Cognitive Behavioral Therapy for Insomnia as a Preparatory Treatment for Exposure Therapy for Posttraumatic Stress Disorder

Introduction

Insomnia is one of the most frequently reported symptoms of posttraumatic stress disorder (PTSD); it is present in as many as 70% of individuals with PTSD (Ohayon & Shapiro 2000). Various investigators have hypothesized that insomnia, which arises secondary to other symptoms of PTSD (e.g., nightmares and hypervigilance), should remit after a successful course of PTSD treatment. Unfortunately, after successful evidence-based treatment of PTSD, insomnia remains a problem in about half of cases, even when nightmares and hypervigilance have remitted (Zayfert & DeViva, 2004).

Evidence is mixed regarding the optimal order in which to treat comorbid insomnia and PTSD. This highlights a growing concern in the general literature regarding the ordering of evidence-based psychothera-pies for co-occurring disorders (e.g., McCauley, Killeen, Gros, Brady, & Back, 2012). The treatment manual for cognitive behavioral therapy for insomnia (CBTi) warns that an unresolved comorbid psychiatric illness may complicate—and may even be a contraindication for—CBTi (Perlis, Jungquist, Smith & Posner, 2005). There is, however, some evidence that CBTi can produce sleep improvements in individuals with comorbid conditions. Two studies that tested CBTi in patients with comorbid psychiatric disorders (including PTSD) found that in both veteran (N = 8, Gellis & Gehrman, 2011) and civilian (N = 41; Edinger et al., 2009) populations, patients experienced statistically significant reductions in insomnia symptoms by the end of treatment. Although the effect sizes in these studies for the treatment outcomes were in the medium-large range for many of the treatment outcome measures (CBTi outperformed the sleep hygiene condition, ds = .26 to .95, in Edinger et al., 2009; veterans reported subjective sleep improvements in the d = 0.6 to 3.2 range in Gellis & Gehrman, 2011), the percent change in symptoms was relatively modest (Edinger et al., 2009; Gellis & Gehrman, 2011). Based on the idea that comorbid PTSD may reduce the effectiveness of treatment for insomnia, DeViva and colleagues (2005) assessed the effectiveness of CBT for residual insomnia subsequent to PTSD treatment. Their results were similar to previous findings with modest percent symptom change. Thus, when it comes to the care of patients with comorbid PTSD and insomnia, there is little clear evidence to recommend administering CBT for PTSD prior to CBT for insomnia or vice versa. Additionally, while each treatment individually provides some benefit, there is certainly room for more clinically significant change. There may be more benefit to patients when both treatments are used together or in immediate succession.

Despite the effectiveness of exposure therapy in reducing symptoms of PTSD (Gros, Yoder, Tuerk, Lozano, & Acierno, 2011), many veterans with PTSD are reluctant to engage in such treatment. Indeed, as many as 50% of patients drop out either before the initial appointment or during exposure-based treatment (Schottenbauer, Glass, Arnkoff, Tendick, & Gray, 2008). Certainly, exposure therapy for PTSD can be uncomfortable, involving techniques that encourage patients to confront situations and memories that make them most anxious. In some cases, patients may not view their PTSD symptoms (e.g., anger, hypervigilance, and isolation) as a problem, but may see such symptoms as appropriately self-protective (Murphy et al., 2004). Together, these factors may work together to undermine patients’ willingness to participate in exposure therapy for PTSD.

In the present case, we used a combination of CBT for insomnia (CBTi) and psychoeducation about PTSD to address simultaneously a very problematic set of insomnia symptoms (as reported by the patient) and to prepare and motivate the patient for exposure therapy for PTSD (treatment that the patient initially refused).

Method

Case: Mr. M

Mr. M was a 70-year-old, Caucasian male, a veteran who had served in the Navy as a fireman for more than 25 years. Symptoms of PTSD and insomnia were identified at Mr. M’s initial appointment at a Veteran Affairs Medical Center in the Southeastern United States. Mr. M reported trying several trials (with variable compliance) of sleep medications, which had little benefit. Mr. M noted that he was reluctant “drug himself” to sleep in part because he wanted to be able to wake up if an intruder were to break into his home. Mr. M agreed to referral for psychotherapy to address these symptoms. No medications were prescribed for his symptoms during his initial appointments.

Assessment

Mr. M completed the initial clinical intake within the psychotherapy clinic; assessment included structured (i.e., the Mini International Neuro-psychiatric Interview (MINI; Sheehan et al., 1998)) and unstructured clinical interviews and several self-report measures. Mr. M met criteria for PTSD on the MINI. He also met criteria for primary insomnia consistent with the diagnostic criteria in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR; American Psychiatric Association, 2001).

His index trauma involved witnessing an unconscious sailor being dragged out of a compartment in the dangerously hot underside of a ship’s boiler room and then being pressured to enter the compartment and complete the sailor’s unfinished duties. Mr. M recalled feeling a sense of helplessness and fear in this situation, and still felt he narrowly escaped serious injury or even death. Mr. M reported intrusive thoughts (both during the daytime and nightmares) of his naval experiences of being in enclosed spaces. He also described having thoughts and nightmares about news stories (e.g., stories of people trapped in small spaces) that reminded him of his upsetting experiences. He acknowledged attempts to suppress these disturbing thoughts and of being unable to remember the details of some of his disturbing experiences in the Navy. After leaving the Navy, he avoided small spaces; he quit a good job that required him to go under the building to do maintenance work and he was currently avoiding the work he needed to do under his house. In addition to his difficulty staying asleep, he reported being watchful and vigilant at night, for example, before going to bed, he would check multiple times to make sure the doors to his home were locked. He also reported being irritable and having difficulty concentrating.

Mr. M stated he had no trouble getting to sleep at night, but reported that he had not slept through the night since he joined the Navy. On average, he reported, he was in bed for eight hours and asleep for about four and a half of them. In terms of sleep hygiene, Mr M. reported some good sleep hygiene habits, including dedicated relaxation time before bed, minimal use of alcohol, caffeine use only in the morning, use of the bed only for sleep and sex. He avoided naps and kept his bedroom comfortable, cool, and dark during sleeping hours. However, when Mr. M awoke in the middle of the night, he would plan and prepare for the next day, a longstanding habit that had, he felt, served him well during his working life. On some nights, intrusive, trauma-related thoughts would keep him awake for hours. After returning to bed, he would “toss and turn” until he fell back to sleep.

As presented in Table 1, Mr. M filled out self-report questionnaires during his intake appointment, which tapped symptom severity and impact. Mr. M endorsed mild symptoms of PTSD on the Posttraumatic Stress Disorder Checklist (PCL; Blanchard, Jones-Alexander, Buckley, & Forneris, 1996; Weathers, Litz, Herman, Huska, & Keane, 1993). On the Depression, Anxiety, and Stress Scales (DASS-21; Lovibond & Lovibond, 1995), he reported severe symptoms of anxiety, normal symptoms of depression, moderate stress and severe insomnia symptoms on the Insomnia Severity Index (ISI; Bastien, Vallieres, & Morin, 2001). His score on the Dysfunctional Beliefs and Attitudes about Sleep Scale (DBAS-16; Morin, Vallières & Ivers, 2007) was in the moderate range. He reported mild overall interference on the Illness Intrusiveness Rating Scale (IIRS; Devins et al., 1983), with the most impairments noted in his work, relationship with his spouse, and family relations categories.

Treatment

During the first treatment session, the diagnoses and case conceptualization were reviewed with Mr. M., and a recommendation made for a treatment involving a combination of

Mr. M was willing to work on his insomnia but expressed skepticism and nervousness regarding the techniques involved in exposure therapy. The collaborative decision was made to start only CBTi, assess symptom improvement weekly, and reassess his need and readiness for exposure therapy later in treatment.

We began CBTi with psychoeducation about sleep hygiene and sleep restriction. Mr. M was encouraged to get out of bed if he woke up and could not fall asleep within 15 minutes. He was also encouraged, while out of bed, to engage in a relaxing activity rather than doing work or planning for the day ahead. Mr. M agreed to postpone his planning and preparation for his day and to do this in the morning after rising for the day. Mr. M was assigned to keep a daily sleep diary in which he recorded time into bed and out of bed, total number of awakenings and total time awake, subjective sleep quality, and the thoughts and feelings he was having after waking up at night.

Given that Mr. M’s total sleep time was 4.5 hours and only 56.3% efficient, he was assigned to restrict his time to sleep to 4.5 hours each night. Due to Mr. M’s preference for being awake during the morning hours and because he already maintained a roughly consistent bedtime of 9:30 P.M. to 10:00 P.M, he decided on a 9:30 P.M bedtime and a 2:30 A.M wake time and we improved his efficiency to 74%, with an average of 3.7 hours of total sleep time per night. Mr. M attempted to further increase his sleep efficiency based on these initial improvements. During the next weeks he increased time in bed to five hours. The following week (after session 2), Mr. M’s sleep efficiency improved to 83%, with an average of 4.2 hours of total sleep time per night. Over the next several weeks, Mr. M steadily increased his allotted sleep time, while targeting and maintaining sleep efficiency between 80-85%. By the end of the week 6, Mr. M had increased his average sleep to 5 hours per night, with a minimum of 83% efficiency.

Treatment also involved identification and modification of sleep-interfering thoughts. To test whether anxious thoughts, trauma-related thoughts, and/or dysfunctional beliefs about sleep might be getting in the way of his sleep, Mr. M recorded in his sleep diaries his thoughts when he woke up in the middle of the night. His sleep diaries indicated that he often kept himself awake by thinking about unresolved problems or difficult situations in his life, for example, his wife’s recovery from a minor surgery, and he often reviewed his plans and repeatedly made mental to-do lists for the next day. Mr. M also reported dysfunctional thoughts about sleep, for example, he would wonder if his inability to sleep would cause him to be “unable to function” in his volunteer work, or would it cause serious health problems. We advised him to shift the task building to an alternate time in the morning to reduce thoughts interfering with his sleep. Cognitive restructuring was reviewed and practiced to reduce his dysfunctional sleep-related beliefs. We also applied cognitive restructuring to Mr. M’s significant frustrations about his inability to sleep and his ruminations about the origins and meaning of his distressing symptoms.

Less frequently (i.e., on one or two nights a week) he began thinking about his traumatic experiences in the Navy and feeling anxious and physiologically aroused. Although these thoughts occurred less frequently than his day-to-day concerns, they were also associated with much longer periods of sleeplessness during the night (i.e., roughly 2 hours vs. 1 hour) and worse sleep quality. The relationship between his trauma-related thoughts and his length of nighttime wakefulness and sleep quality was discussed, and Mr. M began to recognize the role of his trauma-related thoughts in maintaining his insomnia. Mr. M was assigned psychoeduca-tional readings about exposure-based PTSD treatment (Rothbaum, Foa, & Hembree, 2007) to further his understanding of these treatments.

During the final session of CBTi, treatment progress was reviewed and relapse prevention strategies were introduced. Additional information about exposure treatment was provided and discussed in detail, and the guidelines of exposure therapy explained (Foa et al., 2007). Given his improved understanding of exposures, in addition to his recent success with another CBT-based therapy, Mr. M expressed willingness to try exposure therapy.

Prolonged Exposure Therapy for PTSD

Mr. M’s exposure therapy for PTSD involved both situational and imaginal exposures. In terms of situational exposures, Mr. M completed a hierarchy that included (from least distressing to most distressing): spending time in very small rooms (particularly if other people were between him and the exit); working in the crawl space under his house; reading news stories about people stuck in caves, wells, and other tight spaces; visiting the boiler room on a naval ship. Mr. M’s first situational exposure took place during a treatment session, in a closet-sized room with the therapist sitting between him and the door. During the 25-minute exposure, Mr. M reported a decrease in his anxiety, which was rated on a ten-point scale, with ten representing the highest anxiety. It dropped from a 6 of 10 to a 2 of 10. This gave Mr. M first-hand experience of the anxiety reduction that exposures can produce.

In the twelfth treatment session, Mr. M stated that he was ready to face the situation at the top of his hierarchy: exposure to the boiler room of decommissioned naval ship that was open to the public as a museum. He visited the ship four times, each time moving closer to the boiler room. During the last two visits, he was able to make himself stand in front of the boiler, looking into it. He stated, “I couldn’t see much but I could remember a lot.” Each time he stayed next to the boiler for more than hour (the boiler room was closed to visitors, so he was unable to enter into it). After his final visit, he realized, “I can almost enjoy going back to the maritime museums now.” Mr. M was also asked to reduce safety behaviors (i.e., multiple times checking his locks before bed & upon getting out of bed). He was able to do this.

In his imaginal exposures, he revisited his memory of the boiler room incident (his index trauma). He told and retold the story in session and listened to a tape-recording of the session 4 to 5 times a week for homework. He experienced benefit from the imaginal exposures quickly. During his first imaginal exposure in session, his anxiety went from an 8/10 to a 5/10. During the week, he listened to the recording for homework. He remarked,

During the second week of listening to his imaginal exposure recordings for homework, Mr. M’s anxiety peaked at a 2/10.

As Table 1 shows, Mr. M’s symptoms of PTSD, anxiety, depression, and stress, as well as his illness-related impairment, had improved markedly by treatment termination at session 13. In fact, Mr. M demonstrated reliable change scores, as computed using Jacobson’s reliable change index (Jacobson & Truax, 1991), on the majority of these measures.

Throughout, he continued to practice sleep restriction and to fill out sleep diaries. As Table 2 shows, his sleep continued to improve throughout treatment. Mr. M’s sleep efficiency and quality significantly improved by session 6 compared to baseline (Table 2). Mr. M attributed this improvement primarily to the sleep restriction. He recorded in one of his sleep diaries, “I feel like the sleep restriction is helping because I get sleep when I go to sleep versus tossing and turning.” The week after beginning exposure therapy (session 6), Mr. M experienced what he reported was his first uninterrupted night of sleep in 50 years. By the end of treatment, he had had had three or four uninterrupted nights’ sleep each week for three weeks.

Mr. M’s sleep quality, efficiency, and time asleep continued to improve, showing significant gains in between session 6 and the end of treatment (see Table 2). His PTSD symptoms also improved considerably (see Table 1). Mr. M left treatment having found considerable relief. In the last session, he stated, “A weight on my chest that has been here for years has been lifted.”

Follow Up

Mr. M was seen for follow up 90 days after his final treatment session. He reported that he encountered a few “difficult” nights a couple of weeks after the final treatment session. On these nights, he woke up multiple times “in a cold sweat.” He stated, “I knew what I had to do: I went back to the maritime museum.” He continued his own situational exposure therapy. He reported that this helped him put his traumatic experiences into perspective. He stated, “I’ve been through [the traumatic experience] and I survived. I will never have to [go through the same experience] again.”

At follow-up, Mr. M had sustained his improvements, showing lasting reliable change in important symptom areas (see Table 1). His insomnia, in the severe range at pretreatment, was minimal at follow up. His PTSD symptom severity, anxiety, and stress also had decreased significantly. Mr. M’s impairment was also changed significantly; he reached a level of minimal impairment at follow up. Interestingly, Mr. M’s score on the dysfunctional beliefs and attitudes about sleep scale remained roughly the same at follow up, suggesting that he retained many of the dysfunctional beliefs he had at baseline. For example, at both time points, he endorsed fears that chronic insomnia would have serious consequences for his health (at baseline, 10/10 and at follow-up, 6/10). However, based on his other reports, these beliefs were having minimal effect on his overall sleep quality.

Discussion

There has been some controversy regarding the order in which to treat PTSD vs insomnia in individuals for whom both are present. The CBTi treatment manual suggests that CBTi may not be indicated for individuals with comorbid disorders such as PTSD because they may interfere with CBTi, and because in some cases insomnia will resolve through treatment of the PTSD (Perlis et al., 2005). Now evidence suggests that insomnia may linger after completion of PTSD treatment (Zayfert & DeViva, 2004) and that the insomnia can be treated successfully with CBTi (DeViva et al., 2005). Other evidence suggests CBTi may lead to sleep improvement in patients who have insomnia and PTSD (Edinger et al., 2009; Gellis & Gehrman, 2011). However, these studies did not explore the use of CBTi to facilitate entry into PTSD treatment.

Not all patients with PTSD are ready to address this condition in treatment (Murphy et al., 2004). Despite the effectiveness of exposure treatments, it is particularly difficult to engage patients in exposure treatments, and dropout rates are high (Schottenbauer et al., 2008). Efforts are underway to find approaches that can increase treatment engagement. Motivational enhancement is one method that has been shown to increase patient readiness for exposure treatment (Murphy et al., 2009). In the present case CBTi (in addition to psychoeducation) was a brief, active treatment that provided symptom relief and facilitated initiation of exposure therapy.

Specifically, we found that the combination of sleep restriction, self-monitoring, and psychoeducation about PTSD prepared Mr. M for PTSD treatment by: (1) improving insomnia symptoms, helping him gain confidence in the beneficial effects of psychosocial treatment, (2) providing the patient with evidence (from his sleep diaries) of the link between his nightmares (a PTSD symptom) and his worst nights of sleep; thereby helping the patient make an informed decision to pursue PTSD treatment.

The current results support previous research suggesting that individuals with comorbid Axis I disorders can be successfully treated for insomnia (Edinger et al., 2009). Furthermore, they suggest that insomnia treatment can be combined with psychoeducation about PTSD to enhance the patient’s readiness for PTSD treatment. In this case, pursuing a plan to complete the treatments in the reverse order would likely have led this patient to feel that his main concerns were not being addressed.

This case study has implications for the treatment of individuals with comorbid PTSD and insomnia. It suggests insomnia treatment can perform a motivational enhancement function to prepare patients with a primary complaint of insomnia (but who also have underlying PTSD) for entry into a more challenging, more aversive, but ultimately, effective PTSD treatment. Clinicians might consider CBTi (for patients with co-morbid insomnia and PTSD) as an alternative to motivational enhancement groups (e.g., Murphy et al., 2009), which are sometimes used to good effect to prepare patients for treatment entry. As an alternative, CBTi has the advantage of directly producing symptom reduction.

This study does have some limitations. No objective measures of sleep quality were obtained, and these tend to improve less than the subjective measures from CBT (Edinger et al., 2009). Given the single-case, uncontrolled nature of the current study, it is not possible to determine how well this treatment combination would generalize to other patients. However, these findings suggest the potential fruitfulness of larger-scale investigations of the use of CBTi as a treatment for exposure-resistant patients with PTSD who present with comorbid insomnia.

In summary, the current findings provide initial support for the utility of CBTi as a preliminary treatment for significantly reducing insomnia symptoms and promoting PTSD treatment engagement in veterans with comorbid PTSD and insomnia. The findings have implications for the ordering of CBTi and exposure therapy, suggesting that CBTi can be used prior to exposure therapy when patients have higher readiness to work on their insomnia than their PTSD. Finally, the findings lend further support to emerging evidence for the effectiveness of integrative treatments that allow for the treatment of comorbid symptoms of depression (Gros, Price, Strachan, Yuen, Milanak, & Acierno, 2012) or substance use problems (Back, Killeen, Foa, Santa Ana, Gros, & Brady, 2012) to help prepare patients for engagement in exposure treatment.