Main outcome measure
Balliett et al. (2015) 
4 sessions of individual, in pairs, or individual and in pairs ERRT-M
ERRT-M was found to significantly improve sleep disturbances and related psychopathy. No significant change in PTSD symptoms was reported
Berlin et al. (2010) 
1 male Vietnam veteran
6 sessions of individual CBT-I and IRT (1 assessment session, 1 session of CBT-I, and 4 sessions of IRT)
Sleep and nightmare diaries
Number of nightmares per week decreased from 17 at pretreatment to 5 at posttreatment. Sleep quality and nighttime awakening improved
Cook et al. (2010) 
124 male Vietnam war veterans
6 group sessions of IRT vs. 6 group sessions of sleep and nightmare management control group
Participants in the IRT condition did not improve significantly more than the comparison condition on primary and secondary outcome measures
Gellis and Gehrman (2011) 
Uncontrolled pilot study
11 male veterans older than 50 years of age
5 sessions of individual CBT-I
Improvements were demonstrated in self-reported wake time after sleep onset, total sleep time, sleep efficiency, and overall insomnia. There were no pre-post differences for the actigraphy, PTSD symptoms, or nightmare frequency
Harb et al. (2009) 
11 male Iraq war veterans
7–8 individuals sessions of CBT-I combined with IRT (3 sessions of CBT-I, 4–5 sessions of IRT)
Pre- to posttreatment, moderate effect sizes were found for improvements in nightmare frequency, global sleep quality, and PTSD symptom severity. Sleep diaries found large effects for increased sleep time and sleep onset latency. Sleep diaries found no change in nightmare frequency but a trend toward less intense dreams
Sleep and nightmare diaries
Long et al. (2011) 
37 male veterans
6 sessions of imagery rescripting and exposure therapy (IRET) – group treatment
IRET significantly (statistically and clinically) decreased nightmare frequency, increased quantity of sleep, and reduced PTSD symptoms
Daily sleep activity log
Lu et al. (2009) 
15 male veterans
6 group sessions of IRT
Nightmare effects survey
There were no improvements at posttreatment. At 3- and 6-month follow-up, trauma-related nightmare frequency (nights/week) significantly decreased. Non-trauma-related nightmares did not decrease. The number of trauma-related nightmares and total nightmares per week and PTSD symptoms significantly decreased at 3-month follow-up but was not maintained at the 6-month follow-up. The impact of nightmares, sleep quality, and depression did not improve
PTSD dream rating scale
Margolies et al. (2013) 
40 combat veterans
4 sessions of CBT-I with adjunctive IRT or a waitlist control
The CBT-I and IRT group displayed improvements in insomnia and PTSD symptoms compared to the waitlist control
Moore and Krakow (2007) 
11 active-duty Iraq war soldiers
4 sessions of IRT
Number of nightmares
The majority of soldiers (7/11) experienced large clinical improvements in nightmares, posttraumatic stress symptoms, and insomnia at the 1-month follow-up. Among the four nonresponders, one had an increase in nightmares and two had an increase in posttraumatic stress symptoms
Nappi et al. (2010) 
58 male and female veterans
5 sessions of individual or group IRT
Daily nightmare log
Treatment completers experienced significant decreases in weekly frequency and maximum intensity of nightmares, insomnia severity, and PTSD symptoms. No change was found for the PSQI. 23% of treatment completers experienced one or fewer nightmares per week and 11% experienced no nightmares at the end of treatment. Posttreatment, PCL and ISI were below cutoffs. Participants in the individual treatment had significantly greater reductions on the ISI
Swanson et al. (2009) 
10 male Vietnam and Gulf War veterans
10 sessions of group treatment (5 sessions of CBT-I, 2 sessions of ERRT, 2 sessions of discussion/trouble-shooting, and 2 sessions of relapse prevention)
Nightmare frequency and distress decreased by half. Following treatment, average insomnia was in the subthreshold range. Although at posttreatment, sleep quality had significantly improved, the mean score was still in the clinically significant range. PTSD symptom reduction was nonsignificant
Sleep and dream diaries
Talbot et al. (2014) 
45 adults with PTSD
8 sessions of individual CBT-I vs. monitor-only waitlist control
CBT-I significantly improved sleep and psychosocial functioning in adults with PTSD
Ulmer et al. (2011) 
Randomized parallel group experiment
22 male and female veterans
Usual care condition vs. 6 individual sessions of sleep intervention for PTSD (SIP) (3 sessions of CBT-I followed by 3 sessions of IRT)
Intent-to-treat analyses found that individuals in SIP had substantial improvements on insomnia severity, PTSD symptoms, and sleep quality. The SIP group had improvements on all sleep diary outcomes. Groups did not differ on depression or the PTSD-specific sleep measure
Wanner et al. (2010) 
2 male Vietnam veterans
4–5 individual sessions of ERRT
Veteran 1 experienced clinically significant reductions in PTSD and depressive symptoms from pre- to posttreatment; these improvements remained at 3-month follow-up. However, depressive symptoms remained in the severe range. At 3-month follow-up, the frequency of trauma-related nightmares decreased by half. Quantity of sleep increased by an hour from pre- to follow-up. Veteran 2 achieved clinically significant reductions in PTSD symptoms from pre- to 3-month follow-up. Depressive symptoms clinically decreased between pre- and post- but were not maintained at follow-up. From pre- to follow-up, trauma-related nightmares decreased by one third and he gained an average of 1 h of sleep per night
Ho et al. (2016) 
Meta-analysis of 11 RCTs
Trauma-exposed individuals, veterans, sexual assault survivors, and adults with PTSD and insomnia
Sleep-specific CBT vs. waitlist control
Various PTSD, depression, and sleep measures
Sleep-specific CBT showed significant reduction in PTSD symptoms, depressive symptoms, and insomnia severity
Sleep intervention followed by CBT for PTSD
CBT for PTSD followed by sleep intervention
Forbes et al. (2001) 
12 male Australian Vietnam veterans
6 group sessions of IRT at least 6 months following inpatient treatment of PTSD
Significant improvements from pre- to posttreatment on measures of nightmares, PTSD, and depression. 7 patients reported cessation of the target nightmare and 11 patients reported improved frequency and/or intensity of the target nightmare
Galovski et al. (2016) 
108 female assault survivors
Either 3 weeks of hypnosis or symptom monitoring control before standard CPT
Hypnosis significantly improved sleep disturbances but, however, did not augment PTSD recovery during CPT
Forbes et al. (2003) 
12-month follow-up to Forbes et al. (2001) study
Treatment gains (reported above) were maintained at 12-month follow-up. At the 12-month follow-up, 60% reported cessation of the nightmare targeted in treatment
Integrated PTSD and sleep intervention
Among civilians, noncontrolled trials and case studies have been conducted among crime victims with PTSD [35, 55], in a large group of fire evacuees as part of a larger treatment protocol , adjudicated adolescent girls in a residential facility , and in a community sample of individuals with idiopathic nightmares and individuals with trauma-related nightmares . In general, results suggest IRT has a positive impact on nightmares and related distress, although findings are limited by lack of controlled design.
Exposure, Relaxation, and Rescripting Therapy
ERRT for trauma-related nightmares was developed to incorporate imagery rehearsal with evidence-based techniques from treatments for insomnia and PTSD. Specifically, ERRT combines rescripting and imagery rehearsal with sleep hygiene education and relaxation from anxiety and insomnia treatments. A key difference between ERRT and IRT is that ERRT emphasizes exposure to the nightmare and trauma-related material and integrates identification of trauma-related themes. Exposure and trauma-related themes are key components of the evidence-based PTSD treatments discussed above . To incorporate exposure, the ERRT protocol instructs participants to write out their most frequent or distressing nightmare (regardless of similarity to a traumatic event) in the present tense and with as much sensory detail as possible. Participants next read the nightmares aloud, and trauma-related themes are discussed prior to rescripting. Participants then are instructed to write their rescription according to the most prominent trauma-related theme or themes. For example, if the nightmare includes a prominent sense of powerlessness, the rescripted content would specifically include increased power in whichever way the participant chooses. In ERRT, several mechanisms of change are purported, including emotional processing through exposure to the nightmare (similar to PE for PTSD), cognitive changes through discussing trauma-related themes inherent in nightmare content (similar to CPT for PTSD), and rescription of the nightmare content. Nightmares may also change through modification of sleep behaviors. Taking steps to improve sleep quality and quantity may increase the individual’s coping resources and decrease distress. Utilization of relaxation techniques may reduce cognitive and physiological arousal close to bedtime, which may reduce the likelihood of experiencing a nightmare and also may improve sleep.
To date, two RCTs comparing ERRT to a waitlist control group have been conducted in community-based samples, and a pilot study has been conducted in a veteran sample. In the first study examining a community sample of traumatized adults , the ERRT group (n = 21) demonstrated positive effects on self-report measures of nightmares, sleep, and PTSD compared to control (n = 22) . A second study was conducted to replicate these findings and to assess changes in physiological responses to nightmare content through a script-driven imagery paradigm [49, 50], a method shown to adequately evoke emotional and psychological reactions to nightmare cues . In addition to replicating the findings from the first RCT, results showed the treatment group demonstrated significant reductions on all measures of physiological arousal (skin conductance, heart rate, and facial electromyogram) associated with the personal nightmare scripts at 1-week posttreatment, whereas the control group did not change . The pilot study evaluating ERRT among veterans  (N = 19) also found improvements in nightmare frequency and severity, depression, sleep quality, and insomnia severity at the 1-week and 2-month follow-ups with 50% of the sample reporting no nightmares in the previous week at the follow-up. Thus, evidence suggests that treatments specifically targeting nightmares have positive impacts on nightmares, sleep problems, and daytime PTSD symptoms in relatively few sessions. However, only limited research has been conducted to examine the extent to which these treatments are effective in military populations.
Among veterans, two case studies and an open trial have been reported on a variant of ERRT, called imagery rescripting and exposure therapy (IRET) [61, 62] (see Table 24.1). IRET was developed specifically to address the needs of the veteran population by allocating more time to sleep management, exposure to the nightmare, and rescripting for a total of six sessions. Wanner and colleagues  report on two case studies for veterans participating in the open trial reported by Long et al. . The open trial examined IRET in a sample of 33 veterans, primarily from the Vietnam Era. Both studies found significant and clinically meaningful improvements in nightmare frequency, sleep, and PTSD symptoms.
Cognitive Behavioral Therapy for Insomnia Combined with Nightmare and PTSD Treatments
Additional studies combining nightmare treatments with insomnia treatments also have been conducted. Research supports the use of several behavioral techniques for treating primary and secondary insomnia . The American Academy of Sleep Medicine provides a review of 48 studies utilizing these approaches from 1970 to 1997 , 37 studies from 1998 to 2004 , and practice parameters for psychological and behavioral approaches based on the literature . The most widely studied and supported approaches for insomnia include stimulus control (using the bed and bedroom for sleep and sexual activity only), relaxation training, sleep restriction (limiting time in bed to more closely approximate the individual’s sleep ability), sleep hygiene education, and cognitive therapy. Treatment packages of cognitive behavioral therapy for insomnia (CBT-I) may include any combination of these elements. Research has examined CBT-I combined with nightmare treatment and broader treatments for PTSD.
CBT-I in PTSD
Gellis and Gehrman  conducted a pilot study of five sessions of CBT-I in eight veterans with chronic PTSD. The authors found that insomnia severity and sleep parameters measured with the sleep diary demonstrated significant improvements. Objectively measured sleep with actigraphy (small, wrist-worn devices that monitor movement and provide estimates of sleep continuity) did not show improvements. Although statistically significant improvements in depression were observed, there were no improvements in nightmares or other PTSD symptoms. The authors hypothesize that although subjectively reported sleep improved, the presence of PTSD and nightmares may limit the effectiveness of CBT-I in individuals with PTSD.
In an RCT with 45 veterans with PTSD, Talbot and colleagues  compared CBT-I with a monitor-only waitlist control. They found that CBT-I improved insomnia among these veterans with PTSD. Large changes were found with the sleep diary, polysomnography with regard to total sleep time, several other self-report sleep measures, as well as social and work functioning. Although PTSD improved it did not improve more than the monitoring group. All participants were required to be enrolled in treatment for PTSD (medications or psychotherapy), and so the impact of CBT-I alone is unclear.
CBT-I Combined with Nightmare Treatments
More recently, researchers have proposed that treatment approaches that combine elements of nightmare treatments (i.e., exposure, rescripting, and imagery rehearsal) and CBT-I (i.e., sleep restriction ) may be optimal for treating sleep disturbances in trauma-exposed adults. One case study , one noncontrolled trial , and a randomized parallel group experiment  of IRT combined with CBT-I in veterans have been reported. Another study implemented CBT-I and included IRT only as an adjunctive intervention for some participants  (see Table 24.1). In the case study , a 69-year-old Vietnam veteran who had been experiencing nightmares for 35 years completed two sessions of CBT-I followed by three sessions of IRT . At the end of treatment, his nightmare frequency reduced from 17 per week to 5 per week. He also reported decreased nightmare intensity and improved sleep. In an open trial, Harb et al.  examined a six-session treatment protocol that combined CBT-I and IRT in seven veterans of the Iraq war. The treatment included three sessions of psychoeducation, progressive muscle relaxation, and sleep hygiene (including stimulus control) followed by three sessions of IRT (one session for writing the target nightmare, one session for brainstorming changes to the nightmare, and one session for rescripting the nightmare). Overall, improvements were found in nightmare frequency, sleep quality, and PTSD symptom severity. Further, the authors divided the sample into responders (n = 4) and nonresponders (n = 3). The authors noted that the nonresponders were experiencing certain issues that may have impacted treatment outcome including pending redeployment, traumatic brain injury, alcohol use, nightmares of perpetration, and guilt. Thus, research is needed to examine potential predictors of response to IRT.
In the randomized parallel group study , 22 veterans with PTSD from a variety of eras were randomized to either the treatment condition (three sessions of CBT-I followed by three sessions of IRT) or to care as usual (including a variety of mental health treatments such as medication management/supportive therapy and unspecified individual therapy sessions). From baseline to immediate posttreatment, veterans in the CBT-I with IRT condition demonstrated greater improvements in nightmare frequency, sleep parameters, insomnia severity, sleep quality, and PTSD symptoms than the control group, providing additional support for the efficacy of treatments targeting sleep disturbances in veterans with PTSD.
ERRT also has been combined with CBT-I to treat trauma-related nightmares. From its inception, the ERRT manual has included components to improve sleep directly, including relaxation training (e.g., diaphragmatic breathing and progressive muscle relaxation), sleep hygiene, and stimulus control [71, 72]. However, participants typically have been asked to select any poor sleep hygiene habit to modify during the course of treatment and any positive sleep hygiene habit to maintain. Thus, stimulus control and sleep restriction, which are considered integral components of CBT-I , are not always selected as an integral component of ERRT .
Swanson et al.  modified ERRT to include five sessions of CBT-I (including stimulus control and sleep restriction for every patient) prior to nightmare rescripting for a total of ten sessions of therapy. This protocol was evaluated in an open trial with a sample of nine combat veterans with PTSD (see Table 24.1). Results demonstrated an average reduction in nightmares per week of 50% and an average reduction in total nightmare distress of 46%. Interestingly, data indicate a substantial decrease in nightmare frequency prior to implementation of exposure or rescription of the nightmare followed by continuing improvements. Furthermore, the average insomnia ratings were in the subthreshold range at the conclusion of treatment. However, PTSD symptom ratings did not demonstrate a significant decrease. Further research is needed to determine the optimal combinations of treatment components and to determine for which symptom presentations the treatments are needed .
Another study conducted by Margolies et al.  randomized 40 veterans with PTSD into either CBT-I with adjunctive IRT or a waitlist control . IRT was implemented with 13 participants who reported nightmares but only 6 of those participants opted to continue IRT in subsequent sessions. Veterans who were placed into the CBT-I/IRT group reported significant increases in overall sleep quality and decreased nighttime PTSD symptoms, PTSD symptom severity, and depressed mood when compared to the control. These results support the use of sleep-based treatments for individuals experiencing sleep and PTSD disturbances and support the notion that sleep disturbances may be exacerbating symptoms of PTSD .
CBT-I Combined with PTSD Treatment
Few studies have examined treatment approaches that combine sleep interventions with broader PTSD treatments. Because nightmares and problems falling and staying asleep are symptoms of PTSD, it is thought that these symptoms will remit along with the treatment of the constellation of re-experiencing, avoidance, and hyperarousal symptoms. As discussed further below, this approach has been the topic of some debate.