The diagnosis of insomnia is derived primarily from a detailed clinical evaluation of the patient’s subjective complaint (see Table 37.2). The sleep history should cover the type of complaint (initial, middle, late insomnia), its duration (acute vs. chronic), and course (recurrent, persistent); typical sleep schedule; functional analysis of precipitating, perpetuating, and alleviating factors; perceived consequences and functional impairments; and the presence of medical, psychiatric, or environmental contributing factors. A complete history of alcohol and drug use and prescribed and over-the-counter medications is also essential [52–54].

The use of a sleep diary is essential in the evaluation of insomnia (see Table 37.3). A daily sleep diary is very helpful to document the nature and severity of insomnia, identify behavioral and scheduling factors that may perpetuate insomnia, and monitor treatment compliance and progress [55]. The Insomnia Severity Index [56] is a brief questionnaire that provides a global measure of the patient’s perception of insomnia severity and its impact on daytime functioning (see Table 37.4). Several additional measures of insomnia symptoms, fatigue, anxiety, and depressive symptomatology may also provide useful complementary information in the evaluation of insomnia [53]. A more comprehensive psychological evaluation may be necessary for patients with suspected psychiatric disorders.

Although polysomnography is not indicated for the routine evaluation of insomnia, it is often necessary to rule out other sleep disorders that might contribute to the insomnia complaint (e.g., periodic movements during sleep and sleep apnea) [57]. PSG can also be particularly useful in suspected case of paradoxical insomnia or when a patient is unresponsive to treatment. The role of actigraphy in insomnia evaluation and treatment monitoring is not well established. Although it may represent a useful adjunct, actigraphy is not clinically indicated for routine assessment, diagnosis, or management of insomnia. Nonetheless, it is useful for examining night-to-night variability and for identifying individuals with circadian rhythm disorders. It has also been used to document treatment adherence and outcome in clinical trials of behavioral therapies for insomnia [58]. Although a potentially useful complement to self-report and PSG measures, actigraphy devices and algorithms are not all equivalent and there may be significant variability in the reliability and validity of sleep-wake data derived from different devices.

Treatment goals and indications. Psychological and behavioral therapies for insomnia include sleep restriction, stimulus control therapy, relaxation-based interventions, cognitive strategies, sleep hygiene education, and combined cognitive behavioral therapy. A summary of those interventions is provided below and in Table 37.5; more extensive descriptions are available in other sources [52, 59]. The main objectives of psychological and behavioral approaches are to alter factors that perpetuate or exacerbate sleep disturbances. Such features may include hyperarousal, sleep scheduling factors, poor sleep habits, and misconceptions about sleep and the consequences of insomnia. Although numerous factors can precipitate insomnia, when it becomes a persistent problem, psychological and behavioral factors are almost always involved in perpetuating it over time, hence the need to target those factors directly in treatment. The primary indication for behavioral treatment is in the management of persistent insomnia, with evidence available for both primary and comorbid insomnia.

Sleep restriction. Poor sleepers often increase their time in bed in a misguided effort to provide more opportunity for sleep, a strategy that is more likely to result in fragmented and poor quality sleep. Sleep restriction consists of curtailing the amount of time spent in bed to the actual amount of sleep [60]. For example, if a person reports sleeping an average of 6 h per night out of 8 h spent in bed, the initial sleep window (i.e., from initial bedtime to final arising time) would be set at 6 h. Subsequent adjustments to this “sleep window” are based on sleep efficiency (SE) for a given period of time (usually the preceding week); time in bed is increased by about 20 min for a given week when SE exceeds 85 %, decreased by the same amount of time when SE is lower than 80 %, and kept stable when SE falls between 80 and 85 %. Periodic (weekly) adjustments are made until optimal sleep duration is achieved. Changes to the prescribed sleep window can be made at the beginning of the night (i.e., postponing bedtime), at the end of the sleep period (i.e., advancing arising time), or at both ends. To prevent excessive daytime sleepiness, time in bed should not be reduced to less than 5 h per night in bed. This procedure leads to improvements of sleep continuity through a mild sleep deprivation and reduction of sleep anticipatory anxiety. Due to potential residual daytime sedation, sleep restriction should be used with caution with patients operating heavy equipment or required to drive long distances [61]. Sleep restriction is contraindicated in patients with a bipolar disorder, with seizures, or with some parasomnias (sleep walking, night terrors).

Stimulus control therapy. Individuals with insomnia may develop apprehension around bedtime and come to associate the bedroom with frustration and arousal rather than with sleep. Stimulus control therapy [62] consists of a set of instructions designed to strengthen the association between temporal (bedtime) and environmental (bed and bedroom) stimuli and rapid sleep onset and to establish a regular circadian sleep-wake rhythm. These instructions are as follows: (a) going to bed only when sleepy, (b) getting out of bed when unable to sleep (e.g., after 20 min), going to another room, and returning to bed only when sleep is imminent; (c) curtailing all sleep-incompatible activities (i.e., no TV watching and problem solving in bed); (d) arising at a regular time every morning regardless of the amount of sleep the night before; and (e) avoiding daytime napping. Despite the straightforward nature of these recommendations, the main challenge for most patients is to comply with all of them, which is essential to reverse the conditioning processes perpetuating insomnia. Caution is advised in using some of these procedures (e.g., getting out of bed when unable to sleep) with the frail elderly who may be at risk for falls.

Relaxation-based interventions. Relaxation is probably the most commonly used nondrug therapy for insomnia. Some relaxation methods (e.g., progressive muscle relaxation) focus primarily on reducing somatic arousal (e.g., muscle tension), whereas attention-focusing procedures (e.g., imagery training and meditation) target mental arousal in the forms of worries or intrusive thoughts. Mindfulness-based stress reduction [63] is another relaxation variant, and all these methods are fairly equivalent in terms of efficacy for insomnia. The most critical issue is to practice diligently and daily the selected method for at least two to four weeks. Professional guidance is often necessary in the initial phase of training.

Cognitive therapy. This psychotherapeutic method seeks to alter dysfunctional sleep cognitions (e.g., beliefs and expectations) and maladaptive cognitive processes (e.g., excessive self-monitoring) through Socratic questioning and behavioral experiments. The basic premise of this approach is that appraisal of a given situation (sleeplessness) and excessive monitoring of sleep-related cues (e.g., fatigue and time left for sleep) can trigger an emotional response (fear, anxiety) that is incompatible with sleep. For example, when a person is unable to sleep at night and worries about the possible consequences of sleep loss on the next day’s performance, this can set off a spiral reaction and feed into the vicious cycle of insomnia, emotional distress, and more sleep disturbances. Cognitive therapy is designed to identify dysfunctional cognitions and reframe them into more adaptive substitutes in order to short-circuit the self-fulfilling nature of this vicious cycle. Treatment targets may include unrealistic expectations (“I must get my 8 h of sleep every night”) and amplification of the consequences of insomnia (“Insomnia may have serious consequences on my health”) [52]. Cognitive therapy is particularly useful to modify these maladaptive cognitions and to teach patients more adaptive skills to cope with insomnia [64].

Multicomponent therapies. Despite some unique features, the interventions described above can be effectively combined together. There is a general preference among investigators and clinicians for combining multiple interventions, with cognitive behavioral therapy (CBT) becoming the standard approach in the field [58]. The most common combination involves a behavioral (stimulus control, sleep restriction, and, sometimes, relaxation), a cognitive, and an educational (sleep hygiene) component, usually referred to as CBT. Such combination is often preferred to address the different components presumed to perpetuate insomnia.

Evidence for efficacy. Several meta-analyses [65–67] and systematic reviews commissioned by the American Academy of Sleep Medicine [58, 68] have summarized the findings from clinical trials evaluating the efficacy of psychological and behavioral therapies for persistent insomnia. Evidence from these sources shows that treatment produces reliable changes in several sleep parameters, including sleep-onset latency (effect sizes ranging from 0.41 to 1.05), number of awakenings (0.25–0.83), duration of awakenings (0.61–1.03), total sleep time (0.15–0.49), and sleep quality ratings (0.94–1.14). Based on Cohen’s criteria, the magnitude of those therapeutic effects is large (i.e., d > 0.8) for sleep latency and sleep quality and moderate (i.e., d > 0.5) for other sleep parameters. When transformed into a percentile rank, these data indicate that approximately 70–80 % of patients with insomnia achieve a therapeutic response with psychological and behavioral therapies.

In terms of absolute changes, treatment reduces subjective sleep-onset latency and time awake after sleep onset from averages of 60–70 min at baseline to about 35 min at post-treatment, and total sleep time is increased by 30 min, from 6 to 6.5 h after treatment. Thus, for the average insomnia patient, treatment effects may be expected to reduce sleep latency and time awake after sleep onset by about 50 % and to bring the absolute values of those sleep parameters below or near the 30-min cutoff criterion initially used to define insomnia. Treatment effects are similar for sleep-onset and sleep-maintenance problems, although fewer studies have targeted early morning awakening problems. Overall, findings from meta-analyses represent fairly conservative estimates of treatment effects as they are based on averages computed across all nonpharmacological interventions and insomnia diagnoses (i.e., primary and comorbid). On the other hand, although the majority of patients benefit from treatment, only about 40 % achieves clinical remission [58].

Treatment outcome has been documented primarily with prospective daily sleep diaries, although several studies have also complemented those findings with data from polysomnography [69–71] and with wrist actigraphy [72, 73]. In general, the magnitude of improvements is smaller on PSG measures, but those changes tend to parallel sleep improvements reported on daily sleep diaries. PSG findings indicate that treatment does not only alter sleep perception, as measured by patient-reported outcomes, but also produce objective changes on EEG sleep continuity measures. Except for a modest increase in stages 3–4 following sleep restriction, there is little evidence of changes in sleep architecture with psychological and behavioral treatment. In addition to improving sleep continuity parameters, there is also some evidence showing improvements on several secondary endpoints including measures of daytime fatigue, quality of life, and psychological symptoms [58, 73].

Treatment of Comorbid Insomnia. Insomnia is often a pervasive problem among patients suffering from other medical and psychiatric conditions [74]. Although sleep may improve with appropriate treatment of the comorbid condition, sleep disturbances are also likely to persist. Thus, the presence of a comorbid medical or psychiatric disorder should not preclude using a behavioral intervention concomitantly as behavioral factors are often involved in perpetuating or even exacerbating the sleep problem. Evidence from small clinical trials [75, 76] suggests that patients with medical and psychiatric conditions can also benefit from insomnia-specific treatment [77]. Controlled studies have also shown that behavioral treatment is effective for insomnia associated with chronic pain [78], fibromyalgia [79], cancer [80, 81], and various medical conditions in older adults [82, 83]. In general, insomnia symptoms are more severe among patients with comorbid disorders, but the absolute changes on those outcomes during treatment are comparable to those obtained with primary insomnia.

Insomnia in older adults is more likely to be comorbid with another medical or another sleep disorder than to be primary in nature. Recent studies have shown that older adults respond to insomnia treatment, particularly when they are screened for other sleep disorders that increase in incidence in older age (e.g., restless legs syndrome and sleep apnea). A meta-analysis [84] suggested that effect sizes were comparable (moderate to large) for middle-aged and older adults on subjective measures of sleep latency, wake after sleep onset, and sleep quality. Older adults with either comorbid medical or psychological conditions can benefit from sleep-specific treatment [82, 83, 85, 86]. Three clinical trials have shown that a supervised and time-limited withdrawal program, with or without behavioral treatment for insomnia, can facilitate discontinuation of hypnotics among older adults with insomnia who are prolonged users [87–89].

Which insomnia therapies work best? Although there has been no complete dismantling of cognitive behavioral therapies to isolate the relative efficacy of each component, direct comparisons of some of those components indicate that sleep restriction, alone or combined with stimulus control therapy, is more effective than relaxation which, in turn, is more effective than sleep hygiene education alone [58]. Sleep restriction tends to produce better outcome than stimulus control for improving sleep efficiency and sleep continuity, but it also decreases total sleep time during the initial intervention. Although some basic education about sleep hygiene is incorporated to most insomnia treatments, sleep hygiene education produces little impact on sleep when used as the only intervention. A recent study has shown that cognitive therapy alone can be effective in the management of insomnia [64].

There is no strong evidence that a multicomponent approach is more effective than any of its single component. However, the appeal for this multimodal approach may come from the fact that it addresses different facets presumed to perpetuate sleep disturbances. While little information is available about the active treatment mechanisms of cognitive behavioral therapy, some evidence suggests that stimulus control and sleep restriction are particularly effective for improving sleep continuity, whereas changes in sleep-related cognitions are associated with better maintenance of sleep changes over time [90]. With increasing evidence that hyperarousal is implicated in primary insomnia, there is a need for greater attention to identify the biologic as well as psychological mechanisms responsible for sleep changes.

Behavioral and pharmacological therapies can play a complementary role in the management of insomnia. No single treatment is effective with all forms of insomnia or acceptable to all patients. Even among treatment responders, few patients reach complete remission and some residual sleep disturbances often persist even after treatment. Thus, combined approaches should theoretically optimize outcome by capitalizing on the more immediate and potent effects of hypnotics and the more sustained effects of behavioral interventions.

Only a few studies have directly compared the effects of behavioral and pharmacological therapies for insomnia. Three studies compared triazolam to relaxation [91, 92] or sleep hygiene [93], and five investigations compared CBT to temazepam [68], zolpidem [71, 94], or zopiclone [95, 96]. Collectively, findings from these studies indicate that both therapies are effective in the short term, with medication producing faster results in the acute phase (first week) of treatment, whereas both treatments are equally effective in the short-term interval (4–8 weeks). Combined interventions appear to have a slight advantage over single-treatment modality during the initial course of treatment, but it is unclear whether this advantage persists over time. Long-term effects are consistent for the single-treatment modalities; patients treated with CBT maintain their improvements, whereas therapeutic effects are typically lost after discontinuation of medication. Long-term effects of combined interventions are more equivocal. Some studies indicate that a combined intervention (i.e., triazolam plus relaxation) produces more sustained benefits than medication alone [91, 92], whereas others report more variable long-term outcomes [69, 93]. Some patients retain their initial sleep improvements, but others return to their baseline values. As behavioral and attitudinal changes are often essential to sustain sleep improvements, patients’ attributions of the initial benefits may be critical in determining long-term outcomes. Attribution of therapeutic benefits to the hypnotic alone, without integration of self-management skills, may place a patient at greater risk for recurrence of insomnia once medication is discontinued. Thus, despite the intuitive appeal of combining behavioral and medication therapies, it is not entirely clear when, how, and for whom it is indicated to combine these treatment modalities for insomnia. Additional research is needed to evaluate the effects of combined treatments and to examine optimal methods for integrating these therapies.

Comparisons of effect sizes from meta-analyses [65, 67, 97] on different sleep variables indicate that behavioral therapy may have a slight advantage on measures of sleep-onset latency and sleep quality and pharmacotherapy (benzodiazepine receptor agonists), a more favorable outcome on total sleep time. One study examined different sequences of CBT and medication therapies [96]. The best results were obtained when CBT was introduced first in the sequence, but medication was found helpful to improve total sleep time, which may be an important advantage given that one component of CBT (i.e., sleep restriction) reduces total sleep time during the initial course of therapy and could lead some patients to premature therapy discontinuation.

Until more evidence-based treatment guidelines become available, several strategies can be considered for selecting the most appropriate treatment in the clinical management of insomnia. The use of hypnotic medication may be particularly indicated in the initial stage of therapy to break the vicious cycle of insomnia and to provide some rapid relief. On the other hand, CBT is essential to alter perpetuating factors and to teach coping skills. As such, it is an essential treatment component to maximize durability of sleep improvements. Ideally, medications should be discontinued, under supervision, after an initial treatment course of a few weeks. However, given that insomnia may be a recurrent problem, even among those who benefit from treatment initially, it may be necessary to use medications intermittently after the initial acute treatment.

Several different classes of medications are used for insomnia (Table 37.6), including both over-the-counter (OTC) and prescription agents; however, many of these are not approved by the US Food and Drug Administration (FDA) for the treatment of insomnia. Current FDA-approved insomnia medications include a group of benzodiazepine receptor agonists (BZRAs), one melatonin receptor agonist (ramelteon), and one tricyclic antidepressant (doxepin). Although not FDA-approved for the treatment of insomnia, sedating antidepressants have been prescribed widely; other classes of prescription medications used with increasing frequency for their potential sleep-inducing side effects include anticonvulsants and atypical antipsychotics.