (MCV; (Price & Bushnell, 1994): The MCV questionnaire asks patients to provide visual analogue scal e (VAS) ratings on pain, mood and function dimensions. These dimensions include measures of the pain experience itself, specifically pain intensity and pain unpleasantness, rated in reference to current levels, as well as highest, lowest, and usual levels dur ing the preceding week. In addition, negative feelings associated with the pain experience (i.e., depression anxiety, frustration, fear, and anger) are also rated, in reference to the previ ous week. Ratings were also provided regarding the extent to which pain prevented an individu al from doing what he/s he wanted to do, how

PAGE 44

44 difficult an individual found it to endure pain over time, and how concerned an individual is with his/her health. Beck Depression Inventory (BDI; (Beck et al., 1961): Th e BDI is a self-report measure of depression consisting of 21 items, which assess cognitive and affective, and neurovegetative symptoms of depression. This measure is designe d to determine the extent to which individuals currently exhibit or experience each of these sympto ms. Participants are instructed to indicate the statement in each item group that is most descri ptive of how they have been feeling during the past week, including the current day. Each item is scored on a scale that ranges from 0 to 3. The use of the BDI has been evaluated in psyc hiatric and nonpsychiatri c populations, and alpha coefficients ranging from .73 to .95 have been reported (Beck, Steer, & Ga rbin, 1988). The BDI is a well-validated assessment instrument for depression and it is an extensively used measure in experimental pain research. Pain Anxiety Symptom Scale (PASS; (McCracken et al., 1992): The PASS is a selfreport questionnaire consisting of 40 items assessing four dimensi ons of pain-related anxiety cognitive anxiety, escape/avoidance, fearful appraisal, and physio logical anxiety (McCracken, et al., 1992). Participants indicate to what extent the items are an accurate description of them on a 6-point scale, which ranges from never (0) to al ways (5). There are five items that are reverse scored. Previous studies have examined the relia bility and validity of this measure (McCracken & Dhingra, 2002; Roelofs et al., 200 4), and results have shown the PASS to be psychometrically sound. Cronbachs alpha coefficients of 0.94 have been reported in various samples of chronic pain patients (fibromyalgia, low back pain) (Roelofs et al., 2004). State-Trait Anger Expression Inventory (STAXI; (Spielberger, 1988): The STAXI is used to assess both state anger symptoms and mo re general trait-like or constitutional anger

PAGE 45

45 symptoms. The factor structur e of the STAXI has been suppor ted in analyses with a large sample of college students (D. G. Forgays, Fo rgays, & Spielberger, 1997) and a sample of middle-aged men and women (D. K. Forgays, Spielberger, Ottaway, & Forgays, 1998). The measurement properties of this assessment inst rument have been shown to be acceptable, including good reliability and adequa te validity (Kramer & Conoley, 1992). Pennebaker Inventory of Limbic Languidness (PILL; (Pennebaker, 1982): The PILL is a self-report measure of the occurrence and frequency of non-specific common physical complaints, and is used as a measure of somatic focus. It consists of 54 items, such as upset stomach, sore throat, headache, and nausea. The following response categories are provided: have never or almost never expe rienced the symptom, less than 3 or 4 times per year, every month or so, every week or so, and more th an once every week, and responses are indicated by a five-point Likert scale. This measure ev aluates commonly experienced symptoms over an unspecified time period in the past and assesses a general tendency to experience and report symptoms instead of the persons specific sympto m experience (Gijsbers van Wijk, van Vliet, & Kolk, 1996). Therefore, the PILL is conceptualized as a trait-like symptom scale that evaluates somatization or a general propensity to report physical symptoms (Pennebaker, 1982). When used with healthy subjects, a hi gh score is indicative of somati zation. Internal consistency for this measure is high (Cronbachs alpha = 0.91) (G ijsbers van Wijk et al., 1996). The PILL also has sufficient test-retest reliabil ity (r = 0.83) and was shown to co rrelate moderately with similar symptom scales (Pennebaker, 1982). Coping Strategies QuestionnaireRevised (CSQ-R; (Riley & Robinson, 1997): The CSQ-R is a reformulation of the original CSQ (Rosenstiel & Keefe, 1983), which was a rationally constructed instrument designed to a ssess pain coping and was formulated to measure

PAGE 46

46 the extent to which patients used six different cognitive coping strategies and two behavioral coping strategies. The CSQ-R retains 27 of the original 48 items of the CSQ, and proposes a 6factor solution. The items that were discarded from the original CSQ did not appear to possess good factor discrimination across se veral studies examining the fact or structure of this measure (Riley & Robinson, 1997; Swartzman, Gwadry, Shap iro, & Teasell, 1994; Tuttle, Shutty, & DeGood, 1991). The six-factor solution for the CS Q-R has been replicated, both in samples of chronic pain patients (Riley & Robinson, 1997; Robi nson et al., 1997) as well as in an ethnically diverse sample of healthy individuals (Hastie, Riley, & Fillingim, 2004) Additionally, this measure was found to have acceptable internal consistency (Cronbachs alpha = 0.72 to 0.91) across ethnic groups (Hastie et al., 2004). Pain Disability Inventory (PDI; (Pollard, 1984): The PDI is a 7-item measure of the degree to which chronic pain interferes with patients functioning in the following areas of life: family/home responsibilities, re creation, social activity, occupati on, sexual behavior, self care, and life-support activity (Pollard, 1984). An 11point scale ranging from 0 (no disability) to 10 (total disability) is used to i ndicate the amount of disability ex perienced in each of the domains listed above. The seven ratings are summed to compute a total score (0 70). The PDI has adequate psychometric properties w ith an internal consistency co efficient of .86 (Tait, Pollard, Margolis, Duckro, & Krause, 1987). Additionally, results reported by Tait, Chibnall, and Krause (1990) demonstrated the construct validity of the PD I in a large sample of chronic pain patients, and also indicated that this m easure had adequate test-retest re liability with a smaller group of pain patients undergoing inpatient treatm ent (Tait, Chibnall, & Krause, 1990). Graded Thermal Stimulation or RAMP and HOLD (RH) All the rmal stimuli were delivered using a computer-controlled Medoc Thermal Sensory Analyzer (TSA-2001, Ramat Yishai, Israel), whic h is a peltier-element-based stimulator. The

PAGE 47

47 temperature of the probe was calibrated immediatel y prior to each session. Visual Analog Scale (VAS) ratings of 4 graded intensities (45, 47, 49, 51 o C) of 3 second temperature stimuli were obtained in the following fashion. Stimuli were applied in random order to the forearm by a contact thermode and were 3 sec onds in duration. Several sites located on the forearms of both arms were employed. Stimulus presentation was ti med such that no site was stimulated with less than a 3-minute interval to avoid sensitization of the site. Participants rated 8 stimuli (2 at each intensity) using a VAS for pain intensity anchored at the right end by t he most intense pain imaginable. A second random sequence of 8 stimu li (2 at each intensity) was rated by VAS for pain unpleasantness (anchored at the right end by the most unpleas ant sensation imaginable.) This method of pain assessment has been shown to yield ratio scale measurement of clinical pain that is both internally consis tent and provides independent se nsory intensity and affective dimensions of experimentally induced pain (Price, Harkins, & Baker, 1987). Temporal Summation (Wind-up) Another m ethod of eliciting second pain was employed that mimics natural conditions of nociceptive thermal stimulation such as when one touches a hot object. Trains of 8 stimuli with an inter-stimulus interval of 3 seconds were used. The stimuli were pulsed from a baseline temperature of 45 Co to 52 Co. When rating sensory magnitude, th e participants were instructed to attend to the peak of late se nsations that occur approximately 1.5 to 2 seconds after the probe leaves the skin on each presentation. This type of stimulus presentation results in a temporal summation believed to be primarily C-fiber determined. Sleep Diary and Actigraphy Although actigraphy does not corr elate perfectly with polys omnographic m easurement of sleep, the use of actigraphy provi ded an objective measure of ce rtain sleep parameters, which enhanced the methodology of the present research study. As described previously, sleep can be

PAGE 48

48 measured in various ways, and current recomm endations involve using a multi-modal assessment of sleep patterns (meaning a combination of subj ective, behavioral, and physiological measures of sleep). Actigraphy data prov ided an objective measure of pa rticipants sleep without the burden inherent in polysomnography, while the sl eep diary captured pa rticipants subjective perception of certain sleep parameters. Both t ypes of information are important and can provide useful information during both assessment and treatment activities. Furthermore, assessment of the degree of agreement between these measures can illustrate the parameters for which these methodologies overlap, and the parameters about which each provided unique, and potentially important, information. Actigraphy assessment used a high-sensitivity algorithm. Variables obtained from actigraphy included: time in bed, total sleep ti me (TST), wake after sleep onset (WASO), number of awakenings, sleep effici ency, and sleep onset latency. For the sleep diary, participants were asked to complete the diary befo re they went to bed at night, and then again the following morning. The variables that were included in the morning sleep diary are listed below: Time participant got into bed Approximate time participant fell asleep (SOL) Wake time Time participant got out of bed Number and duration of any awakenings dur ing the night (number of awakenings and WASO) Rating of overall sleep quality using a visual analogue scale Participants were also asked to list any other factors that interfered with their sleep, such as pain or worries Visual analogue scale (VAS) for pain in tensity and pain unpleasantness upon waking The portion of the sleep diary th at participants completed in the evening (i.e., before bed) contained a different set of variables, including: Number and amount of caffeinated beverages consumed during the day

PAGE 49

49 Number and amount of alcoholic be verages consumed during the day Any medications taken during the day Amount and type of physical activity Number and duration of naps taken during the daytime or early evening Visual analogue scale (VAS) for pain intens ity and pain unpleasantness at bedtime Statistical Analyses Analysis 1: To determ ine if there were differences in sleep quality variables across type of chronic pain group, Kruskal-Wallis nonparametr ic tests for 3 independent samples were conducted for the following component scores from the PSQI across each of the three pain groups: sleep quality, sleep latenc y, and sleep disturbances. This nonparametric test was used due to the ordinal scale of the dependent variables in these analyses. Analysis 2: Structural Equation Modeling (SEM) was used to examine the role of negative affect in the relationship between sleep and pain. Mediation was indicated if there was a lower or non-significant path co efficient between the latent factor of sleep and the latent factor of pain, after the latent factor of negative affect was entered into the model. Analysis 3: A General Linear Model approach was used to conduct a mixed model analysis of variance to asse ss group (good sleeper vs. poor sleeper), heat pulse number, and group by heat pulse number interactions on pa in report in the te mporal summation/windup protocol. Analysis 4: A General Linear Model approach was used to conduct a mixed model analysis of variance to assess group (good sleeper vs. poor sleeper), temperature level, and group by temperature level interactions on pain report in the ramp and hold procedure. Analysis 5: A General Linear Model approach was used to conduct a mixed model analysis of variance to assess chronic pain group, heat pulse nu mber, and chronic pain group by heat pulse number interactions on pain re port in the temporal summation/windup protocol.

PAGE 50

50 Analysis 6: A General Linear Model approach was used to conduct a mixed model analysis of variance to assess chronic pain group, heat pulse nu mber, and chronic pain group by heat pulse number interactions on pain report in the ramp and hold procedure Analysis 7: To determine if there were differences in actigraphy and sleep diary variables for participants with and without complaints of sleep disturbances, 2 multivariate analyses of variance (MANOVA) were conducted. One MANOVA examined SOL, WASO, sleep efficiency, as measured by actigraphy, between pa rticipants who did and did not report sleep disturbances. The second MANOVA examined SOL, WASO, sleep efficiency, as measured by sleep diary, between participants who did and did not report sleep disturbances. Analysis 8: Pearson correlations were computed for SOL, WASO, TST, and sleep efficiency, as measured by actigraphy and sleep di ary, for participants who did and did not report sleep disturbances. Differences between groups were then tested using z-tests.

PAGE 51

51 CHAPTER 3 RESULTS Sam ple characteristics were examined to determine whether there were any significant demographic differences among the three groups of chronic pain patients. Significant differences were found across gr oups in years of education, ( F (2,157)=5.57, p<.01), with post hoc analyses indicating that facial pain (FP) patients had signif icantly more years of education than back pain (BP) patients. Significant differences in sex (2(2)=36.96, p <.001) and use of narcotic medications (2(2)=37.35, p <.001) were also found across the three chronic pain groups. No other significant differences in sample characteristics were found. Demographic characteristics of the sample and results of these analyses are provi ded in Table 3-1. Differences in Subjective Sleep Across the Three Pain Groups To examine whether there were differences in subjective sleep quality variables across the three chronic pain groups, Kruskal-Wallis nonparametric tests for 3 independent samples were performed. The dependent va riables in this analysis were the sleep quality, sleep latency, and sleep disturbance component scores from the PSQI; the Kruskal-Wallis nonparametric test was used due to the ordinal scale of these variab les. Results revealed significant differences among the three groups on PSQI sleep quality, 2(2) = 33.74, p<.001; PSQI sleep latency, 2(2) = 16.63, p<.001; and PSQI sleep disturbances, 2(2) = 44.28, p<.001. For each of the three PSQI components, post hoc analyses were conducte d using Mann-Whitney U tests in order to determine which groups were significantly differe nt from one another. For sleep quality component scores, facial pain patients had si gnificantly lower scores (indicating better sleep quality) compared to back pain patients, z = 3.89, p<.001; and fibromyalgia patients, z = 5.53, p<.001. For sleep latency component scores, facial pain patients had significantly lower scores (indicating shorter sleep latency) co mpared to back pain patients, z = 2.87, p<.01; and

PAGE 52

52 fibromyalgia patients, z = 3.83, p<.001. For sleep disturbances component scores, facial pain patients had significantly lower sc ores (indicating fewer sleep dist urbances) compared to back pain patients, z = 3.68, p<.001; and fibromyalgia patients, z = 6.33, p<.001; there was also a trend for back pain patients to have lower scores compared to fibromyalgia patients, z = 1.73, p<.10. These results are shown in Figure 3-1. Thes e findings are consistent with results from a one-way ANOVA comparing PSQI global scores for each of the three pain groups, F (2,262) = 25.86, p<.001, 2 = 0.17; post hoc Tukey tests indicated the same pattern of results (FP
PAGE 53

53 its influence on the sleep-pain relationship. Following the procedures of Baron and Kenny (1986), mediation was indicated if there was a reduced or non-significant path coefficient between the latent factor of sleep and the latent factor of pain, af ter the latent f actor of negative mood entered into the m odel (Baron & Kenny, 1986). The sleep factor was indicated by three obs erved variables: PSQI global score, sleep quantity (raw score) from the PSQI, and the sleep quality component score from the PSQI. The negative mood factor was indicated by four observe d variables, BDI total score, depression score from the MCV, PASS total score, and anxiety score from the MCV. The pain factor was indicated by three observed vari ables, MPQ total score, VAS for average pain intensity, and usual pain intensity from the MCV. The relationships between the latent factors were tested sequentially, in accordance with the procedures outlined by Baron and Kenny (1986) and path coefficients were examined to determine whether significant causa l relationships existed. The fi rst step involved testing the relationship between the sleep f actor and the pain factor. A significant relationship was found, path coefficient = -0.46, t = -3.72, indicating that poorer sleep predicts greater pain; the model was also found to provide a good fit for the data, 2(8)=3.85, p=0.87, RMSEA=0.00. Next, the relationship between the sleep factor and the negative mood factor was tested. A significant relationship was found, path coefficient = -0.57, t = -7.66, such that poore r sleep also predicts higher levels of negative affect; the model was also found to provide a good fit for the data, 2(13)=18.74, p=0.13, RMSEA=0.04. Third, the relationshi p between the negative mood factor and the pain factor was tested. A signifi cant relationship was found, path coefficient = 0.69, t = 3.63, indicating that higher levels of negative affect predict greater pain; howev er, the fit of this model did not provide an excellent fit for the data, 2(12)=40.59, p=0.00, RMSEA=0.09.

PAGE 54

54 Finally, the full model involving all 3 latent f actors was tested. Examination of the path coefficients in this final model, revealed a non -significant path between the sleep factor and the pain factor, path coefficient = -0.02, t = -0.18; a significant path betw een the sleep factor and the negative mood factor, path coefficient = -0.57, t = -7.72; and significant path between the negative mood factor and the pain factor, path coefficient = 0.99, t = 4.34. Results indicated that the full model fit the data well, 2(32)=41.77, p =0.12, RMSEA=0.03. This final model indicates that negative affect mediates the effect of sleep on pain in this sample of chronic pain patients. Poorer sleep also continued to predict higher levels of negative affect, and increased levels of negative affect continued to predict increased pain in the final model. The final structural equation model is shown in Figure 3-2. As stated above, mediation of the relationship between sleep and pain by ne gative mood was indicated by this final model. Specifically, the path coefficient between the slee p factor and the pain factor was reduced when negative mood was entered into th e model (-0.02), compared to th e path coefficient between the sleep factor and the pain factor without negative mood in the mode l (-0.46). This indicates that the direct relationship between sl eep and pain was significantly reduced when negative affect was also included in the analysis, indicating th at negative mood mediates the role of sleep on pain. Psychophysical TestingTemporal Summa tion/Wind up and Ramp and Hold A General L inear Model approach was used to conduct a series of mixed model analyses of variance. For the temporal summation proced ures, these analyses examined group (either: good sleeper vs. poor sleeper, or chronic pain gr oup), heat pulse number, and group by heat pulse number interactions on pain re port among facial pain, back pain, and fibromyalgia subgroup participants. For the ramp and hold procedur es, these analyses examined group (either: good

PAGE 55

55 sleeper vs. poor sleeper, or chronic pain group), temperature level, and group by temperature level interactions on pain inte nsity ratings and pain unpleasant ness ratings among facial pain, back pain, and fibromyalgia subgroup participants. Omnibus multivariate tests for the temporal summation procedures indicated no significant differences in pain re port across either good vs. poor sleepers, or across chronic pain groups. Omnibus multivariate tests for the ramp and hold procedures indicated no significant differences in pain intensity ratings or pain unpleasantness ratings across either good vs. poor sleepers, or across chronic pain groups. Si gnificantly higher pain intensity ratings and significantly higher pain unpleasantness ratings were found at higher temperatures, in all ramp and hold analyses. Results of these mixed model analyses of variance are provided in Table 3-3. Actigraphy Data and Sleep Diary Data A MANOVA examined sleep onset latency (SOL), wake after sleep onset (WASO), sleep efficiency, and total sleep time (TST), as meas ured by actigraphy, between subgroup participants who did and did not report sleep disturbances. The overall omnibus test was non-significant, F (4,17) = 0.79, p>.05, 2 = 0.16. A second MANOVA examined sleep onset latency (SOL), wake after sleep onset (WASO), sleep efficiency, and total sleep time (TST), as measured by participants sleep diaries, between subgroup participants wh o did and did not report sleep disturbances. The overall omnibus test fo r this analysis was also non-significant, F (4,17) = 1.47, p>.05, 2 = 0.26. Finally, separate Pearson correlations were computed between actigraphy and diary measurement of SOL, WASO, sleep efficiency (SE), and total sleep time (TST) for good and poor sleepers in the subgroup (Table 3-4). The correlations for each variable (SOL, WASO, SE, TST) were compared between good and poor sleepers. Results revealed significantly stronger

PAGE 56

56 correlations between diary and actigraphy m easurements of WASO for good sleepers. Correlations between diary and actigraphy measurement of SOL, SE, and TST were also higher among good sleepers; however, the difference in the magnitude of the correlations between good and poor sleepers for each of these variab les did not reach significance (Table 3-4).

PAGE 57

57 0 0.5 1 1.5 2 2.5 3 sleep qualitysleep latencysleep disturbances PSQI subscalesAverage ratings Facial Pain Back Pain Fibromyalgia *** *** *** *** ** *** Figure 3-1. Self-reported sleep (as reported by PSQI) for 3 chronic pain groups. **p<.01; ***p<.001.

PAGE 58

58 Figure 3-2. Structural equati on model for the relationship among sleep, pain, and negative mood. 2(32)=41.77, p=0.12, RMSEA=0.03. PSQI Sleep quantity MPQ VAS avg pain PASS BDI Sleep Negative Mood P a in MCV anxiety MCV depression MCV usual pain -0.02 -0.57 0.99 Sleep quality

PAGE 59

59 Table 3-1. Demographic information fo r the 3 chronic pain patient groups Facial pain Back pain Fibromyalgia N = 116 N = 55 N = 121 M SD M SD M SD F df p Age 46.28 14.12 46.44 10.28 47.53 10.82 0.35 2,291 0.706 Pain duration 91.32 109.91 128.70 111.57 130.29 70.80 2.37 2,175 0.097 Years of education 14.20 2.76 12.89 2.32 15.86 4.71 5.57 2,157 0.005 Facial pain Back pain Fibromyalgia N (%) N (%) N (%) 2 p Sex Female 95 (80.5) 34 (61.8) 147 ( 95.5) 36.96 0.000 Male 23 (19.5) 21 (38.2) 7 ( 4.5) Race Caucasian 107 (90.7) 46 (83.6) 6 ( 85.7) 9.30 0.318 Black/AfricanAmerican 4 ( 3.4) 8 (14.5) 1 ( 14.3) Asian 1 ( 0.8) 0 ( 0.0) 0 ( 0.0) Hispanic 4 ( 3.4) 1 ( 1.8) 0 ( 0.0) Other 2 ( 1.7) 0 ( 0.0) 0 ( 0.0) Work status Full-time 23 (19.5) 8 (15.1) 0 ( 0.0) 15.32 0.053 Part-time 8 ( 6.8) 2 ( 3.8) 0 ( 0.0) Working, unspecified 25 (21.2) 7 (13.2) 0 ( 0.0) Student 13 (11.0) 1 ( 1.9) 1 ( 14.3) Not employed 49 (41.5) 35 (66.0) 6 ( 85.7) Narcotic med use Yes 29 (26.1) 41 (75.9) 4 ( 57.1) 37.35 0.000 No 82 (73.9) 13 (24.1) 3 ( 42.9) Antidepressant med use Yes 41 (36.9) 22 (42.3) 4 ( 66.7) 2.33 0.313 No 70 (63.1) 30 (57.7) 2 ( 33.3) Sleep med use Yes 47 (41.2) 20 (45.5) 3 (100.0) 4.51 0.341 No 62 (54.4) 23 (52.3) 0 ( 0.0)

PAGE 60

60Table 3-2. Comparison of diaryand actigraphy-measured sleep variables across 3 pain groups Facial pain Back pain Fibromyalgia M SD M SD M SD F p 2 Sleep diary variables df(2,19) SOL 21.74 13.24 42.42 24.91 32.73 19.08 2.20 0.138 0.19 WASO 13.17 12.35 47.54 79.53 54.40 39.83 1.29 0.298 0.12 SE 91.12 3.89 86.87 7.23 85.22 5.77 2.00 0.163 0.17 VAS sleep quality 5.46 1.22 5.04 1.31 4.49 1.24 1.02 0.380 0.10 M SD M SD M SD F p 2 Actigraphy variables df(2,19) SOL 16.93 13.37 29.68 23.47 23.46 16.67 0.95 0.404 0.09 WASO 47.26 13.06 57.58 14.77 67.25 27.92 1.99 0.164 0.17 SE 83.66 3.56 78.48 9.26 79.05 6.00 1.37 0.278 0.13 Note. SOL=Sleep onset latency; WASO=Wake after sleep on set; SE=Sleep efficiency; VAS=Visual analogue scale.

PAGE 61

61 Table 3-3. Multivariate mixed model MANOVA results for pain ratings in temporal summation and ramp and hold procedures Temporal summation Source Hypothesis df Error df F p 2 Pulse number 3 18 1.24 0.326 0.17 Pulse number x type of sleeper 3 18 1.24 0.325 0.17 Pulse number 3 17 1.28 0.312 0.19 Pulse number x chronic pain group 6 36 0.90 0.509 0.13 Ramp and hold pain intensity ratings Source Hypothesis df Error df F p 2 Temperature 3 18 24.70 0.000 0.81 Temperature x type of sleeper 3 18 0.26 0.855 0.04 Temperature 3 17 24.10 0.000 0.81 Temperature x chronic pain group 6 36 0.99 0.448 0.14 Ramp and hold pain unpleasantness ratings Source Hypothesis df Error df F p 2 Temperature 3 18 18.82 0.000 0.76 Temperature x type of Sleeper 3 18 0.26 0.852 0.04 Temperature 3 17 20.14 0.000 0.78 Temperature x chronic pain group 6 36 1.08 0.392 0.15

PAGE 62

62 Table 3-4. Correlation between actigraphyand diary-measured sleep variables for good and poor sleepers SE actigraphy WASO actigraphy SOL actigraphy TST actigraphy Good sleepers ( N =9) SE diary 0.47 WASO diary 0.91** SOL diary 0.77* TST diary 0.92*** Poor sleepers ( N =13) SE diary 0.34 WASO diary 0.19 SOL diary 0.47 TST diary 0.82** Difference z-score 0.30 2.59** 0.99 0.84 Note. SOL=Sleep onset latency; WASO=Wake af ter sleep onset; SE=Sleep efficiency; TST=Total sleep time. p<.05; **p<.01; ***p<.001.

PAGE 63

63 CHAPTER 4 DISCUSSION Results revealed significant differences in subjective sl eep reports across the three chronic pain populations in the current sample. Most notably, the facial pain patients reported better sleep quality, shorter latenc y to sleep onset, and fewer sleep disturbances, compared to the patients in the back pain and fibromyalgia groups. This suggests that the facial pain patients were experiencing less disturbed sleep overall, compar ed to the other two groups. The back pain patients and the fibromyalgia patients endorsed si milar reports regarding sleep quality, latency to sleep onset, and sleep disturbances overall. Th ey reported higher levels of subjective sleep problems on a validated self-report measure of sleep quality, compared to the facial pain patients. Descriptive analyses indicated significant differences in sex, education, and narcotic medication use across chronic pain groups. No si gnificant differences on any of the three sleep variables were found across sex. Significant di fferences were found for sleep quality and sleep latency across years of education, with poorer sleep being related to fewer years of education. Facial pain patients also had si gnificantly higher education compar ed to back pain patients. Finally, significant differences in narcotic medication use was found across all three sleep variables, with users having higher scores on each of the three sleep variables (indicating poorer sleep) compared to non-users. The highest rates of narcotic use were found among back pain patients, followed by fibromyalgia patients, with facial pain patients endorsing the lowest rates of narcotic medication use. As narcotic medications are known to affect sleep continuity and sleep architecture (reduced sleep quantity, suppression of REM and st age 3-4 sleep) (Kay, Eisenstein, & Jasinski, 1969), it is possible that higher rates of use of these medica tions among the back pain and fibromyalgia patients is rela ted to the poorer sleep reported by these patients. However,

PAGE 64

64 fibromyalgia patients reported the most impaired sleep across subscales, and back pain patients had the greatest rates of narcot ic medication use. Further, st udies have demonstrated that individuals typically develop a to lerance to the sleep effects of these medications within a few weeks after initiating use (Kay et al., 1969). Additionally, average pain levels were compared across pain groups, and fibromyalgia patients were found to have higher le vels of average pain compared to facial pain patients; back pain pati ents did not differ from either of the other two groups. It is possible that higher levels of average pain among fibromyalgia patients were related to the increased subjective sleep problems re ported by these patients; however, back pain patients also reported similar levels of sleep problems and similar le vels of pain, so it is unclear how important of a role average pain ratings pl ayed in the subjective sleep reports of these patients. Thus, while some support was found for the hypothe sis that there would be differences in the distribution of sleep problems across the thre e chronic pain groups, it was not the case that each group of chronic pain patients exhibited increased levels of specific types of sleep problems. The differences in sleep that emerge d between the three chronic pain groups were in quantity rather than type of sl eep disturbance. Within each of the three chronic pain groups, multiple types of sleep problems (difficulties with initiation of sleep, frequent disturbances during sleep, and poor subjective sl eep quality) were reported. Howe ver, it appears that patients with back pain and fibromyalgia endorse greate r levels of sleep problems when compared to patients suffering from chronic facial pain. Back pain and fibrom yalgia groups reported significantly greater problems within all three areas: initiation of sl eep, frequency of sleep disturbances, and subjective sleep quality, compared to the group of facial pain patients in this study.

PAGE 65

65 It may be that the sleep problems of chroni c pain patients are similar overall, or it is possible that the component scores from the Pittsburgh Sleep Quality Index (PSQI) were not sensitive enough to differentiate each of the specific sleep problems being investigated. Therefore, the sleep onset latency (SOL), wake after sleep onset (WASO), and sleep quality VAS ratings from subgroup participants diaries (or actigraphy and diar ies for SOL and WASO) were also examined to determine whether any differences existed between the three chronic pain groups on any of these characteristics. No signifi cant differences emerged in these analyses; this may be due in part to the small sample size of the subgroup as large effect sizes were seen in these analyses. The overall pattern of results that emerged from analysis of the diary and actigraphy variables in the subgroup is generally consistent with th e findings from the self-report measures in the larger sample of participants. The structural equation m odeling analysis sought to in crease understanding of the relationship between sleep, pain, an d negative mood. In particular, th e role of sleep in predicting pain was of interest, as was the potential mediat ing effect of negative mo od on this relationship. Results from this analysis supported a direct re lationship between sleep a nd pain, when negative mood was not considered in the model. This supports and extends previous research by demonstrating that increased sl eep problems predict greater pai n, using validated and reliable measures in a large group of chronic pain patients Poorer sleep was also shown to be predictive of higher levels of negative mood, which again supports previous resear ch and highlights the importance of addressing sleep problems when pr esent in chronic pain patients. Additionally, when negative mood was included in the mode l, it was shown to mediate the relationship between sleep and pain in this sample. Thus, the importance of assessi ng and treating both sleep problems and negative mood in chronic pain patients cannot be underestimated.

PAGE 66

66 While the relationship between sleep and pain is complex, these results suggest that there are multiple pathways by which sleep is related to pain. First, sleep disturbance may lead directly to increased pa in among chronic pain patients. Seco nd, sleep disturbance may lead to increased pain in chronic pain patients through sleeps influence on negative mood. While higher levels of negative mood are found among chr onic pain patients, no t all patients with chronic pain have increased levels of negative mood. This model demonstrates that the influence of sleep on pain may take multiple routes, and is also related to whether increased levels of negative mood are present. This suggests th at assessment of both sleep and negative mood should routinely be undertaken in chronic pain populations. This model also suggests that treatment efforts can take multiple routes, although research is needed to determine whether interventions that improve sleep also lead to improvements in mood and/or pain as well. No differences were found across the three ch ronic pain groups in their response to the psychophysical testing. All participants noted higher ratings of pain intensity and pain unpleasantness at higher temperatures in the ra mp and hold procedure, but no other significant results were found between pain groups in eith er the ramp and hold or the temporal summation procedures. Additionally, th e results of the psychophysical testing in the subgroup of participants did not supp ort the hypothesis of greater sensitiv ity to painful stimuli among those patients with concurrent sleep dist urbance. However, it should be noted that these analyses were exploratory in nature, and a ge neral pattern of larg e effect sizes was demonstrated across analyses. The results of this study demonstrated th e high prevalence of subjective sleep problems in chronic pain patients. By including different groups of chronic pain patients, and examining differences in sleep reports across these differe nt groups of patients, this study allows for a

PAGE 67

67 greater understanding of the similarities and di fferences in the sleep problems reported by different populations of chronic pa in patients. Contrary to the hypothesis that different types of sleep problems would be more prevalent in specific chronic pa in groups, results suggested the emergence of similar subjective sleep reports across all three chr onic pain groups. One significant difference noted across groups, was that the facial pain patients reported fewer sleep disturbances overall compared to the other two groups. This patte rn of results suggests that a broad-based sleep treatment, within interventions for chronic pain, may be the most appropriate approach given the current evidence. Since no specific sleep problems emerged within each of the chronic pain groups, there is no evidence to suggest targeting sleep treatments for specific sleep problems within a ny chronic pain populations. Examination of the relationships among slee p, pain, and negative mood revealed several important findings, and underscored the role of both sleep and negative mood in the clinical picture of chronic pain. First, poor sleep wa s found to directly pred ict both negative mood and pain in this sample of chronic pain patient s. This highlights the importance of thoroughly assessing for and treating sleep problems within chronic pa in populations. The direct relationship between sleep and pain suggests that addressing sleep pr oblems in pain patients will likely have a beneficial impact on these patients pain experience. However, since negative mood mediated the relationship between sleep and pa in in this sample of chronic pain patients, addressing negative mood may have a greater im pact on patients pain experience among those chronic pain patients who are experienci ng both sleep disturbance and negative mood. Within the subgroup of participants w ho completed the psychophysical testing, no significant differences emerged across the three ch ronic pain groups on either the ramp and hold or temporal summation protocols. Participants from all three groups responded similarly to the

PAGE 68

68 thermal pain stimuli in each protocol. Additionally, no significant di fferences were noted between subgroup participants with and without c oncurrent sleep disturbances. These results did not support the hypothesis that chronic pain pa tients with concurrent sleep problems would experience greater sensitivity to the painful ther mal stimuli, and fail to support previous findings of increased pain sensitivity among individuals who experienced reduced or disrupted sleep (Kundermann et al., 2004; Moldof sky & Scarisbrick, 1976; Onen, A lloui, Gross et al., 2001). Comparison of Study Results to Previous Findings Similar to previous studies, the current results support the ubiquit ous nature of sleep problems in chronic pain populations. Overall, 6 3.5% of participants reported subjectively poor sleep using a modified cut-off sc ore on the PSQI, within this samp le of back pain, facial pain, and fibromyalgia patients. This supports th e hypothesis of sleep problems as a pervasive condition among chronic pain patients, and sugges ts that sleep problems should be routinely assessed for and treated in this population. This study adds to the existing knowledge by examining the relative distribution of specific sleep complaints reported by patients with different chronic pain conditions allowing for a comparison of findings across different chronic pain patient samples. In particular, while sim ilarly disrupted sleep was noted across the entire sample, results indicated that reported sleep pr oblems were more seve re among back pain and fibromyalgia patients, as compared to the facial pain patients. The back pain and facial pain patients did not appear to differ from one another in the type or severity of sleep problems reported. Further study employing polysomnography would be useful in identifying whether objective sleep differences emerged among differen t groups of patients, but the present findings suggest that subjective e xperience of sleep is similar for back pain and fibromyalgia patients. While it was hypothesized that the three ch ronic pain groups would differ in their subjective sleep reports, results supported the presence of a more generalized pattern of sleep

PAGE 69

69 disturbance across groups. Specifi cally, all three chroni c pain groups reported difficulties with both sleep onset and frequent sl eep disturbances, as well as s ubjectively poor quality sleep. Differences among groups emerged in the severity of these sleep problems, with fibromyalgia patients endorsing significantly gr eater levels of these problems compared to the facial pain patients, and back pain patients endorsing sleep problems that fell intermediate of these other two groups. Thus, it appears that sleep is more broadly disturbed within chronic pain conditions than initially hypothesize d, and no specific differences regard ing the sleep problems reported by various chronic pain populations were evident in this sample. The experience of chronic pain, either alone or in conjunction with negative mood, may simply se rve as a gross disruptive factor for sleep among these patients. No significant relationships were found betw een subjective reports of good vs. poor sleep and sensitivity to painful thermal stimuli in the current study. Previous studies have reported increased sensitivity to painful stimuli following conditions of interrupted or restricted sleep (Kundermann et al., 2004; Moldof sky & Scarisbrick, 1976; Onen, Alloui, Gross et al., 2001). Thus, the present findings are somewhat inconsis tent with these previ ous reports from the literature. Interestingly, th e effect sizes found in the subgr oup analyses for the temporal summation procedure were large, suggesting that the presence or absence of sleep disturbance may be an important predictor of experimental pain response in chronic pain populations. Prior studies have demonstrated that chronic pain pati ents have an increased likelihood of sensitization to painful stimuli (Staud et al., 2003), and it ma y be that this is further compounded by the experience of disturbed sleep. However, this cannot be determined based on the findings from the present analyses and needs to be ex amined in future studies.

PAGE 70

70 Role for Negative Mood in the Sleep-Pain Relationship Higher levels of negative mood are co mmonly reported, both among chronic pain populations and among individuals with sleep problems (Breivik et al., 2005; Breslau, Roth, Rosenthal, & Andreski, 1996; B uysse et al., 1994; Duquesnoy et al., 1998; Ford & Kamerow, 1989; Ohayon, 1997; Verhaak et al., 1998). Howeve r, the interrelations hips among sleep, pain, and negative mood have been inconsistent in the existing literature (Atkinson et al., 1988; Moffitt et al., 1991; Morin et al., 1998; Pilowsky et al., 1985; Sayar et al., 2002). Some studies have reported higher levels of depression and/or anxiet y among chronic pain patients with sleep disturbances; other studies have not found these relationships. The current results support a mediating role for negative mood in the relations hip between sleep and pain. In other words, while significant relationships ha ve been reported between slee p and pain, these relationships must take into account the presen ce of negative mood. In particular among chronic pain patients endorsing high levels of negative mood, pain is more strongly predicte d by negative mood than by sleep. However, it is not difficult to see how negative mood, sleep disturbances, and pain can each act to increase the other tw o, leading to a cycle that perp etuates itself if there is no intervening action to disrupt it. Addressing negative mood will likel y have a beneficial effect on patients pain experience or perceptions regardin g their ability to cope with their pain, and may also improve sleep as well. Further, am ong patients with low le vels of negative mood, intervening to improve sleep disturbances will likely have a positive impact on patients pain experience as well. As effective interventions have been developed for both negative mood and sleep, implementing these treatment will be importan t as it is likely to improve patients pain experience and lead to signifi cant improvements in functioning.

PAGE 71

71 Implications for Conceptualization and Treatment of Chronic Pain The comorbidity between chronic pain and sl eep disturbances has been widely reported in the literature (Atkin son et al., 1988; Pilowsky et al., 1985 ; M. T. Smith & Haythornthwaite, 2004). Similarly, numerous studies have report ed increased reports of negative mood among both chronic pain patients (Breivik et al., 2005; Duquesnoy et al., 1998; Robinson & Riley III, 1999; Verhaak et al., 1998), and patients with sl eep problems, particularly those reporting insomnia complaints (Breslau et al., 1996; Chang, Ford, Mead, Cooper-Patrick, & Klag, 1997; Ford & Kamerow, 1989; Katz & McHorney, 2002) The model investigated in this study examined the role of sleep and negative mood in predicting pain among chronic pain patients, and revealed that negative mood mediated the re lationship between sleep disturbance and pain. Thus, previous findings suggesti ng a causal link between sleep dist urbances and pain in chronic pain patients may be due, in part, to the eff ects of negative mood. High levels of negative mood may increase or perpetuate the impact of sleep disturbances on patient s pain experiences, possibly through the interruption of sleep, which is a common report in the clinical pictures of both depression and anxiety. Additionally, certain areas in th e central nervous system have been implicated in both pain perception/regulation, as well as the re gulation/facilita tion of sleep, such as the hypothalamus (Montagna, 2006). Abnormalities of the hypothalamic-pituitary-adrenal (HPA) axis functioning have been noted among 30-70% of individuals suffering from major depression, and good treatment response has been demonstrated for individuals who evidence normalization of HPA functioning with use of antidepressant medications (Takahashi, 2002). This lends further supports to the importance of this pa thway in mood regulation, and given the overlap with functions related to both sleep and pain, this pathwa y may be a common physiological link between these oft related conditions. Taken togeth er, these findings support the argument for the

PAGE 72

72 involvement of common pathophysiol ogical pathways in sleep disturbance, mood disorders, and chronic pain. Interestingly, no clear relationship was found between sleep disturbanc e and sensitivity to painful stimuli in the present study. This is so mewhat contradictory to findings from previous studies, which have reported incr eased sensitivity to painful stimuli or increased pain reports among healthy individuals followi ng conditions of sleep restri ction or interrupted sleep (Kundermann et al., 2004; Moldof sky & Scarisbrick, 1976; Onen, A lloui, Gross et al., 2001). Similarly, studies examining pain sensitivity among ch ronic pain patients have also suggested an inverse relationship between pain response to ex perimental stimuli and sleep (Agargun et al., 1999). The present results are incons istent with these findings. It should be noted that these analyses util ized participants se lf-reports about their subjective sleep to determine good and poor sleepe rs, rather than experimentally manipulating participants sleep. When sleep has been manipulated experimentally among healthy participants, increased pain sensitivity has been reported (Kundermann et al., 2004; Moldofsky & Scarisbrick, 1976; Onen, Alloui, Gross et al ., 2001). Additionally, a recent study by Smith and colleagues demonstrated differential effects on pain inhibition for restricting total sleep time to a shorter duration compared to repeatedly disrupting sleep using forced awakenings to produce a similar shorter total duration of sleep, w ithin a sample of healthy participants (M. T. Smith, Edwards, McCann, & Haythornthwaite, 2007). These results suggest that differential effects on pain processing may be produced by disr uptions in sleep contin uity as compared to reduced sleep quantity produced by a shortened sche dule. The pattern of more generalized sleep disturbances reported by the subgroup of chroni c pain patients in the present study may be qualitatively different than the sleep deprivation produced by re stricting total sleep time in

PAGE 73

73 healthy participants. Thus, the increased pain sensitivity related to sleep disturbances reported by previous studies may vary as a function of the t ype of sleep disturbance, as well as the type of population (healthy volunteers versus chronic pain pa tients). In other word s, differential effects on pain sensitivity may be seen following experimentally-produced sleep restriction in healthy volunteers as compared to the effect of more ch ronic or cumulative sleep interruptions on pain response in chronic pain patients. If this hypothesis is true, it coul d help to explain the lack of significant findings regarding pain sensitivity in the present analyses. Current evidence supports cognitive-behaviora l approaches to the treatment of chronic pain as having the most consistent empirica l support (Robinson & O'Brien, in press), with improvements seen in pain, mood, and coping, as well as reductions in interference and improved functional outcomes (such as fewer abse nces from work, reduced medication use and health care visits) (Linton, Boersma, Janss on, Svard, & Botvalde, 2005; Nash, Park, Walker, Gordon, & Nicholson, 2004; Turner, Mancl, & Aaron, 2006). Within CBT approaches for chronic pain, negative mood is often addressed either directly or indi rectly, and the results of this study suggest that addressing nega tive mood within chronic pain treatments is important and should continue to be a central part of these tr eatments. However, the assessment and treatment of sleep is a less standard component in CBT trea tments for chronic pain, and most often consists of education about sleep hygiene when it is incl uded. While this may be useful information, sleep hygiene alone has not been demonstrated to be effectiv e as a means of implementing changes to improve sleep (Engle-Friedman, B ootzin, Hazlewood, & Tsao, 1992; Guilleminault et al., 1995). Given the high comorbidity of sleep distur bance and chronic pain, it would appear prudent to include a formal a ssessment of patients sleep with in chronic pain assessments.

PAGE 74

74 Additionally, incorporating established techniques (stimulus control, sleep restriction) from CBT for Insomnia (CBT-I) treatments for addressing sleep disturbances into existing CBT treatments for chronic pain would likely be more effective for addressing sleep problems being experienced by chronic pain patients. The sk ill-based, behavioral nature of these techniques allows for their inclusion within existing cogniti ve-behavioral treatments for chr onic pain. Incorporating these sleep treatment components with existing techniques for addressi ng mood and activity, as well as increasing patients knowledge regarding chronic pain, will likely lead to better outcomes and more fully address the full spectrum of impa irment experienced by chronic pain patients. Limitations Some limitations to the above findings shoul d be noted. Analyses involving the subgroup of participants completing sleep diaries, actigraphy monitorin g, and the psychophysical testing procedures were intended to be exploratory, in order to id entify the presence of potentially provocative findings that could subsequently be e xplored in a larger a nd more representative sample. As such, the sample size of this subgro up of participants was small and may have made it difficult to detect significant findings using traditional statistical criteria. Results from the analyses of participants response to psychophysical testing revealed large effect sizes between good and poor sleepers (temporal summa tion), and across chronic pain groups (temporal summation and ramp and hold), in these analyses. This suggests that sleep disturbances may be an important of determinant of chronic pain patients responses to painful stimuli and that type of pain condition may al so be an important in fluence on individuals response to painful stimuli. However, the pr esent analyses may not have been able to demonstrate these relationships at traditional statistical significance levels, due to the small sample size of the subgroup reducin g the power of the analyses. There was no pain-free sample

PAGE 75

75 of participants in the present study, so results are not availa ble to examine the role of sleep disturbance on psychophysical pain response am ong individuals withou t chronic pain. Similarly, when sleep diary and actigraphy measurement of several sleep parameters across good and poor sleepers were examined, no significant differences emerged. The large effect sizes of these analyses again suggested that the samp le size of the subgroup may have impacted the ability for significant findings to emer ge. Thus, consistent with the general pattern of correlations found between sleep diary and actigraphy meas ures found among good and poor sleepers, it appears that these measures ma y be more strongly asso ciated in good sleepers compared to poor sleepers. Examination of these relationships in a larger sample will increase the power of the analyses and improve the ability to detect such relationships, if they are present. The large effect sizes found across subgroup anal yses suggests avenues for additional studies using larger numbers of particip ants, in order to elucidate the nature of the relationships among sleep disturbance, type of chronic pain, and psychophysical response. Additionally, the number of female participan ts was much greater than the number of male participants in the sample. This is cons istent with the literature on chronic pain, where females tend to be over-represented in most chronic pain populations (Moulin et al., 2002; Verhaak et al., 1998). While the unequal distribution of males and females makes it difficult to directly examine sex differences in these findings, the preponderance of da ta has demonstrated a lack of sex differences in clinical pain (R obinson, Wise, Riley III, & Atchison, 1998). This allows for increased confidence in the applicab ility of the current findings to both male and female chronic pain populations. Future Directions It would be interesting to examine whether incorporating CBT-I techniques into existing CBT treatments for chronic pain affords any benefit for chronic pain patients. Specifically,

PAGE 76

76 examination of patients mood, sleep, and pain reports following standard CBT treatment for pain, CBT for pain plus sleep hygiene information alone, and CBT for pa in plus sleep hygiene information and CBT-I techniques, would provide in formation about the additive benefits of the addition of sleep treatment to existing CBT treat ments for chronic pain. While it is likely that the addition of the sleep techniqu es would confer added utility to chronic pain treatments, it would also increase the duration and thus, the co st, of these treatments. Thus, it would be important to provide a justifica tion for the additional time and cost of adding these techniques into existing treatment paradigms, by demons trating improved patient outcomes when these components are part of the treatment package. Additionally, longitudinal evaluations would be useful to examine any improvement in outcomes that patients experience following treatment.

PAGE 77

77 LIST OF REFERENCES Affleck, G., Urrows, S., Tennen, H., Higgins, P., & Abeles, M. (1996). S equential daily relations of sleep, pain intensity, and attention to pain among women with fibromyalgia. Pain, 68(2-3), 363-368. Agargun, M. Y., Tekeoglu, I., Gunes, A., Adak, B ., Kara, H., & Ercan, M. (1999). Sleep quality and pain threshold in patients with fibromyalgia. Compr Psychiatry, 40 (3), 226-228. American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (4th edition ed.). Washington, DC: American Psychiatric Association. Anch, A. M., Lue, F. A., MacLean, A. W ., & Moldofsky, H. (1991). Sleep physiology and psychological aspects of the fibr ositis (fibromyalgia) syndrome. Can J Psychol, 45(2), 179-184. Ancoli-Israel, S., Cole, R., Alessi, C., Chambers M., Moorcroft, W., & Pollak, C. P. (2003). The role of actigraphy in the study of sleep and circadian rhythms. Sleep, 26(3), 342-392. Andersson, G. B. (1999). Epidemiological features of chronic low-back pain. Lancet, 354(9178), 581-585. Andersson, H. I., Ejlertsson, G., Leden, I., & Sc hersten, B. (1999). Impact of chronic pain on health care seeking, self car e, and medication. Results from a population-based Swedish study. J Epidemiol Community Health, 53 (8), 503-509. Arima, T., Svensson, P., Rasmussen, C., Nielsen, K. D., Drewes, A. M., & Arendt-Nielsen, L. (2001). The relationship between selective sleep depriva tion, nocturnal jaw-muscle activity and pain in healthy men. J Oral Rehabil, 28 (2), 140-148. Atkinson, J. H., Ancoli-Israel, S ., Slater, M., Garfin, S. R., & Gillin, J. C. (1988). Subjective sleep disturbance in chronic pain. Clinical Journal of Pain, 4 225-232. Bailey, D. R. (1997). Sleep disorders. Overview and relationship to orofacial pain. Dent Clin North Am, 41(2), 189-209. Barabas, G., Ferrari, M., & Matthews, W. S. (1983). Childhood migraine and somnambulism. Neurology, 33 (7), 948-949. Baron, R. M., & Kenny, D. A. (1986). The moderato r-mediator variable di stinction in social psychological research: con ceptual, strategic, and st atistical considerations. J Pers Soc Psychol, 51(6), 1173-1182. Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric properties of the Beck Depression Inventory: Twenty -five years of evaluation. Clinical Psychology Review, 8 77-100.

PAGE 78

78 Beck, A. T., Ward, C. H., Mendelson, M., Moc k, J., & Erbaugh, J. (1961). An inventory for measuring depression. Arch Gen Psychiatry, 4 561-571. Bennett, R. M. (1993). The origin of myopain: An integrated hypothesis of focal muscle changes and sleep disturbance in patients with the fibromyalgia syndrome. J Musculoskel Pain, 1 105. Binkley, J., Finch, E., Hall, J., Black, T., & Go wland, C. (1993). Diagnos tic classification of patients with low back pain: report on a survey of physical therapy experts. Phys Ther, 73(3), 138-150; discussion 150-135. Blau, J. N. (1982). Resolution of migraine attacks: sleep and the recovery phase. J Neurol Neurosurg Psychiatry, 45 (3), 223-226. Blau, J. N. (1990). Sleep deprivation headache. Cephalalgia, 10 (4), 157-160. Bragdon, E. E., Light, K. C., Costello, N. L., Sigurdsson, A., Bunting, S., Bhalang, K., et al. (2002). Group differences in pain modulation: pain-free women compared to pain-free men and to women with TMD. Pain, 96(3), 227-237. Branco, J., Atalaia, A., & Paiva, T. (1994). Sleep cycles and alpha -delta sleep in fibromyalgia syndrome. J Rheumatol, 21 (6), 1113-1117. Breivik, H., Collett, B., Ventafridda, V., Cohen, R., & Gallacher, D. (2005). Survey of chronic pain in Europe: Prevalence, impact on daily life, and treatment. Eur J Pain Breslau, N., Roth, T., Rosenthal, L., & Andreski P. (1996). Sleep disturbance and psychiatric disorders: a longitudinal epid emiological study of young adults. Biol Psychiatry, 39 (6), 411-418. Buysse, D. J., Reynolds, C. F., 3rd, Kupfer, D. J ., Thorpy, M. J., Bixler, E. Manfredi, R., et al. (1994). Clinical diagnoses in 216 insomnia pa tients using the Intern ational Classification of Sleep Disorders (ICSD), DSM-IV and ICD-10 categories: a report from the APA/NIMH DSM-IV Field Trial. Sleep, 17 (7), 630-637. Buysse, D. J., Reynolds, C. F., 3rd, Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res, 28 (2), 193-213. Campbell, S. M., Clark, S., Tindall, E. A., Fore hand, M. E., & Bennett, R. M. (1983). Clinical characteristics of fibrositis. I. A "blind ed," controlled study of symptoms and tender points. Arthritis Rheum, 26 (7), 817-824.

PAGE 79

79 Carette, S., McCain, G. A., Bell, D. A., & Fam, A. G. (1986). Evaluation of amitriptyline in primary fibrositis. A double-blind, placebo-controlled study. Arthritis Rheum, 29(5), 655659. Carli, G., Suman, A. L., Biasi, G., & Marcolongo R. (2002). Reactivity to superficial and deep stimuli in patients with chronic musculoskeletal pain. Pain, 100(3), 259-269. Carpenter, J. S., & Andrykowski, M. A. (1998). Psychometric evaluation of the Pittsburgh Sleep Quality Index. J Psychosom Res, 45 (1 Spec No), 5-13. Casey, K. L., Morrow, T. J., Lorenz, J., & Minos hima, S. (2001). Temporal and spatial dynamics of human forebrain activity during heat pain: analysis by positron emission tomography. J Neurophysiol, 85 (2), 951-959. Chang, P. P., Ford, D. E., Mead, L. A., Cooper-P atrick, L., & Klag, M. J. (1997). Insomnia in young men and subsequent depression. The Johns Hopkins Precursors Study. Am J Epidemiol, 146(2), 105-114. Chase, M. A., & Morales, F. R. (1994). The control of motorneurons during sleep. In M. H. Kryger, T. Roth, D. W. C. & e. al. (Eds.), Principles and Practi ce of Sleep Medicine (2nd ed., pp. 163-176). Philadelphia, PA: W. B. Saunders. Chiu, Y. H., Silman, A. J., Macfarlane, G. J., Ra y, D., Gupta, A., Dickens, C., et al. (2005). Poor sleep and depression are indepe ndently associated with a reduced pain threshold. Results of a population based study. Pain, 115 (3), 316-321. Clauw, D. J., Williams, D., Lauerman, W., Dahl man, M., Aslami, A., Nachemson, A. L., et al. (1999). Pain sensitivity as a correlate of clin ical status in indivi duals with chronic low back pain. Spine, 24(19), 2035-2041. Cohen, M. J. M., Menefee, L. A., Doghramji, K., Anderson, W. R., & Frank, E. D. (2000). Sleep in chronic pain: Problems and treatments. International Review of Psychiatry, 12 115126. Culebras, A., & Miller, M. (1984) Dissociated patterns of noctu rnal prolactin, cortisol, and growth hormone secretion after stroke. Neurology, 34 (5), 631-636. Curran, S. L., Carlson, C. R., & Okeson, J. P. (1996). Emotional and physiologic responses to laboratory challenges: patients with temporom andibular disorders versus matched control subjects. J Orofac Pain, 10 (2), 141-150. Demarco, G. J., Baghdoyan, H. A., & Lydic, R. ( 2003). Differential cholin ergic activation of G proteins in rat and mouse brainstem: relevance for sleep and nociception. J Comp Neurol, 457(2), 175-184.

PAGE 80

80 de Souza, L., Benedito-Silva, A. A., Pires, M. L., Poyares, D., Tufik, S., & Calil, H. M. (2003). Further validation of actigraphy for sleep studies. Sleep, 26(1), 81-85. Diagnostic Classification Steering Co mmittee, T. M., Chairman. (1990). International Classification of Sleep Disorders: Diagnostic and Coding Manual (ICSD) Rochester, MN: American Sleep Disorders Association. Drewes, A. M., Nielsen, K. D., Arendt-Nielse n, L., Birket-Smith, L., & Hansen, L. M. (1997). The effect of cutaneous and deep pain on the electroencephalogram during sleep--an experimental study. Sleep, 20 (8), 632-640. Drewes, A. M., Nielsen, K. D., Taagholt, S. J ., Bjerregard, K., Svendsen, L., & Gade, J. (1995). Sleep intensity in fibromyalgia: focus on the microstructure of the sleep process. Br J Rheumatol, 34(7), 629-635. Duquesnoy, B., Allaert, F. A., & Verdoncq, B. ( 1998). Psychosocial and o ccupational impact of chronic low back pain. Rev Rhum Engl Ed, 65 (1), 33-40. Engle-Friedman, M., Bootzin, R. R., Hazlewood, L., & Tsao, C. (1992). An evaluation of behavioral treatments for inso mnia in the older adult. J Clin Psychol, 48 (1), 77-90. Espie, C. A., Brooks, D. N., & Lindsay, W. R. (1989). An evaluation of tailored psychological treatment of insomnia. J Behav Ther Exp Psychiatry, 20 (2), 143-153. Farasyn, A., & Meeusen, R. (2005). The influenc e of non-specific low back pain on pressure pain thresholds and disability. Eur J Pain, 9 (4), 375-381. Flor, H., Diers, M., & Birbaumer, N. (2004). Peri pheral and electrocortical responses to painful and non-painful stimulation in chronic pain patients, tension h eadache patients and healthy controls. Neurosci Lett, 361(1-3), 147-150. Foo, H., & Mason, P. (2003). Brainstem modulation of pain during sleep and waking. Sleep Med Rev, 7 (2), 145-154. Ford, D. E., & Kamerow, D. B. (1989). Epid emiologic study of sleep disturbances and psychiatric disorders. An opportunity for prevention? Jama, 262(11), 1479-1484. Forgays, D. G., Forgays, D. K., & Spielberger, C. D. (1997). Factor structure of the State-Trait Anger Expression Inventory. J Pers Assess, 69(3), 497-507. Forgays, D. K., Spielberger, C. D., Ottaway, S. A., & Forgays, D. G. (1998). Factor structure of the State-Trait Anger Expression Inve ntory for middle-ag ed men and women. Assessment, 5(2), 141-155. Gallagher, R. M., & Verma, S. (1999). Managing pa in and comorbid depression: A public health challenge. Semin Clin Neuropsychiatry, 4 (3), 203-220.

PAGE 81

81 Gijsbers van Wijk, C. M., van Vliet, K. P., & Kolk, A. M. (1996). Gender perspectives and quality of care: towards appropriate and adequate health care for women. Soc Sci Med, 43(5), 707-720. Glaros, A. G. (1981). Incidence of diurnal and nocturnal bruxism. J Prosthet Dent, 45 (5), 545549. Grabois, M. (2005). Management of chronic low back pain. Am J Phys Med Rehabil, 84 (3 Suppl), S29-41. Guilleminault, C., Clerk, A., Black, J., Laba nowski, M., Pelayo, R., & Claman, D. (1995). Nondrug treatment trials in psychophysiologic insomnia. Arch Intern Med, 155 (8), 838844. Harding, S. M. (1998). Sleep in fibromyalgia patients: subjective and objective findings. Am J Med Sci, 315(6), 367-376. Harvey, A. G. (2000). Sleep hygiene and sleep-onset insomnia. J Nerv Ment Dis, 188(1), 53-55. Hastie, B. A., Riley, J. L., 3rd, & Fillingim, R. B. (2004). Ethnic differe nces in pain coping: factor structure of the coping strategies questionnaire and co ping strategies questionnairerevised. J Pain, 5 (6), 304-316. Hauri, P. J., & Wisbey, J. (1992). Wrist actigraphy in insomnia. Sleep, 15 (4), 293-301. Headache Classification Committee of the Internationa l Headache Society. (1988). Classification and diagnostic criteria for h eadache disorders, cranial neuralgias and facial pain. Cephalalgia, 8 (suppl 7), 10-73. Horne, J. A., & Shackell, B. S. (1991). Alpha-like EEG activity in non-REM sleep and the fibromyalgia (fibrositis) syndrome. Electroencephalogr Clin Neurophysiol, 79 (4), 271276. Hurtig, I. M., Raak, R. I., Kendall, S. A., Ge rdle, B., & Wahren, L. K. (2001). Quantitative sensory testing in fibromyalgia patients and in healthy subjects: identification of subgroups. Clin J Pain, 17(4), 316-322. Jennum, P., Drewes, A. M., Andreasen, A., & Niel sen, K. D. (1993). Sleep and other symptoms in primary fibromyalgia and in healthy controls. J Rheumatol, 20 (10), 1756-1759. Jones, B. E. (1994). Basic mechanisms of sleep-wake states. In M. H. Kryger, T. Roth, D. W. C. & e. al. (Eds.), Principles and Practice of Sleep Medicine (2nd ed., pp. 145-162). Philadelphia, PA: W. B. Saunders. Katz, D. A., & McHorney, C. A. (2002). The re lationship between insomnia and health-related quality of life in patients with chronic illness. J Fam Pract, 51(3), 229-235.

PAGE 82

82 Kay, D. C., Eisenstein, R. B., & Jasinski, D. R. (1969). Morphine effects on human REM state, waking state and NREM sleep. Psychopharmacologia, 14 (5), 404-416. Kramer, J. J., & Conoley, J. C. (Eds.). (1992). The eleventh mental measurements yearbook. Lincoln, NE: Buros Institute of Mental Measurements. Kriegler, J. S., & Ashenberg, Z. S. (1987). Management of chronic low back pain: a comprehensive approach. Semin Neurol, 7(4), 303-312. Kundermann, B., Spernal, J., Huber, M. T., Kr ieg, J. C., & Lautenbacher, S. (2004). Sleep deprivation affects thermal pain thresholds but not somatosensory th resholds in healthy volunteers. Psychosom Med, 66 (6), 932-937. Kushida, C. A., Chang, A., Gadkary, C., Guille minault, C., Carrillo, O., & Dement, W. C. (2001). Comparison of actigraphic, polysom nographic, and subjective assessment of sleep parameters in slee p-disordered patients. Sleep Med, 2 (5), 389-396. Landis, C. A., Frey, C. A., Lentz, M. J., Roth ermel, J., Buchwald, D., & Shaver, J. L. (2003). Self-reported sleep quality and fatigue correlates with actig raphy in midlife women with fibromyalgia. Nurs Res, 52(3), 140-147. Lanes, T. C., Gauron, E. F., Spratt, K. F., Wernimont, T. J., Found, E. M., & Weinstein, J. N. (1995). Long-term follow-up of patients w ith chronic back pain treated in a multidisciplinary rehabilitation program. Spine, 20 (7), 801-806. Laursen, B. S., Bajaj, P., Olesen, A. S., Delmar C., & Arendt-Nielsen, L. (2005). Health related quality of life and quantitative pain measurement in females with chronic non-malignant pain. Eur J Pain, 9 (3), 267-275. Lichstein, K. L., Stone, K. C., Donaldson, J., Nau, S. D., Soeffing, J. P., Murray, D., et al. (2006). Actigraphy validation with insomnia. Sleep, 29(2), 232-239. Linton, S. J., Boersma, K., Jansson, M., Svar d, L., & Botvalde, M. (2005). The effects of cognitive-behavioral and physical therapy prev entive interventions on pain-related sick leave: a randomized controlled trial. Clin J Pain, 21 (2), 109-119. Lobbezoo, F., Visscher, C. M., & Naeije, M. (2004) Impaired health status, sleep disorders, and pain in the craniomandibular and cervical spinal regions. Eur J Pain, 8 (1), 23-30. Lockley, S. W., Skene, D. J., & Arendt, J. (1999). Comparison between subjective and actigraphic measurement of sleep and sleep rhythms. J Sleep Res, 8 (3), 175-183. Love, A., Leboeuf, C., & Crisp, T. C. (1989). Chir opractic chronic low back pain sufferers and self-report assessment methods. Part I. A reli ability study of the Visual Analogue Scale, the Pain Drawing and the McGill Pain Questionnaire. J Manipulative Physiol Ther, 12(1), 21-25.

PAGE 83

83 Lowe, N. K., Walker, S. N., & McCallum, R. C. (1991). Confirming the theoretical structure of the McGill Pain Questionnaire in acute clinical pain. Pain, 46 53-60. Macfarlane, T. V., Blinkhorn, A. S., Davies, R. M., Kincey, J., & Worthington, H. V. (2002). Oro-facial pain in the community: prevalence and associated impact. Community Dent Oral Epidemiol, 30(1), 52-60. Macfarlane, T. V., Kincey, J., & Worthingt on, H. V. (2002). The association between psychological factors and oro-faci al pain: a community-based study. Eur J Pain, 6 (6), 427-434. Maixner, W., Fillingim, R., Booker, D., & Sigur dsson, A. (1995). Sensitivity of patients with painful temporomandibular disorders to experimentally evoked pain. Pain, 63 (3), 341351. Malanga, G. A., & Nadler, S. F. (1999). Nonoperative treatment of low back pain. Mayo Clin Proc, 74(11), 1135-1148. McAlpine, T. H. (1987). Sleep, Divine and Human in the Old Testament Sheffield: Academic Press. McCracken, L. M., & Dhingra, L. (2002). A short version of the Pain Anxiety Symptoms Scale (PASS-20): preliminary development and validity. Pain Res Manag, 7 (1), 45-50. McCracken, L. M., Zayfert, C., & Gross, R. T. (1992). The Pain Anxi ety Symptoms Scale: development and validation of a scale to measure fear of pain. Pain, 50(1), 67-73. Melzack, R. (1975). The McGill Pain Questionna ire: major properties and scoring methods. Pain, 1(3), 277-299. Melzack, R., & Katz, J. (1992). The McGill Pain Questionnaire: Appraisal a nd current status. In D. C. Turk & R. Melzack (Eds.), Handbook of pain assessment (pp. 152-168). New York, NY: Guilford Press. Menefee, L. A., Frank, E. D., Doghramji, K., Picar ello, K., Park, J. J., Ja lali, S., et al. (2000). Self-reported sleep quality and quality of life for individuals with chronic pain conditions. Clin J Pain, 16(4), 290-297. Moffitt, P. F., Kalucy, E. C., Kalucy, R. S., Baum, F. E., & Cooke, R. D. (1991). Sleep difficulties, pain and other correlates. J Intern Med, 230 (3), 245-249. Moldofsky, H. (1989). Sleep and fibrositis syndrome. Rheum Dis Clin North Am, 15 (1), 91-103. Moldofsky, H. (1994). Central nervous system a nd peripheral immune functions and the sleepwake system. J Psychiatry Neurosci, 19 (5), 368-374.

PAGE 84

84 Moldofsky, H., & Lue, F. A. (1980). The relationship of alpha and delta EEG frequencies to pain and mood in 'fibrositis' patients treated with chlorpromazine and L-tryptophan. Electroencephalogr Clin Neurophysiol, 50 (1-2), 71-80. Moldofsky, H., Lue, F. A., & Smythe, H. A. ( 1983). Alpha EEG sleep and morning symptoms in rheumatoid arthritis. J Rheumatol, 10 (3), 373-379. Moldofsky, H., & Scarisbrick, P. (1976). Induction of neurasthenic musculoskeletal pain syndrome by selective sleep stage deprivation. Psychosom Med, 38 (1), 35-44. Moldofsky, H., Scarisbrick, P., England, R., & Sm ythe, H. (1975). Musculosketal symptoms and non-REM sleep disturbance in patients with "f ibrositis syndrome" and healthy subjects. Psychosom Med, 37 (4), 341-351. Montagna, P. (2006). Hypothalamus, sleep and headaches. Neurol Sci, 27 Suppl 2, S138-143. Morin, C. M. (2003). Measuring outcomes in random ized clinical trials of insomnia treatments. Sleep Med Rev, 7 (3), 263-279. Morin, C. M., Gibson, D., & Wade, J. (1998). Se lf-reported sleep and mood disturbance in chronic pain patients. Clin J Pain, 14 (4), 311-314. Moulin, D. E., Clark, A. J., Speechley, M., & Mo rley-Forster, P. K. (2002). Chronic pain in Canada--prevalence, treatment, impact and the role of opioid analgesia. Pain Res Manag, 7(4), 179-184. Mountz, J. M., Bradley, L. A., Modell, J. G., Al exander, R. W., Triana-A lexander, M., Aaron, L. A., et al. (1995). Fibromyalgia in wome n. Abnormalities of regional cerebral blood flow in the thalamus and the caudate nucleus are associated with low pain threshold levels. Arthritis Rheum, 38(7), 926-938. Nash, J. M., Park, E. R., Walker, B. B., Gordon, N., & Nicholson, R. A. (2004). Cognitivebehavioral group treatment for disabling headache. Pain Med, 5 (2), 178-186. Nicassio, P. M., & Wallston, K. A. (1992). L ongitudinal relationships among pain, sleep problems, and depression in rheumatoid arthritis. J Abnorm Psychol, 101 (3), 514-520. Ohayon, M. M. (1997). Prevalence of DSM-IV diagnostic criteria of in somnia: distinguishing insomnia related to mental disorders from sleep disorders. J Psychiatr Res, 31 (3), 333346. Ohayon, M. M. (2002). Epidemiology of insomnia: what we know and what we still need to learn. Sleep Med Rev, 6 (2), 97-111.

PAGE 85

85 Onen, S. H., Alloui, A., Gross, A., Eschallier, A ., & Dubray, C. (2001). The effects of total sleep deprivation, selective sleep inte rruption and sleep recovery on pain tolerance thresholds in healthy subjects. J Sleep Res, 10 (1), 35-42. Onen, S. H., Alloui, A., Jourdan, D., Eschalier, A., & Dubray, C. (2001). Effects of rapid eye movement (REM) sleep deprivation on pain sensitivity in the rat. Brain Research, 900 261-267. Paiva, T., Batista, A., Martins, P., & Martins, A. (1995). The re lationship between headaches and sleep disturbances. Headache, 35 (10), 590-596. Paiva, T., Esperanca, P., Martins, I., Batista A., & Martins, P. ( 1992). Sleep disorders in headache patients. Headache Quarterly, 3 438-442. Paiva, T., Farinha, A., Martins, A., Batista, A ., & Guilleminault, C. (1997). Chronic headaches and sleep disorders. Arch Intern Med, 157 (15), 1701-1705. Paiva, T., Martins, P., Batista, A., Esperanca, P., & Martins, I. (1994). Sleep disturbances in chronic headache patients: A comparison with healthy controls. Headache Quarterly, 5 135-141. Pau, A. K., Croucher, R., & Marcenes, W. (2003). Prevalence estimates and associated factors for dental pain: a review. Oral Health Prev Dent, 1 (3), 209-220. Paulson, P. E., Casey, K. L., & Morrow, T. J. (2002). Long-term changes in behavior and regional cerebral blood flow associated w ith painful peripheral mononeuropathy in the rat. Pain, 95 (1-2), 31-40. Pearce, J., & Morley, S. (1989). An experimental investigation of the construct validity of the McGill Pain Questionnaire. Pain, 39(1), 115-121. Pennebaker, J. W. (1982). The psychology of physical symptoms New York: Springer Verlag. Pillemer, S. R., Bradley, L. A., Crofford, L. J., Moldofsky, H., & Chrousos, G. P. (1997). The neuroscience and endocrinology of fibromyalgia. Arthritis Rheum, 40(11), 1928-1939. Pilowsky, I., Crettenden, I., & Townley, M. (1985) Sleep disturbance in pain clinic patients. Pain, 23(1), 27-33. Pollard, C. A. (1984). Preliminary validity study of the pain disability index. Percept Mot Skills, 59(3), 974. Price, D. D., & Bushnell, M. C. (1994). Overview of pain dimensions and their psychological modulation. In D. D. Price & M. C. Bushnell (Eds.), Psychological methods of pain control: Basic science and clinical perspectives (pp. 3-17). Seattle: IASP Press.

PAGE 86

86 Price, D. D., Harkins, S. W., & Baker, C. (1987) Sensory-affective relati onships among different types of clinical a nd experimental pain. Pain, 28(3), 297-307. Reite, M., Buysse, D., Reynolds, C., & Mendels on, W. (1995). The use of polysomnography in the evaluation of insomnia. Sleep, 18 (1), 58-70. Riley, J. L., 3rd, Benson, M. B., Gremillion, H. A., Myers, C. D., Robinson, M. E., Smith, C. L., Jr., et al. (2001). Sleep disturbance in orofacial pain patient s: pain-related or emotional distress? Cranio, 19(2), 106-113. Riley, J. L., 3rd, & Robinson, M. E. ( 1997). CSQ: five factors or fiction? Clin J Pain, 13 (2), 156162. Robinson, M. E., & O'Brien, E. M. (in press). Chronic pain. In Handbook of rehabilitation psychology (2nd edition ed.). Robinson, M. E., & Riley III, J. L. (1999). The role of emotion in pain. In R. J. Gatchel & D. C. Turk (Eds.), Psychosocial factors in pain (pp. 74-88). New York: Guilford Press. Robinson, M. E., Riley, J. L., 3rd, Myers, C. D., Sa dler, I. J., Kvaal, S. A., Geisser, M. E., et al. (1997). The Coping Strategies Qu estionnaire: a large sample, item level factor analysis. Clin J Pain, 13 (1), 43-49. Robinson, M. E., Wise, E. A., Riley III, J. L ., & Atchison, J. W. (1998). Sex differences in clinical pain: A multisample study. Journal of Clinical Psychol ogy in Medical Settings, 5(4), 413-424. Roelofs, J., McCracken, L., Peters, M. L., Crom bez, G., van Breukelen, G., & Vlaeyen, J. W. (2004). Psychometric evaluation of the Pain Anxiety Symptoms Scale (PASS) in chronic pain patients. J Behav Med, 27 (2), 167-183. Rosenstiel, A. K., & Keefe, F. J. (1983). The use of coping strategies in chronic low back pain patients: relationship to patient ch aracteristics and current adjustment. Pain, 17(1), 33-44. Russell, I. J., Michalek, J. E., Vipraio, G. A., Fl etcher, E. M., Javors, M. A., & Bowden, C. A. (1992). Platelet 3H-imipramine uptake recepto r density and serum serotonin levels in patients with fibromyalgia/fibrositis syndrome. J Rheumatol, 19 (1), 104-109. Sadeh, A., Hauri, P. J., Kripke, D. F., & Lavi e, P. (1995). The role of actigraphy in the evaluation of sleep disorders. Sleep, 18 (4), 288-302. Sahota, P. K., & Dexter, J. D. (1990). Sleep and headache syndromes: a clinical review. Headache, 30(2), 80-84.

PAGE 87

87 Sastre, J. P., Buda, C., Kitahama, K., & Jouvet, M. (1996). Importance of the ventrolateral region of the periaqueductal gray and adjacent tegmen tum in the control of paradoxical sleep as studied by muscimol micr oinjections in the cat. Neuroscience, 74 (2), 415-426. Sateia, M. J., Doghramji, K., Hauri, P. J., & Morin, C. M. (2000). Evaluation of chronic insomnia. An American Academy of Sleep Medicine review. Sleep, 23(2), 243-308. Sayar, K., Arikan, M., & Yontem, T. (2002). Sleep quality in chronic pain patients. Can J Psychiatry, 47 (9), 844-848. Schaefer, K. M. (1995). Sleep disturbances and fa tigue in women with fibromyalgia and chronic fatigue syndrome. J Obstet Gynecol Neonatal Nurs, 24 (3), 229-233. Seigel, J. M. (1994). Brainstem mechanism genera ting REM sleep. In M. H. Kryger, T. Roth, D. W. C. & e. al. (Eds.), Principles and Practic e of Sleep Medicine (2nd ed., pp. 122-144). Philadelphia, PA: W. B. Saunders. Shapiro, C. M., Devins, G. M., & Hussain, M. R. (1993). ABC of sleep disorders. Sleep problems in patients with medical illness. Bmj, 306(6891), 1532-1535. Shaver, J. L., Lentz, M., Landis, C. A., Heitkem per, M. M., Buchwald, D. S., & Woods, N. F. (1997). Sleep, psychological distress, and st ress arousal in women with fibromyalgia. Res Nurs Health, 20 (3), 247-257. Smith, B. H., Elliott, A. M., Chambers, W. A ., Smith, W. C., Hannaford, P. C., & Penny, K. (2001). The impact of chronic pain in the community. Fam Pract, 18 (3), 292-299. Smith, M. T., Edwards, R. R., McCann, U. D., & Haythornthwaite, J. A. (2007). The effects of sleep deprivation on pain inhibiti on and spontaneous pain in women. Sleep, 30(4), 494505. Smith, M. T., & Haythornthwaite, J. A. (2004). Ho w do sleep disturbance a nd chronic pain interrelate? Insights from the longitudinal and cogni tive-behavioral clinical trials literature. Sleep Med Rev, 8 (2), 119-132. Smythe, H. A. (1995). Studies of sleep in fibrom yalgia; techniques, clin ical significance, and future directions. Br J Rheumatol, 34 (10), 897-899. Spielberger, C. (1988). State-Trait Anger Expression Invento ry, research edition. Professional manual. Odessa, FL: Psychological Assessment Resources. Staud, R., Cannon, R. C., Mauderli, A. P., Robinson, M. E., Price, D. D., & Vierck, C. J., Jr. (2003). Temporal summation of pain from mech anical stimulation of muscle tissue in normal controls and subjects with fibromyalgia syndrome. Pain, 102(1-2), 87-95.

PAGE 88

88 Steriade, M., & Llinas, R. R. (1988). The functiona l states of the thalamus and the associated neuronal interplay. Physiol Rev, 68 (3), 649-742. Swartzman, L. C., Gwadry, F. G., Shapiro, A. P., & Teasell, R. W. (1994) The factor structure of the Coping Strategies Questionnaire. Pain, 57(3), 311-316. Tait, R. C., Chibnall, J. T., & Krause, S. ( 1990). The Pain Disability Index: psychometric properties. Pain, 40(2), 171-182. Tait, R. C., Pollard, C. A., Margolis, R. B., Du ckro, P. N., & Krause, S. J. (1987). The Pain Disability Index: psychometric and validity data. Arch Phys Med Rehabil, 68(7), 438441. Takahashi, L. K. (2002). Neurobi ology of schizophrenia, mood diso rders, and anxiety disorders. In K. L. McCance & S. E. Huether (Eds.), Pathophysiology: The biologic basis for disease in adults and children (4th edition ed., pp. 550565). St. Louis: Mosby. Turk, D. C., Rudy, T. E., & Salovey, P. (1985). The McGill Pain Questi onnaire reconsidered: confirming the factor structure and examining appropriate uses. Pain, 21(4), 385-397. Turner, J. A., Mancl, L., & Aaron, L. A. (2006). Shortand long-term efficacy of brief cognitivebehavioral therapy for patients with chr onic temporomandibular disorder pain: a randomized, controlled trial. Pain, 121(3), 181-194. Tuttle, D. H., Shutty, M. S., & DeGood, D. E. (1991). Empirical dimensions of coping pain patients: A factorial analysis. Rehabilitation Psychology, 36 (3), 179-188. Vaeroy, H., Helle, R., Forre, O., Kass, E., & Terenius, L. (1988). Cerebrospinal fluid levels of beta-endorphin in patients with fibr omyalgia (fibrositis syndrome). J Rheumatol, 15 (12), 1804-1806. Vallieres, A., & Morin, C. M. (2003). Actig raphy in the assessment of insomnia. Sleep, 26(7), 902-906. Verhaak, P. F., Kerssens, J. J., Dekker, J., Sorb i, M. J., & Bensing, J. M. (1998). Prevalence of chronic benign pain disorder among a dults: a review of the literature. Pain, 77 (3), 231239. Verma, S., & Gallagher, R. M. (2002). The ps ychopharmacologic treatmen t of depression and anxiety in the contex t of chronic pain. Curr Pain Headache Rep, 6 (1), 30-39. Waters, W. F., Hurry, M. J., Binks, P. G., Carney, C. E., Lajos, L. E., Fuller, K. H., et al. (2003). Behavioral and hypnotic treatments for insomnia subtypes. Behav Sleep Med, 1 (2), 81101.

PAGE 89

89 Widerstrm-Noga, E., Dyrehag, L. E., Brglum-Jensen, L., Aslund, P. G., Wenneberg, B., & Andersson, S. A. (1998). Pain threshold responses to two different modes of sensory stimulation in patients with orofacial mu scular pain: psycholo gic considerations. J Orofac Pain, 12 (1), 27-34. Widerstrom-Noga, E. G., Felipe-Cuervo, E., & Yezier ski, R. P. (2001). Chronic pain after spinal injury: interference with sleep and daily activities. Arch Phys Med Rehabil, 82(11), 15711577. Wilson, K. G., Watson, S. T., & Currie, S. R. (1998). Daily diary and ambulatory activity monitoring of sleep in patients with insomnia associated with chronic musculoskeletal pain. Pain, 75 (1), 75-84. Wittig, R. M., Zorick, F. J., Blumer, D., Heilbronn, M., & Roth, T. (1982). Disturbed sleep in patients complaining of chronic pain. J Nerv Ment Dis, 170 (7), 429-431. Wolfe, F., Anderson, J., Harkness, D., Bennett, R. M., Caro, X. J., Goldenberg, D. L., et al. (1997). A prospective, longitudinal, multicente r study of service utilization and costs in fibromyalgia. Arthritis Rheum, 40(9), 1560-1570. Wolfe, F., Smythe, H. A., Yunus, M. B., Bennett, R. M., Bombardier, C., Goldenberg, D. L., et al. (1990). The American College of Rheuma tology 1990 Criteria for the Classification of Fibromyalgia. Report of the Multicenter Criteria Committee. Arthritis Rheum, 33(2), 160172. Yatani, H., Studts, J., Cordova, M., Carlson, C. R., & Okeson, J. P. (2002). Comparison of sleep quality and clinical and psyc hologic characteristics in patie nts with temporomandibular disorders. J Orofac Pain, 16 (3), 221-228.

PAGE 90

90 BIOGRAPHICAL SKETCH Erin Maureen OBrien was born in 1978 in Pisc ataway, New Jersey. The oldest of three daughters, she grew up in Marl ton, New Jersey and graduated from Cherokee High School in 1996. She attended Saint Josephs University in Philadelphia, Pennsylvania, and was accepted into a 5-year B.S./M.S. program, where she earned a B.S. in psychology and an M.S. in experimental psychology in 2000 and 2001, respectively. Her masters thesis was titled, Sleep and Risk-Taking Behavior in Adolescents and this work was published in 2005. Upon graduating in 2001 with her M.S. in experimental psyc hology, Erin obtained a position as a clinical research coordinator in the Center for Sleep and Respiratory Neurobiology at the University of Pennsylvania. Here she coordinated the progressi on of several large NIHfunded clinical research protoc ols under the direction of Dr. Allan Pack. This experience solidified Erins interest in pursuing additional study in the fi eld of psychology and resulted in her application and acceptance into the doctoral progr am at the University of Florida so that she could earn her Ph.D. in clinical psychology. He re she pursued her dual in terests in the study of sleep disturbances and chroni c pain, which culminated in her dissertation research. As part of her doctoral program, Erin comple ted a year-long clinical internship at the Warren Alpert Medical School at Brown University where she pursued additional training in the area of behavioral medicine. Upon completion of her Ph.D. program, Erin will be continuing her research and clinical work as a post-doctoral fellow in the Methods to Improve Diagnostic Assessment and Services (MIDAS) clinical-research program at Brown University, with a focus on working with patients presenting with insomnia and other sleep disorders.