行政院國家科學委員會補助國內專家學者出席國際學術會議報告 - 失眠的過度激發病理機制：慢性失眠患者與高、低暫時性失眠脆弱特質者之比較 (III)

101 年 6 月 29 日

報告人姓名楊建銘服務機構

及職稱國立政治大學心理學系教授

時間會議地點

June 9-13, 2012 Boston, MA, USA

本會核定補助文號

臺會綜二字第 1010024839 號會議

名稱

(中文) 睡眠 2012 第二十六屆專業睡眠學會聯合年會

(英文) SLEEP 2012 – 26^th Annual Meeting of the Associated Professional Sleep Societies

發表論文題目

The Association between Treatment Effect and Changes of Cognitive and Behavioral Factors Following CBT-I Treatment in Primary and Comorbid Insomnia

一、參加會議經過

此次參加會議我一共帶領了兩位博士班學生、一位碩士班學生以及一位大學部學生與會，參與論文發表。在正式會議開始前，先在6/9至6/10參與了行為睡眠醫學學會（Society of Behavioral Sleep Medicine）的第一次年會，會中由發展出失眠治療的刺激控制法的Dr. Richard Bootzin給予主題演講，針對行為睡眠醫學取向的失眠治療，回顧了發展的歷史，並指引了未來的方向；之後由失眠認知行為治療著名的學者Dr. Manber及Dr. Perlis進行失眠認知行為治療的工作坊。

正式會議在6/10下午陸續展開，6/11由Dr. Mark Rosekind與Dr. Robert Stickgold分別針對睡眠與交通運輸安全以及睡眠、記憶與夢境進行兩場的大會主題演講，從不同的角度強調睡眠與日常紐約Cornell Medical Center睡眠中心的心理師Dr. Matthew Ebben討論研究合作案，希望能開啟更多國際研究合作。

-1-

The Association between Treatment Effects and Changes of Cognitive and Behavioral Factors Following CBT-I Treatment in Primary and Comorbid Insomnia

Chien-Ming Yang, Ph.D.

Department of Psychology and the Research Center for Mind, Brain, and Learning, National Chengchi University, Taipei, Taiwan

Tsung-Tsair Yang, M.D., Ph.D.

Department of Psychiatry, Cardinal Tien Hospital; School of Medicine, Fu-Jen University, Taipei, Taiwan

Ya-Wen Jan, M.S.

Department of Psychology, National Chengchi University, Taipei, Taiwan

-2-

Abstract

Introduction: Cognitive Behavioral Therapy for Insomnia (CBT-I) consists of multiple components that are designed to change sleep-related psychological and behavioral factors for insomnia. The goals of this study is to exam the association between treatment outcome and changes in beliefs about sleep, sleep-related practices, and pre-sleep arousal levels following CBTI treatment in both primary insomnia (PI) and insomnia comorbid with other sleep disorders and/or psychiatric conditions (CI).

Methods: Participants were 101 insomnia patients, including 21 PI received CBT-I only, 55 PI received CBT-I and hypnotic, and 25 CI received CBT-I and medication.

Dsyfunctional Beliefs and Attitude about Sleep Scale (DBAS), Sleep Hygiene Practice Scale (SHPS), Presleep Arousal Scale (PSAS), and Insomnia Severity Index (ISI) were administered before and after treatment.

Results: ANOVAs showed that all outcome variables (ISI, SOL, WASO, SE) improved significantly after treatment except for TST, which reached a near-significant increase.

For PI received CBT-I only, Pearson correlations showed that improvement in ISI correlated with decrease of both somatic (r=.645, p<.005) and cognitive (r=.659, p=.001) pre-sleep arousal and reduction of arousal-related behaviors (r=.592, p=.005);

for PI received CBTI and hypnotic, reduction of ISI score correlated with reduction of factor II (predictability of sleep) subtests of the DBAS (r=.325, p<.05) and reduction of arousal-related behavior (r=.330, p<.05); for CI group, ISI improvement correlated with change of scores on the factor I (perceived consequences of insomnia, r=.438, p<.05) and factor II (r=.752, p<.001) subtests of the DBAS, both somatic (r=.530, p<.005) and cognitive (r=.772, p<.001) pre-sleep arousal, arousal-related behavior (r=.605, p<.005), and poor sleep environment (r=.450, p<.05).

Conclusion: Treatment effect of CBT-I is primarily associated reduction of arousal and changes in dysfunctional sleep beliefs. The association however differs in patients with primary insomnia and insomnia with comorbid conditions.

Keywords: Insomnia, Cognitive Behavioral Therapy for Insomnia, Treatment Efficacy, Sleep Belief, Sleep Hygiene Practice, Hyperarousal

-3-

Introduction

Insomnia is among the most common health-related complaints. About

one-third of the adult population reported insomnia symptoms, 9–15% showed sleep difficulties as well as daytime consequences, and about 6% showed diagnosable insomnia disorder (Ohayon, 2002). Insomnia can have great impacts on not only nighttime sleep but also daytime functioning (e.g. Godet-Cayré et al., 2006; Roth &

Ancoli-Israel, 1999; Ustinov et al., 2009; see Riedel & Lichstein, 2000, for review), and can generate great costs of the individuals and the society (Chilcott, Shapiro, 1996;

Martin 2004; Stoller, 1994; Walsh & Engelhart, 1999).

It has been well recognized that insomnia may result from some psychological and behavioral factors, such as cognitive and somatic hyperarousal (Bonnet & Arand, 1997; Perlis, Giles, Mendelson, Bootzin, Wyatt, 1997), dysfunctional beliefs and attitudes about sleep (Morin, Stone, Trinkle, Mercer, Remsberg, 1993), association learning (Bootzin, 1972), maladaptive sleep hygiene, and misalignment of

endogenous circadian rhythm with environmental time (Czeisler, Richardson, Coleman, et al. 1981). Hyperarousal model has been supported by the findings that insomnia showed indications of higher activation of the autonomic nervous system (Bonnet & Arand, 1992, 1995, 1996, 1998; Freedman & Sattler, 1982; Vgontzas et al., 1998, 2001). Increased cortical arousal has also been evidenced by higher power of high frequency EEG (Freedman, 1986; Merica H, Blois R, Gaillard J-M., 1998; Perlis, Kehr, Smith, et al., 2001) that was shown to be reduced after treatment with

cognitive behavioral therapy for insomnia (CBTI; Jacobs, Benson, & Friedman, 1993).

Although the hyperarousal may be partially innate, it is usually elevated by

behavioral and psychological factors to maintain increase the insomnia. For example, the association of arousal with bedtime cues may lead to prolonged sleep difficulties (Bootzin, 1972). In addition, dysfunctional cognition about sleep, such as overly emphasizing the impacts of sleeplessness, may result in increased arousal and/or maladaptive sleep habit that interfering sleep (Harvey, 2002; Morin, 1993; Morin, Stone, Trinkle, Mercer, & Remsberg, 1993). Research has found that insomnia patients have more dysfunctional beliefs about sleep and the level of dysfunctional beliefs is associated with the severity of their sleep problem (Morin et al., 1993;

Fichten, Creti, Amsel, Brender, Weinstein, & Libman, 1995; Edinger, Fins, Glenn, et al., 2000). Therefore, techniques that address the psychological and behavioral factors may be effective in the treatment of insomnia.

Another factor that is commonly associated with insomnia is emotional

disturbances. Previous studies have shown that poor emotion regulation may lead to sleep disruption; sleep disturbance can also impair next-day affect control. The

-4-

relationship between insomnia and emotional disturbances is more complicated than just a one-way causal effect (for review, see Baglioni et al., 2009; Schmidt, Harvey, &

Van der Linden, 2011). Also, insomnia is a key symptom of many psychiatric disorders.

Mood disorder is the most common condition comorbid with insomnia associated with insomnia. On the other hand, insomnia has been found to be a risk factor for depression and anxiety disorder (Ford and Kamerow, 1989). A longitudinal follow-up study has also reported that insomnia can occur prior to the onset of 40% of

participants who developed depression and about 20% of participants who

developed anxiety disorders (Ohayon & Roth, 2003). The insomnia can therefore be a consequence as well as a precursor or trigger of emotional disorders. It is therefore of interest to compare the differences between primary insomnia and insomnia comorbid with depression.

With the understanding of the affects of psychological and behavior factors on insomnia, many techniques that aim to reduce arousal, adjust sleep cognition or modify maladaptive sleep practices were developed or employed for the treatment of insomnia. For example, relaxation trainings with progressive muscle relaxation, autogenic training, meditation, and biofeedback were shown to be effective in empirical studies (Borkovec, Grayson, O’Brien, et al., 1979; Nicassio & Bootzin, 1974;

Woolfolk, Carr-Kaffashan, McNulty, 1976; Haynes, Sides, Lockwood, 1977; Hauri, 1981). Also, Stimulus Control was devised to reduced the association between sleep incompatible behaviors with sleep environment (Bootzin, 1972). Sleep Restriction therapy was developed to enhance sleep drive by reducing time in bed (Spielman, Saskin, Thorpy, 1987; Friedman, Bliwise, Yesavage, et al, 1991). In addition, cognitive reconstructuring was used to change sleep-related cognitions and was found to be more effective than some relaxation techniques aiming to reduce cognitive arousal (Morin, 1993; Morin, 1999; Edinger, Hoelscher, Marsh, Lipper, & Ionescue-Pioggia, 1992; Edinger et al., 2001a). The treatment effects of these techniques have been proven with empirical studies and meta-analysis (e.g. Morin et al., 1999; Murtagh, 1995; Morin, Culbert, & Schwartz, 1994).

Because of the multi-factual nature of the etiology of insomnia, the above- mentioned techniques are usually combined in clinical setting, known as CBTI. It has been demonstrated to be an effective treatment in empirical studies (Edinger, 1992;

Morin, 1993; Morin, Stone, Trinkle, Mercer, & Remsberg, 1993). Previous studies have also shown that, comparing to pharmacological treatment, CBTI are at least equally effective (e.g. Smith, Perlis, Park, et al, 2002), and may be more effective with long-term follow-up (Morin, Colecchi, Stone, et al. 1999; Jacobs, Pace-Schott,

Stickgold, et al. 2004). Therefore, CBTI has been recommended to be a first-line treatment for insomnia by the American Academy of Sleep Medicine and the

-5-

consensus by National Institute of Health (NIH, 2005; Schutte-Rodin, Broch, Buysse, Dorsey, Sateia, 2008). CBTI has not only been shown to have sustainable therapeutic effects on primary insomnia, but also to be helpful to patients with comorbid

insomnia and depression (Kuo, Manber, & Loewy, 2001; Lichstein, Wilson, & Johnson, 2000; Taylor, Lichstein, Weinstock, Sanford, & Temple, 2007). Manber and colleagues studied the effect of CBTI combined with antidepressants on 30 insomnia patients comorbid with major depressive disorder, and found that CBTI is helpful to the mood and insomnia symptoms of these patients (Manber et al., 2008).

Although CBTI has been shown to be an effective treatment for insomnia, the changes occurred during the treatment have not been well studied. The effective components of the changes have not been identified. Very few studies look into the changes that are associated with treatment effect after CBTI. It has been reported that change in sleep cognition was associated with sleep improvement (Edinger, et al., 2001b; Morin, Blais & Savard, 2002). An earlier study also reported that the cognitive component (concentration) is more beneficial than somatic component of relaxation when using progressive muscle relaxation in the treatment of insomnia (Borkovec &

Hennings, 1978). These study, however, focused more on specific components of treatment target. A more recent study examined the degree of practice of treatment techniques on treatment effect. The results showed that stimulus control, sleep restriction and cognitive reconstructuring are more effective components. Relaxation training and sleep hygiene, on the other hand, did not show significant predictability for sleep improvement (Harvey, Inglis & Espie, 2002). Therefore, it is possible that the changes of some psychological/behavioral factors are more related to treatment outcome than the others. Therefore, the current study is to investigate association between changes in psychological and behavioral factors after CBTI treatment and treatment outcomes. Also, the importance of different treatment factors may differ for different patient populations. We therefore compared the treatment outcome associated changes in primary insomnia patients and patients with comorbid insomnia and depression.

Methods

Participants were 101 insomnia patients, including 21 PI received CBT-I only, 55 PI received CBT-I and hypnotic, and 25 CI received CBT-I and medication.

Participants were 101 insomnia patients. They were divided into three groups:

21 patients (14 females, aged 38.14±12.54) with primary insomnia who received CBTI only, 55 patients (38 females, aged 42.58±12.02) with primary insomnia who

-6-

received CBTI and continued their hypnotic treatment, and 25 patients (19 females, aged 41.44±9.82) with comorbid insomnia and depression who received CBTI with their psychiatric medication. Inclusive criteria were: 1) aged between 18 to 60 years old, 2) fulfilled primary insomnia diagnosis in DSM-IV, 3) significant distress or impairment in daytime functioning associated with sleep difficulty, 4) no history or current medical condition, other psychiatric condition, substance use that may lead to sleep disturbance, and 5) none shift-worker with regular sleep/wake schedule.

Subjects who passed a clinical interview were enrolled into the study. A

polysomnographic recording was conducted to screen for other sleep disorders. They then came to a university based sleep center six times for a CBTI program.

Dsyfunctional Beliefs and Attitude about Sleep Scale (DBAS; Morin, 1993), Sleep Hygiene Practice Scale (SHPS; Lin, Cheng, Yang, & Hsu, 2007), and Presleep Arousal Scale (PSAS; Nicassio, Mendlowitz, Fussell, 1985) were administered before and after treatment to assess changes in sleep-related cognition, maladaptive sleep-related behaviors, and level of pre-sleep arousal, respectively. Insomnia Severity Index (ISI) and sleep logs were also given to assess treatment effects.

Measurements

Insomnia Severity Index (ISI). The ISI is a 7-item Likert-type self-rating scale designed to assess the subjective perception of the severity of insomnia (Morin, 1993). The scale contains items that measure the symptoms, associated features, and impacts of insomnia, including difficulty falling asleep, difficulty maintaining sleep, early morning awakening, satisfaction with sleep, concerns about insomnia, and functional impacts of insomnia. The total score could be used to differentiate patients with and without insomnia, and to categorize patients to different levels of insomnia severity. The scale was found to have adequate internal consistency (Cronbach’s α = .74; Bastien, Vallieres, & Morin, 2001). It was also shown to have small to moderate correlations with polysomnographic indices of sleep quality (r = 0.32~0.55).

Pre-sleep Arousal Scale (PSAS). The PSAS consists of 16 items designed to measure the level of arousal prior to sleep (Nicassio, Mendlowitz, Fussell, & Petras, 1985). The items are further divided into two subscales: a somatic arousal subscale and a cognitive arousal subscale. Previous study has shown good internal consistency (Chronbach’s α= .79-.84 for somatic subscale; Chronbach’s α= .67-.88 for cognitive subscale) and test-rest reliability with 3-month interval (r=.76 for somatic arousal;

r=.72 for cognitive subscale).

Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS). The DBAS consists

-7-

of 30 items concerning the belifs, attitudes, expectation, and attributions about sleep and insomnia (Morin, 1993). Subjects had to rate their level of agreement on a 10-point scale for each item. A brief version of the DBAS has recently been developed and validated by Morin and colleagues (Morin,Vallieres & Ivers, 2007), which was demonstrated to be comprise of four factors: (a) perceived consequences of insomnia, (b) worry/helplessness about insomnia, (c) sleep expectations, and (d) medication. The scale was shown to have good internal consistency (Chronbach’s α= .79) and test-retest reliability (r= .83).

Beck Anxiety Inventory (BAI; Beck, Epstein, Brown, Steer, 1988; Beck & Steer, 1993). The BAI is a 21-item self-rating scale that measures the severity of subjective anxiety. It consists of descriptive statements of anxiety symptoms that are rated on a 4-point Likert scale. It has been shown to have good internal consistency with

Cronbach’s alpha coefficients ranged from .92 to .94 for adults and good test-retest reliability with 1-week interval (r = .75). Concurrent validity was also documented by showing significant correlations with different scales for anxiety, including Hamilton Anxiety Rating Scale, anxiety subscale of the Cognition Check List, and State-Trait Anxiety Inventory. The BAI total score is the sum of the ratings for the 21 symptoms.

A total score of 13 has been recommended to be the cut-off score for screening of anxiety problem.

Results

The three groups showed no significant differences in age, year of education, sleep parameters, and all the psychological and behavioral variables, except for the ratings on BAI and BDI. The comorbid group, as expected, showed higher depression and anxiety levels than the other two groups at baseline. Table 1 shows the baseline data of the three groups.

Baseline Data

-8-

Table 1. Comparison of the variables at baseline

PI_CBT

+Hypnotic Comorbid F p Gender

(M:F)

7:14 17:38 6:19 -- --

Age 38.14±12.54 42.58±12.02 41.44±9.82 1.106 .335 BAI 6.84±6.34 10.36±8.29 15.26±10.49 5.242 .007 BDI 6.84±5.95 9.62±7.22 15.00±11.04 5.433 .006 Edu.

(year)

16.26±3.26 14.52±3.51 14.84±3.77 1.726 .184 ISI 17.48±3.60 17.40±5.47 18.40±5.70 .334 .717 DBAS 149.52±37.91 160.49±38.40 174.16±30.46 2.653 .075 SHPS 75.29±13.88 75.55±15.87 76.13±13.59 .020 .981 PSAS 36.38±11.53 37.09±11.50 42.67±12.38 2.240 .112 TST 383.21±76.55 392.26±77.16 384.14±70.73 .161 .851 SOL 46.02±44.26 37.79±33.00 28.30±17.64 1.689 .190 WASO 17.18±22.18 17.86±17.71 18.71±17.44 .039 .961 SE .82±.13 .83±.11 .86±.06 .964 .385

2 (pre-treatment vs post-treatment) x 3 (groups) ANOVA showed significant treatment main effects and no group main effect and interaction for the ratings on the ISI. The scores of ISI showed that insomnia severity reduced significantly for all the three groups (F=156.84, p<.001), indicating significant treatment effect of CBTI.

Sleep log shows similar results and revealed that sleep onset latency (F=11.70, p=.001), wake after sleep onset (F=23.89, p<.001),, and sleep efficiency (F=16.98, p<.001), all showed significant improvement, except for total sleep time that reached a near-significant increase after treatment (F=3.63, p=.060). None of the group main effects and interactions were significant.

Treatment Effect

For primary insomnia patients received CBT-I only, Pearson correlations showed that improvement in ISI correlated with decrease of both somatic (r=6.45, p<.005) and cognitive (r=6.59, p=.001) pre-sleep arousal and reduction of arousal-related behaviors (r=5.92, p=.005). For primary insomnia received CBTI and hypnotic, reduction of ISI score correlated with reduction of factor II (predictability of sleep) Correlation between changes in psychological and behavioral factors and

improvement

-9-

subtests of the DBAS (r=3.25, p<.05) and reduction of arousal-related behavior (r=.330, p<.05). For comorbid insomnia group, ISI improvement correlated with change of scores on the factor I (perceived consequences of insomnia, r=.438, p<.05) and factor II (predictability of sleep, r=.752, p<.001) subtests of the DBAS, both somatic (r=.530, p<.005) and cognitive (r=.772, p<.001) pre-sleep arousal, arousal-related behavior (r=.605, p<.005), and poor sleep environment (r=.450, p<.05). Table 2 shows the results of the correlations.

Table 2. Correlations between changes in psychological/behavioral factors and sleep improvements following treatment

Change score ISI

PI CBTI PI CBTI+PT Comorbid

DBAS Total Score .035 .217 .670***

F1: causes of insomnia .022 .150 .438*

F2: consequences of insomnia .352 .325* .752***

F3: sleep expectations -.386 -.019 .124 F4: control and predictability -.250 -.054 .191 F5: sleep-promoting practices -.097 .074 .582**

PSAS Somatic Arousal .645** .161 .530**

Cognitive Arousal .659** .160 .772***

SHPS Sleep Schedule .166 .141 .328 Arousal Behaviors .592** .330* .605**

Eating Behaviors -.029 -.180 .282 Sleep Environment .130 .197 .450*

*: p < .05; **: p < .01; ***: p < .001 Conclusion

The purpose of this study is to explore the associations between changes in psychological and behavioral factors and symptom improvement following CBTI in both primary insomnia patients and patients with insomnia comorbid with other

-10-

conditions, in order to identify the critical components of CBTI program. Treatment effect was firstly confirmed by showing significant improvements after treatment in the overall rating of insomnia severity as well as most sleep parameters on sleep logs for both primary insomnia patients and insomnia with a comorbid condition. The changes in psychological and behavioral parameters are also evidenced. The treatment did, as intended, lower the dysfunctional beliefs about sleep, the maladaptive sleep hygiene behaviors, and the pre-sleep arousal.

Among the changes in psychological and behavioral factors that are associated with treatment effect, both somatic and cognitive pre-sleep arousal as well as the arousal-related behaviors are the most consistent ones across the three groups. This confirms the important role of hyperarousal in chronic insomnia as proposed in the etiological models for insomnia (Bonnet and Arand 1997; Perlis et al. 1997). Many techniques in CBTI, such as cognitive reconstructuring, relaxation training, and stimulus control were all aimed to reduce arousal level.

In terms of sleep hygiene practice, it is interested to see that only

arousal-related behaviors are associated with sleep improvement. Although sleep hygiene education has long been considered to be an important part of CBTI, this result is consistent with previous finding that sleep hygiene education alone may not be sufficient for insomnia treatment (Morin et al., 1999; Murtagh, 1995; Morin, Culbert, & Schwartz, 1994). In addition, previous study has shown that only

在文檔中失眠的過度激發病理機制：慢性失眠患者與高、低暫時性失眠脆弱特質者之比較 (III) (頁 37-56)