An insomnia diagnosis needs to be differentiated from normal sleep variants, circadian rhythm sleep disorders, insufficient sleep, and other sleep disorders. As there is a continuum of sleep need, short sleepers can be differentiated from insomnia by the absence of sleep initiation and maintenance complaints and daytime impairments. Two circadian rhythm sleep disorders, delayed sleep phase type and advanced sleep phase type, are characterized by sleep difficulties arising from misalignment between preferred sleep schedules and the endogenous circadian rhythm. Delayed sleep phase type is differentiated from insomnia because individuals with insomnia feel sleepy at their desired bedtime but are unable to sleep regardless of their bed and rise times whereas an individual with delayed sleep phase type will report normal sleep duration with little to no sleep initiation or maintenance complaints when allowed to sleep on their internally preferred schedule. Delayed sleep phase type is more common among adolescents and young adults. Individuals with insomnia report sleep maintenance problems and early morning rise times regardless of sleep schedule whereas sleep duration and continuity is normal for an individual with advanced sleep phase type when sleeping in phase (i.e., early bed and wake time). Advanced sleep phase type is more common in older adults than children or young adults. A differential diagnosis may be enhanced by having patients complete self-reported circadian rhythm characteristics such as the Horne and Östberg Morningness-Eveningness Questionnaire (MEQ) [55] or the Munich ChronoType Questionnaire (MCTQ) [56]. Insufficient sleep syndrome is differentiated from insomnias by the presence of excessive daytime somnolence, which is not typical in chronic insomnia. In addition, the chronic sleep restriction associated with insufficient sleep syndrome is volitional whereas those with chronic insomnia have reduced total sleep times due to sleep initiation and maintenance problems in the context of adequate opportunity and circumstances for sleep. Finally, as discussed above, other sleep disorders including sleep-disordered breathing and periodic limb movement disorder may manifest as sleep maintenance problems, and restless legs syndrome may contribute to sleep initiation or maintenance complaints.

Nationally representative samples of adults in the USA indicate use of prescription medication to treat insomnia has increased over time [15, 57] with approximately 3 % of adults reporting the use of medication to treat insomnia symptoms [15] and 25 % of these individuals treating their insomnia for 4 months or longer [6]. Hypnotics remain first-line treatments for chronic insomnia in primary care because they are widely available, are easy to prescribe, provide immediate symptomatic relief, and are efficacious in the short term [58]. When addressed in primary care settings, insomnia is typically treated with FDA-approved hypnotics including benzodiazepine receptor agonists, nonbenzodiazepine receptor agonists, melatonin receptor agonists, antihistamines, and antidepressants [59–64] (Table 10.1). Benzodiazepine and nonbenzodiazepine receptor agonists generally have similar efficacy on PSG and sleep diary outcomes [65]. Several medications with sedating properties used to treat anxiety, depression, and seizures are commonly used off-label to treat insomnia, however, efficacy data is limited [59]. Patients often seek over-the-counter medications containing antihistamines or natural products (e.g., melatonin, l-tryptophan, valerian), but the FDA does not regulate herbs and supplements; therefore, patients may be unaware that the dose and purity of a given product is not monitored. Over-the-counter medications and supplements are generally not recommended due to limited efficacy and safety data [27]. Available data on antihistamines suggests that tolerance to these medications likely develops quickly [66]; therefore, use should be short term. The limited evidence for evaluating exogenous melatonin for insomnia indicates relatively modest benefit as reflected in relatively small effect sizes for reductions in sleep onset latency and increased sleep efficiency and total sleep time relative to benzodiazepine and nonbenzodiazepine receptor agonists [60]. Nonbenzodiazepine receptor agonists (1.23 % of the US population) are the most commonly prescribed medications for insomnia followed by trazodone (0.97 %), benzodiazepines (0.40 %), quetiapine (0.32 %), and doxepin (0.12 %). It is not uncommon to be prescribed additional sedative medications, with one study indicating that up to 55 % of patients take an additional sedative hypnotic (e.g., opioid or benzodiazepine) and 10 % take three or more [15]. It is noteworthy that there is no medication prescribed in the USA with FDA approval for use in pediatric patients for the treatment of insomnia, but pediatric patients frequently are prescribed medications off-label to treat insomnia, with antihistamines (33 %), benzodiazepine (26 %), alpha-2 agonists (e.g., clonidine, guanfacine; 15 %), and antidepressants (6 %) commonly recommended [67]. The timing of medication is also critical for optimal treatment, with minimal side effects occurring when taken within close proximity to bedtime. However, estimates indicate that up to 20 % of individuals use their hypnotic medication in the middle of the night [68]. This practice can be quite problematic given that an adequate opportunity to sleep is recommended when using hypnotics given their long half-life (e.g., 1.5–100 h for benzodiazepines and 1–6 h for nonbenzodiazepine receptor agonists). Individuals utilizing medication in the middle of the night are vulnerable to next-day hangover effects, including cognitive impairment and psychomotor impairment. Furthermore, benzodiazepines carry the risk of tolerance, dependence, and withdrawal effects such as rebound insomnia [69]. Relative to placebo, nonbenzodiazepine receptor agonists and sedating antidepressants are associated with higher rates of headache, somnolence, dizziness, and nausea [65]. Both benzodiazepine and nonbenzodiazepine receptor agonists are associated with complex sleep-related behaviors such as sleep walking, aggressive behavior, eating, and driving. Therefore, care is required in coordinating pharmacological interventions to minimize potentially sedating side effects, which may increase risk of injury.

In the short term hypnotics serve as an appropriate measure for treating insomnia complaints. However, insomnia is often chronic, making short-term use of hypnotics a less favorable treatment option. In addition, there is limited evidence to support the long-term use of hypnotics with many placebo-controlled studies lasting for only 4 weeks or less in non-elderly adults [65, 70] and very few studies lasting 12 months [71–73]. Hypnotics may be contraindicated in individuals with medical histories including alcohol or sedative abuse/dependence and untreated sleep apnea [74]. Many patients report a strong desire to discontinue use of hypnotic medications [75], and this should be done gradually (e.g., 25 % reduction/week) under the supervision of a physician to minimize withdrawal effects (e.g., rebound insomnia).

Non-pharmacological interventions for insomnia are intended to promote greater control over sleep, reduce emotional distress, reduce variability in sleep schedules, consolidate sleep, and enhance sleep efficiency. These interventions have well-established short-term and long-term efficacy [76, 77] and seem to have fewer side effects [78] than pharmacological interventions for insomnia. Behavioral strategies including sleep restriction therapy and stimulus control therapy (Table 10.2) meet established “standard” treatment criteria as defined by the American Psychological Association criteria for empirically validated treatments [79] or consensus recommendations made by the American Academy of Sleep Medicine as effective and evidenced-based interventions for insomnia [80]. These strategies are typically combined and supplemented with cognitive therapy and sleep education as a multicomponent cognitive behavioral therapy for insomnia (CBT-I), which is often considered more acceptable than pharmacological treatment by patients [81]. The efficacy of non-pharmacological interventions has been reviewed in several meta-analyses [76, 77, 82–85] which have consistently indicated that 50–70 % of treated patients benefit from intervention with medium to large effect sizes observed for sleep onset latency, wake time after sleep onset, sleep efficiency, and sleep quality ratings (Fig. 10.2). Effect sizes for total sleep time at the end of intervention are small but improve during follow-up periods with medium effect sizes [82, 84, 85]. Initial treatment gains maintained during follow-up periods for as long as 2–3 years [78, 86, 87]. Although CBT-I yields effective changes in sleep, categorical response and remission treatment outcomes are difficult to determine because no universally accepted criteria have been established [30]. Several trials have utilized the ISI [88–90], sleep efficiency >85 % [87], or a combination of sleep efficiency and sleep quality ratings [91], with overall report of response rates of 52–65 % and remission rates of 32–55 % [89–91]. Indeed the establishment of guidelines will guide clinical practice and future effectiveness trials. The evaluation of treatment outcomes has also unfortunately focused primarily on sleep-related variables rather than daytime functioning, a cardinal feature of insomnia and a principal factor influencing those seeking treatment. It is critical to evaluate daytime functioning to further assess response and remission, and the use of an instrument like the PROMIS Sleep-Related Impairment questionnaire may help evaluate response and remission multidimensionally.

The efficacy and safety of CBT-I have been documented not only in numerous controlled trials of patients with primary insomnia but also in patients with insomnia and co-occurring conditions commonly seen in primary care, such as depression and other psychiatric disorders [92, 93], chronic pain [94–96], breast cancer [97, 98], and hypnotic dependence [75]. Head-to-head comparisons of CBT-I and pharmacotherapy found that ratings of sleep quality were equivalent or better for CBT-I treated patients at posttreatment and superior 1 and 6 months following treatment [78, 99]. Although smaller in magnitude, improvements in objective sleep quality measured by actigraphy [91] and polysomnography [87, 90] have also been consistently found with CBT-I. CBT-I offers several potential advantages over pharmacotherapy, including greater patient preference [81, 100], equal to superior outcomes after 6–8 weeks of therapy [78, 99, 101], more evidence of sustained treatment efficacy (e.g., [86, 87]), fewer adverse side effects and contraindications, and no concerns about drug interactions. However, sleep restriction therapy, a common component of CBT-I, may increase somnolence and decrease psychomotor vigilance [102]. Therefore, it is recommended that providers advise patients to exercise caution when engaging in sleep-sensitive tasks (e.g., driving, operating machinery). Moreover, providers are also encouraged to consider the limitations of CBT-I’s primary behavioral components, sleep restriction, and stimulus control for individuals with conditions that may be exacerbated by sleep resection, including bipolar disorder, epilepsy, parasomnias, and psychotic disorders. Also stimulus control may not be appropriate for individuals who are disabled, cannot get out of bed easily unassisted, or are at risk of accidental injury.

Despite several strengths, dissemination of CBT-I is limited by the dearth of clinicians trained to deliver these interventions [103], access to sleep disorders centers, and cost. Relative to pharmacotherapy, CBT for insomnia is perceived to be more burdensome, with a typical course requiring 5–6 face-to-face sessions over a 6–7-week period [77]. Another significant barrier is that providers trained to deliver CBT-I are typically not located in primary care. Instead, patients who prefer CBT-I over hypnotics must find one of the few qualified providers in the country (fewer than 200 providers are certified in Behavioral Sleep Medicine nationwide [http://​www.​absm.​org/​BSMSpecialists.​aspx], a designation that denotes expertise in CBT for insomnia), most of whom are located in departments of psychology/psychiatry or affiliated with sleep disorders centers in major urban areas. Finally, CBT-I is often criticized as being more costly than pharmacotherapy. Using 2002 reimbursement rates, one study estimated that 8 weeks of group (4–6 patients) CBT-I costs $279.92 compared with $255.30 for 100 tablets of zolpidem 10 mg. These estimates predate the availability of zolpidem in generic form, meaning that short-term medication costs are substantially lower. On the other hand, they also fail to consider that CBT-I is time limited whereas hypnotic use frequently exceeds 8 weeks [104]. No study has conducted prospective comparative cost-effectiveness analyses of available insomnia therapies despite the importance of such information [105]. Addressing these major barriers—treatment modality, provider, and cost—is likely critical for CBT-I to be adopted as a first-line insomnia therapy.