- Email: [email protected]
Are changes in beliefs and attitudes about sleep related to sleep improvements in the treatment of insomnia?
Behaviour Research and Therapy 40 (2002) 741–752 www.elsevier.com/locate/brat
Are changes in beliefs and attitudes about sleep related to sleep improvements in the treatment of insomnia? C.M. Morin *, F. Blais, J. Savard E´cole de Psychologie, Universite´ Laval, Sainte-Foy, Quebec, Canada G1K 7P4 Accepted 30 March 2001
Abstract Dysfunctional beliefs and attitudes about sleep are presumed to play an important mediating role in perpetuating insomnia. The present study evaluated the impact of cognitive-behavioral and pharmacological treatments for insomnia on sleep-related beliefs and attitudes and the relationship between those changes and sleep improvements. The participants were older adults with chronic and primary insomnia. They received cognitive-behavior therapy (CBT), pharmacotherapy (PCT), combined CBT+PCT (COMB), or a medication placebo (PLA). In addition to daily sleep diaries and sleep laboratory measures, the participants completed the dysfunctional beliefs and attitudes about sleep scale (DBAS) at baseline and posttreatment, and at 3-, 12- and 24-month follow-up assessments. The results showed that CBT and COMB treatments produced greater improvements of beliefs and attitudes about sleep at posttreatment than PCT and PLA. Reductions of DBAS scores were significantly correlated with improvements of sleep efficiency as measured by daily sleep diaries and by polysomnography. In addition, more adaptive beliefs and attitudes about sleep at posttreatment were associated with better maintenance of sleep improvements at follow-ups. These findings highlight the importance of targeting sleep-related beliefs and attitudes in the treatment of insomnia. 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction Insomnia is a prevalent condition affecting between 9 and 15% of the adult population on a chronic basis and over 25% of people aged over 60 years old (Ford & Kamerow, 1989; Mellinger, Balter, & Uhlenhuth, 1985). Sleep disturbances are frequently associated with fatigue, problems of attention and concentration, and with mood disturbances. Chronic insomnia is also a risk factor
* Corresponding author. E-mail address: [email protected] (C.M. Morin).
0005-7967/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 5 - 7 9 6 7 ( 0 1 ) 0 0 0 5 5 - 9
742
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
for major depression and it carries important health-care costs (Ford & Kamerow, 1989; Simon & VonKorff, 1997). Insomnia can be secondary to medical, psychiatric, and other sleep disorders (American Sleep Disorders Association, 1990). For primary insomnia, the etiology is often multifactorial involving such factors as hyperarousal, maladaptive sleep habits (e.g. irregular schedule, excessive time spent in bed), and dysfunctional beliefs and attitudes about sleep (e.g. worry over sleep loss and its consequences) (Espie, 1991; Lichstein & Morin, 2000; Morin, 1993). Of those contributing actors, the role of cognitive variables (e.g. intrusive thoughts, beliefs, attitudes, expectations) has received little research attention, despite increasing recognition that they play a critical role in the development and maintenance of insomnia. An example of the influence of cognitions on insomnia is the typical reaction of excessive worrying over sleep loss and its potential consequences. This reaction can increase physiological and emotional arousal (i.e. performance anxiety) and, in a vicious cycle, perpetuate the underlying sleep disturbance. The belief concerning the absolute need for 8 h of sleep each night to feel refreshed and function well during the day is another form of faulty belief. As sleep needs vary across individuals, this belief can lead to emotional distress and exacerbate insomnia when this requirement is not met. Most studies investigating the role of cognitive processes on insomnia have focused on the impact of pre-sleep or nocturnal cognitions on sleep. These studies have shown that individuals with insomnia report more negative thoughts during nocturnal awakenings than good sleepers (Watts, Coyle, & East, 1994) and that negative presleep cognitions (e.g. thoughts about not falling asleep) are associated with increased subjective (but not objective) sleep difficulties (Van Egeren, Haynes, Franzen, & Hamilton, 1983). Among older adults, highly distressed poor sleepers endorse more negative thoughts about insomnia and other issues (e.g. health, work, and family) during nocturnal awakenings compared to good sleepers and low distressed poor sleepers (Fichten et al., 1998). Although these findings support the view that negative cognitions are related to sleep disturbance, it is not clear whether these situational and automatic thoughts cause insomnia or simply represent an epiphenomenon of poor sleep. Underlying beliefs and attitudes about sleep (i.e. schemas), which are more deeply ingrained than automatic thoughts, are likely to play an even more important role in the development of insomnia. In a study of older adults, participants with chronic insomnia endorsed stronger dysfunctional beliefs and attitudes about sleep than selfdefined good sleepers (Morin, Stone, Trinkle, Mercer, & Remsberg, 1993b). Specifically, poor sleepers held stronger beliefs about the potential consequences of insomnia, and worried more about losing control and about the unpredictability of sleep. Edinger et al. (2000) reported similar findings and also found that such faulty beliefs and attitudes about sleep were more closely related to subjective than objective insomnia. Together, these results suggest that sleep-related maladaptive attitudes and beliefs are possibly involved in chronic insomnia and that they represent an important target for treatment. It is only recently that psychological interventions for insomnia have incorporated cognitive restructuring strategies to directly target faulty beliefs and attitudes about sleep (Edinger, Hoelscher, Marsh, Lipper, & Ionescu-Pioggia, 1992; Edinger, Wohlgemuth, Radtke, Marsh, & Quillian, 2001; Espie, Inglis, Tessier, & Harvey, 2000; Morin, Colecchi, Stone, Sood, & Brink, 1999a; Morin, Kowatch, Barry, & Walton, 1993a; Morin, Stone, McDonald & Jones, 1994; Riedel, Lichstein, & Dwyer, 1995; Sanavio, Vidotto, Bettinardi, Rolleto, & Zorzi, 1990; Verbeek, Schreuder, & Declerk, 1999). Results of these studies suggest that adding cognitive restructuring
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
743
to behavioral strategies (e.g. stimulus control, sleep restriction) may produce slightly greater benefits than behavioral treatment alone (Lichstein & Riedel, 1994; Morin, Hauri et al., 1999a). However, because cognitive therapy has always been evaluated in the context of multifaceted treatments, the specific contribution of altering maladaptive sleep cognitions in the treatment of insomnia remains unknown. In a study examining the role of attitudinal change in the treatment of chronic users of hypnotics, Pat-Horenczyk (1998) implemented a medication withdrawal program, followed by CBT focusing on the correction of faulty sleep beliefs and attitudes. Although improvements were noted during withdrawal, it is only with the addition of cognitive therapy that significant changes occurred in attitudes and beliefs. Also, the magnitude of those changes was correlated with subjective sleep measures. While these findings are interesting, they were based on a small sample (n=19) and it is unclear whether the attitudinal changes initiated during withdrawal would have continued without the addition of CBT. It is also intriguing that only subjective sleep measures were associated with the attitudinal change. Together, those findings suggest that beliefs and attitudes about sleep are involved in the etiology of primary insomnia and that such cognitive variables may represent an important treatment target. Accordingly, the objectives of this study were to compare changes in sleep-related beliefs and attitudes obtained with cognitive-behavioral and pharmacological therapies for insomnia, and to evaluate the relationship between those changes and sleep improvements.
2. Method 2.1. Participants Data collection was part of a larger study comparing the efficacy of CBT and PCT for latelife insomnia (Morin et al., 1999a). Participants were older adults, recruited through newspaper advertisements, who met DSM-IV criteria for primary insomnia (APA, 1994). Minimal entry criteria were that participants reported chronic (greater than a 6-month duration) difficulties initiating (sleep-onset latency greater than 30 min per night) and/or maintaining sleep (wake after sleep onset greater than 30 min per night) for more than three nights per week. Participants were excluded if there was evidence of a progressive medical illness (e.g. congestive heart failure, cancer), a major psychiatric disorder (e.g. major depression), or another sleep disorder (e.g. sleep apnea or periodic limb movements during sleep). All participants were free from psychotropic drugs for at least two weeks prior to entering the study. Of the 78 patients enrolled in the study, 72 (47 women and 25 men) completed treatment. Their mean age was 64.7 years (SD=6.9 years). Sixty-nine percent of the participants were married, 18.1% widowed, and 12.5% were divorced. Also, 45.8% were retired, 37.5% were employed, 15.3% were homemakers, and 1.4% were unemployed. The mean education level was 14.4 years (SD=2.5). The average duration of insomnia was 17.0 years (SD=17.2 years). Sixty four percent of the participants had mixed difficulties initiating and maintaining sleep, while 25% had sleepmaintenance and 6.9% sleep-onset difficulties only.
744
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
2.2. Procedure After an initial telephone screening, all prospective subjects underwent a multistep evaluation involving a clinical interview with a board-certified sleep specialist, a physical examination, a structured clinical interview for DSM-III-R diagnoses (SCID), and several additional self-report measures including the dysfunctional beliefs and attitudes about sleep (described later). In addition, all participants completed daily sleep diaries for the duration of the study and three consecutive nights of polysomnography PSG at baseline and posttreatment assessment. Subjects were randomly assigned to one of the following conditions: (a) cognitive-behavior therapy (CBT; n=18); (b) pharmacotherapy (PCT; temazepam; n=20); (c) combined CBT and PCT (COMB; n=20); or (d) a medication placebo (PLA; n=20). All treatments lasted eight weeks and were conducted on an outpatient basis. More detailed information on the assessment and treatment procedures of this study is available in a previously published report (Morin et al., 1999a). 2.3. Measures 2.3.1. Sleep diary All participants completed a sleep diary every morning at breakfast time for a 2-week baseline period, during the 8-week treatment period, and for an additional 2-week period at 3-, 12- and 24-month follow-up assessments. Several measures were derived from the diary, including sleeponset latency, wake after sleep onset, total wake time, total sleep time, and sleep efficiency. For the purpose of the present study, only the sleep efficiency variable was used. This variable was selected because it represents the best composite score reflecting the severity of insomnia. Its value is computed as the ratio of total sleep time to time spent in bed and multiplied by 100. Sleep variables were coded each night and a weekly mean was computed. Although sleep diary data do not reflect absolute values obtained with PSG, they are often the most practical modality of obtaining repeated measurements of sleep patterns and still represent a reliable index of insomnia (Coates et al., 1982; Lacks & Morin, 1992). Sleep diary data were available for 72 participants at posttreatment, and for 59, 51, and 49 participants at the 3-, 12- and 24-month follow-up assessments, respectively. 2.3.2. Polysomnography The participants underwent three consecutive nights of sleep laboratory evaluation both before and after treatment. Bedtime and arising time in the sleep laboratory were kept as close as possible (i.e. within 30 min) to the subject’s habitual sleep schedule. A standard PSG montage that included electroencephalographic (EEG), electromyographic (EMG) and electrooculographic (EOG) monitoring was used. Sleep stages were scored by an experienced technician according to standardized criteria (Rechtschaffen & Kales, 1968). As was the case for the sleep diaries, several dependent variables were derived from sleep laboratory assessment, but only the sleep efficiency (total sleep time divided by total time spent in bed and multiplied by 100) variable was used in the present study. There were 72 patients completing PSG evaluation at baseline and 70 subjects at posttreatment.
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
745
2.4. Dysfunctional beliefs and attitudes about sleep scale The dysfunctional beliefs and attitudes about sleep scale (DBAS) (Morin 1993, 1994) is a 30item scale tapping various beliefs, attitudes, expectations and attributions about sleep and insomnia. The nature of these cognitions cluster around five conceptually derived themes: (a) misconceptions about the causes of insomnia (e.g. “I believe insomnia is essentially the result of aging and there isn’t much that can be done about this problem”); (b) diminished perception of control and predictability of sleep (e.g. “When I sleep poorly on one night, I know it will disturb my sleep schedule for the whole week”); (c) unrealistic sleep expectations (e.g. “I must get 8 h of sleep to feel refreshed and function well during the day”); (d) misattribution or amplification of the consequences of insomnia (e.g. “I am concerned that chronic insomnia may have serious consequences on my physical health”); and (e) faulty beliefs about sleep-promoting practices (e.g. “When I have trouble sleeping, I should stay in bed and try harder”). For each statement, the person rates his or her level of agreement/disagreement on a 100-mm visual analog scale ranging from 0 (strongly disagree) to 100 (strongly agree). Except for item 23 for which the score is reversed, a higher score indicates more dysfunctional beliefs and attitudes about sleep. The total score is based on the average score of all items. Likewise, subscale scores are computed by adding the sum of scores for the items and dividing by the number of items making up each subscale. Initial psychometric data indicate that the DBAS has an excellent internal consistency (Cronbach’s Alpha=0.80) and an average item-total correlation of 0.37 (Morin, 1994; Morin et al., 1993b). Additional data have confirmed the acceptable psychometric properties of this scale (Espie et al., 2000). In the present study, completed DBAS were available for 66 participants at baseline, 69 at posttreatment, and 56, 55, and 53 participants at the 3-, 12- and 24-month follow-up assessments, respectively. 2.5. Treatment 2.5.1. Cognitive behavior therapy CBT consisted of a multifaceted intervention involving behavioral, cognitive and educational components. The behavioral component involves a combination of sleep restriction (Spielman, Saskin, and Thorpy, 1987) and stimulus control procedures (Bootzin, Epstein, & Wood, 1991). The cognitive therapy component was modeled after the work of Beck, Ellis, and Meichenbaum and was aimed at altering dysfunctional beliefs and attitudes about sleep. The same five themes as those covered in the DBAS scale were addressed during cognitive therapy sessions: misconceptions about the causes of insomnia, misattributions or amplification of the consequences of insomnia, diminished perception of control and predictability of sleep, unrealistic sleep expectations, and faulty beliefs about sleep-promoting practices. Cognitive therapy sessions followed a threestep process: (a) identification of patient-specific dysfunctional thoughts (using both the DBAS scale and self-monitoring); (b) confrontation and challenging of those thoughts; and (c) implementation of methods for replacing these thoughts with more rationale substitutes. Three therapy sessions were devoted to cognitive therapy and two others were used to reinforce the main principles and procedures covered by this therapy component. Finally, the educational component covered basic facts about changes in sleep occurring with aging and about sleep hygiene principles regarding the effects of caffeine, alcohol, exercise, and diet on sleep (Hauri, 1991).
746
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
2.5.2. Pharmacotherapy PCT consisted of 7.5–30 mg temazepam recommended for 2–3 nights per week but made available for nightly use if the patients so desired. The dosage was adjusted weekly, based on treatment response and side effects. Both active and placebo drugs were provided in identical gelatin capsules, filled with an equivalent amount of powder. Participants in all three conditions involving medications (PCT, PLA, and COMB) were seen by a psychiatry resident on a weekly basis for 20–30 min sessions devoted to reviewing treatment progress, side effects, and insomnia symptoms. Urine drug screens for benzodiazepines were conducted at baseline to ensure that subjects were free from these agents and at late treatment to ensure compliance with prescribed medication regimen. At the end of the 8-week treatment phase, all participants receiving medication began a 1–2 week tapering schedule. 2.5.3. Combined (COMB) CBT plus PCT Participants in this condition received both CBT and active medication. They were seen weekly by both a clinical psychologist for CBT and a physician for the medication management. 2.5.4. Medication-placebo Participants in this condition received medication which consisted of an inert lactose solution. However, it was provided in an identical manner as those participants in the active medication condition. 3. Results In order to understand better the current results, this section begins with a brief summary of the main sleep findings from the original study (Morin et al., 1999a). All three active treatments produced significantly greater increases in sleep efficiency from baseline to posttreatment than the placebo condition. Baseline to posttreatment changes on the sleep diary measure were as follows: CBT, 68–85%; PCT, 72–83%; COMB, 64–85% and PLA, 69–74%. The average effect size pooled across the three treatments was 1.16. Changes on the polysomnographic measure were as follows: CBT, 78–86%; PCT, 77–85%; COMB, 76–87% and PLA, 79–80%. The average effect size pooled across the three treatments was 0.80. Although there was a trend for the combined condition to produce greater improvement rates, there was no significant difference among the three active treatment conditions. There were respectively 78% (CBT), 56% (PCT), 75% (COMB), and 14% (PLA) of the participants who reached a clinically meaningful outcome on sleep patterns at posttreatment. Also, follow-up data showed that CBT sustained its sleep improvements very well over time, the drug therapy alone condition returned near baseline, and the combined condition yielded an intermediate outcome falling between those two single therapy conditions. 3.1. Attitudinal changes across conditions Because of some missing data on the DBAS and attrition over time, the sample size varied from 69 subjects at posttreatment to 46 subjects at the 24-month follow-up. Missing data were
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
747
not replaced nor estimated for those analyses. Group means were computed for total DBAS scores at each assessment period. These data are presented in Fig. 1. An analysis of variance (ANOVA) revealed no significant difference between groups on pretreatment DBAS total score. An ANOVA with repeated measures on baseline and posttreatment scores yielded significant Time (F(1,59)=31.61, p⬍0.0001), Group (F(3,59)=6.31, p⬍0.001), and Group×Time interaction effects (F(3,59)=21.32, p⬍0.0001). Post hoc comparisons, using the Tukey test, showed that the CBT and COMB groups had significantly lower DBAS scores at posttreatment compared to PCT and PLA conditions (ps⬍0.05). A visual inspection of Fig. 1 shows that DBAS scores decreased sharply from baseline to posttreatment among patients receiving CBT, alone or in combination with medications, whereas those scores increased slightly in the two drug only conditions. No significant difference was obtained between CBT and COMB or between PCT and PLA groups. Average effect sizes, based on change scores from baseline to posttreatment, were 1.84 for CBT, 1.89 for COMB, and 0.31 for PCT. Posttreatment DBAS scores obtained for the CBT and COMB conditions fell below the average score (i.e. 30) obtained in another study by self-defined good sleepers of the same age (Morin et al., 1993b). Analyses of variances computed for each of the five separate DBAS themes (expectations, causal attributions, etc.) generally yielded the same results as those obtained for the total DBAS score and are not reported here. A repeated measures ANOVA was conducted on posttreatment, 3-, and 12-month follow-up DBAS scores. The 24-month follow-up assessment was not included because of too many missing data. A significant effect for Group, (F(3,42)=8.72, p⬍0.001), was obtained but there was no significant effect for Time or Group×Time interaction. All three active treatment conditions (CBT, PCT and COMB) endorsed less dysfunctional beliefs and attitudes about sleep than the PLA group (ps⬍0.05), but there was no significant difference among them. 3.2. Relationship between DBAS and sleep efficiency The relationship between attitudinal and sleep pattern changes was examined by correlating pre to posttreatment change scores on the DBAS with change scores on sleep efficiency as meas-
Fig. 1. Changes in beliefs and attitudes over time (total score).
748
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
ured by daily sleep diaries and PSG. These analyses were computed for all participants in the four conditions combined. Pearson correlations (see Table 1) indicate that reduced DBAS scores were significantly correlated with improved sleep efficiency as measured by daily sleep diaries (r(63)=-0.28, p⬍0.05); the correlation was nearly significant for the sleep efficiency variable based on PSG (r(62)=-0.24, p⬍0.065). Correlational analyses were also computed between change scores on the five separate themes of the DBAS and change scores on sleep efficiency from the pre to posttreatment period. The pattern of results was very similar for each of those themes (see Table 1). For example, improvements (i.e. reduction of score) of beliefs related to theme 1 (causal attributions of insomnia) were significantly associated with improved sleep efficiency as measured by daily sleep diaries (r(63)=0.28, p⬍0.05) but not with PSG. Changes on theme 2 (control and predictability of sleep) were significantly correlated with improved sleep efficiency as measured by daily sleep diaries (r(63)=0.27, p⬍0.05) and by PSG (r(62)=-0.40, p⬍0.005). Changes on theme 3 (sleep requirement expectations) were significantly associated with improved sleep efficiency as measured by daily sleep diaries only (r(63)=-0.27, p⬍0.05). Changes on theme 4 (consequences of insomnia) were correlated with improved sleep efficiency based on sleep diaries (r(63)=-0.28, p⬍0.01) and PSG (r(62)=-0.25, p⬍0.053. Changes on theme 5 (sleep-promoting practices) were not significantly correlated with sleep efficiency. 3.3. Relationship between DBAS at posttreatment and maintenance of therapeutic gains As shown in Table 2, posttreatment DBAS scores were significantly correlated with sleep efficiency (based on sleep diary assessment) at each of the three follow-up assessments: 3-month (r(56)=-0.22, p⬍0.05), 12-month (r(52)=-0.36, p⬍0.01), and 24-month (r(46)=-0.51, p⬍0.0005). Analyses for separate themes showed that posttreatment scores for theme 1 (causal attributions of insomnia) were significantly correlated with sleep efficiency at 12-month (r(52)=-0.37, p⬍0.01) and 24-month follow-ups (r(46)=-0.46, p⬍0.005)). DBAS scores for theme 2 (control and predictability of sleep) and 4 (consequences of insomnia) were significantly correlated with sleep efficiency at 12- and 24-month follow-ups (-0.48⬎rs⬎-0.28, ps⬍0.05). Posttreatment scores for theme 3 (sleep expectations) were significantly correlated with sleep efficiency at 24-month folTable 1 Correlations between change scores from baseline to posttreatment on DBAS and sleep efficiency measures (note: * p⬍0.05 **p⬍0.01; ***p⬍0.005) Sleep efficiency
DBAS total score Causal attributions Control/predictability Sleep expectations Consequences Sleep-promoting practices
Polysomnography (N=63)
Sleep diary (N=62)
⫺0.24 ⫺0.09 ⫺0.40*** ⫺0.19 ⫺0.25* ⫺0.03
⫺0.28* ⫺0.28* ⫺0.27* ⫺0.27* ⫺0.29** ⫺0.19
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
749
Table 2 Correlations between DBAS posttreatment scores and sleep efficiency (diary) at follow-ups (note: *p ⬍0.05, ** p⬍0.01, *** p ⬍0.005) Sleep efficiency
DBAS postreatment scores Causal attributions Control/predictability Sleep expectations Consequences Sleep-promoting practices
3-month follow-up (N=56)
12-month follow-up (N=52)
24-month follow-up (N=46)
⫺0.28*
⫺0.36**
⫺0.51***
⫺0.20 ⫺0.24 ⫺0.19 ⫺0.24
⫺0.37** ⫺0.40*** ⫺0.14 ⫺0.28*
⫺0.46*** ⫺0.48*** ⫺0.35** ⫺0.36**
⫺0.24
⫺0.32*
⫺0.45***
low-ups (r(46)=-0.35, whereas scores for theme 5 were significantly correlated with sleep efficiency at 12-month, r(52)=-0.32, p⬍0.05, and 24-month follow-up, r(46)=-0.45, p⬍0.005). Thus, a lower posttreatment score on sleep-related cognitions was related to a higher sleep efficiency at follow-ups. 4. Discussion The results indicate that dysfunctional beliefs and attitudes endorsed by older adults with chronic insomnia are improved with CBT, implemented singly or in combination with drug therapy. These changes are specific to CBT as no such improvements were obtained among patients treated with drug therapy alone. The magnitude of changes on sleep-related cognitions was associated with the degree of sleep improvements at posttreatment. In addition, more adaptive beliefs and attitudes about sleep (i.e. lower DBAS scores) at posttreatment were associated with better maintenance of sleep improvements at follow-ups. These findings raise an important question as to whether changes in beliefs and attitudes are produced by CBT per se, or simply represent a by-product of sleep improvements, regardless of the type of treatment that produced those changes. As all three active treatments produced equivalent increases of sleep efficiency from baseline to posttreatment (Morin et al., 1999a,b), it can be argued that improved sleep pattern in itself was not sufficient to produce changes in beliefs and attitudes. Consistent with previous findings (Espie, Inglis, Tessier, & Harvey, 2001; Harvey, Inglis, & Espie, 2001; Pat-Horenczyk, 1998), it appears that psychological treatment (CBT) is a necessary component to reduce dysfunctional sleep-related attitudes and beliefs. A related question is whether cognitive therapy is an essential ingredient to alter sleep-related cognitions or whether similar changes would have been obtained with behavioral treatment alone. Because cognitive therapy was combined with behavioral procedures in the two conditions that yielded significant changes in beliefs and attitudes, it is not possible to answer this question. Another important issue is whether changes in beliefs and attitudes are essential to effectively treat insomnia. The present results, although correlational, suggest that those changes are instru-
750
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
mental to successful treatment outcome. Significant correlations were found between the magnitude of changes on sleep-related beliefs and attitudes and the degree of sleep improvements at posttreatment. Moreover, the strength of the relationship between cognitive changes and sleep improvements increased over time. Although this finding has to be interpreted cautiously because of the smaller number of participants available at 24-month follow-up, this trend suggests that cognitive changes occurring during the course of treatment are important for the long-term maintenance of therapeutic gains. Participants in all three active treatments, including those receiving PCT alone, endorsed fewer dysfunctional sleep cognitions at follow-ups compared to those receiving placebo. These results indicate that changes in beliefs and attitudes achieved at posttreatment with CBT, either alone or in combination with medication, were well sustained over time. While these data also suggest that PCT alone improved sleep cognitions, participants in that condition had consistently higher DBAS scores at each follow-up assessment compared to the two treatments incorporating CBT (see Fig. 1). In addition, there was no significant change over time in DBAS scores within the drug therapy condition alone. Because of attrition at follow-ups, it is possible that the absence of difference among the three active treatments was due to low statistical power. An interesting finding is that participants treated with CBT, alone or in combination, obtained DBAS scores at posttreatment that were comparable to those obtained by older adults without insomnia complaints in two other studies (Edinger et al., 2000; Morin et al., 1993b). Although one could argue that a lower DBAS score is to be expected following cognitive therapy targeting specific beliefs included on that scale, it is still interesting to note that posttreatment status of CBT-treated subjects reached a level previously observed in non-complaining individuals. Although additional normative data are needed to further validate this scale, these preliminary findings provide evidence highlighting the clinical significance of changes in beliefs and attitudes about sleep. Consistent with previous findings (Edinger et al., 2000; Pat-Horenczyk, 1998; Van Egeren et al., 1983), changes in sleep cognitions were more strongly associated with subjective (sleep diaries) than with objective (PSG) sleep improvements. Although it is common to obtain higher correlations between variables from a similar assessment modality (i.e. subjective vs. subjective) than for variables measured with different modalities (e.g. subjective vs. physiological), another possible explanation for this finding is that untreated insomniacs tend to overestimate their sleep problem and to underestimate their sleep duration (Coates et al., 1982; Morin, 1993). As cognitive therapy addressed such misperceptions, it is plausible that patients treated with CBT learned to perceive their sleep more accurately over time which, in turn, contributed to a stronger relationship between cognitive changes and subjectively assessed sleep improvement. Previous studies of psychological interventions for insomnia have used predominantly behavioral approaches (e.g. stimulus control, sleep restriction) that focused on modifying maladaptive sleep habits. These procedures have been shown to be effective with both younger and older adults (Lichstein & Riedel, 1994; Lichstein & Morin, 2000; Morin et al., 1994; Pallesen, Nordhus, & Kvale, 1998; Schramm, Hohagen, Backhaus, Lis, & Berger, 1995), although not all patients have been responsive nor compliant with such behavioral procedures. The present findings highlight the importance of targeting sleep-related cognitions in the management of insomnia. The addition of cognitive restructuring procedures that target perpetuating factors (i.e. fear of the consequences of insomnia) or that prevent compliance with behavioral instructions could enhance
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
751
treatment efficacy. Future research will need to isolate the cognitive therapy component in order to determine its unique contribution to treatment outcome, both on cognitive and sleep variables. Acknowledgements Research supported by National Institute of Mental Health grants #MH47020 and #MH55469. References American Psychiatric Association (APA). (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: APA. American Sleep Disorders Association. (1990). International classification of sleep disorders (ICSD): diagnostic and coding manual. Rochester, MN: APA. Bootzin, R. R., Epstein, D., & Wood, J. M. (1991). Stimulus control instructions. In P. Hauri (Ed.), Case studies in insomnia (pp. 19–28). New York: Plenum Press. Coates, T. J., Killen, J. D., George, J., Silverman, S., Marchini, E., & Thoresen, C. E. (1982). Estimating sleep parameters: a multi-trait–multi-method analysis. Journal of Consulting and Clinical Psychology, 50, 345–352. Edinger, J. D., Hoelscher, T. J., Marsh, G. R., Lipper, S., & Ionescu-Pioggia, M. (1992). A cognitive-behavioral therapy for sleep-maintenance insomnia in older adults. Psychology and Aging, 7, 282–289. Edinger, J. D., Fins, A. I., Glenn, D. M., Sullivan, R. J., Bastian, L. A., Marsh, G. R., Dailey, D., Hope, T. V., Young, M., Shaw, E., & Vasilas, D. (2000). Insomnia in the eye of the beholder: are there clinical markers of objective sleep disturbances among adults with and without chronic insomnia complaints? Journal of Consulting and Clinical Psychology, 68, 586–593. Edinger, J. D., Wohlgemuth, W. K., Radtke, R. A., Marsh, G. R., & Quillian, R. E. (2001). Treating sleep-maintenance insomnia without sleeping pills: a randomized controlled trial to test cognitive-behavioral therapy. Journal of the American Medical Association, 285, 1856–1864. Espie, C. A. (1991). The psychological treatment of insomnia. Chichester, England: Wiley. Espie, C. A., Inglis, S. J., Harvey, L., & Tessier, S. (2000). Insomniac’s attributions: psychometric properties of the Dysfunctional Beliefs and Attitudes about Sleep scale and the Sleep Disturbance Questionnaire. Journal of Psychosomatic Research, 48, 141–148. Espie, C. A., Inglis, S. J., Tessier, S., & Harvey, L. (2001). The clinical effectiveness of cognitive behaviour therapy for chronic insomnia: implementation and evaluation of a sleep clinic in general medical practice. Behaviour Research and Therapy, 39, 45–60. Fichten, C. S., Libman, E., Creti, L., Amsel, R., Tagalakis, V., & Brender, W. (1998). Thoughts during awake times in older good and poor sleepers: the Self-Statement Test 60+. Cognitive Therapy and Research, 22, 1–20. Ford, D. E., & Kamerow, D. B. (1989). Epidemiologic study of sleep disturbances and psychiatric disorders: an opportunity for prevention? Journal of American Medical Association, 262, 1479–1484. Harvey, L., Inglis, S. J., & Espie, C. A. (2001). Insomniacs’ reported use of CBT components and relationship to longterm clinical outcome. Behaviour Research and Therapy. Hauri, P. J. (1991). Case studies in insomnia. New York: Plenum Press. Lichstein, K. L., & Morin, C. M. (Eds.), (2000). Treatment of late-life insomnia. Thousand Oaks, CA: Sage Publications. Lichstein, K. L., & Riedel, B. W. (1994). Behavioral assessment and treatment of insomnia: a review with an emphasis on clinical application. Behavior Therapy, 25, 659–688. Mellinger, G. D., Balter, M. B., & Uhlenhuth, E. H. (1985). Insomnia and its treatment: prevalence and correlates. Archives of General Psychiatry, 42, 225–232. Morin, C. M. (1993). Insomnia: psychological assessment and management. New York: Guilford Press. Morin, C. M. (1994). Dysfunctional beliefs and attitudes about sleep: preliminary scale development and description. The Behavior Therapist, 163–164.
752
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
Morin, C. M., Kowatch, R. A., Barry, T., & Walton, E. (1993a). Cognitive behavior therapy for late-life insomnia. Journal of Consulting and Clinical Psychology, 61, 137–146. Morin, C. M., Stone, J., Trinkle, D., Mercer, J., & Remsberg, S. (1993b). Dysfunctional beliefs and attitudes about sleep among older adults with and without insomnia complaints. Psychology and Aging, 8, 463–467. Morin, C. M., Colecchi, C. A., Stone, J., Sood, R. K., & Brink, D. (1999a). Behavioral and pharmacological therapies for late-life insomnia: a randomized controlled trial. Journal of the American Medical Association, 781, 991–999. Morin, C. M., Hauri, P. J., Espie, C., Spielman, A., Buysse, D. J., & Bootzin, R. R. (1999b). Nonpharmacologic treatment of chronic insomnia. Sleep, 22, 1134–1156. Pallesen, S., Nordhus, I. H., & Kvale, G. (1998). Nonpharmacological interventions for insomnia in older adults: a meta-analysis of treatment efficacy. Psychotherapy, 35, 472–482. Pat-Horenczyk, R. (1998). Changes in attitudes toward insomnia following cognitive intervention as part of a withdrawal treatment from hypnotic. Behavioural and Cognitive Psychotherapy, 26, 345–357. Rechtschaffen, A., & Kales, A. (1968). A manual of standardized terminology, techniques, and scoring system for sleep stages of human subjects. Los Angeles, CA: Brain Information Service/Brain Research Institute, UCLA. Riedel, B. W., Lichstein, K. L., & Dwyer, W. O. (1995). Sleep compression and sleep education for older insomniacs: self-help versus therapist guidance. Psychology and Aging, 10, 54–63. Sanavio, E., Vidotto, G., Bettinardi, O., Roletto, T., & Zorzi, M. (1990). Behavior therapy for DIMS: comparison of three treatment procedures with follow-up. Behavioral Psychotherapy, 18, 151–167. Schramm, E., Hohagen, F., Backhaus, J., Lis, S., & Berger, M. (1995). Effectiveness of a multicomponent group treatment for insomnia. Behavioural and Cognitive Psychotherapy, 23, 109–127. Simon, G., & VonKorff, M. (1997). Prevalence, burden, and treatment of insomnia in primary care. American Journal of Psychiatry, 154, 1417–1423. Spielman, A. J., Saskin, P., & Thorpy, M. J. (1987). Treatment of chronic insomnia by restriction of time in bed. Sleep, 10, 45–56. Van Egeren, L., Haynes, S. N., Franzen, M., & Hamilton, J. (1983). Presleep cognitions and attributions in sleep-onset insomnia. Journal of Behavioral Medicine, 6, 217–232. Verbeek, I., Schreuder, K., & DeClerk, G. (1999). Evaluation of short-term nonpharmacological treatment of insomnia in a clinical setting. Journal of Psychosomatic Research, 47, 369–383.
Are changes in beliefs and attitudes about sleep related to sleep improvements in the treatment of insomnia? C.M. Morin *, F. Blais, J. Savard E´cole de Psychologie, Universite´ Laval, Sainte-Foy, Quebec, Canada G1K 7P4 Accepted 30 March 2001
Abstract Dysfunctional beliefs and attitudes about sleep are presumed to play an important mediating role in perpetuating insomnia. The present study evaluated the impact of cognitive-behavioral and pharmacological treatments for insomnia on sleep-related beliefs and attitudes and the relationship between those changes and sleep improvements. The participants were older adults with chronic and primary insomnia. They received cognitive-behavior therapy (CBT), pharmacotherapy (PCT), combined CBT+PCT (COMB), or a medication placebo (PLA). In addition to daily sleep diaries and sleep laboratory measures, the participants completed the dysfunctional beliefs and attitudes about sleep scale (DBAS) at baseline and posttreatment, and at 3-, 12- and 24-month follow-up assessments. The results showed that CBT and COMB treatments produced greater improvements of beliefs and attitudes about sleep at posttreatment than PCT and PLA. Reductions of DBAS scores were significantly correlated with improvements of sleep efficiency as measured by daily sleep diaries and by polysomnography. In addition, more adaptive beliefs and attitudes about sleep at posttreatment were associated with better maintenance of sleep improvements at follow-ups. These findings highlight the importance of targeting sleep-related beliefs and attitudes in the treatment of insomnia. 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction Insomnia is a prevalent condition affecting between 9 and 15% of the adult population on a chronic basis and over 25% of people aged over 60 years old (Ford & Kamerow, 1989; Mellinger, Balter, & Uhlenhuth, 1985). Sleep disturbances are frequently associated with fatigue, problems of attention and concentration, and with mood disturbances. Chronic insomnia is also a risk factor
* Corresponding author. E-mail address: [email protected] (C.M. Morin).
0005-7967/02/$ - see front matter 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 0 5 - 7 9 6 7 ( 0 1 ) 0 0 0 5 5 - 9
742
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
for major depression and it carries important health-care costs (Ford & Kamerow, 1989; Simon & VonKorff, 1997). Insomnia can be secondary to medical, psychiatric, and other sleep disorders (American Sleep Disorders Association, 1990). For primary insomnia, the etiology is often multifactorial involving such factors as hyperarousal, maladaptive sleep habits (e.g. irregular schedule, excessive time spent in bed), and dysfunctional beliefs and attitudes about sleep (e.g. worry over sleep loss and its consequences) (Espie, 1991; Lichstein & Morin, 2000; Morin, 1993). Of those contributing actors, the role of cognitive variables (e.g. intrusive thoughts, beliefs, attitudes, expectations) has received little research attention, despite increasing recognition that they play a critical role in the development and maintenance of insomnia. An example of the influence of cognitions on insomnia is the typical reaction of excessive worrying over sleep loss and its potential consequences. This reaction can increase physiological and emotional arousal (i.e. performance anxiety) and, in a vicious cycle, perpetuate the underlying sleep disturbance. The belief concerning the absolute need for 8 h of sleep each night to feel refreshed and function well during the day is another form of faulty belief. As sleep needs vary across individuals, this belief can lead to emotional distress and exacerbate insomnia when this requirement is not met. Most studies investigating the role of cognitive processes on insomnia have focused on the impact of pre-sleep or nocturnal cognitions on sleep. These studies have shown that individuals with insomnia report more negative thoughts during nocturnal awakenings than good sleepers (Watts, Coyle, & East, 1994) and that negative presleep cognitions (e.g. thoughts about not falling asleep) are associated with increased subjective (but not objective) sleep difficulties (Van Egeren, Haynes, Franzen, & Hamilton, 1983). Among older adults, highly distressed poor sleepers endorse more negative thoughts about insomnia and other issues (e.g. health, work, and family) during nocturnal awakenings compared to good sleepers and low distressed poor sleepers (Fichten et al., 1998). Although these findings support the view that negative cognitions are related to sleep disturbance, it is not clear whether these situational and automatic thoughts cause insomnia or simply represent an epiphenomenon of poor sleep. Underlying beliefs and attitudes about sleep (i.e. schemas), which are more deeply ingrained than automatic thoughts, are likely to play an even more important role in the development of insomnia. In a study of older adults, participants with chronic insomnia endorsed stronger dysfunctional beliefs and attitudes about sleep than selfdefined good sleepers (Morin, Stone, Trinkle, Mercer, & Remsberg, 1993b). Specifically, poor sleepers held stronger beliefs about the potential consequences of insomnia, and worried more about losing control and about the unpredictability of sleep. Edinger et al. (2000) reported similar findings and also found that such faulty beliefs and attitudes about sleep were more closely related to subjective than objective insomnia. Together, these results suggest that sleep-related maladaptive attitudes and beliefs are possibly involved in chronic insomnia and that they represent an important target for treatment. It is only recently that psychological interventions for insomnia have incorporated cognitive restructuring strategies to directly target faulty beliefs and attitudes about sleep (Edinger, Hoelscher, Marsh, Lipper, & Ionescu-Pioggia, 1992; Edinger, Wohlgemuth, Radtke, Marsh, & Quillian, 2001; Espie, Inglis, Tessier, & Harvey, 2000; Morin, Colecchi, Stone, Sood, & Brink, 1999a; Morin, Kowatch, Barry, & Walton, 1993a; Morin, Stone, McDonald & Jones, 1994; Riedel, Lichstein, & Dwyer, 1995; Sanavio, Vidotto, Bettinardi, Rolleto, & Zorzi, 1990; Verbeek, Schreuder, & Declerk, 1999). Results of these studies suggest that adding cognitive restructuring
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
743
to behavioral strategies (e.g. stimulus control, sleep restriction) may produce slightly greater benefits than behavioral treatment alone (Lichstein & Riedel, 1994; Morin, Hauri et al., 1999a). However, because cognitive therapy has always been evaluated in the context of multifaceted treatments, the specific contribution of altering maladaptive sleep cognitions in the treatment of insomnia remains unknown. In a study examining the role of attitudinal change in the treatment of chronic users of hypnotics, Pat-Horenczyk (1998) implemented a medication withdrawal program, followed by CBT focusing on the correction of faulty sleep beliefs and attitudes. Although improvements were noted during withdrawal, it is only with the addition of cognitive therapy that significant changes occurred in attitudes and beliefs. Also, the magnitude of those changes was correlated with subjective sleep measures. While these findings are interesting, they were based on a small sample (n=19) and it is unclear whether the attitudinal changes initiated during withdrawal would have continued without the addition of CBT. It is also intriguing that only subjective sleep measures were associated with the attitudinal change. Together, those findings suggest that beliefs and attitudes about sleep are involved in the etiology of primary insomnia and that such cognitive variables may represent an important treatment target. Accordingly, the objectives of this study were to compare changes in sleep-related beliefs and attitudes obtained with cognitive-behavioral and pharmacological therapies for insomnia, and to evaluate the relationship between those changes and sleep improvements.
2. Method 2.1. Participants Data collection was part of a larger study comparing the efficacy of CBT and PCT for latelife insomnia (Morin et al., 1999a). Participants were older adults, recruited through newspaper advertisements, who met DSM-IV criteria for primary insomnia (APA, 1994). Minimal entry criteria were that participants reported chronic (greater than a 6-month duration) difficulties initiating (sleep-onset latency greater than 30 min per night) and/or maintaining sleep (wake after sleep onset greater than 30 min per night) for more than three nights per week. Participants were excluded if there was evidence of a progressive medical illness (e.g. congestive heart failure, cancer), a major psychiatric disorder (e.g. major depression), or another sleep disorder (e.g. sleep apnea or periodic limb movements during sleep). All participants were free from psychotropic drugs for at least two weeks prior to entering the study. Of the 78 patients enrolled in the study, 72 (47 women and 25 men) completed treatment. Their mean age was 64.7 years (SD=6.9 years). Sixty-nine percent of the participants were married, 18.1% widowed, and 12.5% were divorced. Also, 45.8% were retired, 37.5% were employed, 15.3% were homemakers, and 1.4% were unemployed. The mean education level was 14.4 years (SD=2.5). The average duration of insomnia was 17.0 years (SD=17.2 years). Sixty four percent of the participants had mixed difficulties initiating and maintaining sleep, while 25% had sleepmaintenance and 6.9% sleep-onset difficulties only.
744
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
2.2. Procedure After an initial telephone screening, all prospective subjects underwent a multistep evaluation involving a clinical interview with a board-certified sleep specialist, a physical examination, a structured clinical interview for DSM-III-R diagnoses (SCID), and several additional self-report measures including the dysfunctional beliefs and attitudes about sleep (described later). In addition, all participants completed daily sleep diaries for the duration of the study and three consecutive nights of polysomnography PSG at baseline and posttreatment assessment. Subjects were randomly assigned to one of the following conditions: (a) cognitive-behavior therapy (CBT; n=18); (b) pharmacotherapy (PCT; temazepam; n=20); (c) combined CBT and PCT (COMB; n=20); or (d) a medication placebo (PLA; n=20). All treatments lasted eight weeks and were conducted on an outpatient basis. More detailed information on the assessment and treatment procedures of this study is available in a previously published report (Morin et al., 1999a). 2.3. Measures 2.3.1. Sleep diary All participants completed a sleep diary every morning at breakfast time for a 2-week baseline period, during the 8-week treatment period, and for an additional 2-week period at 3-, 12- and 24-month follow-up assessments. Several measures were derived from the diary, including sleeponset latency, wake after sleep onset, total wake time, total sleep time, and sleep efficiency. For the purpose of the present study, only the sleep efficiency variable was used. This variable was selected because it represents the best composite score reflecting the severity of insomnia. Its value is computed as the ratio of total sleep time to time spent in bed and multiplied by 100. Sleep variables were coded each night and a weekly mean was computed. Although sleep diary data do not reflect absolute values obtained with PSG, they are often the most practical modality of obtaining repeated measurements of sleep patterns and still represent a reliable index of insomnia (Coates et al., 1982; Lacks & Morin, 1992). Sleep diary data were available for 72 participants at posttreatment, and for 59, 51, and 49 participants at the 3-, 12- and 24-month follow-up assessments, respectively. 2.3.2. Polysomnography The participants underwent three consecutive nights of sleep laboratory evaluation both before and after treatment. Bedtime and arising time in the sleep laboratory were kept as close as possible (i.e. within 30 min) to the subject’s habitual sleep schedule. A standard PSG montage that included electroencephalographic (EEG), electromyographic (EMG) and electrooculographic (EOG) monitoring was used. Sleep stages were scored by an experienced technician according to standardized criteria (Rechtschaffen & Kales, 1968). As was the case for the sleep diaries, several dependent variables were derived from sleep laboratory assessment, but only the sleep efficiency (total sleep time divided by total time spent in bed and multiplied by 100) variable was used in the present study. There were 72 patients completing PSG evaluation at baseline and 70 subjects at posttreatment.
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
745
2.4. Dysfunctional beliefs and attitudes about sleep scale The dysfunctional beliefs and attitudes about sleep scale (DBAS) (Morin 1993, 1994) is a 30item scale tapping various beliefs, attitudes, expectations and attributions about sleep and insomnia. The nature of these cognitions cluster around five conceptually derived themes: (a) misconceptions about the causes of insomnia (e.g. “I believe insomnia is essentially the result of aging and there isn’t much that can be done about this problem”); (b) diminished perception of control and predictability of sleep (e.g. “When I sleep poorly on one night, I know it will disturb my sleep schedule for the whole week”); (c) unrealistic sleep expectations (e.g. “I must get 8 h of sleep to feel refreshed and function well during the day”); (d) misattribution or amplification of the consequences of insomnia (e.g. “I am concerned that chronic insomnia may have serious consequences on my physical health”); and (e) faulty beliefs about sleep-promoting practices (e.g. “When I have trouble sleeping, I should stay in bed and try harder”). For each statement, the person rates his or her level of agreement/disagreement on a 100-mm visual analog scale ranging from 0 (strongly disagree) to 100 (strongly agree). Except for item 23 for which the score is reversed, a higher score indicates more dysfunctional beliefs and attitudes about sleep. The total score is based on the average score of all items. Likewise, subscale scores are computed by adding the sum of scores for the items and dividing by the number of items making up each subscale. Initial psychometric data indicate that the DBAS has an excellent internal consistency (Cronbach’s Alpha=0.80) and an average item-total correlation of 0.37 (Morin, 1994; Morin et al., 1993b). Additional data have confirmed the acceptable psychometric properties of this scale (Espie et al., 2000). In the present study, completed DBAS were available for 66 participants at baseline, 69 at posttreatment, and 56, 55, and 53 participants at the 3-, 12- and 24-month follow-up assessments, respectively. 2.5. Treatment 2.5.1. Cognitive behavior therapy CBT consisted of a multifaceted intervention involving behavioral, cognitive and educational components. The behavioral component involves a combination of sleep restriction (Spielman, Saskin, and Thorpy, 1987) and stimulus control procedures (Bootzin, Epstein, & Wood, 1991). The cognitive therapy component was modeled after the work of Beck, Ellis, and Meichenbaum and was aimed at altering dysfunctional beliefs and attitudes about sleep. The same five themes as those covered in the DBAS scale were addressed during cognitive therapy sessions: misconceptions about the causes of insomnia, misattributions or amplification of the consequences of insomnia, diminished perception of control and predictability of sleep, unrealistic sleep expectations, and faulty beliefs about sleep-promoting practices. Cognitive therapy sessions followed a threestep process: (a) identification of patient-specific dysfunctional thoughts (using both the DBAS scale and self-monitoring); (b) confrontation and challenging of those thoughts; and (c) implementation of methods for replacing these thoughts with more rationale substitutes. Three therapy sessions were devoted to cognitive therapy and two others were used to reinforce the main principles and procedures covered by this therapy component. Finally, the educational component covered basic facts about changes in sleep occurring with aging and about sleep hygiene principles regarding the effects of caffeine, alcohol, exercise, and diet on sleep (Hauri, 1991).
746
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
2.5.2. Pharmacotherapy PCT consisted of 7.5–30 mg temazepam recommended for 2–3 nights per week but made available for nightly use if the patients so desired. The dosage was adjusted weekly, based on treatment response and side effects. Both active and placebo drugs were provided in identical gelatin capsules, filled with an equivalent amount of powder. Participants in all three conditions involving medications (PCT, PLA, and COMB) were seen by a psychiatry resident on a weekly basis for 20–30 min sessions devoted to reviewing treatment progress, side effects, and insomnia symptoms. Urine drug screens for benzodiazepines were conducted at baseline to ensure that subjects were free from these agents and at late treatment to ensure compliance with prescribed medication regimen. At the end of the 8-week treatment phase, all participants receiving medication began a 1–2 week tapering schedule. 2.5.3. Combined (COMB) CBT plus PCT Participants in this condition received both CBT and active medication. They were seen weekly by both a clinical psychologist for CBT and a physician for the medication management. 2.5.4. Medication-placebo Participants in this condition received medication which consisted of an inert lactose solution. However, it was provided in an identical manner as those participants in the active medication condition. 3. Results In order to understand better the current results, this section begins with a brief summary of the main sleep findings from the original study (Morin et al., 1999a). All three active treatments produced significantly greater increases in sleep efficiency from baseline to posttreatment than the placebo condition. Baseline to posttreatment changes on the sleep diary measure were as follows: CBT, 68–85%; PCT, 72–83%; COMB, 64–85% and PLA, 69–74%. The average effect size pooled across the three treatments was 1.16. Changes on the polysomnographic measure were as follows: CBT, 78–86%; PCT, 77–85%; COMB, 76–87% and PLA, 79–80%. The average effect size pooled across the three treatments was 0.80. Although there was a trend for the combined condition to produce greater improvement rates, there was no significant difference among the three active treatment conditions. There were respectively 78% (CBT), 56% (PCT), 75% (COMB), and 14% (PLA) of the participants who reached a clinically meaningful outcome on sleep patterns at posttreatment. Also, follow-up data showed that CBT sustained its sleep improvements very well over time, the drug therapy alone condition returned near baseline, and the combined condition yielded an intermediate outcome falling between those two single therapy conditions. 3.1. Attitudinal changes across conditions Because of some missing data on the DBAS and attrition over time, the sample size varied from 69 subjects at posttreatment to 46 subjects at the 24-month follow-up. Missing data were
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
747
not replaced nor estimated for those analyses. Group means were computed for total DBAS scores at each assessment period. These data are presented in Fig. 1. An analysis of variance (ANOVA) revealed no significant difference between groups on pretreatment DBAS total score. An ANOVA with repeated measures on baseline and posttreatment scores yielded significant Time (F(1,59)=31.61, p⬍0.0001), Group (F(3,59)=6.31, p⬍0.001), and Group×Time interaction effects (F(3,59)=21.32, p⬍0.0001). Post hoc comparisons, using the Tukey test, showed that the CBT and COMB groups had significantly lower DBAS scores at posttreatment compared to PCT and PLA conditions (ps⬍0.05). A visual inspection of Fig. 1 shows that DBAS scores decreased sharply from baseline to posttreatment among patients receiving CBT, alone or in combination with medications, whereas those scores increased slightly in the two drug only conditions. No significant difference was obtained between CBT and COMB or between PCT and PLA groups. Average effect sizes, based on change scores from baseline to posttreatment, were 1.84 for CBT, 1.89 for COMB, and 0.31 for PCT. Posttreatment DBAS scores obtained for the CBT and COMB conditions fell below the average score (i.e. 30) obtained in another study by self-defined good sleepers of the same age (Morin et al., 1993b). Analyses of variances computed for each of the five separate DBAS themes (expectations, causal attributions, etc.) generally yielded the same results as those obtained for the total DBAS score and are not reported here. A repeated measures ANOVA was conducted on posttreatment, 3-, and 12-month follow-up DBAS scores. The 24-month follow-up assessment was not included because of too many missing data. A significant effect for Group, (F(3,42)=8.72, p⬍0.001), was obtained but there was no significant effect for Time or Group×Time interaction. All three active treatment conditions (CBT, PCT and COMB) endorsed less dysfunctional beliefs and attitudes about sleep than the PLA group (ps⬍0.05), but there was no significant difference among them. 3.2. Relationship between DBAS and sleep efficiency The relationship between attitudinal and sleep pattern changes was examined by correlating pre to posttreatment change scores on the DBAS with change scores on sleep efficiency as meas-
Fig. 1. Changes in beliefs and attitudes over time (total score).
748
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
ured by daily sleep diaries and PSG. These analyses were computed for all participants in the four conditions combined. Pearson correlations (see Table 1) indicate that reduced DBAS scores were significantly correlated with improved sleep efficiency as measured by daily sleep diaries (r(63)=-0.28, p⬍0.05); the correlation was nearly significant for the sleep efficiency variable based on PSG (r(62)=-0.24, p⬍0.065). Correlational analyses were also computed between change scores on the five separate themes of the DBAS and change scores on sleep efficiency from the pre to posttreatment period. The pattern of results was very similar for each of those themes (see Table 1). For example, improvements (i.e. reduction of score) of beliefs related to theme 1 (causal attributions of insomnia) were significantly associated with improved sleep efficiency as measured by daily sleep diaries (r(63)=0.28, p⬍0.05) but not with PSG. Changes on theme 2 (control and predictability of sleep) were significantly correlated with improved sleep efficiency as measured by daily sleep diaries (r(63)=0.27, p⬍0.05) and by PSG (r(62)=-0.40, p⬍0.005). Changes on theme 3 (sleep requirement expectations) were significantly associated with improved sleep efficiency as measured by daily sleep diaries only (r(63)=-0.27, p⬍0.05). Changes on theme 4 (consequences of insomnia) were correlated with improved sleep efficiency based on sleep diaries (r(63)=-0.28, p⬍0.01) and PSG (r(62)=-0.25, p⬍0.053. Changes on theme 5 (sleep-promoting practices) were not significantly correlated with sleep efficiency. 3.3. Relationship between DBAS at posttreatment and maintenance of therapeutic gains As shown in Table 2, posttreatment DBAS scores were significantly correlated with sleep efficiency (based on sleep diary assessment) at each of the three follow-up assessments: 3-month (r(56)=-0.22, p⬍0.05), 12-month (r(52)=-0.36, p⬍0.01), and 24-month (r(46)=-0.51, p⬍0.0005). Analyses for separate themes showed that posttreatment scores for theme 1 (causal attributions of insomnia) were significantly correlated with sleep efficiency at 12-month (r(52)=-0.37, p⬍0.01) and 24-month follow-ups (r(46)=-0.46, p⬍0.005)). DBAS scores for theme 2 (control and predictability of sleep) and 4 (consequences of insomnia) were significantly correlated with sleep efficiency at 12- and 24-month follow-ups (-0.48⬎rs⬎-0.28, ps⬍0.05). Posttreatment scores for theme 3 (sleep expectations) were significantly correlated with sleep efficiency at 24-month folTable 1 Correlations between change scores from baseline to posttreatment on DBAS and sleep efficiency measures (note: * p⬍0.05 **p⬍0.01; ***p⬍0.005) Sleep efficiency
DBAS total score Causal attributions Control/predictability Sleep expectations Consequences Sleep-promoting practices
Polysomnography (N=63)
Sleep diary (N=62)
⫺0.24 ⫺0.09 ⫺0.40*** ⫺0.19 ⫺0.25* ⫺0.03
⫺0.28* ⫺0.28* ⫺0.27* ⫺0.27* ⫺0.29** ⫺0.19
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
749
Table 2 Correlations between DBAS posttreatment scores and sleep efficiency (diary) at follow-ups (note: *p ⬍0.05, ** p⬍0.01, *** p ⬍0.005) Sleep efficiency
DBAS postreatment scores Causal attributions Control/predictability Sleep expectations Consequences Sleep-promoting practices
3-month follow-up (N=56)
12-month follow-up (N=52)
24-month follow-up (N=46)
⫺0.28*
⫺0.36**
⫺0.51***
⫺0.20 ⫺0.24 ⫺0.19 ⫺0.24
⫺0.37** ⫺0.40*** ⫺0.14 ⫺0.28*
⫺0.46*** ⫺0.48*** ⫺0.35** ⫺0.36**
⫺0.24
⫺0.32*
⫺0.45***
low-ups (r(46)=-0.35, whereas scores for theme 5 were significantly correlated with sleep efficiency at 12-month, r(52)=-0.32, p⬍0.05, and 24-month follow-up, r(46)=-0.45, p⬍0.005). Thus, a lower posttreatment score on sleep-related cognitions was related to a higher sleep efficiency at follow-ups. 4. Discussion The results indicate that dysfunctional beliefs and attitudes endorsed by older adults with chronic insomnia are improved with CBT, implemented singly or in combination with drug therapy. These changes are specific to CBT as no such improvements were obtained among patients treated with drug therapy alone. The magnitude of changes on sleep-related cognitions was associated with the degree of sleep improvements at posttreatment. In addition, more adaptive beliefs and attitudes about sleep (i.e. lower DBAS scores) at posttreatment were associated with better maintenance of sleep improvements at follow-ups. These findings raise an important question as to whether changes in beliefs and attitudes are produced by CBT per se, or simply represent a by-product of sleep improvements, regardless of the type of treatment that produced those changes. As all three active treatments produced equivalent increases of sleep efficiency from baseline to posttreatment (Morin et al., 1999a,b), it can be argued that improved sleep pattern in itself was not sufficient to produce changes in beliefs and attitudes. Consistent with previous findings (Espie, Inglis, Tessier, & Harvey, 2001; Harvey, Inglis, & Espie, 2001; Pat-Horenczyk, 1998), it appears that psychological treatment (CBT) is a necessary component to reduce dysfunctional sleep-related attitudes and beliefs. A related question is whether cognitive therapy is an essential ingredient to alter sleep-related cognitions or whether similar changes would have been obtained with behavioral treatment alone. Because cognitive therapy was combined with behavioral procedures in the two conditions that yielded significant changes in beliefs and attitudes, it is not possible to answer this question. Another important issue is whether changes in beliefs and attitudes are essential to effectively treat insomnia. The present results, although correlational, suggest that those changes are instru-
750
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
mental to successful treatment outcome. Significant correlations were found between the magnitude of changes on sleep-related beliefs and attitudes and the degree of sleep improvements at posttreatment. Moreover, the strength of the relationship between cognitive changes and sleep improvements increased over time. Although this finding has to be interpreted cautiously because of the smaller number of participants available at 24-month follow-up, this trend suggests that cognitive changes occurring during the course of treatment are important for the long-term maintenance of therapeutic gains. Participants in all three active treatments, including those receiving PCT alone, endorsed fewer dysfunctional sleep cognitions at follow-ups compared to those receiving placebo. These results indicate that changes in beliefs and attitudes achieved at posttreatment with CBT, either alone or in combination with medication, were well sustained over time. While these data also suggest that PCT alone improved sleep cognitions, participants in that condition had consistently higher DBAS scores at each follow-up assessment compared to the two treatments incorporating CBT (see Fig. 1). In addition, there was no significant change over time in DBAS scores within the drug therapy condition alone. Because of attrition at follow-ups, it is possible that the absence of difference among the three active treatments was due to low statistical power. An interesting finding is that participants treated with CBT, alone or in combination, obtained DBAS scores at posttreatment that were comparable to those obtained by older adults without insomnia complaints in two other studies (Edinger et al., 2000; Morin et al., 1993b). Although one could argue that a lower DBAS score is to be expected following cognitive therapy targeting specific beliefs included on that scale, it is still interesting to note that posttreatment status of CBT-treated subjects reached a level previously observed in non-complaining individuals. Although additional normative data are needed to further validate this scale, these preliminary findings provide evidence highlighting the clinical significance of changes in beliefs and attitudes about sleep. Consistent with previous findings (Edinger et al., 2000; Pat-Horenczyk, 1998; Van Egeren et al., 1983), changes in sleep cognitions were more strongly associated with subjective (sleep diaries) than with objective (PSG) sleep improvements. Although it is common to obtain higher correlations between variables from a similar assessment modality (i.e. subjective vs. subjective) than for variables measured with different modalities (e.g. subjective vs. physiological), another possible explanation for this finding is that untreated insomniacs tend to overestimate their sleep problem and to underestimate their sleep duration (Coates et al., 1982; Morin, 1993). As cognitive therapy addressed such misperceptions, it is plausible that patients treated with CBT learned to perceive their sleep more accurately over time which, in turn, contributed to a stronger relationship between cognitive changes and subjectively assessed sleep improvement. Previous studies of psychological interventions for insomnia have used predominantly behavioral approaches (e.g. stimulus control, sleep restriction) that focused on modifying maladaptive sleep habits. These procedures have been shown to be effective with both younger and older adults (Lichstein & Riedel, 1994; Lichstein & Morin, 2000; Morin et al., 1994; Pallesen, Nordhus, & Kvale, 1998; Schramm, Hohagen, Backhaus, Lis, & Berger, 1995), although not all patients have been responsive nor compliant with such behavioral procedures. The present findings highlight the importance of targeting sleep-related cognitions in the management of insomnia. The addition of cognitive restructuring procedures that target perpetuating factors (i.e. fear of the consequences of insomnia) or that prevent compliance with behavioral instructions could enhance
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
751
treatment efficacy. Future research will need to isolate the cognitive therapy component in order to determine its unique contribution to treatment outcome, both on cognitive and sleep variables. Acknowledgements Research supported by National Institute of Mental Health grants #MH47020 and #MH55469. References American Psychiatric Association (APA). (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: APA. American Sleep Disorders Association. (1990). International classification of sleep disorders (ICSD): diagnostic and coding manual. Rochester, MN: APA. Bootzin, R. R., Epstein, D., & Wood, J. M. (1991). Stimulus control instructions. In P. Hauri (Ed.), Case studies in insomnia (pp. 19–28). New York: Plenum Press. Coates, T. J., Killen, J. D., George, J., Silverman, S., Marchini, E., & Thoresen, C. E. (1982). Estimating sleep parameters: a multi-trait–multi-method analysis. Journal of Consulting and Clinical Psychology, 50, 345–352. Edinger, J. D., Hoelscher, T. J., Marsh, G. R., Lipper, S., & Ionescu-Pioggia, M. (1992). A cognitive-behavioral therapy for sleep-maintenance insomnia in older adults. Psychology and Aging, 7, 282–289. Edinger, J. D., Fins, A. I., Glenn, D. M., Sullivan, R. J., Bastian, L. A., Marsh, G. R., Dailey, D., Hope, T. V., Young, M., Shaw, E., & Vasilas, D. (2000). Insomnia in the eye of the beholder: are there clinical markers of objective sleep disturbances among adults with and without chronic insomnia complaints? Journal of Consulting and Clinical Psychology, 68, 586–593. Edinger, J. D., Wohlgemuth, W. K., Radtke, R. A., Marsh, G. R., & Quillian, R. E. (2001). Treating sleep-maintenance insomnia without sleeping pills: a randomized controlled trial to test cognitive-behavioral therapy. Journal of the American Medical Association, 285, 1856–1864. Espie, C. A. (1991). The psychological treatment of insomnia. Chichester, England: Wiley. Espie, C. A., Inglis, S. J., Harvey, L., & Tessier, S. (2000). Insomniac’s attributions: psychometric properties of the Dysfunctional Beliefs and Attitudes about Sleep scale and the Sleep Disturbance Questionnaire. Journal of Psychosomatic Research, 48, 141–148. Espie, C. A., Inglis, S. J., Tessier, S., & Harvey, L. (2001). The clinical effectiveness of cognitive behaviour therapy for chronic insomnia: implementation and evaluation of a sleep clinic in general medical practice. Behaviour Research and Therapy, 39, 45–60. Fichten, C. S., Libman, E., Creti, L., Amsel, R., Tagalakis, V., & Brender, W. (1998). Thoughts during awake times in older good and poor sleepers: the Self-Statement Test 60+. Cognitive Therapy and Research, 22, 1–20. Ford, D. E., & Kamerow, D. B. (1989). Epidemiologic study of sleep disturbances and psychiatric disorders: an opportunity for prevention? Journal of American Medical Association, 262, 1479–1484. Harvey, L., Inglis, S. J., & Espie, C. A. (2001). Insomniacs’ reported use of CBT components and relationship to longterm clinical outcome. Behaviour Research and Therapy. Hauri, P. J. (1991). Case studies in insomnia. New York: Plenum Press. Lichstein, K. L., & Morin, C. M. (Eds.), (2000). Treatment of late-life insomnia. Thousand Oaks, CA: Sage Publications. Lichstein, K. L., & Riedel, B. W. (1994). Behavioral assessment and treatment of insomnia: a review with an emphasis on clinical application. Behavior Therapy, 25, 659–688. Mellinger, G. D., Balter, M. B., & Uhlenhuth, E. H. (1985). Insomnia and its treatment: prevalence and correlates. Archives of General Psychiatry, 42, 225–232. Morin, C. M. (1993). Insomnia: psychological assessment and management. New York: Guilford Press. Morin, C. M. (1994). Dysfunctional beliefs and attitudes about sleep: preliminary scale development and description. The Behavior Therapist, 163–164.
752
C.M. Morin et al. / Behaviour Research and Therapy 40 (2002) 741–752
Morin, C. M., Kowatch, R. A., Barry, T., & Walton, E. (1993a). Cognitive behavior therapy for late-life insomnia. Journal of Consulting and Clinical Psychology, 61, 137–146. Morin, C. M., Stone, J., Trinkle, D., Mercer, J., & Remsberg, S. (1993b). Dysfunctional beliefs and attitudes about sleep among older adults with and without insomnia complaints. Psychology and Aging, 8, 463–467. Morin, C. M., Colecchi, C. A., Stone, J., Sood, R. K., & Brink, D. (1999a). Behavioral and pharmacological therapies for late-life insomnia: a randomized controlled trial. Journal of the American Medical Association, 781, 991–999. Morin, C. M., Hauri, P. J., Espie, C., Spielman, A., Buysse, D. J., & Bootzin, R. R. (1999b). Nonpharmacologic treatment of chronic insomnia. Sleep, 22, 1134–1156. Pallesen, S., Nordhus, I. H., & Kvale, G. (1998). Nonpharmacological interventions for insomnia in older adults: a meta-analysis of treatment efficacy. Psychotherapy, 35, 472–482. Pat-Horenczyk, R. (1998). Changes in attitudes toward insomnia following cognitive intervention as part of a withdrawal treatment from hypnotic. Behavioural and Cognitive Psychotherapy, 26, 345–357. Rechtschaffen, A., & Kales, A. (1968). A manual of standardized terminology, techniques, and scoring system for sleep stages of human subjects. Los Angeles, CA: Brain Information Service/Brain Research Institute, UCLA. Riedel, B. W., Lichstein, K. L., & Dwyer, W. O. (1995). Sleep compression and sleep education for older insomniacs: self-help versus therapist guidance. Psychology and Aging, 10, 54–63. Sanavio, E., Vidotto, G., Bettinardi, O., Roletto, T., & Zorzi, M. (1990). Behavior therapy for DIMS: comparison of three treatment procedures with follow-up. Behavioral Psychotherapy, 18, 151–167. Schramm, E., Hohagen, F., Backhaus, J., Lis, S., & Berger, M. (1995). Effectiveness of a multicomponent group treatment for insomnia. Behavioural and Cognitive Psychotherapy, 23, 109–127. Simon, G., & VonKorff, M. (1997). Prevalence, burden, and treatment of insomnia in primary care. American Journal of Psychiatry, 154, 1417–1423. Spielman, A. J., Saskin, P., & Thorpy, M. J. (1987). Treatment of chronic insomnia by restriction of time in bed. Sleep, 10, 45–56. Van Egeren, L., Haynes, S. N., Franzen, M., & Hamilton, J. (1983). Presleep cognitions and attributions in sleep-onset insomnia. Journal of Behavioral Medicine, 6, 217–232. Verbeek, I., Schreuder, K., & DeClerk, G. (1999). Evaluation of short-term nonpharmacological treatment of insomnia in a clinical setting. Journal of Psychosomatic Research, 47, 369–383.