A clinical approach to nightmares and bad dreams in cognitive–behavioural therapy

Summary

Clinicians trained in cognitive–behavioural therapy (CBT) are frequently not trained to work with dreams. Given the high prevalence and impact of nightmares and bad dreams, empowering CBT therapists to effectively work with these sleep phenomena is crucial to improve therapeutic outcomes. This article briefly outlines a cognitive–behavioural model of dreams and reviews some clinical guidelines for directly and indirectly addressing nightmares and bad dreams in CBT practice.

Nightmares are typically defined as lengthy, elaborate and well-remembered dreams that (a) seem real and provoke anxiety, fear or other dysphoric emotions; (b) generally occur during the second half of the major sleep episode when dreaming is longer and more intense; and (c) usually terminate with awakening and quick return of full alertness. Clinically, nightmares are likely to trigger dysphoric emotions that may persist and contribute to difficulty returning to sleep or even long-lasting daytime distress (American Psychiatric Association 2013). When working with these sleep phenomena, two conceptual distinctions should be noted: first, nightmares that do not lead to awakening are best described as ‘bad dreams’; and second, nightmare frequency does not necessarily equate with nightmare-induced distress (Zadra 2000).

Nightmares/bad dreams and mental health

Nightmares may be understood in three essential ways (Zadra 2000; Spoormaker 2008): as normative sleep experiences (with up to 70% of the general population reporting occasional nightmares); as specific diagnostic criteria (e.g. repeated nightmares in nightmare disorder, recurrent ‘replicative nightmares’ in post-traumatic stress disorder); or as phenomena comorbid with several medical conditions (e.g. cancer, coronary heart disease) and mental disorders (e.g. depressive disorders (including grief during bereavement), anxiety disorders, conduct disorder, psychosis).

A neurocognitive model of dreams argues that dreaming provides a dynamic sequence of contexts to elicit the formation of fear extinction memories. From this point of view, bad dreams are assumed to be examples of resolved fear extinctions (i.e. a fear memory is neutralised), whereas nightmares are thought to be examples of ineffective fear extinctions (Levin 2007; Spoormaker 2008). In fact, cognitive neuroscience research (Siclari 2020) has suggested that the emotional intensity of a waking event is one of the strongest predictors of incorporation into dreams (e.g. traumatic experiences usually manifest as post-traumatic nightmares, which can persist for decades). Moreover, the empirically supported stress acceleration hypothesis of nightmares asserts that idiopathic nightmares originate in early adverse experiences (particularly during the ‘infantile amnesia period’, which ends around age 3 years 6 months), through the disruption of an affect network (namely, a fear circuit that includes the amygdala, hippocampus and medial prefrontal cortex) hypothesised to regulate fear extinction during rapid eye movement (REM) sleep (Nielsen 2017).

Accordingly, nightmare frequency appears to be more highly correlated than bad dream frequency with measures of psychological well-being (e.g. psychopathological symptoms, stress and personal adjustment), suggesting that nightmares represent a rarer and more severe expression of the same basic phenomenon (Zadra 2000). Furthermore, global nightmare distress seems to result from an interaction between nightmare frequency and heightened emotional reactivity measured as neuroticism (Schredl 2019). Also, chronic nightmares seem to develop through the interplay of affective hyperarousal and impaired fear extinction, where trait distress is likely to be triggered by trauma, adversity, thought suppression and, potentially, sleep-disordered breathing (Gieselmann 2019).

Despite such evidence, a therapist trained in cognitive–behavioural therapy (CBT) is frequently not used to working with dreams, and may overlook valuable opportunities to address the richness of thematic content provided by dreams (Freeman 2002). To prevent such disregard and empower CBT therapists to work more effectively with patients’ bad dreams and nightmares, in the next paragraphs a cognitive–behavioural approach to dreams is reviewed and its clinical implications are discussed.

A CBT perspective on dreams

Challenging traditional psychoanalytic assumptions to therapeutic dream work (e.g. ‘dreams are wish fulfilment’, ‘dreams are attempts at problem-solving’), Beck (1971, 2002) outlined a first cognitive model of dreams. According to Beck, some dreams reflect an individual's dysfunctional attitudes by bringing unrealistic thoughts and expectations to consciousness; in therapy, these can be used to clarify their problem and to help them recognise their idiosyncratic patterns of distorted thinking. Therefore, just like unrealistic cognitions are salient in waking ideation (e.g. in depression, people ruminate about their inadequacies, losses and desolate future; in anxiety, they worry about hypothetical dangers; in paranoia, they struggle with the fear of being abused by others), these dominant patterns exert their influence in nightmares and bad dreams, with depression, anxiety and paranoia bringing a tendency to dream respectively of themes of failure, danger or abuse.

Following the same line of thought, Freeman & White (2002) conceptualised dream content and themes as opportunities to understand and challenge the patient's distressing cognitions, with a resultant positive affect shift. Specifically, they drew up a series of guidelines to assist the clinician in utilising dreams within the context of CBT (for a detailed description, see Freeman & White, 2002). Some of those guidelines are particularly informative of the clinical rationale and procedures for working with patients’ nightmares and bad dreams from a CBT perspective:

• • ‘The thematic content of the dream is idiosyncratic to the dreamer and must be viewed within the context of the dreamer's life.’ The thematic meaning-making of a dream should be explored within the context of the patient's current concerns, life experience and cognitive schemas (e.g. emotional deprivation, loss/abandonment, defectiveness/shame, mistrust/abuse). In fact, dream content can illustrate a patient's schema (assessment phase) and provide powerful imagery for use in experiential intervention (Young 2003).

• • ‘The specific language and imagery of the dream are important to the meaning.’ The voice tone, visual images and quality of verbal and non-verbal language experienced in nightmares or bad dreams influence the patient's expressed affect.

• • ‘The affective responses to the dreams can be seen as similar to the dreamer's affective responses in waking situations.’ When recalling a nightmare or bad dream in therapy, the patient should be guided to identify where and how the emotions were being generated. Likewise, the patient may wake with an affective residue that needs to be assessed as a potential trigger for maladaptive emotion-driven behaviours.

• • ‘Dreams can be used when the patient appears “stuck” in therapy.’ When a particular direction of therapy seems lost or blurred, sensitive use of dream material (e.g. a dream image) can be a valuable adjunctive tool for re-engagement, possibly because dreams are so personal and such a familiar part of their living experience. For instance, the image of losing one's voice may be reframed to recall the patient's need to develop assertiveness and cultivate a self-determined ‘inner voice’; or the image of a lost loved one may be cherished to recall the therapeutic intention of delicately facing grief-related memories and/or avoidance.

• • ‘The patient should try to capsule and to draw a “moral” from the dream.’ The patient should be encouraged to draw a helpful and realistic conclusion from the dream. This personal ‘learning’ can then be used in therapy to further develop the patient's coping repertoire. For example, nightmares of falling, losing one's voice, being chased or crying alone may depict a patient's concern regarding their blocked emotional needs and perceived lack of control in a given context of their lives.

Addressing nightmares and bad dreams in clinical practice

Nightmares and bad dreams are common complaints in people with diverse mental disorders that should be tackled in CBT, especially during initial assessment and while building rapport (i.e. ‘if you don't ask meaningful questions, you will hardly get meaningful answers’). From a CBT perspective (Beck 1971, 2002; Freeman 2002), looking at the themes of patients’ nightmares or bad dreams through guided discovery facilitates the identification of core fears and related defensive strategies. Likewise, the images evoked by such dreams may provide inspiring metaphors to compassionately engage with distress (e.g. a patient suffering from panic disorder who has a recurrent nightmare of a big truck coming in their direction to run them over but wakes up before it reaches them may find that image useful in exposure experiments by recalling that although extremely scaring and intense, panic symptoms may be faced with no catastrophic consequence). Note that coping-based interventions may be needed to prevent or reduce the negative impact of nightmare-induced distress on daily functioning. And finally, when seeking to reduce nightmare frequency, indirect CBT (e.g. recording of nightmares, relaxation) may be useful, but nightmare-focused CBT (exposure/systematic desensitisation, and cognitive restructuring/imagery rehearsal therapy) appears to be more effective (Lancee 2008). Box 1 lists further reading on working with nightmares and dreams in therapy.

Although evidence is lacking on the relative effectiveness of CBT in reducing nightmare frequency and distress in comparison with other treatment approaches, a meta-analysis of randomised controlled trials of psychological (mostly including CBT-based techniques such as exposure, relaxation, recording and image rehearsal therapy with sleep hygiene) and pharmacological treatments for nightmares found an overall medium improvement (e.g. reduction of sleep complaints, nights with nightmare, and nightmare frequency, distress or intensity) for all studies combined, as well as for psychological interventions alone and for prazosin alone (Augedal 2013; for a review of pharmacological treatments of nightmares and dream vividness, see Siclari et al, 2020). Nevertheless, experimental longitudinal research is warranted to identify the mechanisms of change by which different interventions may operate in reducing nightmare frequency and nightmare-induced distress in patients with different diagnoses.