The quarter of an hour rule: a simplified cognitive-behavioural intervention for insomnia improves sleep (2024)

Malaffo, Marina (2006) The quarter of an hour rule: a simplified

cognitive-behavioural intervention for insomnia improves sleep. PhD

thesis.

http://theses.gla.ac.uk/1529/

Copyright and moral rights for this thesis are retained by the author A copy can be downloaded for personal non-commercial research or study, without prior permission or charge

This thesis cannot be reproduced or quoted extensively from without first obtaining permission in writing from the Author

The content must not be changed in any way or sold commercially in any format or medium without the formal permission of the Author

When referring to this work, full bibliographic details including the author, title, awarding institution and date of the thesis must be given

Glasgow Theses Service

http://theses.gla.ac.uk/

, em

I

ß'

1 I.

UNIVERSITY

of

GLASGOW

The Quarter of an Hour Rule:

a simplified cognitive-behavioural

SLEEP RESEARCH GROUP

intervention for insomnia improves sleep

Author:

Marina Malaffo

Submitted

for the degree of Ph. D. to the

higher degrees committee of the Faculty of Medicine,

Department

_{of Psychological}

Medicine,

University of Glasgow

Acknowledgements

I would like to take the opportunity to thank my supervisor Prof Colin

A. Espie for the time he devoted to helping me with this thesis and for

his suggestions, support and enthusiasm throughout this research.

My gratitude goes also to the sleep research group for providing a space

to discuss ideas constructively, to my parents for their constant

encouragement and to Gregor for kindly proof-reading my writing.

Thanks also to the GPs who referred their patients to the trial and,

especially, to the participants who kindly devoted considerable amount

of their time to be tested and provide daily data.

Declaration

I, Marina

Malaffo,

declare that this is my own work,

carried out under the normal term of supervision.

'The copyright of this thesis belongs to the author under the terms of the

United Kingdom Copyright Acts as qualified by University of Glasgow.

Due acknowledgement must always be made of the use of any material

contained in, or derived from, this thesis.

'

Abstract

Stimulus control (SC) is a core component of cognitive behavioural therapy (CBT) for insomnia and is the single intervention for which there is most empirical evidence. Nonetheless, little is known about whether all of the elements within SC are critical to sleep improvement. This study, therefore, investigated the impact on sleep of the Quarter of an Hour Rule (QHR) a single, situational element considered central to SC for insomnia. The mechanisms of effect of SC intervention remain also unclear. An associated aim of the present study, therefore, was to contrast two forms

of administration of the QHR to test aspects of the learning theory presumed to underlie the SC model. In addition adherence to the behavioural intervention was investigated and the possibility of using actigraphy to measure adherence objectively was explored.

Prior to the randomised controlled trial (primary study), two preliminary studies were conducted. The first preliminary study aimed at determining the optimal cut-off to represent normalcy in sleep onset latency (SOL). The results indicated it to be fifteen minutes and, therefore, participants in studies two and three were asked to apply the QHR if they were not asleep within a quarter of an hour. Study two comprised three single cases and tested the feasibility of the QHR as a standalone therapy for insomnia. Visual inspection of the data and interrupted time series analyses evidenced SOL, wake after sleep onset (WASO) and sleep efficiency (S. E. ) improvements in two out of three participants. Their Pittsburgh Sleep Quality Index (PSQI) score at the end of the intervention was reduced by 50% compared to baseline. The participant, whose sleep was not improved, following the intervention, had not applied the QHR. The results of this exploratory, single case, study warranted further investigation of the QHR.

In study three forty-one GP and self referred volunteers, aged 18-72 years, with SOL and/or WASO complaints, formed 3 randomised groups: QHRin bed, QHRout of bed and control. Both QHR conditions required to `read if not asleep within a quarter of an hour', with groups differing only with the location (in bed versus out of bed) where to apply the QHR. Sleep diary pre-treatment (two weeks) and post- treatment (three weeks), home polysomnography (PSG) (two nights pre-, two post-

treatment) and sleep related questionnaire (pre and post) data were collected. Adherence with the QHR was measured objectively (actigraphy + light monitoring) and subjectively (adherence diary).

Following QHR treatments, statistically significant reductions in SOL (QHRout) and WASO (QHRin and QHRout), an increase in S. E. (QHRin and QHRout) and a decrease in PSQI score (QHRin and QHRout) were found. Trends also indicated increased total sleep time (TST). Clinically significant improvements (SOL and WASO S 31 minutes or reduced by 50%, PSQI : 55 or reduced by 50%) were obtained in 33-57% of active groups participants. Furthermore, 43% of QHRin and 66% of QHRout participants achieved S. E.

_85%. Magnitude of effect was greater for the QHRout than the QHRin and changes were achieved within the first week of treatment. No differences were found in PSG defined sleep variables. Nevertheless,

subjective-objective discrepancies (i. e. differences between sleep diary and PSG defined sleep) were reduced following intervention. Sleep related questionnaires indicated fewer sleep pattern problems, less cognitive and somatic arousal and less cognitive activity (sleep effort, problem solving and sensory engagement) following QHR interventions. Participants tended to adhere in categorical terms (either adhering most times (majority) or not adhering at all). Adherence was adequate indicating that the QHR is well tolerated and accepted. Importantly, the combination of actigraphy and light provided a tool to measure adherence objectively.

The QHR, a single instruction administered in the form of brief therapy, was found to improve subjectively reported sleep both statistically and clinically. The finding that both QHRin and QHRout were efficacious for WASO sheds doubt on the argument that for difficulties maintaining sleep, the QHR component works

solely via conditioning principles. However it is possible that re-conditioning effects apply not only to external (environmental) factors as posited by SC theory but, also, to internal (cognitive) factors. The finding that sub-ob discrepancies decreased

following QHR intervention is discussed in terms of levels of arousal and stimuli engagement at the time of falling asleep and during PSG defined sleep. Further research is warranted.

Declaration

I, Marina Malaffo, declare that this is my own work, carried out under the normal terms of supervision.

'The copyright of this thesis belongs to the author under the terms of the United Kingdom Copyright Acts as qualified by University of Glasgow. Due

acknowledgement must always be made of the use of any material contained in, or derived from, this thesis. '

Acknowledgements

I would like to take the opportunity to thank my supervisor Prof Colin A. Espie for the time he devoted to helping me with this thesis and for his suggestions, support and enthusiasm throughout this research.

My gratitude goes also to the sleep research group for providing a space to discuss ideas constructively, to my parents for their constant encouragement and to Gregor for kindly proof-reading my writing.

Thanks also to the GPs who referred their patients to the trial and, especially, to the participants who kindly devoted considerable amount of their time to be tested and

provide daily data.

Table of Contents

Abstract ... _.. P. 1 Declaration ... _.. P. 3 Acknowledgments ... _.. P. 4 Table of contents ... _.. P. 5 Chapter 1- Introduction 1.1 Overview ... _.. P. 13 1.2 Insomnia ... _.. P. 14 1.2.1 Basic Facts about Sleep _{... .. P. 14} 1.2.2 Definition and Classification of Insomnia _{... .. P. 17}

1.2.3 Epidemiology of Insomnia _{... .. P.} 21

1.3 Assessment of Insomnia and Monitoring Treatment Efficacy _{. .. P. 23} 1.3.1 Subjective Measures ... .. P. 24 1.3.1.1 Sleep Questionnaires ... . P. 24 1.3.1.2 Sleep Diary ... . P. 25

1.3.2 Objective Measures of Sleep _...

. P. 27

1.3.2.1 Polysomnography

... _. P. 27 1.3.2.2 Actigraphy

... _. P. 30 1.4 Theoretical Perspectives on Insomnia _...

. P. 33 1.4.1 Physiological Perspective of Insomnia _{... . P.} 33 1.4.2 Environmental/Behavioural Perspective of Insomnia _{... . P.} 35

1.4.2.1 Stimulus Control

... _. P. 37 1.4.2.2 The Three Factor Model

... _. P. 39 1.4.3 Cognitive Perspective of Insomnia _...

. P. 41 1.4.3.1 Cognitive Model of the Maintenance of Insomnia _{.. . P.} 44

1.4.4 Integrative Models of Insomnia _...

. P. 47

1.4.4.1 Insomnia as a Psychobehavioural Disorder _...

. P. 48 1.4.4.2 An Integrative Model of Insomnia _...

. P. 49 1.4.4.3 The Neurocognitive Model of Insomnia _...

. P. 52 1.4.4.4 The Psychobiological Model of Insomnia _...

. P. 55 1.5 Therapeutic Approaches to the Treatment of Insomnia _{... . P.} 57

1.5.1 Treatments Related to the Physiological Perspective .... P. 58 1.5.1.1 Pharmacotherapy ... P. 58 1.5.1.2 Sleep Hygiene ... P. 59 1.5.1.3 Relaxation ... P. 60

1.5.2 Treatments Related to the Behavioural Perspective

... P. 60 1.5.2.1 Sleep Restriction

... P. 60 1.5.2.2 Stimulus Control Therapy for Insomnia

... P. 61

1.5.3 Cognitive Treatments for Insomnia

... P. 62

1.5.4 Multifaceted Cognitive Behavioural Therapy

... P. 63 1.6 Treatment Outcome Evidence

... P. 63 1.6.1 Meta-Analytic Procedures ... P. 64 1.6.2 Meta-Analytic Findings ... P. 64 1.6.3 Clinical Studies ... P. 68

1.6.4 Is CBT the Treatment of Choice? _{. ...} P. 68 1.6.5 Considerations regarding Meta-Analytic Findings _.... P. 69

1.7 Stimulus Control: a detailed Description and Critical Appraisal ... P. 72 1.7.1 Stimulus Control Therapy: its Elements

... P. 72

1.7.2 Stimulus Control Therapy: Outcome Studies

... P. 76 1.7.3 `Get up and go to another room if not asleep': a feasible therapy? P. 87 1.8 Treatment Response and Moderating Variables

... P. 89 1.8.1 Treatment Adherence

... P. 89

1.8.2 Credibility and Expectancy _{... P. 92} 1.8.3 Other Moderating Variables

... P. 93 1.9 Summary of the Introduction _{... P. 94} Chapter 2- Overview of the Present Research

2.1 Rationale

... p. 96

2.2 Aims

... P. 99

2.3 Ethical Issues and Approval _{... P. 100}

Chapter 3- Study One: A Cross-Sectional Comparison of Sleep Behaviour in Normal and Poor Sleepers

3.1 Rationale

... _... p. 102 3.2 Aims and Hypotheses _{... .. P. 104} 3.3 Methods ... .. P. 104 3.3.1 Design ... .. P. 105 3.3.2 Participants ... .. P. 105 3.3.3 Materials and Apparatus _{... ..} P. 105

3.3.3.1 Diary ... .. P. 106 3.3.3.2 Sleep-Questionnaire ... .. P. 107 3.3.4 Data Processing ... .. P. 107 3.3.5 Procedure ... .. P. 107 3.4 Results ... .. P. 107 3.4.1 Participants' Characteristics ... .. P. 107

3.4.2 Sleep Continuity and Sleep Quality Data ... .. P. 108 3.4.3 Upper Limit of Sleep Onset for Normal Sleepers ... .. P. 110 3.4.4 Sleep Behaviour at Bedtime and in the Morning _{... ..} P. 111

3.4.5 Reasons for Waking Up in the Morning

... .. P. 112 3.5 Discussion

... .. P. 112

3.5.1 Summary of the Results _{... ..} P. 112 3.5.2 Interpretations of the Results _{... ..} P. 113 3.5.3 Considerations for study two _{... ..} P. 115 Chapter 4- Study Two: Could a Single Element of the Stimulus Control

Package Improve Sleep? An Exploratory Study 4.1 Rationale

... . p. 117 4.2 Aims and Hypotheses _{... .} P. 118 4.3 Methods ... . P. 119 4.3.1 Design ... . P. 119 4.3.2 Participants ... . P. 120 4.3.3 Materials and Apparatus _{... . P.} 121

4.3.3.1 Sleep Diary

... . P. 121 4.3.3.2 Sleep Assessment and Anxiety and Depression Scale _{. . P.} 121

4.3.3.2a The Pittsburgh Sleep Quality Index

... . P. 121

4.3.3.2 b The Screening Interview

... _. P. 121 4.3.3.2c The Hospital Anxiety and Depression Scale _{... . P. 122}

4.3.3.3 Objective Measurement of Sleep _...

. P. 122

4.3.3.3a Ambulatory EEG/PSG Recording Device

.. . P. 122 4.3.3.3b Analysis of the Electroencephalographic Recording _{. . P.} 123 4.3.3.4 Measures of Adherence _{... . P. 124} 4.3.3.4a Actigraphy ... . P. 125 4.3.3.4b Adherence Diary ... . P. 125 4.3.3.4c Adherence Score ... . P. 126 4.3.3.5 The Quarter of an Hour Rule _{... .} P. 126 4.3.4 Procedure ... . P. 127 4.3.5 Data Analysis ... . P. 129 4.4 Results ... . P. 130 4.4.1 Participants' Characteristics ... . P. 130

4.4.2 Sleep Pattern Changes: Visual Inspection of Sleep Diary Data

and ITSA ... . P. 131 4.4.2.1 Complaint of Difficulty of Falling Asleep _{... .} P. 131 4.4.2.2 Complaint of Difficulty Staying Asleep _{.... . P.} 132 4.4.3 Clinical Significance

... . P. 134 4.4.4 Objective Sleep Continuity Data

... . P. 135 4.4.5 Adherence with the Quarter of an Hour Rule _{... .} P. 136 4.4.6 Adherence Comments

... . P. 137 4.5 Discussion

... _. P. 138 4.5.1 Summary of the Results _{... .} P. 138 4.5.2 Interpretation of the Results _{... .} P. 139 4.5.3 Considerations for Study Three

... P. 142

Chapter 5- Study Three: Impact on Sleep Parameters of the Quarter of

an Hour Rule Intervention Implemented In Bed and Out of Bed 5.1 Rationale

... p. 145

5.2 Aims and Hypotheses _{... p. 146} 5.3 Methods

... p. 148

5.3.1 Design

... p. 148

5.3.2 Participants

... P. 150

5.3.3 Materials and Apparatus ... P. 151

5.3.3.1 The Screening Interview

... P. 151

5.3.3.2 Subjective Measurement of Sleep and Sleep Questionnaires p. 152 5.3.3.2a Sleep Diary

... _. P. 152 5.3.3.2b Sleep Related Questionnaires

... P. 152 5.3.3.3 Objective Measurement of Sleep _... P. 155

5.3.3.3a Objectively Measured Sleep

... P. 156 5.3.3.3b Ambulatory EEG/PSG Recording Device

.... P. 156 5.3.3.3c Analysis of the Polygraphic Recording ... P. 157 5.3.3.3d Sleep Scoring

... P. 158 5.3.3.4 Subjective Measurement of Adherence ... p. 159 5.3.3.5 Objective Measurement of Adherence _... P. 159

5.3.3.5a Actigraphy

... P. 159 5.3.3.5. b Adherence Algorithm and Adherence Score _.. P. 160

5.3.3.6 Material for the Intervention

... P. 162

5.3.3.6a The Single Component Interventions

... P. 162 5.3.3.6b Self-Monitoring: The Control Group

... P. 163 5.3.4 Data Processing ... P. 163 5.3.5 Procedure ... P. 165 5.4 Results ... P. 169 5.4.1 Participants' Characteristics ... P. 170 5.4.2 Subjective Sleep Continuity Data and PSQI Total Score _.. P. 171 5.4.2.1 Credibility of the QHR Treatment ... P. 172 5.4.2.2 Descriptive Sleep Continuity Data Gathered via Sleep Diary p. 172 5.4.2.3 Sleep Onset Latency

... P. 173 5.4.2.4 Wake After Sleep Onset

... P. 174 5.4.2.5 Total Sleep Time

... P. 175 5.4.2.6 Sleep Efficiency

... P. 176 5.4.2.7 Pittsburgh Sleep Quality Index Total Score

... P. 177 5.4.2.8 Summary of Subjectively Measured Sleep Continuity _. P. 178

5.4.3 Sleep Quality Data

... _. P. 179 5.4.3.1 Descriptive Sleep Quality Data

... . P. 179 5.4.3.2 Sleep Quality

... _. P. 180 5.4.3.3 Restedness Upon Awakening

... . P. 180 5.4.3.4 Alertness Upon Awakening

... . P. 181 5.4.3.5 Summary of the Sleep Quality Results _{... . P.} 183 5.4.4 Clinical Changes Associated to the QHRin and QHRout _{. .} P. 183

5.4.4.1 Clinical Significance

... . P. 183

5.4.4.2 Therapeutics Effects at week one, two and three ... . P. 184 5.4.4.2a Response Curve for Sleep Onset Latency

.. . P. 185 5.4.4.2b Response Curve for Wake After Sleep Onset

.. . P. 186 5.4.4.2c Response Curve for Total Sleep Time _{... . P. 187}

5.4.4.2d Response Curve for Sleep Efficiency.

.... . P. 188

5.4.4.2e Summary of the Response Curve Findings ... . P. 189 5.4.5 Sleep Related Questionnaires

... . P. 190 5.4.5.1 Descriptive Data

... . P. 190 5.4.5.2 Q1: Sleep Related Self Rating Scale-R

.... . P. 190 5.4.5.3 Q2: Glasgow Sleep Effort Scale

... . P. 191 5.4.5.4 Q3: Sleep Disturbance Questionnaire

... . P. 192 5.4.5.4a Cognitive Arousal

... . P. 192 5.4.5.4b Physical Tension

... . P. 192 5.4.5.4c Sleep Effort

... . P. 193 5.4.5.4d Sleep Pattern Problems

... . P. 193 5.4.5.5 Q4: Glasgow Content of Thoughts Inventory _{... .} P. 194

5.4.5.5a Active Problem Solving

... . P. 195 5.4.5.5b Sleep and Wakefulness _{... .} P. 195

5.4.5.5c Somatic and Sensory Engagement _{.... .} P. 195

5.4.5.6 Q5: Pre-Sleep Arousal Scale

... . P. 196 5.4.5.6a Cognitive Arousal

... . P. 196 5.4.5.6b Somatic Arousal

... . P. 197 5.4.5.7 Q6: The DBAS-10

... _{. P.} 197 5.4.5.8 Summary of the Sleep Related Questionnaires Results _{. P.} 197

5.4.6 Objective Sleep Continuity Data

... P. 199 5.4.6.1 Sleep Continuity Data Gathered via Polysomnography _{. P. 199} 5.4.6.2 Sleep Architecture Data

... P. 200

5.4.7 Subjective and Objective Sleep Continuity: Association

... P. 201

5.4.7.1 Descriptive Subjective and Objective Sleep Continuity _{.. P. 201} 5.4.7.2 Associations between Objective and Subjective Measures

of Sleep Continuity Variables _{... P. 204}

5.4.8 Adherence with the Quarter of an Hour Rule _{... p. 205} 5.5 Discussion

... P. 207

5.5.1 Summary of Results _{... P. 208} 5.5.2 Interpretation of Results _{... P. 210}

5.5.2.1 Subjective Sleep Continuity Data

... P. 210 5.5.2.2 The PSQI Total Score

... P. 213 5.5.2.3 Sleep Quality Data

... P. 214 5.5.2.4 Clinical Changes Associated with the Quarter of an Hour Ru le p. 215

5.5.2.5 Sleep Related Questionnaires

... P. 217

5.5.2.6 Objective Sleep Continuity Data

... P. 220 5.5.2.7 Subjective and Objective Sleep Continuity: Association _{. P. 227} 5.5.2.8 Adherence with the Quarter of an Hour Rule _.... P. 227 Chapter 6- General Issues and Future Research

6.1 Theoretical Implications

... _... P. 230 6.2 Concluding Comments

... _... P. 233 6.3 Limitations of the Present Study _{... ... P. 235} 6.4 Future Research ... ... P. 239 6.5 Concluding Comments ... ... P. 241 References ... ... P. 244 List of Tables and Figures _{... ... P. 264} Appendixes Appendix 1... ... P. 267 Appendix 2... ... P. 268 Appendix 3... ... P. 273 Appendix 4... ... P. 274

11

CHAPTER 1

Introduction

4

1.1 Overview

This chapter presents an overview of insomnia, its classifications, characteristics and epidemiology. Insomnia assessment and diagnostic issues are also considered

with a view to what questions each of these methods can help to answer. Thereafter

the three main perspectives of insomnia and some theoretical models within which it

is possible to understand the emergence, maintenance and exacerbation of insomnia are outlined and evidence supporting and / or disconfirming these perspectives and

models are presented. By taking into account the psychological and behavioural

factors presumed to be involved in insomnia, a rationale for the use of psychological treatments is proposed. Psychological interventions for insomnia are briefly outlined and treatment outcome studies are critically reviewed so as to highlight what is still to be done. Furthermore, variables moderating treatment efficacy are discussed and a call for investigation of adherence to treatment is made. Drawing on the models and current limitations in insomnia treatment research, and the recent guidelines on

insomnia, it is proposed that the investigation of one single component of the behavioural treatment for insomnia stimulus control, namely the quarter of an hour rule (QHR), can be employed not only to mitigate sleep difficulties but, importantly,

to shed light on mechanisms of effects and the minimum dose needed to achieve

therapeutic gains. In addition, employing only one single element permits a closer

look at adherence and to the testing of an objective methodology to measure and monitor adherence and its effects on therapy outcomes.

1.2 Insomnia

Lay-person definitions of insomnia include "not being able to sleep well", "taking ages to fall asleep", "spending hours awake during the night" and, as a result, `feeling very tired during the day with difficulties concentrating", `feeling

lethargic and with snappy moods". The self-labelled `insomniacs' will probably go on to say that they would love to have a good night's sleep, that they have tried

everything (but nothing worked), that they don't want drugs or that drugs are not

working anymore. Some will remember how easy it was to sleep once, and how good

it feels when they have the occasional good night.

Insomnia is so widespread that most people have an idea of what insomnia is, although the `insomnia ingredients' vary according to personal (one's own or

someone else's) experiences. But what is insomnia really? Perhaps, as proposed by Espie (1991,2002) the best way to start answering this question is to understand what normal sleep is and what regulates it.

1.2.1 Basic Facts about Sleep

Sleep is not a unitary state: while the observer can only see the two opposite states, that is wakefulness and sleep, the individual experiences different types of sleep such as deep and light sleep and dreaming. Research has corroborated the

subjective experience of sleep as a non-unitary state by the measuring of the

electrical currents of the brain.

The first division of sleep is that of non-rapid eye movement (NREM) sleep and REM sleep. As illustrated in figure 1, NREM sleep is further subdivided into four distinct stages.

Figure 1: _{An example of EEG traces during NREM and REM stages} Stage 1 Stage 2_L, ý ýýýrVlff ý'ý Slow Waves (stages 3& 4) r ýtNyw, ý/y, vM REM sleep

Stage 1 is a very light sleep characterised by a relatively low-voltage and mixed frequency activity where theta rhythm (3 to 7 cycles per second - cps) and vertex sharps are common. Stage 2 represents light sleep as well and the EEG shows a

pattern of relatively low voltage with mixed frequency activity. Two specific EEG

patterns enable the distinction of stage 2 from stage 1: the sleep spindles (12-14 cps,

with a duration of 0.5 to 1.5 seconds) and k-complex (a well delineated sharp wave

of duration exceeding 0.5 seconds which is immediately followed by a positive

component and stands out from the ongoing background activity). Each epoch (30

seconds) in stage 2 comprises up to 20% of delta waves (delta waves have high

voltage, i. e. amplitudes greater than 75 millivolts, and there are maximum 2 for cps). Deep sleep is seen in stage 3 and 4. Stage 3 is defined by the presence of delta waves in a percentage between 20 and 50 per each epoch of EEG recording. Stage 4, finally, is characterised by delta waves in more than 50% of the epoch. REM sleep is characterized by EEG activation: relatively low voltage, mixed frequency (typical of

relaxed wakefulness) at the scalp level coupled to rapid eye movement and motor

atonia (i. e. loss of muscle tone). Detailed guidelines and criteria for staging sleep are

provided by the standard sleep staging manual (Rechtschaffen and Kales, 1968).

Sleep is, then, a highly organised activity and it follows a cyclic pattern. A normal adult enters sleep via NREM stages (from 1 to 4); this sequence is then reversed from stage 4 to 2 leading to the first REM episode. This first cycle of sleep takes about 70

to 120 minutes and on average, during one night, four to five of these cycles happen. In Table 1 generalisations about sleep in normal and primary insomnia adults are provided. These PSG values for sleep continuity and sleep architecture measures

were estimated by Smith, Nowakowski, Soeffing, Orff and Perlis (2003) from

literature.

Table 1- Average PSG Sleep variables for Healthy Normal Sleepers and Primary Insomnia

can Measures .'i le _{onto ujý}

SOL 10 (minutes) 56 (minutes) WASO 10 (minutes) 45 (minutes) TST 420 (minutes) 352 (minutes) SE _>90% 79% Wake 5% 11% Stage 1 5% 10% Stage 2 60% 56% Stage 3&4 15% 12% Stage REM 15% 22%

The question is: what regulates sleep?

In the 1970s Borbely and his colleagues in Zurich, began the exploration of a two- factor theory of the regulation of sleep based on experimental work with both animals and humans, which was then formulated in 1984 (Daan, Beersma and

Borbely, 1984). In brief, they proposed that two `opponent processes' regulate sleep: on one hand a homeostatic sleep drive, which is constantly active, promotes and

maintains sleep (the longer one is without sleep the greater the sleep debt and as a

consequence the drive to sleep); on the other hand wakefulness is fostered by a

biological clock (regulated by a circadian timer), active during the day and inactive at night which acts as a kind of arousal system (Carskadon and Dement, 1981;

Dement and Vaughan, 1999). In addition environmental and social factors influence sleep (e. g. shift work) (Webb, 1988).

An important point was made by Espie (2002): normal sleep is a `passive' state that does not require effort, the 'good' sleeper falls asleep automatically without

forcefully trying, in fact without trying at all.

Having discussed normal sleep, its regulation and `passivity' it is easier to define and understand insomnia.

1.2.2 Definition and Classification of Insomnia

Broadly speaking insomnia refers to difficulty initiating and/or maintaining sleep. A more specific set of sleep disorders, all of which have in common a complaint of

insomnia, has been provided by classification systems such as the Diagnostic and Statistical Manual of Mental Disorders (DSM) and the International Classification of Sleep Disorders (ICSD). Insomnia can be broadly subdivided into two categories: primary insomnia (i. e. independent, no comorbid symptoms) and secondary insomnia

(i. e. insomnia is instigated by other sleep, psychiatric or medical conditions). It is noteworthy that determining causality is often difficult (Smith, Smith, Nowakosky

and Perlis, 2003) and that a diagnosis of primary insomnia is often made by

exclusion (Morin, 1993).

Insomnia is also characterized by its duration: in the DSM insomnia is defined as `acute' if it lasts for less than a month and it is accompanied by specific events

precipitating the sleep problems (e. g. pain, bereavement). Insomnia is, instead,

considered `chronic' when the symptoms last longer than one month despite the

stressor having been resolved or the person having adjusted to it. According to the

ICSD-R, instead, insomnia is considered chronic if it lasts for at least six months.

Typically if the individual presents with secondary insomnia the treatment will focus on the disorder (for example depression) deemed to give rise to the insomnia

symptoms. In instances of acute insomnia the treatment choice, if treatment is

provided, is pharmacotherapy. When the individual presents with primary insomnia

symptoms, instead, behavioural interventions for insomnia are provided, if available,

as either standalone treatment or in addition to pharmacotherapy.

The present research investigates psychological interventions for insomnia, hence attention will focus on insomnia when it is chronic and a primary disorder.

In short, primary insomnia is defined by DSM-IV as a complaint of initiating and/or maintaining sleep or non-restorative sleep lasting for at least one month

associated to decreased functioning during the day (e. g. lack of concentration,

tiredness, decreased productivity) (American Psychiatric Association, 2000). Table

2, here below, provides the DSM-IV diagnostic criteria for primary insomnia.

Table 2- DSM-IV - Diagnostic Criteria for 307.42 Primary Insomnia

A. The predominant complaint is difficulty initiating or maintaining sleep or non-restorative sleep, for at least one month.

B. The sleep disturbance (or associated daytime fatigue) causes

clinically significant distress or impairment in social, occupational, or other important areas of functioning.

C. The sleep disturbance does not occur exclusively during the course of narcolepsy, breathing related sleep disorder, circadian rhythm sleep disorder, or a parasomnia.

D. The disturbance does not occur exclusively during the course of another mental disorder (e. g. major depressive disorder,

generalised anxiety disorder, a delirium).

E. The disturbance is not due to the direct physiological effects of

a substance (e. g. a drug of abuse, a medication) or a general medical condition.

Primary insomnia as defined by the DSM-IV subsumes to a number of insomnia categories in the ICSD. The category most closely resembling the DSM-IV definition

of primary insomnia is that of psychophysiologic insomnia. According to the ICSD-

R (American Sleep Disorder Association, 1997) psychophysiologic insomnia is a disorder of somatized tension and learned sleep-preventing associations resulting in a complaint of insomnia and associated decreased functioning during wakefulness

(table 3 reports diagnostic criteria).

It should be noted that this definition, as compared to that of primary insomnia, is more tied up with the underpinning of the disorder (i. e. how insomnia started and it is

maintained). Somatized tension refers to somatic hyperarousal, either perceived or

objectively measured. Learned sleep-preventing associations refer to either internal

cognitions (mainly worry about the inability to sleep) or external stimuli (conditioned

association of sleeplessness _{with situations and behaviours related to sleep such as}

lying in bed) both provoking pre-sleep arousal. The patient's focused attention on sleep is of paramount importance in psychophysiologic insomnia.

Table 3- ICSD-R - Diagnostic Criteria: Psychophysiologic Insomnia 307.42-0

a. A complaint of insomnia is present and is combined with a complaint of decreased functioning during wakefulness.

b. Indication of learned sleep-preventing associations are found and include the following:

1. Trying too hard to sleep, suggested by an inability to fall asleep when desired, but ease to fall asleep during other relatively monotonous pursuits, such as watching television or reading.

2. Conditioned arousal to bedroom or sleep-related activities, indicated by sleeping poorly at home but sleeping better away from the home or when not carrying out bedtime routines.

c. There is evidence that the patient has increased somatized tension (e. g. agitation, muscle tension, or increased vasoconstriction)

d. Polysomnographic monitoring demonstrates all of the following: 1. An increased sleep latency

2. Reduced sleep efficiency

3. An increased number and duration of awakenings

e. No other medical or mental disorders accounts for the sleep disturbance.

f. _{Other sleep disorders can coexist with the insomnia, e. g. inadequate sleep} hygiene, obstructive sleep apnoea syndrome, etc.

Minimal criteria: A+B

The ICSD-R has recently been revised and a new version has been published (ICSD-2, American Academy of Sleep Medicine, 2005). This new version, informed by current research and guidelines, incorporates changes with regard to sleep disorders. In particular psychophysiologic insomnia, is renamed psychophysiological

insomnia in the ICSD-2 and described as a complaint of primary insomnia with the essential features of heightened arousal and sleep preventing associations (appendix

1 presents the revised criteria). It is noteworthy that although the characteristics pertaining to psychophysiological insomnia have hardly changed from one version to

the next, in the latest version cognitive activity (in the form of intrusive, ruminative

presleep cognitions, often related to the inability to sleep and its consequences) are

explicitly mentioned.

Having _clearly defined _primary insomnia (as per DSM-IV) _and psychophysiological insomnia (as per ICSD-R, ICSD-2), attention can be turned to

the epidemiology of insomnia worldwide.

1.2.3 Epidemiology of Insomnia

The prevalence of primary insomnia in the adult population, as evidenced by epidemiologic surveys, is around 10-15 % increasing to around 20 % in older adults

with a 'women to men' ratio of roughly 3: 2 (Ancoli-Israel and Roth, 1999; Foley,

Monjan, Brown, Simonsick, Wallace et al, 1995; Ford and Kamerow, 1989; Gallup Organisation, 1995; Morgan, 2003; Ohayon, 2002). Although there are variations in percentage of insomnia incidence, it is a worldwide complaint (Cirignotta, Mondini,

Zucconi, Lenzi and Lugaresi, 1985; Gislason and Almqvist, 1987; Kim, Uchiyama, Okawa, Liu and Ogihara, 2000; Leger, Guilleminault, Dreyfus, Delahaye and Paillard, 2000; McGhie and Russell, 1962; Ohahion and Partinen, 2002, Ohahion and Hong, 2002). For example, an epidemiological survey of the French population found that 29% of the respondents had insomnia symptoms at least three times per week during the previous month (Leger et al, 2000). Cirignotta and colleagues

(1985) found it in around 15% of the population of San Marino. Similarly a survey of the South Korean general population found insomnia symptoms at least three times per week in 17% of those surveyed (Ohahion and Hong, 2002) and the overall prevalence of insomnia in the Japanese general population has been reported as 21 % (Kim, Uchiyama, Okawa, Lin and Oshihara, 2000). A survey carried out in the UK evidenced that sleep problems are by far the most'common mental health complaint

(Singleton, Bumpstead, O'Brien, Lee, Meltzer, 2001) and primary insomnia is the most common of all sleep complaints (Bixler, Kales, Soldatos, Kales, and Healy,

1979). It is noteworthy that around 2% of the general population meet criteria for psychophysiological insomnia (American Academy of Sleep Medicine, 2005).

Insomnia is often dismissed as less of a serious problem (Simon and vonKorff, 1997) and often ignored at the clinical level despite being a considerable public health problem which is reported by around 20% of patients seen in general practice (Shocat, Umphress, Israel and Ancoli-Israel, 1999) and being associated with increased economic costs both in health care and industry (Simon et al, 1997). Indeed, insomnia is a contributing factor to reduced productivity, absenteeism, and accidents and, once established, it is difficult to extinguish (Brassington, King and Bliwise, 2000; Katz and McHorney, 1998; Mendelson, 1995; Roth and Ancoli-Israel,

1999).

In sum, primary insomnia is a persistent, common and difficult to treat condition associated with reduced daytime functioning, distress and sizeable economic costs.

In addition insomnia has been found in two meta-analyses to be a key risk factor in the first episode and recurrence of depression (Cole and Dendukuri, 2003; Riemann and Voderholzer, 2003). It seems, therefore, important to understand what

mechanisms underlie the development and maintenance of insomnia and effective ways of treating it. First of all, however, ways of assessing insomnia and treatment outcomes need to be discussed.

1.3 Assessment of Insomnia and Monitoring Treatment Efficacy

The main methods for measuring sleep are subjective measures, in the form of retrospective and prospective self-reports and objective measures in the form of

actigraphy and polysomnography (PSG). Each of these will be briefly described and

evaluated so as to outline their strengths and weaknesses.

Typically, the sleep data collected by means of prospective self-reports and objective measures are i) sleep onset latency (SOL) that is the time taken to fall

asleep at the beginning of the night; ii) wake after sleep onset (WASO) that is the

total time spent awake during the night after sleep onset; iii) total time in bed (TIB) that is the amount of time spent in bed; iv) total sleep time (TST, TST=TIB -

[SOL+WASO]) that is the amount of time actually spent asleep while in bed, v) sleep efficiency (S. E., S. E. =[TST/TTB]*100) that is the percentage of time spent asleep while in bed, and vi) number and length of arousals (only objective data). It should

be noted that WASO can comprise all awakenings including early morning awakenings or it can be subdivided into WASO during the night and `early

wakening' that is the amount of time from the last wakening in the morning until

rising.

1.3.1 Subjective Measures

The two most commonly used self-report measures of sleep are sleep questionnaires (retrospective self reports) and sleep diaries (prospective self-reports).

1.3.1.1 Sleep Questionnaires

Sleep questionnaires have sufficient diagnostic sensitivity in discriminating between normal and poor sleepers and they provide an easy, quick and inexpensive way to measure sleep patterns. They are, therefore, particularly in use where a large

number of people are to be screened or surveyed, although they are also used as pre-

and post- treatment measures of the individual's perception of their sleep quality and

quantity. For example The Pittsburgh Sleep Quality Index (PSQI: Buysse, Reynolds,

Monk, Berman, and Kupfer; 1989) was expressly designed for this latter purpose (Smith, Nowakowski, et al, 2003).

One important limitation of sleep questionnaires is their validity: they usually require the individual to provide estimates in terms of average which are most likely

influenced by the particular mood of the individual at the time of completion and the most salient or recent experience of sleep (e. g. the particularly bad night). In

addition, an average figure raises validity issues given that in insomnia night-to-night

variability is high and, importantly, a feature of insomnia: most researchers and clinicians will have come across questionnaires with replies such as "it varies" or "5

minutes to 4 hours", to the questionnaire item: `average time to fall asleep in the last

month'.

In sum, sleep questionnaires provide a quick, convenient and inexpensive tool to measure sleep but have the disadvantage of being based on estimates drawn from

memory of a number of past nights and of giving one global figure for each sleep

variable under scrutiny. Sleep diaries, critically outlined below, partly avoid these

pitfalls.

1.3.1.2 Sleep Diary

This measurement is the standard of practice for behavioural sleep medicine. Unlike sleep questionnaires, the sleep diary is completed each morning upon awakening for a given number of days so that sleep can be sampled across time.

Therefore, although sleep measures are taken retrospectively (i. e. the previous night), they are applied prospectively in that they are gathered for a number of (future)

consecutive nights. Similarly to retrospective sleep measures the individual is

required to estimate variables such as `time to fall a sleep' and `duration of

wakenings during the night'. However, sleep diaries require such estimates to only

one time point (the previous night) hence the response biases (e. g. using averages)

discussed above are avoided. Of course sleep diaries collect subjective judgements of sleep and, therefore, are not as precise as objective measures (section 1.3.2 here

below). Nonetheless a very important feature of subjective measures is that, unlike PSG or actigraphy, they enable the gathering of information about the perception both of sleep disturbance (i. e. the complaint) and of recovery following treatment.

An issue with the diary is adherence to the instruction `fill it in every morning': people might fill the diary in later on during the day or even worse a few days later.

Another issue with sleep diaries, worth addressing especially for research purposes,

relates to standardisation: although there is a general agreement across sleep laboratories regarding the sleep parameters to be gathered, a standardised sleep diary is not in use and this can result in variations in the characterisation of sleep continuity across studies. These problems could be resolved by employing simple ways to store the daily sleep diary entries (e. g. I-pod type) and by agreeing on a

minimal standard form, both in terms of sleep parameters to be collected and

wording of questions, to be used across all sleep laboratories.

As it has been shown in the preceding discussion the sleep diary is an inexpensive tool for measuring sleep continuity and sleep quality variables night after night. The fact that it relies on the individual's report of the previous night is a weakness but also a major strength of sleep diaries. On one hand the reported values are not free of

the memory biases typical of retrospective measures but on the other hand sleep diaries capture the subjective perception of the sleep difficulties. Given that the essential feature of insomnia is the individual's complaint of poor sleep (minimum

three days per week) and its consequences during the day, it can be argued that the sleep diary is the most valid tool to measure this feature of insomnia and it is, therefore, not surprising that it is standard practice in behavioural sleep medicine.

Sleep diaries measure sleep continuity and collect the individual's perception of being either asleep or awake. As already mentioned, the subjective report of sleep argues against a unitary sleep state and the different `layers' of sleep can be

monitored via PSG, an objective measurement of sleep and wakefulness.

1.3.2 Objective Measures of Sleep

In this section polysomnography and actigraphy will be described and critically evaluated.

1.3.2.1 Polysomnography

Polysomnography _{(PSG) permits researchers and clinicians to measure sleep} objectively, to obtain an exact temporal (second-by-second) resolution of sleep, to

have direct and quantitative measures of brain and muscle activity and to differentiate sleep into substates (stages 1,2,3,4 and REM) which are not detectable with other sleep measurements (Smith, Nowakowski et al, 2003).

Three _{measures are sufficient} for _{monitoring and scoring sleep:}

electroencephalography (EEG) in particular the recording of brain wave activity

from central and occipital areas of the scalp, electrooculography (EOG) which measures a differential of electrical potential, generated by each eye movement,

between the cornea and the retina; and electromyography (I3MG), which monitors muscle tone and is generally recorded at the chin site. These electrical currents can be represented continuously and plotted over time and their combination permits the detection and categorization of sleep stages and wakefulness. In order to obtain reliable PSG data the electrodes need to be placed accurately: the 10-20 International

System (Jasper, 1958) is the most commonly used system for electrode attachment.

The PSG data are divided into sleep stages employing criteria for staging normal human sleep provided in the Rechtschaffen and Kales' (1968) manual (section 1.2.1

and figure 1 give explanation and example of sleep stages, and figure 15a shows a

PSG recording device).

An important issue is that PSG recording usually takes place in sleep laboratories and several studies have noted that PSG sleep differences between poor and normal

sleepers are at best modest (Means, Edinger, Glenn and Fins, 2003; Sugarman, Stem

and Walsh, 1985). It could be argued that in laboratory settings the conditioned cues

for wakefulness typical of the poor sleeper's bedroom are absent. Additionally the laboratory setting may provide a pre-bed routine that obviates to sleep-disruptive habits and the novelty might distract the individual from the usual pattern of worrying in bed. The fact that poor sleep is expected can paradoxically promote a

good night's sleep by removing sleep anxiety performance. The opposite is true for

the normal sleeper: habitual cues for sleep are absent, the usual pre-bed routine is

disrupted and they might question their ability to sleep in a novel environment.

In other words, it can be argued that a short series of sleep recordings conducted in laboratory settings might not be representative of individuals' typical sleep. For example, Edinger and colleagues conducted a series of PSG studies where PSG

recordings were conducted both at participants' homes and in laboratory settings.

One study showed that individuals, regardless of their sleep status, acclimatise more readily to PSG recording when in their home as compared to the laboratory setting

and that PSG recording at home enabled capture of the relatively high night-to-night

variability typical of people with insomnia (Edinger, Fins, Sullivan, Marsh, Dailey,

Hope, Young, Shaw, Carlson and Vasilas, 1997). In a subsequent study Edinger and colleagues found that people with insomnia who had a bed partner showed sleep

differences with regard to normal sleepers only when PSGs were recorded at home,

once again suggesting that home PSG is a suitable option if capturing `typical' sleep

patterns is important (Edinger, Glenn, Bastian, Marsh, Daile, Hope, Young, Shaw

and Meeks, 2001).

It seems, therefore, worthwhile to measure PSG sleep at home as this increases ecological validity in comparison with laboratory based trials. Simple to use, non-

intrusive, relatively cheap, portable EEG systems have been recently devised which make it feasible to conduct home PSG so as to monitor objective sleep changes

following insomnia interventions whilst the individuals sleep in the `reality' of their own usual sleep environment.

It is noteworthy that subjective measures of sleep and PSG sleep are very often mismatched in both young adults and the elderly (Carskadon, Dement, Mitler,

Guilleminault, _{Zarcone and Spiegel, 1976; Mercer, Bootzin and Lacks, 2002;} Schramm, Hohagen, Backhaus, Lis and Berger, 1995; Webb and Schneider-Helmert,

1988). In order to explain such discrepancies, Borkovec (1979) suggested that when an overly active reticular system (which _{governs arousal) is present or a}

neurotransmitter-dependant sleep system fails to switch on then PSG confirms

insomnia. In contrast, when reported insomnia is not confirmed by PSG defined sleep the individual has an overly active cognitive processing of anxiety-laden thoughts. It could be argued that, in the latter case, sleep staging does not capture

features of insomnia such as alertness. Alertness in the form of EEG fast activity can be detected by performing power spectral analysis to the PSG recorded night (Perlis, Smith, Orff, Andrews and Gill, 2001).

Interestingly, several studies highlighted that despite positive treatment effects being evidenced by sleep diary data, there were only very small treatment effects or

none at all if PSG data were considered (e. g. ' Morin, Kowatch, Barry and Walton,

1993; Engle-Friedman, Bootzin, Hazlewood and Tsao, 1992; Morin, Stone, Trinkle, Mercer and Remsberg, 1993; Schramm et al, 1995).

Another important point, related to the difference between subjective and objective measures of sleep, could be made by taking into account Espie's (1991)

observation that the subjective complaint normally incorporates an element of

dissatisfaction with sleep quality variables such as `not feeling refreshed in the morning', or `sleep was of very poor quality'. Such complaints are not invalidated by

an absence of EEG evidence of sleep pattern disruption: indeed not being sensitive to

sleep quality variables weakens the role of PSG as a measurement criterion.

1.3.2.2 Actigraphy

Actigraphy measures movement and its absence i. e. this measurement, rather than measuring sleep directly, infers sleep by the absence of movement (Sadeh and

Acebo, 2002). The movement detector, a small piezoelectric accelerometer that generates voltages when there is movement, is housed in a wristwatch-like device.

The detected movement is translated into digital counts accumulated across epoch intervals, whose length can range from 2 seconds to 15 minutes, and stored in the internal memory of the watch. The epoch data collected are transferred to a computer that has the set up and the reader interface and each epoch is evaluated for whether it

represents sleep or wakefulness.

Given that actigraphy is a proxy measure of sleep its validity as a measurement of sleep variables is inferior when compared to that of sleep diary and PSG. Indeed a

problem with actigraphy is validity: good agreement (r=0.94) between actigraphy

and PSG for epoch by epoch sleep has been found in normal sleepers (Jean-Louis,

Kripke, Cole, Assmus and Langer, 2001), but epoch by epoch wake have lower validity most probably because people, especially poor sleepers, tend to lie still in an

attempt to fall asleep. In addition, the discrepancy is much higher when the sleep of

people with insomnia is assessed (Chambers, 1994; Hauri and Wisbey, 1992).

Furthermore, it has been shown that there are gender and age differences in actigraph estimates of sleep (for a review see Sadeh and Acebo, 2002). It could be argued that

such variability decreases the usefulness of actigraphy when the focus of research is

sleep. Kushida, Chang, Gadkary, Guilleminault, Carrillo, _{and Dement, (2001)}

compared PSG, actigraphy and sleep diary measures for one night in one hundred

people with sleep difficulties. They found that the accuracy of actigraphy declines as

sleep diminishes: actigraphy was excellent in detecting sleep (sensitivity = 0.92) but

its ability to detect wakefulness (specificity= 0.48) was rather poor and the ability to detect both sleep and wakefulness (accuracy = 0.78) was low. Pollack, Tryon, Nagaraja and Dzwonczyk (2001) investigated the ability of actigraphy to predict PSG defined sleep and wakefulness and argued that actigraphy is not an accurate

sleep-wake detector, although it may be useful as a measure of rest/activity. In

particular, Sadeh and Acebo (2002, p120) cautioned that "actigraphy is not the best

method if interest is in the precise duration of sleep onset" because the largest

discrepancies between actigraphy and PSG measures are typically around sleep-wake and wake-sleep transitions.

It is important to note that actigraphs can collect other measures in addition to movement: this can be very useful if actigraphy is employed for collecting data other

than sleep. For example one model (Actiwatch-L by Cambridge Neurotechnology,

figure 22 in section 6.3.3.5a for an example of actiwatch) incorporates an integral light sensor that monitors light intensity range measured from 0 to 50,000 lux. In this way, both movement and light are detected and stored. A secondary aim of the

present research is the exploration of the impact of adherence on insomnia treatment

outcomes: this extra feature of actigraphy can be exploited to objectively monitor

adherence to a behavioural instruction asking participants to read (study 3, ch. 5).

In sum, there is not an absolute `gold standard measure' of sleep. While sleep questionnaires are an important tool to quickly gather information regarding the

perception of sleep problems, sleep diaries provide nightly information regarding the

perception of sleep continuity and, very importantly, the subjective complaint. PSG

measures objective sleep continuity and sleep architecture and enables power

spectral analysis. However, given that PSG recordings are costly and time consuming

they are not really feasible as a measurement of many consecutive nights. For this reason actigraphy is employed when the measurement of sleep over time is to be

collected objectively; even though it presents the important problem of validity

discussed above.

It can be concluded that subjective and objective measures of sleep might be complementary and, most importantly, the questions one is trying to answer and the

issues and hypotheses under investigation (e. g. endorsed beliefs about sleep, EEG activity during sleep) should inform which are the best to use.

The next section introduces insomnia aetiological systems and related research.

1.4 Theoretical Perspectives on Insomnia

The aim of this section is to critically outline the major perspectives on insomnia (i. e. physiological, behavioural and cognitive) and related research. In addition

'purist' models within a given perspective and models integrating different perspectives are discussed.

Whilst _{there is limited information regarding mechanisms underlying}

pathophysiology and genetic or heritability factors in insomnia (Richardson and

Roth, 2001; Yves, Morin, Cervena, Carlander, Beset and Billiard, 2003), most insomnia models agree on the hypothesis that insomnia is a disorder of arousal: the person complaining of insomnia has a level of arousal inhibiting sleep onset at

bedtime and/or during the night. The debate regarding the origin (physiological, psychological and/or environmental) of such hyperarousal is, however, ongoing.

1.4.1 Physiological Perspective of Insomnia

This perspective proposes that physiological arousal and sleep are incompatible and that individuals with chronic insomnia experience sleep difficulties because of a

high level of physiological arousal either in general or around bedtime and during the night.

In the nervous system arousal can be maintained by a failure of central nervous system (CNS) processes to promote sleep (homeostatic sleep drive) and/or to inhibit

wakefulness (biological clock) (Daan et al, 1984). There is evidence of heightened

cortical arousal during the sleep of people with insomnia. For example, Merica, Blois

and Gaillard's (1998) PSG study found increased beta power (beta waves are

characteristic of alertness) in their insomnia group as compared to the control group.

Similar results were obtained by Perlis, Kehr, Smith, Andrew, Orff and Giles (2001): high frequency EEG activity during the light sleep stages (stages 1 and 2) was increased in individuals with primary insomnia as compared both to controls and individuals with insomnia secondary to major depression.

High levels of autonomic activity associated with tension and problems in physiological relaxation (e. g. muscle tension, respiration rate) can be another form of

arousal that, less directly, inhibits sleep. For example Monroe (1967) conducted an

important study with 16 normal and 16 poor sleepers and found that poor sleepers exhibited heightened autonomic arousal both during and prior to sleep (evidenced by

higher body temperature, body movements per hour, vasoconstriction per minute, perspiration rate and skin conductance). Other studies, however, did not find

differences between normal and poor sleepers in autonomic variables (for a review see Bootzin and Nicassio, 1978; Borkovec, 1982). More recently, Bonnet and

colleagues assessed physiological arousal employing a measure of oxygen

consumption (V02) and found that the insomnia group, as compared to controls,

displayed higher metabolic rate (autonomic arousal) both during the day and at night (Bonnet and Arand, 1995,1997; Bonnet, McNulty and Arand, 1993). A limitation of these studies is that physical fitness and caloric intake greatly influence body oxygen

consumption, hence the observed differences could have been due, for example, to

the insomnia group being less physically fit than the control group.

In sum, although there are findings that physiological hyperarousal (autonomic and/or cortical) correlates with insomnia, research has not unequivocally supported

the physiological perspective. It is noteworthy that this perspective informs on the

interaction between insomnia and arousal but lacks explanatory power regarding the development and exacerbation of insomnia. Although research evidence of physiological arousal is often used to provide support for aspects of some theoretical

models of insomnia, a formal model of insomnia within the physiological perspective

has not been proposed.

The behavioural perspective, discussed in the next section, attempts to provide an analysis of the maintenance and exacerbation of insomnia.

1.4.2 Environmental/Behavioural Perspective of Insomnia

According to the behavioural perspective, acute insomnia might be precipitated by a number of psychosocial events but chronic insomnia has its source in the

environment. The source of the individual's recurrent difficulties with sleep is not

considered to be physiological or psychological in origin, rather the individual's

arousal is caused by situation and temporal factors or the interaction between the

individual and their environment.

This perspective posits that the environment may elicit arousal responses by either failing to predict sleep at the right time for sleeping or predicting wakefulness at the wrong times. It is suggested that sleep disruptive habits (e. g. extending the time spent

in bed, napping) and conditioned responses (e. g. lying in bed elicits wakefulness rather than sleep) interfere with sleep drive and serve as environmental inhibitors of

sleep (Bootzin and Epstein, 2000; Edinger and Wohlgemuth, 1999; Morin, Savard

and Blais, 2000; Spielman, Saskin and Thorpy, 1987).

However, results from the couple of studies investigating, directly or indirectly, conditioning in insomnia call into question the conceptual basis of the behavioural

perspective of insomnia. Kazarian, Howe and Csapo (1979) found support for the

hypothesis that poor sleepers engage in sleep incompatible behaviour (both in the form of overt behaviour such as reading and covert such as thinking). However, two other studies found that normal sleepers, similarly to poor sleepers, engaged in sleep

incompatible activity in the bedroom and that the two groups did not differ in pre- sleep activities (Harvey, 2000a; Haynes, Adams, West, Kamens and Safranek, 1982).

The findings that poor and normal sleepers endorse similar behaviour at bedtime do not automatically rule out the role of conditioning in insomnia: it could be, for

example, that the concept of `preparedness' applies to poor sleepers but not to

normal sleepers with regard to sleep incompatible behaviours and sleep. In learning

theory, the concept of `preparedness' refers to the observation that depending on the

organism's make-up and social modelling, the organism is prepared (predisposed) to

learn some associations more readily than others (Seligman, 1971; Hygge and Ohman, 1978). Learned taste aversion studies, for example, suggest that there is a biological preparedness to learn certain relations more readily than others and similar effects occur in operant learning with some responses more readily strengthened by

some reinforcers than others (Bolles, 1970; Shettleworth, 1972; Bernstein, 1978). As

discussed in section 1.4.2.2, predisposition is one of the three main factors theorised to play a major role in insomnia (Spielman et al, 1987). Additional research, perhaps

employing prospective rather than retrospective measures, is much warranted to

validate this hypothesis.

Two behavioural models of insomnia (stimulus control and the 3 factor model of insomnia) have gained extreme popularity. Perhaps, especially with regard to

stimulus control, this is mainly due to the positive outcomes repeatedly obtained with

stimulus control therapy for insomnia and sleep restriction.

1.4.2.1 Stimulus Control (Bootzin. 1972. Bootzin and Nicassio. 1978)

Bootzin (1972) conceptualised sleep and insomnia in terms of learning theory and developed stimulus control therapy for insomnia. Stimulus control is based on operant conditioning concepts: stimuli can elicit a number of responses depending on

the conditioning history. Stimulus control can be defined as the degree to which an

antecedent stimulus (or class of stimuli) determines the probability that the organism

gives the conditioned response i. e. the antecedent stimulus has some control over that

particular response. Briefly, applying stimulus control concepts to sleep, falling

asleep (operant) is emitted so as to produce sleep (reinforcement) and stimuli

associated with sleep become discriminative stimuli for the occurrence of sleep

(Bootzin and Nicassio, 1978). Stimuli associated with sleep can be temporal (i. e. bedtime) and/ or environmental (i. e. bedroom).

Fig 2 -Operant Analysis _{of Sleep}

Stimuli _{signalling reinforcement}

, Operant Reinforcer Bed

Pillow

Brushing _{teeth Falling}

asleep

p, Sleep

In normal sleet) stimuli such as one's bed are associated to sleep

As shown in figure 2, this model proposes that in normal sleep the bedroom environment provides stimuli signalling sleep and such stimuli elicit falling asleep.

Bootzin and colleagues proposed that in the case of insomnia the discriminative stimuli for sleep are not established (e. g. lack of association sleep-bedroom) or

stimuli for activities that are not compatible with sleep (e. g. television) are present.

Situational triggers (sleep-preventing activities such as watching TV, reading, problem solving and bed aversion) may elicit arousal responses by becoming a cue

for wakefulness or by failing to predict sleep. Espie (2002) suggested that poor stimulus control can inhibit bedtime de-arousal responses that in turn disrupt both the

homeostatic sleep drive and the wakefulness-sleep circadian cycle.

Evidence for this model is provided by the success of the therapy intervention

(section 1.7.2) derived by the stimulus control analysis of insomnia rather than by testing directly the conditioning aspects theorised to be responsible for insomnia. As critically presented in sections 1.6.2 and 1.7.2, many studies found that stimulus

control therapy for insomnia markedly decreased SOL and WASO by 50 to 80% as

compared to pre-treatment values and this findings have been taken as evidence for

the plausibility of this model. Notwithstanding its efficacy, the mechanisms by which stimulus control achieves its effects are unclear and it cannot be ruled out that the

therapy works for reasons other than re-establishing stimulus control. This point has been emphasised by Salkovskis (2002) who argued that treatment evidence does not shed light on the underlying mechanisms of effect. Indeed, as already discussed in

the previous section, only a few studies have investigated pre-bed activities in normal and poor sleepers and no differences have been observed. Contrary to conventional

wisdom (ICSD-R), it has, recently, been reported that people with insomnia sleep

worse when they are away from home and it has been argued that this finding runs

contrary to the hypothesis of conditioning to the bedroom environment (Vallieres,