Metacognitive beliefs in primary insomnia: Developing and validating the Metacognitions Questionnaire – Insomnia (MCQ-I)

J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

Contents lists available at ScienceDirect

Journal of Behavior Therapy and Experimental Psychiatry journal homepage: www.elsevier.com/locate/jbtep

Metacognitive beliefs in primary insomnia: Developing and validating the Metacognitions Questionnaire – Insomnia (MCQ-I) Joanne Waine a, Niall M. Broomfield a, *, Stephen Banham b, Colin A. Espie a a

University of Glasgow Sleep Research Laboratory, Sackler Institute of Psychobiological Research, Southern General Hospital, Glasgow G51 4TF, Scotland, UK Department of Respiratory Medicine, Glasgow Royal Infirmary, 16 Alexandra Parade, Glasgow G31 2ER, Scotland, UK

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 25 April 2007 Received in revised form 18 February 2008 Accepted 4 March 2008

Patients with Primary insomnia often experience intrusive, worrisome cognitive activity in the pre-sleep period. Metacognitive beliefs may explain this yet no valid reliable scale exists. The present study, therefore, developed the Metacognitions Questionnaire – Insomnia (MCQ-I). Following initial metacognitive insomnia profiling interviews, item refinement produced a preliminary 60-item MCQ-I. This was administered to 34 primary insomniacs and 37 normal sleepers. Psychometric data indicate primary insomniac patients score significantly higher than normal sleepers on MCQ-I. Test–retest reliability is good. Face, concurrent, construct and discriminant validity, scale sensitivity and specificity are all acceptable. Further research with larger primary insomnia and normal sleeper samples is now required. Ó 2008 Elsevier Ltd. All rights reserved.

Keywords: Metacognitive beliefs Primary insomnia Scale development Sleep

1. Introduction 1.1. Defining primary insomnia Primary insomnia disorder involves difficulty initiating and/or maintaining sleep despite adequate opportunity, in the absence of medical/psychiatric causes. Found in 1–2% of the general population, and 12–15% of sleep centre patients, primary insomnia is a known, independent risk factor for depression

* Corresponding author. Tel.: þ44 141 232 1300; fax: þ44 141 944 8867. E-mail address: niall.broomfi[email protected] (N.M. Broomfield). 0005-7916/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.jbtep.2008.03.002

16

J. Waine et al. / J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

(Riemann & Voderholzer, 2003). Impact on personal, professional and social functioning can be considerable, with fatigue, cognitive impairments and poor motivation commonly reported. Whilst there is considerable evidence for efficacy of cognitive behavioural therapy in ameliorating primary insomnia (Morin, Culbert, & Schwartz, 1994; Murtagh & Greenwood, 1995; Smith et al., 2002), there remains a lack of research pinpointing how specific cognitive processes might interact, to maintain the disorder (Broomfield, Gumley, & Espie, 2005). One developing area of interest is mechanisms that fuel intrusive and worrisome cognitive activity (Harvey, 2005a). 1.2. Pre-sleep intrusive and worrisome cognitions in primary insomnia Intrusions are spontaneously occurring, non-volitional thoughts associated with negative affect and which are difficult to control (Clark and Rhyno, 2005). Intrusions are likely to persist if appraised as threatening and receive attention focus, and at such times are experienced as worry and/or rumination (Harvey, 2005a). There is widespread acceptance that intrusive thinking at bedtime characterises primary insomnia (e.g. Borkovec, 1982). Reports of mental events disrupting sleep are common place at clinic: ‘‘my mind keeps racing’’. In research studies, this is confirmed. Primary insomnia patients describe their pre-sleep thoughts as intrusive, uncontrollable and negative (Harvey, 2000; Kuisk, Bertelson, & Walsh, 1989), and attribute sleeping difficulties to intrusions (Broman & Hetta, 1994; Espie, Brooks, & Lindsay, 1989; Lichstein & Rosenthal, 1980; Nicassio, Mendlowitz, Fussell, & Petras, 1985). Moreover, experimentally induced cognitive intrusions delay sleep onset, in good and poor sleepers (Ansfield, Wegner, & Bowser, 1996; Gross & Borkovec, 1982, Hall, Buysse, Reynolds, Kupfer, & Baum, 1996). Despite this evidence, theoretical explanations for why intrusive pre-sleep thinking characterises primary insomnia remain lacking. 1.3. The Self-Regulatory Executive Function (S-REF) model One possible explanatory model is Wells’ Self-Regulatory Executive Function (S-REF; Wells, 2000; Wells & Matthews, 1994, 1996). According to Wells, S-REF is switched on when there is a threatening discrepancy between the perceived self-state and the ideal state; for example, wakefulness when the desired goal is sleep. Critically, at the automatic processing level, physical (body state) as well as cognitive and external information is intrusive. So for the sleep-disturbed individual awake at bedtime, noise, thoughts about sleep, planning and being awake can all be intrusive (in theory) and thus lead to processing of the self as a wakeful person (cf. Wells, 2000). S-REF may, therefore, account theoretically for the occurrence of persistent intrusive thinking in primary insomnia: wakefulness being a particularly salient intrusion to primary insomnia sufferers at bedtime, given their desired goal of sleep. Two metacognitive belief types should operate within S-REF in response to such intrusions: (i) beliefs concerning the meaning of the intrusions (e.g. thinking in bed prevents me getting to sleep) and (ii) plans that guide and shape the form that cognition takes (e.g. before I fall asleep, I should try and switch off my thoughts). We have argued elsewhere these belief types and their associated action plans e.g. thought control strategies, sleep effort, attention bias/ focus should characterise insomnia (Broomfield et al., 2005). Importantly, although pre-sleep cognitive activity may not always be experienced as intrusive (e.g. Wicklow & Espie, 2000), any stimulus may do so if it clashes with a person’s metacognitive beliefs, thereby triggering S-REF processing to reach a mental state more conducive to sleep e.g. a calmer mind. 1.4. Evidence for S-REF and primary insomnia Several lines of evidence support metacognitive beliefs and associated action plans as characterising primary insomnia. First, we know primary insomnia patients engage in a range of thought control strategies at night. These include reappraisal, worry, thought suppression and punishment (Ellis & Cropley, 2002; Harvey, 2001; Ree, Harvey, Blake, Tang, & Shawe-Taylor, 2005). These strategies are likely to fuel further intrusions (Wegner, 1994) and maintain sleep disturbance. Primary insomnia patients also endorse more positive belief statements about worry (Harvey, 2003a).

J. Waine et al. / J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

17

Second, we know primary insomnia patients show elevated sleep effort, the anxious desire to control sleep onset (Broomfield & Espie, 2005), and selectively attend to salient internal and external threat cues. Both questionnaire based (Semler & Harvey, 2004a) and experimental work (e.g. Espie, Broomfield, MacMahon, & McPhee, 2006; Jones, Macphee, Broomfield, Jones, & Espie, 2005; Marcetti, Biello, Broomfield, MacMahon, & Espie, 2006) confirm the latter. Finally, thought control, sleep effort and attention bias/focus have all been linked to delayed sleep onset latency (SOL) (Ansfield et al., 1996; Harvey, 2003a; Semler & Harvey, 2004b). Such cognitive ‘action plans’ feature in the metacognitive S-REF hierarchy (Wells, 2000), and support metacognitive belief structures as guiding intrusion appraisal. 1.5. Aim of present study Metacognition has received very little attention in the insomnia literature. No scale exists which can directly assess presence of metacognitive beliefs. The aim of this exploratory study was, therefore, (i) to explore the presence of metacognitive beliefs in primary insomnia, (ii) to develop a metacognitive beliefs of primary insomnia questionnaire and, (iii) to gather initial data on psychometric properties of the scale, using primary insomnia and normal sleeping participants. The exploratory nature of the work meant we did not identify firm hypotheses. 2. Methods and results Methods and results are combined, for the purposes of clarity. 2.1. Stage one: derivation of items for the questionnaire 2.1.1. Rationale Metacognition is under-researched in primary insomnia. Therefore, initially, field interviews with individuals with primary insomnia and normal sleepers were conducted to gain direct insight into salient metacognitive beliefs held by poor sleepers, and to develop a preliminary item pool. 2.1.2. Participants Participants for field interviews were recruited via University e-mail. Individuals with ‘difficulties sleeping at night’ or ‘good sleepers’, interested in assisting sleep research, and between 16 and 65 years old, were asked to contact the investigator and given information and consent forms. Primary insomnia participants were not treatment seeking per se, but met strict research diagnostic criteria for primary insomnia disorder (Edinger et al., 2004): (i) difficulty initiating and/or maintaining sleep or non-restorative sleep (ii) with at least one associated daytime impairment (iii) for at least 1 month (iv) in the absence of a sleep disruptive medical/psychiatric condition, substance abuse (v) and/or other sleep disorder. Primary insomnia participants also had to complain of sleep difficulties at least three nights a week (Morin, 1993) and report a minimum SOL of >30 min on these nights. This follows clinical convention for diagnosing primary insomnia sleep onset type (Morin & Espie, 2003). All primary insomnia participants scored >6 on the Pittsburgh Sleep Quality Index (PSQI; Buysse, Reynolds, Monk, Berman, & Kupfer, 1989). Scores over 5 on PSQI reliably identify clinically significant sleep disturbance with 90% sensitivity and 87% specificity (Buysse et al., 1989). Additionally, participants were only included if they scored mild/normal on HADS depression and anxiety (Hospital Anxiety and Depression Scale; Zigmond & Snaith, 1983) and did not meet DSM-IV criteria for major depression. Any participants currently receiving any psychological treatments were also excluded. Normal sleeping participants met RDC criteria for ‘normal sleepers’ (Edinger et al., 2004): (i) no complaints of sleep disturbance or daytime symptoms attributable to unsatisfactory sleep (ii) evidence of a routine standard sleep/wake schedule (iii) in the absence of a sleep disruptive medical/psychiatric condition, substance abuse (v) and/or a sleep disorder. Normal sleepers scored <5 on PSQI, and viewed themselves as ‘good sleepers’, described their sleep typically as ‘good’ or ‘very good’, and reported a mean SOL of <30 min. These conservative normal sleeping criteria ensured a representative cohort. HADS and DSM-IV depression criteria were per the primary insomnia group.

18

J. Waine et al. / J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

Participant suitability was determined using telephone screening. Of 23 respondents who returned consent forms, five primary insomniac participants were excluded (one withdrew, two failed to meet inclusion criteria, one had migraines which disrupted sleep, one had sleep paralysis and delayed sleep phase syndrome [DSPS]). No normal sleepers were excluded. Nine primary insomnia sufferers (three males, six females; mean age ¼ 25.3 years; mean sleep disturbance duration ¼ 7.9 years, SD ¼ 6.5; mean PSQI ¼ 11.4, SD ¼ 3.5), and nine normal sleepers (four males, five females; mean age ¼ 28.8 years; mean PSQI score ¼ 2.6, SD ¼ 1.0) comprised the sample. No stage one participants took part later in the study. 2.1.3. Measures and procedures Telephone screening followed receipt of consent. Individuals meeting inclusion criteria attended a face-to-face meeting where remaining screening measures (HADS and PSQI) were completed, and the field interview conducted. This was tape-recorded with the participant’s permission and was based on a pre-existing metacognitive profiling interview (MPI; Wells, 2000), adapted for primary insomnia. In line with the original MPI, our adapted version examined recent periods of distress in the pre-sleep period in order to pinpoint salient cognitive processes. Thus, whilst the interview followed the structure and format of the Wells (2000) MPI, it was expanded to include aspects of the pre-sleep period considered intrusive to delineate relevant metacognitive beliefs/processes. Questions pertained to participants’ experience of pre-sleep intrusions, the meaning and appraisals attached to intrusions (negative metacognitive beliefs about intrusions) and the metacognitive beliefs governing response (positive metacognitive beliefs about cognitive strategies). Specific coping strategies and cognitive processes employed to manage intrusions were also examined. The same areas were covered for each participant, the detail of which depended on the perceived relevance to each person. Interview format information is available from the senior author. To further assist item identification, participants completed the MetaCognitions Questionnaire-30 (MCQ-30; Wells & Cartwright-Hatton, 2003) and the Utility of Pre-Sleep Worry Questionnaire (UPWQ; Harvey, 2003b). Finally, participants were debriefed and provided with a copy of the ‘‘Good Sleep Guide’’ to relay information on basic cognitive behavioural principles of good sleep (National Medical Advisory Committee, 1993). An initial item pool was then generated through consensus discussion amongst the authors who have considerable expertise, not only on the psychological basis of insomnia, but also on models of psychopathology in general, including S-REF. The interview materials were available in literal, typed transcript format, and provided the primary source of information. That is, we identified and marked direct quotes wherever possible as potential item stems. We sought to achieve a broad representation of issues that were raised by our primary insomniac participants. Where specific parts of the interview content were similar and terminology appeared synonymous, we sought to construct items that faithfully represented what it was that participants were saying, and what we felt they were meaning by their words. In this context, interview information from normal sleepers provided important comparative data to help us recognise the differences in experience between people with primary insomnia and normal sleepers. In addition to these rich interview data, our discussions were informed by data from other sources, so that we could effectively ‘triangulate’ and so confirm the validity of items. In this regard, we examined individual participant responses to the MCQ-30 and UPWQ. These questionnaires already contain generic (MCQ-30) or specific (UPWQ) items that are relevant to a metacognitive perspective on insomnia. Finally, we reviewed other existing insomnia questionnaires, from the perspective of metacognition (e.g. metacognitive beliefs arising from cognitive intrusions about sleep), and so tried to take into a further body of knowledge already available in the field (Glasgow Content of Thought Inventory GCTI, Harvey & Espie, 2004; Sleep Associated Monitoring Index SAMI, Semler & Harvey, 2004b; Glasgow Sleep Effort Scale GSES, Broomfield & Espie, 2005; Thought Control Questionnaire – Insomnia Revised TCQI-R, Ree et al., 2005; Pre-Sleep Arousal Scale, PSAS, Nicassio et al., 1985). 2.1.4. Data management To ensure consistency with S-REF (Wells, 2000; Wells & Matthews, 1994, 1996) and Broomfield et al. (2005), derived items were then categorised by the consensus group according to metacognitive belief type (negative beliefs about danger and threat related meanings or positive beliefs about cognitive

J. Waine et al. / J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

19

strategies – the plan) and intrusion category (thoughts and images, being awake, physical arousal, environmental factors, or feelings) in a two by five matrix (cf. Rust & Golombok, 1999). 2.2. Stage two: refinement of the item pool and construction of the questionnaire Duplicate items were removed from the matrix. The authors and two behavioural sleep medicine experts considered the remaining items, which were also discussed with a patient with primary insomnia meeting RDC/clinical criteria. In this way, what were felt to be the most appropriate items for each matrix cell were pinpointed. Metacognitive beliefs in response to intrusive thoughts (e.g. Not being able to rest my mind in bed means I won’t perform well the next day; Before I fall asleep, I should try and switch off my thoughts) and wakefulness (e.g. Being awake in bed means I have lost control of my sleep; When awake in bed, I should stay in bed and try harder to sleep) were designated a higher number than other categories, since from interview data these were most relevant to participants. Fifty-four final items were agreed. A final discussion with authors and experts ensured clarity of statements. Lastly, resulting from further discussion and reflection upon the qualitative data, appropriate stem statements were added that appeared consistent with the way in which participants reported their insomnia problems at interview, and with the clinical experience of the authors (e.g. Thinking in bed means I won’t get to sleep, Before I fall asleep I must get things sorted in my mind). A draft scale was created by randomly ordering the 54 items. Instructions asked participants to indicate agreement on a four-point Likert scale (1 ¼ ‘‘do not agree’’, 2 ¼ ‘‘agree slightly’’, 3 ¼ ‘‘agree moderately’’ and 4 ¼ ‘‘agree very much’’) prior to falling asleep at night (cf. MCQ, MCQ-30). No reverse items were included, to avoid potential confusion of respondents. This first draft was then sent to the N ¼ 9 primary insomnia participants from field interviews. A semi-structured response sheet invited feedback and general comments. Based on responses, one item was dropped and seven items added, resulting in the 60-item preliminary Metacognitions Questionnaire – Insomnia (MCQ-I). 2.3. Stage three: MCQ-I field testing and psychometric evaluation To investigate preliminary psychometric properties of MCQ-I, further primary insomnia and normal sleeper participants were recruited. Power calculation based on mean differences data from Cartwright-Hatton and Wells (1997) suggested that 21 participants per group were required assuming a 0.05, power at 0.8 and a large effect size. 2.3.1. Participants The same criterion was applied to select groups. Of 67 ‘poor sleepers’ who returned consent forms, 33 were excluded: 12 withdrew; 7 did not meet inclusion criteria for primary insomnia; 7 met DSM-IV criteria for depression, 2 had Delayed Sleep Phase Syndrome; 3 were substance abusing; 2 had chronic medical conditions. Of 51 normal sleepers who returned consent forms, 14 were excluded (12 did not meet inclusion criteria, 1 met DSM-IV criteria for depression, 1 participant returned incomplete data). The final sample comprised 34 primary insomnia participants and 37 normal sleepers. Primary insomniacs and normal sleepers were equivalent on age and gender (both p > 0.05). People with primary insomnia had significantly longer SOL and sleep disturbance (both p < 0. 001), and higher PSQI, higher HADS-A and higher HADS-D scores (all p < 0. 001), than normal sleepers. Summary data are presented in Table 1. 2.3.2. Procedure Following telephone screening, PSQI, HADS, MCQ-30, and MCQ-I were sent to participants for completion. Three weeks later, participants were sent the MCQ-I again, for test–retest purposes, and received the ‘‘Good Sleep Guide’’ (NMAC, 1993). 2.3.3. Data analysis All continuous data were checked for normality of distribution, using the Kolmogorov–Smirnov (K–S) test. If the findings were significant, data was transformed using Log transformation. The

20

J. Waine et al. / J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

Table 1 Participant characteristics and screening measures (stage 3)

Age Problem duration (years) PSQI Subjective SOL (estimation in minutes from PSQI) HADS (anxiety subscale) HADS (depression scale) a b

Primary insomnia group (n ¼ 37) Mean (SD) median

Normal sleeper group (n ¼ 34) Mean (SD) median

30 (9.2) 31.0 10.2 (7.4) 7.5 11.0 (2.7) 11.0 76.7 (45.4)a 65.0

29 (9.2) 27.0 0.0 (0.0) 0 2.5 (1.0) 2.0 9.6 (5.9)b 8.0

7.6 (3.3) 8.0 3.7 (4.6) 4.0

3.6 (2.7) 3.0 1.1 (1.5) 1.0

Primary insomnia group – missing data for subjective SOL ¼ 2 (excluded from analysis). Normal sleeper group – missing data for subjective SOL ¼ 1 (excluded from analysis).

distribution was then checked again. Based on the outcome, parametric or non-parametric tests were used accordingly. One-tailed tests were employed and are reported here assuming directional hypotheses, with the exception of test–retest validity. 2.3.4. Face validity The scale has good face validity because items were derived directly from participants who were also involved in the iterative three-stage process. 2.3.5. Concurrent validity A significant positive correlation was found between the MCQ-I and MCQ-30 (r ¼ 0.69, p < 0.001) corrected for attenuation. This association represents 47% of explained variance. 2.3.6. Construct validity MCQ-I (total score) and estimated SOL (from PSQI) were positively associated (r ¼ 0.60, p < 0.001), as were MCQ-I total score and PSQI total score (r ¼ 0.59, p < 0.001). 2.3.7. Discriminant validity As is evident in Fig. 1, the median value for primary insomnia participants (124.5) is higher than for normal sleepers (90.1). There is minimal overlap between lowest scoring primary insomnia participants and highest scoring normal sleepers. MCQ-I discriminated primary insomnia (mean ¼ 123.2, SD ¼ 27.6) from normal sleepers (mean ¼ 90.1, SD ¼ 20.0) (t ¼ 5.838, df ¼ 69, p < 0.001), representing a medium effect size r ¼ 0.58. A linear relationship between MCQ-I and HADS-A for primary insomniacs (r ¼ 0.47, p ¼ 0.003) and normal sleepers (r ¼ 0.34, p ¼ 0.02) was observed. Therefore, ANCOVA with HADS-A partialled out was conducted. This was to establish whether differences between types of sleeper could be explained solely by anxiety levels. This indicated primary insomniacs and normal sleepers still differed significantly on MCQ-I (F(1,68) ¼ 10.74, p ¼ 0.001). Sleeper type explained 14% of variance. 2.3.8. Sensitivity and specificity A cut-off of 110 on MCQ-I correctly identified 71% of primary insomniacs and 92% of normal sleepers. Using this cut-off, participants were separated into: ‘high endorsement of metacognitive beliefs’ (score ¼ >111) and ‘low endorsement of metacognitive beliefs’ (score ¼ <110). Mann Whitney U tests examined differences between groups on estimated SOL (n ¼ 68) and PSQI score (n ¼ 71). High endorsers had significantly longer SOL (U ¼ 142.5, p < 0. 001) and higher PSQI (U ¼ 172.5, p < 0. 001) than low endorsers, representing medium to large effect sizes (r ¼ 0.62, 0.60, respectively). 2.3.9. Test–retest reliability Test–retest data for 25 normal sleepers and 18 primary insomnia participants were available (mean interval 28.9 days). Intra-class correlations for all participants (ICC ¼ 0.82, p < 0.001), normal sleepers (ICC ¼ 0.63, p < 0.001) and primary insomniacs (ICC ¼ 0.93, p < 0.001) were all significant.

J. Waine et al. / J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

21

200.00

180.00

5

MCQI - total

160.00

140.00

120.00

100.00

80.00

60.00 normal sleeper

primary insomnia

Type of sleeper Fig. 1. Box and whiskers plot demonstrating MCQ-I’s ability to discriminate between normal sleepers and primary insomnia participants.

2.3.10. Internal consistency Internal consistency data were not calculated due to the high number of items in the preliminary version of the scale relevant to the N of participants (N ¼ 60; cf. Cortina, 1993). 3. Discussion The present study investigated the presence of metacognitive beliefs in primary insomnia through the refinement of a novel scale (MCQ-I), and reported initial MCQ-I psychometric data. Using the pilot scale, significantly higher MCQ-I scores amongst primary insomniacs relative to normal sleepers were observed, an effect that remained when HADS-A was partialled out. This is the first attempt to describe and differentiate metacognition in primary insomnia, and confirms the probable relevance of S-REF to this sleep disorder. A range of metacognitive beliefs may shape pre-sleep cognition in primary insomnia (cf. Broomfield et al., 2005; Harvey, 2005a), although more work is needed. Metacognitive beliefs may reflect a stable trait of primary insomnia – a hypothesis supported by the large correlation (r ¼ 0.82) between the first and second MCQ-I completion for primary insomnia participants. This was lower for normal sleepers (r ¼ 0.63), suggesting items were less consistently meaningful to this group. The current data cannot implicate metacognitive beliefs as a primary insomnia maintaining factor. Following further MCQ-I refinement, an important progression will be to test links between different S-REF components, using experimental methods. It should be possible, for instance, to induce

22

J. Waine et al. / J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

metacognitive beliefs/plans for processing in primary insomnia and normal sleepers, noting effect on pre-sleep intrusions and SOL. MCQ-I would provide a helpful manipulation check in this context. MCQ-I should also prove fruitful in strengthening primary insomnia assessment. Unlike existing scales, MCQ-I can identify metacognitive beliefs in primary insomnia, of relevance for cognitive treatment approaches. Individuals high on MCQ-I may suit a package of cognitive therapy (e.g. Harvey, 2005b), or more tailored cognitive treatment emphasising attention retraining (Wells, 2000), mindfulness (Heidenreich, Tuin, Pflug, Michal, & Michalak, 2006) or metacognitive belief restructuring (Wells, 2000). Treatment approaches such as these, which directly or indirectly disengage S-REF processing, are in our view likely to be productive for certain primary insomnia patients. The validity of the aforementioned work is strengthened given the reasonably promising preliminary psychometric properties of the MCQ-I. The MCQ-I correlated significantly with MCQ-30. Significant associations were observed between MCQ-I and both SOL and PSQI scores. MCQ-I readily discriminated primary insomnia sufferers and normal sleepers (mean scores primary insomnia ¼ 132.2, normal sleepers ¼ 90.1; effect size r ¼ 0.58), even with anxiety controlled out. A score of 110 offered reasonable sensitivity (71%), and good specificity (92%), suggesting MCQ-I may usefully exclude normal sleepers from future research studies. Using this cut-off, high endorsers of metacognitive beliefs showed higher SOL and PSQI scores, relative to low endorsers. Lastly, test–retest reliability was highly acceptable although validation using a larger sample is now necessary. We would acknowledge several limitations. A larger sample size would have enabled more rigorous psychometric examination, including exploration of factor structure, and computation of internal consistency data for subscales. Primary insomnia sufferers were not treatment seeking. However, mean sleep problem duration was over 10 years, and all participants met strict clinical and research diagnostic criteria (Edinger et al., 2004). Parallels can, therefore, be made with poor sleeping individuals approaching their General Practitioner for help. The inclusion of 1–2 nights of sleep data (diary and actigraphy) would have clarified more precisely the relationship between MCQ-I and SOL. And lastly, the MCQ-I refers only to the pre-sleep period; metacognitive beliefs may also relate to nighttime wakefulness after sleep onset. However, the majority of primary insomnia cognitive research to date has examined pre-sleep processes, making it a logical starting point here. Finally, we sought to include obstructive sleep apnea patients, as a highly rigorous control group. Unfortunately, high levels of psychopathology necessitated their exclusion. Future MCQ-I research should endeavour to include a sleep specific control cohort, perhaps snorers, who would be less likely to show such elevated depression. The present study was a preliminary scale development study. Future psychometric research of MCQ-I is required, using a substantially larger sample size, additional ancillary measures of cognitive activity, inclusion of a sleep specific control group e.g. snorers and measures of sleep. Arguably, by maximising MCQ-I sensitivity, the clinical utility of the scale will be augmented. And if findings remain as promising as ours, consideration should also be given to development of a specific metacognitive model of primary insomnia, similar to that for generalised anxiety (Wells, 1995) and obsessive– compulsive disorder (Wells & Matthews, 1994). Certainly, the MCQ-I, the first scale able to directly measure metacognitive beliefs in primary insomnia, shows promising psychometric properties. References Ansfield, M. E., Wegner, D. M., & Bowser, R. (1996). Ironic effects of sleep urgency. Behaviour Research and Therapy, 34(7), 523–531. Borkovec, T. D. (1982). Insomnia. Journal of Consulting and Clinical Psychology, 50, 880–895. Broman, J., & Hetta, J. (1994). Perceived pre-sleep arousal in patients with persistent psychophysiological and psychiatric insomnia. Nordic Journal of Psychiatry, 38(3), 203–207. Broomfield, N. M., & Espie, C. A. (2005). Toward a valid, reliable measure of sleep effort. Journal of Sleep Research, 14, 1–7. Broomfield, N. M., Gumley, A., & Espie, C. A. (2005). Candidate cognitive processes in psychophysiologic insomnia. International Journal of Cognitive Psychotherapy, 19(1), 5–18. Buysse, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Research, 28, 193–213. Cartwright-Hatton, S., & Wells, A. (1997). Beliefs about worry and intrusions: the Meta-Cognitions Questionnaire and its correlates. Journal of Anxiety Disorders, 11(3), 279–296. Clark, D. A., & Rhyno, S. (2005). Unwanted intrusive thoughts in nonclinical individuals: implications for clinical disorders. In D. Clark (Ed.), Intrusive thoughts in clinical disorders theory, research and treatment (pp. 1–29). New York: The Guilford Press. Cortina, J. M. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98–104.

J. Waine et al. / J. Behav. Ther. & Exp. Psychiat. 40 (2009) 15–23

23

Edinger, J. D., Bonnet, M. H., Bootzin, R. R., Doghramji, K., Dorsey, C. M., Espie, C. A., et al. (2004). Derivation of research diagnostic criteria for insomnia. Report of an American Academy of Sleep Medicine Work Group. Ellis, J., & Cropley, C. (2002). An examination of thought control strategies employed by acute and chronic insomniacs. Sleep Medicine, 3, 393–400. Espie, C. A., Brooks, D. N., & Lindsay, W. R. (1989). An evaluation of tailored psychological treatment of insomnia. Journal of Behavior Therapy and Experimental Psychiatry, 20(2), 143–153. Espie, C. A., Broomfield, N. M., MacMahon, K. M., & McPhee, L. M. (2006). The attention–intention-effort pathway in the development of psychophysiologic insomnia: an invited theoretical review. Sleep Medicine Reviews, 10, 215–245. Gross, R. T., & Borkovec, T. D. (1982). Effects of a cognitive intrusion manipulation on sleep-onset latency of good sleepers. Behaviour Therapy, 13, 117–124. Hall, M., Buysse, D. J., Reynolds, C. F., Kupfer, D. J., & Baum, A. (1996). Stress related intrusive thoughts disrupt sleep onset and continuity. Sleep Research, 25, 163. Harvey, A. G. (2000). Pre-sleep cognitive activity: a comparison of sleep-onset insomniacs and good sleepers. British Journal of Clinical Psychology, 39, 275–286. Harvey, A. G. (2001). I can’t sleep, my mind is racing! An investigation of strategies of thought control in insomnia. Behavioural and Cognitive Psychotherapy, 29, 3–11. Harvey, A. G. (2003a). Beliefs about the utility of pre-sleep worry: an investigation of individuals with insomnia and good sleepers. Cognitive Therapy and Research, 27(4), 403–414. Harvey, A. G. (2003b). The attempted suppression of presleep cognitive activity in insomnia. Cognitive Therapy and Research, 27(6), 593–602. Harvey, A. G. (2005a). Unwanted intrusive thoughts in insomnia. In D. Clark (Ed.), Intrusive thoughts in clinical disorders theory, research and treatment (pp. 86–118). New York: The Guilford Press. Harvey, A. G. (2005b). A cognitive theory and therapy for chronic insomnia. Journal of Cognitive Psychotherapy, 19(1), 41–60. Harvey, K. J., & Espie, C. A. (2004). Development and preliminary validation of the Glasgow Contents of Thoughts Inventory (GCTI): a new measure for the assessment of pre-sleep cognitive activity. British Journal of Clinical Psychology, 43, 409–420. Heidenreich, T., Tuin, I., Pflug, B., Michal, M., & Michalak, J. (2006). Mindfulness-based cognitive therapy for persistent insomnia: a pilot study. Psychotherapy and Psychosomatics, 75, 188–189. Jones, B. T., Macphee, L., Broomfield, N. M., Jones, B. C., & Espie, C. A. (2005). Sleep-related attentional bias in good, moderate and poor (primary insomnia) sleepers. Journal of Abnormal Psychology, 114, 249–258. Kuisk, L. A., Bertelson, A. D., & Walsh, J. K. (1989). Presleep cognitive hyperarousal and affect as factors in objective and subjective insomnia. Perceptual and Motor Skills, 69, 1219–1225. Lichstein, K., & Rosenthal, T. (1980). Insomniacs perceptions of cognitive versus somatic determinants of sleep disturbance. Journal of Abnormal Psychology, 89, 105–107. Marcetti, L. M., Biello, S. M., Broomfield, N. M., MacMahon, K. M. A., & Espie, C. A. (2006). Who is pre-occupied with sleep? A comparison of attention bias in people with psychophysiologic insomnia, delayed sleep phase syndrome and good sleepers using the induced change blindness paradigm. Journal of Sleep Research, 15, 212–221. Morin, C. M. (1993). Insomnia: Psychological assessment and management. New York: The Guilford Press. Morin, C. M., Culbert, J. P., & Schwartz, M. S (1994). Non-pharmacological interventions for insomnia: a meta-analysis of treatment efficacy. American Journal of Psychiatry, 151, 1172–1180. Morin, C. M., & Espie, C. A. (2003). Insomnia: A clinical guide to assessment and treatment. New York: Kluwer Academic/Plenum Publishers. Murtagh, D. R., & Greenwood, K. M. (1995). Identifying effective psychological treatments for insomnia: a meta-analysis. Journal of Consulting and Clinical Psychology, 63, 79–89. National Medical Advisory Committee. (1993). The good sleep guide. In Management of anxiety and insomnia. London: HMSO. Nicassio, P. M., Mendlowitz, D. R., Fussell, J. J., & Petras, L. (1985). The phenomenology of the pre-sleep state: the development of the pre-sleep arousal scale, 23(3), 263–271. Ree, M. J., Harvey, A. G., Blake, R., Tang, N., & Shawe-Taylor, M. (2005). Attempts to control unwanted thoughts in the night: development of the thought control questionnaire-insomnia revised (TCQI-R). Behaviour Research & Therapy, 43, 985–988. Riemann, D., & Voderholzer, U. (2003). Primary insomnia: a risk factor to develop depression? Journal of Affective Disorders, 76(1), 255–259. Rust, J., & Golombok, S. (1999). Modern psychometrics: The science of psychological assessment, (2nd ed.). London: Routledge. Semler, C. N., & Harvey, A. G. (2004a). Monitoring for sleep-related threat: a pilot study of the sleep associated monitoring index (SAMI). Psychosomatic Medicine, 66(2), 242–250. Semler, C. N., & Harvey, A. G. (2004b). An investigation of monitoring for sleep-related threat in primary insomnia. Behaviour Research and Therapy, 42(12), 1403–1420. Smith, M. T., Perlis, M. L., Park, A., Giles, D. E., Pennington, J. A., & Buysse, D. J. (2002). Behavioural treatment versus pharmacotherapy for insomnia – a comparative meta-analyses. American Journal of Psychiatry, 159, 5–11. Wegner, D. M. (1994). Ironic processes of mental control. Psychological Review, 101, 34–52. Wells, A. (1995). Meta-cognition and worry: a cognitive model of generalised anxiety disorder. Behavioural and Cognitive Psychotherapy, 23, 301–320. Wells, A. (2000). Emotional disorders and metacognition innovative cognitive therapy. Chichester: John Wiley and Sons, Ltd. Wells, A., & Cartwright-Hatton, S. (2003). A short form of the metacognitions questionnaire: properties of the MCQ-30. Behaviour and Research Therapy, 42(4), 385–396. Wells, A., & Matthews, G. (1994). Attention and emotion: A clinical perspective. Hove: Lawrence Erlbaum Associates. Wells, A., & Matthews, G. (1996). Modelling cognition in emotional disorder: the S-REF model. Behaviour and Research Therapy, 34(11/12), 881–888. Wicklow, A., & Espie, C. A. (2000). Intrusive thoughts and their relationship to actigraphic measurement of sleep: towards a cognitive model of insomnia. Behaviour Research and Therapy, 28, 679–693. Zigmond, A. S., & Snaith, R. P. (1983). The hospital anxiety and depression scale. Acta Psychiatrica Scandinavica, 67(6), 361–370.