COU-2014-1255
RCT of CBT-I for co-morbid insomnia and depression
Supplemental Materials
A Randomized Controlled Trial of Cognitive Behavioral Therapy for Insomnia: An Effective Treatment for Co-morbid Insomnia and Depression
by D. K. Ashworth et al., 2015, Journal of Counseling Psychology
http://dx.doi.org/10.1037/cou0000059
1. Details of CBT-I intervention: CBT-I was delivered across 4 sessions in the following structure:
· Session 1 focused on sleep education (why people sleep, sleep stages, sleep throughout the lifetime, circadian rhythms, homeostatic pressure, sleep inertia, and a model of insomnia). This education provided participants with basic knowledge of sleep, rationale for sleep scheduling, and research evidence to challenge unhelpful beliefs about sleep. Sleep hygiene recommendations were provided to educate participants about environmental factors that can influence their sleep and to help participants identify three areas that they wanted to target for homework. Although not sufficient on their own, sleep hygiene recommendations are considered helpful as part of CBT-I multimodal treatment, especially when tailored to the individual (Ebben & Spielman, 2009). Relaxation was also introduced via a diaphragmatic breathing exercise. As there is support for several relaxation techniques, a different relaxation technique was introduced in each session that was consistent with the theme of the session (1 – education; 2 – behavior; 3 – cognitive; 4 – relapse prevention) with the aim of reducing pre-sleep arousal levels (Edinger & Means, 2005).
· Session 2 focused on the behavioral interventions for insomnia with the most empirical support, including stimulus control (Bootzin, 1972) and sleep restriction (Spielman, Saskin, & Thorpy, 1987). Together, stimulus control and sleep restriction are the greatest predictors in ongoing sleep improvement by 12-month follow-up (Harvey & Payne, 2002). A progressive muscle relaxation exercise (Bernstein, Borkovec, & Hazlett-Stevens, 2000) was conducted to target somatic hyper arousal, and a constructive worry worksheet (Edinger & Carney, 2008) was assigned for homework.
· Session 3 focused on cognitive interventions for insomnia, including constructive worry, unhelpful thinking patterns in insomnia (Edinger & Carney, 2008), and cognitive restructuring of dysfunctional beliefs about sleep (Morin, 1993). Improvements in sleep related cognitions (as measured by reductions in dysfunctional beliefs about sleep) predict lower likelihood of relapse. An imagery (Morin & Azrin, 1987) relaxation exercise was conducted to target cognitive hyper-arousal. Thinking in the form of images has been found to resolve worry more effectively than thinking in the form of words (Nelson & Harvey, 2002). Cognitive arousal is also ten times more likely to be cited as the major cause of insomnia than somatic arousal (Lichstein & Rosenthal, 1980). A thought record (Edinger & Carney, 2008) was assigned for homework.
· Session 4 focused on promoting long-term gains, including troubleshooting common lingering concerns (Edinger & Carney, 2008), self-management (Lundh, 2000), relapse prevention for insomnia (Smith & Posner, 2005), and a breathing-to-connect mindfulness exercise (Harris, 2008, 2009). Mindfulness can effectively improve insomnia (Ong, Shapiro, & Manber, 2008) and has been shown to prevent depressive relapse following remission of symptoms (Ma & Teasdale, 2004). Additional techniques that may be helpful were also included to target specific ongoing perpetuating factors. These included autogenic relaxation if elevated pre-sleep arousal was still a concern, cognitive control (Morin & Espie, 2003) and articulatory suppression (Levey, Aldaz, Watts, & Coyle, 1991) if racing thoughts at night remained problematic, and paradoxical intention (Morin, 1993) if elevated sleep effort was making sleep initiation difficult. Any negative discrepancies between how participants believed they slept (subjective sleep measures) and how they actually slept (objective sleep measures) were discussed with the aim of reducing unhelpful sleep state misperception.
Details of secondary outcome measures:
I. The PSQI (Buysse, Reynolds, Monk, Berman, & Kupfer, 1989) consists of 24 questions over 11 items, and produces 7 domain scores and a total score measuring average self-rated sleep quality over the previous month. Total scores range from 0-21, with 5 or under considered to be good sleep quality, and 6 or above interpreted as poor sleep quality (Buysse et al., 1989).
II. The DASS-21 (Lovibond & Lovibond, 1995) consists of three separate 7-item self-rated scales that aim to delineate the core features of depression, anxiety (DASS-A) and stress. DASS-A scores range from 0-42, with higher scores indicating a greater level of anxiety severity (Lovibond & Lovibond, 1995).
III. The FSS (Krupp, LaRocca, Muir-Nash, & Steinberg, 1989) is a 9 item self-rated scale that asked the participant how much they agreed with each fatigue related statement. Overall FSS ratings range from 9-63, with higher scores indicating a greater level of functional impairment or distress as a result of fatigue (Krupp et al., 1989).
IV. The ESS (Johns, 1991) is an 8 item self-rated scale that asks participants how likely they would be to doze off or fall asleep in several common everyday situations. Overall ESS scores range from 0-24, with scores above 10 indicating excessive daytime sleepiness. Scores above 13 are also considered to be highly indicative of sleep apnea, restless leg syndrome or periodic limb movement disorder (Johns, 1991), and participants who remained above this threshold at the follow-up assessment were informed of their results and encouraged to discuss these findings with a general practitioner or sleep specialist.
V. The sleep diary provided a visual depiction of sleep-related behaviours, which made it easier to assess adherence to different behavioural interventions for CBT-I participants and provide individualised feedback in session. The sleep diary was modified by asking participants four additional questions each day to obtain more specific data on their bedtime (BT), time taken to fall asleep (SOL), duration of night-time awakenings (WASO), and rising time (RT). This also made it possible to calculate subjective time in bed (TIB = difference between BT and RT), TST (= TIB - SOL - WASO) and sleep efficiency (= TST / TIB, expressed as a percentage of time in bed asleep) for later data analysis.
2. Secondary outcome measures results
The CBT-I intervention was more effective at improving sleep quality, anxiety severity and fatigue severity in comparison to the self-help CBT-I intervention (Figure 5). Significant effect of treatment group was observed for PSQI (p < .001), DASS-A (p < .001), and FSS (p < .001). No significant treatment group effect was found for daytime sleepiness (p = .23) and no interaction effects between treatment group and time were found for any of the secondary outcome measures.
Figure S5. Mean ± SEM group (CBT-I, self-help CBT-I) sleep quality (PSQI), anxiety severity (DASS-A), fatigue severity (FSS) and daytime sleepiness (ESS) ratings at baseline (week 0), post-treatment (week 8), and 3-month follow up (week 20). Treatment phase is indicated by the dotted lines (week 2 through to week 8). ** = p < .001.
3. Secondary Outcome Measures Discussion
CBT-I was more effective than self-help CBT-I at improving anxiety. Previous findings support anxiety reductions through CBT-I in individuals with co-morbid insomnia, depression and anxiety (Maroti et al., 2011). Depression has been found to mediate the relationship between social anxiety and insomnia (Buckner, Bernert, Cromer, Joiner, & Schmidt, 2008), so reductions in depression and insomnia through CBT-I could improve anxiety by reducing distress, dissatisfaction with sleep, and concern of daytime functional impairment. Relaxation exercises, which are included in CBT-I to target hyper-arousal symptoms, have also been found to consistently reduce anxiety symptoms (Manzoni, Pagnini, Castelnuovo, & Molinari, 2008). The ongoing reductions in anxiety by follow-up reduce the risk of future insomnia relapse, as anxiety has been found to predict later insomnia onset (Johnson et al., 2006).
In terms of daytime symptoms, CBT-I resulted in improved self-reported fatigue but not sleepiness. This is consistent with the view that fatigue, rather than sleepiness, better reflects the daytime impairments experienced by insomnia patients (Shekleton, Rogers, & Rajaratnam, 2010). It was recently shown that individuals with primary insomnia report higher levels of fatigue compared to healthy controls, but similar levels of daytime sleepiness (Shekleton et al., 2013). Therefore, an intervention that targets insomnia would be expected to relieve daytime fatigue symptoms.
9
COU-2014-1255
RCT of CBT-I for co-morbid insomnia and depression
4. Differences between groups on all outcome measures at baseline, post-treatment and 3-month follow-up assessments
Measure / Baseline / Post-treatment / Follow-upCBT-I
mean (SEM) / Self-help CBT-I mean (SEM) / Mean difference (95% CI) / p / CBT-I mean (SEM) / Self-help CBT-I mean (SEM) / Mean difference (95% CI) / p / d / CBT-I
mean (SEM) / Self-help CBT-I
mean (SEM) / Mean difference (95% CI) / p / d
BDI-II / 26.62 (1.90) / 28.20 (1.77) / 1.58
(-3.69 – 6.85) / .55 / 8.29 (1.85) / 21.80 (2.88) / 13.51
(6.60 – 20.37) / < .001 / 1.24 / 8.83 (1.82) / 22.61 (1.87) / 13.78
(8.50 – 19.06) / < .001 / 1.65
ISI / 18.95 (0.77) / 18.30 (1.12) / -0.65
(-3.37 – 2.07) / .63 / 7.81 (1.35) / 13.75 (1.51) / 5.94
(1.85 – 10.03) / < .01 / 0.92 / 5.50 (0.77) / 14.39 (1.28) / 8.89
(5.90 – 11.88) / < .001 / 1.87
PSQI / 12.71 (0.66) / 12.65 (0.77) / -0.06
(-2.11 – 1.98) / .95 / 6.05 (.61) / 9.90
(0.82) / 3.85
(1.81 – 5.90) / < .001 / 1.19 / 5.00 (0.57) / 10.06 (0.72) / 5.06
(3.22 – 6.90) / < .001 / 1.73
DASS-A / 12.95 (2.15) / 15.90 (1.99) / 2.95
(-2.98 – 8.88) / .32 / 5.24 (1.54) / 13.00 (1.87) / 7.76
(2.88 – 12.64) / < .005 / 1.00 / 5.44 (1.57) / 14.89 (1.52) / 9.45
(5.03 – 13.86) / < .001 / 1.35
ESS / 10.43 (1.37) / 9.10
(1.32) / -1.33
(-5.18 – 2.52) / .49 / 7.33 (1.22) / 6.90
(1.01) / -0.43
(-3.65 – 2.78) / .79 / 0.08 / 7.10 (1.23) / 8.28
(0.79) / 1.17
(-1.81 – 4.16) / .43 / 0.25
FSS / 45.43 (2.76) / 42.55 (3.08) / -2.88
(-11.23 – 5.47) / .49 / 32.10 (3.29) / 40.20 (2.75) / 8.11
(-.62 – 16.83) / .07 / 0.59 / 30.83 (3.07) / 42.94 (2.49) / 12.11
(4.08 – 20.14) / < .005 / 0.96
Subjective SOL (min) / 62.01 (9.30) / 50.99 (9.06) / -11.01
(-37.30 – 15.27) / .40 / 25.23 (2.60) / 32.96 (4.35) / 7.73
(-2.41 – 17.87) / .13 / 0.48 / 24.27 (3.10) / 36.23 (4.67) / 11.96
(0.72 – 23.20) / < .05 / 0.67
Objective SOL (min) / 33.29 (3.95) / 33.78 (6.70) / 0.49
(-15.07 – 16.05) / .95 / 15.02 (2.00) / 22.74 (3.82) / 7.72
(-0.88 – 16.31) / .08 / 0.56 / - / - / - / - / -
Subjective SE (%) / 75.91 (1.98) / 79.20 (2.09) / -3.29
(-9.10 – 2.52) / .26 / 87.65 (1.34) / 82.93 (1.74) / 4.72
(0.31 - 9.13) / < .05 / 0.67 / 89.89 (0.86) / 82.66 (1.88) / 7.23
(3.11 - 11.35) / .001 / 1.10
Objective SE (%) / 76.98 (1.53) / 76.02 (2.02) / 0.97
(-4.13 – 6.06) / .70 / 83.25 (1.30) / 78.41 (1.72) / 4.84
(0.50 - 9.18) / < .05 / 0.70 / - / - / - / - / -
Subjective TST (min) / 393.24 (9.93) / 398.55 (16.26) / -5.31
(-43.43 – 32.81) / .78 / 426.38 (10.58) / 400.84 (10.64) / 25.54
(-4.81 – 55.89) / .10 / 0.53 / 438.75 (10.17) / 399.36 (12.82) / 39.39
(6.47 – 72.31) / < .05 / 0.75
Objective TST (min) / 399.12 (8.72) / 395.10 (11.90) / 4.03
(-25.61 – 33.66) / .79 / 396.32 (10.46) / 367.99 (9.23) / 28.33
(-0.02 – 56.68) / .05 / 0.63 / - / - / - / - / -
Abbreviations: BDI-II Beck Depression Inventory-II; ISI Insomnia Severity Index; PSQI Pittsburgh Sleep Quality Index; DASS-A Depression Anxiety Stress Scales-Anxiety; ESS Epworth Sleepiness Scale; FSS Fatigue Severity Scale; SOL sleep onset latency; SE sleep efficiency; TST total sleep time; CBT-I cognitive behavioral therapy for insomnia; SEM standard error of the mean; CI confidence interval.
9
COU-2014-1255
RCT of CBT-I for co-morbid insomnia and depression
References
Bernstein, D. A., Borkovec, T. D., & Hazlett-Stevens, H. (2000). New directions in progressive relaxation training: A guidebook for helping professionals. Westport, CT: Praeger Publishers.
Bootzin, R. (1972). Stimulus control treatment for insomnia. Paper presented at the Proceedings of the 80th Annual Convention of the American Psychological Association.
Buckner, J. D., Bernert, R. A., Cromer, K. R., Joiner, T. E., & Schmidt, N. B. (2008). Social anxiety and insomnia: the mediating role of depressive symptoms. Depress Anxiety, 25(2), 124-130.
Buysse, D. J., Reynolds, C. F., 3rd, Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Research, 28(2), 193-213.
Ebben, M. R., & Spielman, A. J. (2009). Non-pharmacological treatments for insomnia. J Behav Med, 32(3), 244-254.
Edinger, J. D., & Carney, C. E. (2008). Overcoming Insomnia: A Cognitive-behavioral Therapy Approach Therapist Guide: Oxford University Press.
Edinger, J. D., & Means, M. K. (2005). Cognitive-behavioral therapy for primary insomnia. Clin Psychol Rev, 25(5), 539-558.
Harris, R. (2008). The happiness trap: How to stop struggling and start living: Shambhala Publications.
Harris, R. (2009). ACT made simple. New Harbinger.
Harvey, A. G., & Payne, S. (2002). The management of unwanted pre-sleep thoughts in insomnia: distraction with imagery versus general distraction. Behav Res Ther, 40(3), 267-277.
Johns, M. W. (1991). A new method for measuring daytime sleepiness: The Epworth Sleepiness Scale. Sleep, 14(6), 540-545.
Johnson, E. O., Roth, T., & Breslau, N. (2006). The association of insomnia with anxiety disorders and depression: exploration of the direction of risk. J Psychiatr Res, 40(8), 700-708.
Krupp, L. B., LaRocca, N. G., Muir-Nash, J., & Steinberg, A. D. (1989). The fatigue severity scale. Application to patients with multiple sclerosis and systemic lupus erythematosus. Arch Neurol, 46(10), 1121-1123.
Levey, A. B., Aldaz, J. A., Watts, F. N., & Coyle, K. (1991). Articulatory Suppression and the Treatment of Insomnia. Behaviour Research and Therapy, 29(1), 85-89.