The clinical effectiveness of cognitive behaviour therapy for chronic insomnia: implementation and evaluation of a sleep clinic in general medical practice

https://doi.org/10.1016/S0005-7967(99)00157-6Get rights and content

Abstract

Chronic insomnia is a very common clinical condition which may respond well to non-pharmacological treatment. Indeed, the literature supports the efficacy of cognitive behaviour therapy (CBT). However, there has been no substantial study of clinical effectiveness. Since insomniacs typically present in general medical practice this is a crucial gap in the outcome research. This study, therefore, specifically investigated the clinical effectiveness of CBT delivered by Health Visitors (primary care nurses) trained as therapists. One hundred and thirty-nine insomniacs (mean age 51 yr) were randomised to CBT or Self-Monitoring Control (SMC) in a controlled trial. CBT comprised six group sessions (n=4 to 6 patients). After the controlled phase, SMC patients entered deferred treatment (CBT-DEF), allowing both treatment replication and long-term outcome to be investigated for a sizeable, treated sample. Repeated measures ANOVAs demonstrated superiority of CBT over SMC in substantially reducing sleep latency and wakefulness during the night. CBT-DEF replicated similar effects and maintained improvement was observed in both groups one year later. Furthermore, total sleep increased significantly during follow-up and 84% of patients initially using hypnotics remained drug-free. Results suggest that CBT administered by Health Visitors offers a clinically effective treatment for insomnia.

Introduction

Conservative estimates of the prevalence of chronic insomnia, defined as persistent difficulty in initiating or maintaining sleep (American Sleep Disorders Association, 1997), range from 9 to 12% in adulthood and up to 20% in later life, making sleep disturbance one of the most common complaints in general medical practice (Bixler et al., 1979, Foley et al., 1995, Ford and Kamerow, 1989, Gallup Organisation, 1991, Hoch et al., 1997, Mellinger et al., 1985). One reason for this high prevalence is that sleep is very sensitive to physical or psychological disorder. Thus, “secondary insomnia” presents in a wide range of conditions (e.g. pain, respiratory disorder, depression), often enduring once a disordered pattern has been established, even after the medical or psychiatric problem has resolved. “Primary insomnia”, of course, refers to a sleep disorder constituting the major difficulty at the time of presentation. Age-related developmental change contributes to the increased presentation rates from the middle years onwards (see Espie, 1991, Morin, 1993 for review). Statistics on the prescription of hypnotic medications (particularly benzodiazapines) also demonstrate the scale of sleep complaint within the general population (Gallup Organisation, 1991, Mellinger et al., 1985) and insomnia is now recognised to be associated with significant health-economic cost (Chilcott and Shapiro, 1996, Walsh et al., 1995).

Although benzodiazapines used for short periods of time, or intermittently, can maintain their effectiveness (Kales & Kales, 1984), pharmacological treatment is not recommended for primary or chronic insomnia, or in the elderly. Indeed, the evidence against the prolonged use of such drugs is overwhelming (National Institute of Mental Health, 1984, Russell and Lader, 1992). Over the past 20 yr considerable progress has been made in the psychological assessment and treatment of sleep problems (Bootzin et al., 1978, Espie, 1991, Espie, 1993, Lacks, 1987, Morin, 1993).

There is a sizeable literature on the effectiveness of cognitive behavioural treatment (CBT) for insomnia. Accurate sleep information, the practice of sleep hygiene and the use of specific behavioural and cognitive techniques such as relaxation therapy (see Bootzin & Nicassio, 1978), stimulus control (Bootzin, 1972; Bootzin & Rider, 1997), sleep restriction (Spielman, Saskin & Thorpy, 1983) and cognitive techniques (Espie and Lindsay, 1985, Morin, 1993) has been widely endorsed. Over 50 controlled studies support the clinical efficacy of CBT, with meta-analyses reporting significant effect sizes for improvements in sleep latency (0.87), duration of wakenings (0.65) and sleep quality (0.94) (Morin et al., 1994, Murtagh and Greenwood, 1995). The majority of studies has been North American, although European trials have reported similar results (Espie et al., 1989, Sanavio et al., 1990). Around 70–80% of patients seem to benefit and CBT appears equally efficacious in older adults (Morin et al., 1993, Morin et al., 1999, Morin et al., in press); important because of clear contraindications for hypnotic use in this agegroup (Lader, 1992).

However, despite this evidence and the fact that CBT could be regarded as the treatment of choice for chronic insomnia (Espie, 1999, Morin et al., 1999, Morin et al., in press), impact upon general health care practice so far has been limited. Of course, the scale of the implementation task is such that any realistic alternative to sleeping pills must be capable of being implemented not only effectively but simply, efficiently and locally. The task is also compounded by the likelihood that few general medical practitioners will be aware of the psychological literature; the fact that few psychologists have a special interest in sleep and in any case, may be in short supply; and the continuing tendency for the doctor-patient relationship to raise the patient's expectations of drugs. All of these factors are essentially practical and might be alleviated by greater dissemination of knowledge, educational opportunity for medical students and doctors, increased availability of psychological services and changing the culture surrounding ‘medical’ intervention.

A more fundamental factor, however, is that CBT may be time-consuming, expensive and ultimately less therapeutic to implement in general practice. There is, therefore, also an important empirical question around the feasibility and effectiveness of delivering CBT which has not been answered in efficacy trials, these having generally been carried out at specialised centres rather than in routine practice. Morin et al. (1994) and recently Edinger and Wohlgemuth (1999), have indicated that a major scientific limitation of the available evidence is the lack of any large scale appraisal of clinical effectiveness in ordinary clinical settings. This study, therefore, was designed specifically to investigate this research priority.

In the UK, most General Practitioners (GPs) work in group practices, often based in “health centres”. The primary care teams have GPs, district nurses and health visitors as the core professionals providing community health services, with paramedical staff offering sessional input. The Health Visitor is a primary care nurse with community nursing qualifications who works across the life span, e.g. with infants during early development, with people rehabilitating from illness and with elderly people at home. In planning this study, therefore, we felt that training Health Visitors in CBT for insomnia would provide the most accessible and valid approach to managing insomnia in primary care settings, supported by a psychologist available on a consultancy basis for complex cases e.g. significant co-morbidity, or differential diagnosis. Where required, arrangements were also made for referral to a sleep laboratory. Although polysomnography (PSG) is not usually required for the diagnosis of insomnia (Reite, Buysse, Reynolds & Mendelson, 1995) it is invaluable in differential diagnosis of other suspected sleep problems.

Section snippets

Aims

The aims of the study, therefore, were to evaluate the medium to long-term (12 months post-treatment) clinical effectiveness of a primary care-based insomnia service (the Sleep Clinic), using Health Visitors as trained CBT therapists. In order to ensure validity and generalisability to typical clinical populations, the study required a large number of consecutively referred insomniacs whose sleep problems were both persistent and severe. The intervention was devised to test the feasibility of a

Subjects

Consecutive referrals of adults presenting with chronic insomnia during a 24 month period were received from GPs in Ayrshire. This part of west, central Scotland has both urban and rural communities and comprises a broad socioeconomic span. Selection, based upon International Classification of Sleep Disorders criteria, comprised persistent difficulty initiating or maintaining sleep, occurring 4 or more nights per week, for at least 3 months (American Sleep Disorders Association, 1990). In

Sleep pattern data

Data were analysed by means of SPSS repeated measures ANOVAs, appropriate for the mixed between-within subjects design of this study (Tabachnick & Fidell, 1996). Dependent variables were weekly mean scores for the three principal outcome measures from the Sleep Diary i.e. sleep-onset latency (SOL; min), wake time after sleep-onset (WASO; min) i.e. total time spent awake during night-time wakenings and total sleep time (TST; h). Each variable was considered in a separate ANOVA, with group as the

Discussion

This study set out to evaluate the clinical effectiveness of CBT for chronic insomnia using trained Health Visitors as therapists. With 139 subjects completing a controlled trial, comparing CBT with a self-monitoring control procedure (SMC), this represents the largest published outcome study on the non-pharmacological treatment of insomnia conducted in general practice. Furthermore, since the SMC group was subsequently treated with CBT, as a deferred entry to treatment procedure (CBT-DEF), a

Acknowledgements

This research was funded by a grant from the Chief Scientist Office, Scottish Office Department of Health under the Novel Health Services Research Initiative and by a grant from Ayrshire and Arran Health Board. We particularly thank the Health Visitors Jean McMillan, Gillian Dalziel, Evelyn Berry, Christine McArthur, Maureen McVey and Linda MacDonald and Barbara Gilbert (Secretary) for all their hard work. We are also grateful to Sandra Steele, Eilidh Renwick and the many other GPs who referred

References (45)

  • International classification of sleep disorders: diagnostic and coding manual

    (1990)
  • Classification of sleep disorders: diagnostic and coding manual (revised)

    (1997)
  • A.T Beck et al.

    An inventory for measuring depression

    Archives of General Psychiatry

    (1961)
  • E.O Bixler et al.

    Prevalence of sleep disorders in the Los Angeles metropolitan area

    American Journal of Psychiatry

    (1979)
  • M.L Blood et al.

    A comparison of sleep detection by wrist actigraphy, behavioural response and polysomnography

    Sleep

    (1997)
  • R.R Bootzin et al.

    Behavioral techniques and biofeedback for insomnia

  • L.A Chilcott et al.

    The socioeconomic impact of insomnia: an overview

    Pharmacoeconomics

    (1996)
  • R Downey et al.

    Training insomniacs to accurately perceive sleep onset

    Sleep

    (1992)
  • C.A Espie

    The psychological treatment of insomnia

    (1991)
  • C.A Espie

    The practical management of insomnia: behavioural and cognitive techniques

    British Medical Journal

    (1993)
  • D.J Foley et al.

    Sleep complaints among elderly persons: an epidemiologic study of three communities

    Sleep

    (1995)
  • D.E Ford et al.

    Epidemiologic study of sleep disturbances and psychiatric disorders

    Journal of the American Medical Association

    (1989)
  • Cited by (265)

    • Cognitive therapy for insomnia

      2023, Encyclopedia of Sleep and Circadian Rhythms: Volume 1-6, Second Edition
    • The impact of exercise, sleep, and diet on neurocognitive recovery from mild traumatic brain injury in older adults: A narrative review

      2021, Ageing Research Reviews
      Citation Excerpt :

      Consequently, therapies designed to improve sleep may have downstream effects in protecting cognition and restoring brain health following mTBI. Cognitive behavioural therapy (CBT) focused on providing sleep education, improving sleep hygiene, promoting relaxation and cognitive restructuring to manage intrusive thoughts is the first-line of treatment for insomnia (Ancoli-Israel and Ayalon, 2006; Espie et al., 2001). Sleep-specific CBT has also been applied to other clinical populations where secondary insomnia is common, including TBI patients (Nguyen et al., 2017; Ouellet and Morin, 2007).

    • The effect of cognitive behavioural therapy for insomnia on sedative-hypnotic use: A narrative review

      2021, Sleep Medicine Reviews
      Citation Excerpt :

      A number of observational studies and RCTs published between 1993 and 1996, highlighted the potential benefit of relaxation techniques, and the first multi-component CBTi programs on reduced sedative-hypnotic use [85–89]. Between 1999 and 2003, a number of RCTs investigating relaxation therapy [90], self-administered reading/audio materials with telephone support [91], and multicomponent nurse-administered CBTi in primary care [92] included analyses of changes in sedative-hypnotic use. These early mainly observational and case–control studies which indicated a promising effect of CBTi in reducing sedative-hypnotic use were followed in 2003–2004 by the first larger RCTs investigating the effect of multicomponent CBTi interventions [18,68,69].

    View all citing articles on Scopus
    View full text