Management of Insomnia

The n e w e ng l a n d j o u r na l of m e dic i n e
Clinical Practice
Patrick G. O’Malley, M.D., M.P.H., Editor
Management of Insomnia
Charles M. Morin, Ph.D., and Daniel J. Buysse, M.D.
This Journal feature begins with a case vignette highlighting a common clinical problem. Evidence
supporting various strategies is then presented, followed by a review of formal guidelines, when they exist.
The article ends with the authors’ clinical recommendations.
A 50-year-old woman presents with a 6-month history of difficulty falling asleep and From the School of Psychology and Cen-
staying asleep several nights per week, which affects her work performance. She re- tre de Recherche CERVO–BRAIN Re-
search Center, Université Laval, Quebec,
ports having had mild-to-moderate symptoms of anxiety and depression for the past QC, Canada (C.M.M.); and the Depart-
year. She has hypothyroidism, for which she has received levothyroxine therapy; TSH ment of Psychiatry, University of Pitts-
and thyroid hormone levels were normal when measured the previous month. She burgh Medical Center, Pittsburgh (D.J.B.).
has tried over-the-counter sleep aids (valerian and melatonin), which have had lim- N Engl J Med 2024;391:247-58.
ited effect, and occasionally has tried hypnotic sleep aids (lorazepam and eszopi- DOI: 10.1056/NEJMcp2305655
Copyright © 2024 Massachusetts Medical Society.
clone). She is worried about drug dependence, but also believes that her sleep prob-
lem is getting worse. How would you manage this patient’s insomnia?
CME
The Cl inic a l Probl em
I
nsomnia disorder is characterized by dissatisfaction with sleep
quality or duration associated with difficulty falling or staying asleep and sub-
stantial distress or daytime impairments. The disorder is a sleep disturbance
that occurs 3 nights or more per week, persists for more than 3 months, and is
not the result of inadequate opportunities for sleep.1 It frequently co-occurs with
other medical conditions (e.g., pain) and psychiatric disorders (e.g., depression),
as well as other sleep disorders (e.g., restless legs syndrome and sleep apnea).
Insomnia is the most prevalent sleep disorder in the general population and
among the most frequent issues raised by patients during primary care visits, al-
though it often goes untreated.2 Approximately 10% of adults meet the criteria for
insomnia disorder and another 15 to 20% report occasional insomnia symptoms.3
Insomnia is more prevalent among women and persons with mental or medical
problems, and its incidence increases in middle age and later, as well as during
perimenopause and menopause.3,4 Although the pathophysiological mechanisms
of insomnia disorder are still poorly understood, psychological and physiological
hyperarousal are recognized as core features.
Insomnia can be situational or episodic, but it follows a persistent course in
more than 50% of patients. The first episode typically arises from stressful life
situations, health problems, atypical work schedules, or travel across several time
zones (jet lag). Although most persons resume normal sleep after adjusting to the
precipitating event, chronic insomnia may develop in persons who are vulnerable
to the disorder. Psychological, behavioral, or medical factors often perpetuate
chronic sleep difficulties. For instance, sleeping late in the morning or napping
during the day can initially help persons cope with sleep disturbances; however,
those same practices can exacerbate sleep difficulties over time and become treat-
n engl j med 391;3 nejm.org July 18, 2024 247

The New England Journal of Medicine
Downloaded from nejm.org by JAIRO TABACOW HIDAL on July 17, 2024. For personal use only.
No other uses without permission. Copyright © 2024 Massachusetts Medical Society. All rights reserved.
Key Points
Treatment Approaches to Insomnia

• Insomnia is common, and it frequently occurs when other medical, psychiatric, and other sleep
disorders are present.
• Persistent insomnia is associated with substantial distress, functional impairment, and adverse health
outcomes, including increased risks of major depression, hypertension, and work disability.
• Current guidelines recommend cognitive behavioral therapy for insomnia (CBT-I) as a first-line
treatment for persistent insomnia. CBT-I includes practical strategies for modifying sleep habits,
regulating sleep–wake schedules, reducing arousal from sleep, and reframing unhelpful beliefs about
sleep and insomnia.
• Medications with an indication for insomnia (e.g., benzodiazepine receptor agonists, dual orexin
receptor antagonists, and doxepin) that are approved by the Food and Drug Administration are
recommended as alternative or adjunctive treatments. There is inadequate evidence to support over-
the-counter medications, antipsychotics, or alternative agents for insomnia.
• Recommended therapies for insomnia produce clinically meaningful reductions in insomnia
symptoms, sleep-onset latency, and time awake after sleep onset. CBT-I alone or with medication can
promote rapid and sustained alleviation of insomnia symptoms over time.
ment targets. In perimenopausal women, vaso- sleep–wake behaviors may identify additional be-
motor symptoms may serve as both a precipitating havioral and environmental targets for interven-
and perpetuating factor. Chronic insomnia is tion (Fig. 1). Patient-reported assessment tools and
associated with increased risks of major depres- sleep diaries can provide valuable information
sion,5 hypertension,6 Alzheimer’s disease,7 and about the nature and severity of insomnia symp-
work disability. toms, help screen for other sleep disorders, and
The assessment and diagnosis of insomnia monitor treatment progress (Table 2).
rests on a careful history to document symptoms,
course, co-occurring conditions, and other con- S t r ategie s a nd E v idence
tributing factors (Table 1).8 A 24-hour history of
Current treatment options for insomnia include
Table 1. Key Elements of Assessment. prescribed and over-the-counter medications, psy-
chological and behavioral therapies (also referred
Typical sleep schedule: bedtime, rise time, and daytime napping
to as cognitive behavioral therapy for insomnia
Nature of sleep concern: frequency, duration, course, triggers, and exacerbat- [CBT-I]), and complementary and alternative ther-
ing factors
apies. The usual treatment trajectory involves the
Daytime symptoms and effects: activities that are cancelled or avoided as a
result of sleep problems use of over-the-counter medications and, when
the disorder is brought to the attention of a prac-
Symptoms of other sleep disorders that may produce insomnia
titioner, prescription medication. Few patients
Loud snoring, restless sleep, and excessive daytime sleepiness (sleep
apnea)
receive CBT-I, owing in part to the lack of ade-
quately trained therapists.
Urge to move the legs or unpleasant leg sensations in the evening (rest-
less legs syndrome)
CBTI-I
Unusual or aggressive behaviors during sleep: sleepwalking, rapid-eye
movement (REM)–sleep behavior disorder CBT-I involves a combination of strategies aimed
Medical and psychiatric history: identify contributing medical problems and
at changing the behavioral practices and psycho-
psychiatric conditions logical factors (e.g., excessive worries and un-
Environmental factors helpful beliefs about sleep) that contribute to
Bedroom environment, noise, light level, and temperature
insomnia. The core components of CBT-I include
behavioral and sleep-scheduling strategies (sleep
Sleep hygiene: alcohol use; use of tea, coffee, or nicotine; exercise patterns
restriction and stimulus control instructions),
Previous treatments and outcomes relaxation methods, psychological and cognitive
Prescribed and over-the-counter medications and supplements interventions (or both) aimed at changing un-
Behavioral measures to improve sleep helpful beliefs and excessive worrying about in-
somnia, and sleep hygiene education (Table 3).
248 n engl j med 391;3 nejm.org July 18, 2024

Clinical Pr actice
Bedtime, level
of sleepiness
Activities and mental Time to fall asleep
state before sleep (sleep latency)
Insomnia Assessment
Evening over 24 Hr Awakenings
activities, light level, (number, duration,
and arousal and timing)
Naps and dozing

(number, duration, Response to
timing, and awakenings
situations)
Meal Sleep
and exercise duration
timing
Work, school, Final

and family awakening
activities time
Morning Out-of-bed
light level Ease of time
and activity awakening
Figure 1. Use of 24-Hour History for Insomnia Assessment.

Insomnia manifests as a sleep problem but affects — and is affected by — daytime behaviors. A thorough insomnia history helps evalu-
ate symptoms and behaviors both at night and during the day. Shown are key parts of an insomnia assessment across a 24-hour day.
Additional psychological interventions, such as Evaluation [GRADE] method).17-19 Evidence from

Acceptance and Commitment Therapy and Mind- clinical trials and meta-analyses indicates that
fulness-Based Therapy, have been adapted for CBT-I produces substantial improvements in pa-
insomnia, but fewer data support their efficacy, tient-reported outcomes, typically measured with
and they take more time to yield benefits (Ta- the use of a standardized effect-size method (ei-
ble 3). CBT-I is prescriptive, focused on sleep, and ther Cohen’s d or Hedges’ g). The effect size is a
oriented toward problem solving. It is typically measure of the magnitude of difference between
guided by a mental health therapist (e.g., a psy- groups, with conventional thresholds for the size
chologist) in the context of four to eight consulta- of effect as follows: 0.2, small; 0.5. moderate; and
tion visits. There are several variants in the 0.8, large. In meta-analyses of these trials, CBT-I
methods for implementing CBT-I, including ab- showed improvement in insomnia-symptom se-
breviated and group formats,14 the involvement verity (effect size, 0.98; 95% confidence interval
of other providers (e.g., a nurse practitioner),15 [CI], 0.82 to 1.15), sleep-onset latency (effect size,
and the use of telehealth or digital platforms.16 0.57; 95% CI, 0.50 to 0.65), and time awake after
CBT-I is currently the first-line treatment rec- sleep onset (effect size, 0.63; 95% CI, 0.53 to
ommended in the practice guidelines of several 0.73). Improved sleep continuity was also associ-
professional organizations (labeled as a “strong ated with a corresponding increase in sleep ef-
recommendation” on the basis of the Grading of ficiency (the ratio of time asleep to time spent in
Recommendations Assessment, Development, and bed; effect size, 0.71; 95% CI, 0.61 to 82). Total

Table 2. Tools for the Clinical Assessment of Insomnia.
Domain and Measure Description

Sleep–wake characteristics: sleep Completed daily by the patient to collect information about sleep schedule (bed-
diary time, arising time, napping) and estimates of sleep–wake characteristics
(sleep latency, number and duration of awakenings, and sleep time). Useful
for determining the nature, frequency, and severity of sleep problems and
monitoring progress during treatment.9
Insomnia symptom severity: A 7-item, patient-reported scale for assessing perceived severity of insomnia
Insomnia Severity Index symptoms and daytime distress and impairments. Scores range from 0 to
28; 0 to 7 indicates no significant insomnia, 8 to 14 indicates subthreshold
insomnia, 15 to 21 indicates moderate insomnia, and 22 to 28 indicates severe
insomnia. The scale includes guidelines for defining clinical insomnia and re-
sponse or remission after treatment.10
Sleep quality: Pittsburgh Sleep A 19-item patient-reported scale measuring overall sleep quality and a screening
Quality Index tool for other sleep disorders.11
Screening for sleep apnea and rest-
less legs syndrome
STOP–Bang An 8-item patient-reported questionnaire for evaluating risk of sleep-related
breathing disorders.12
International Restless Legs A 10-item patient-reported questionnaire assessing frequency, severity, and effect
Syndrome Rating Scale of restless legs syndrome (scores range from 0 to 40, with higher scores indi-
cating more severe symptoms).13
sleep time had increased modestly at the end of the important gap between demand and access
treatment (effect size, 0.16; 95% CI, 0.08 to 0.24), to CBT-I. The SHUTi and Sleepio applications
although additional benefits were often seen have substantial published evidence supporting
several weeks or months after the end of thera- their efficacy. A meta-analysis of 11 randomized
py.17,20,21 Effect sizes are strongest for global in- clinical trials involving 1460 participants that
somnia symptom severity. Efficacy does not ap- tested Web-based CBT-I found that eCBT-I had a
pear to be moderated by age, insomnia severity, positive effect on several sleep outcomes (i.e.,
the presence of coexisting conditions, or hypnotic insomnia severity, sleep efficiency, subjective
medication use. Smaller improvements have been sleep quality, wake after sleep onset, sleep-onset
noted for daytime symptoms (e.g., fatigue and latency, total sleep time, and number of noctur-
mood) and quality of life,22,23 which have been nal awakenings), with effect sizes ranging from
attributed in part to the use of generic measure- 0.21 to 1.09. These effects were similar to those
ments not specifically developed for insomnia. observed in trials of face-to-face CBT-I and were
Overall, approximately 60 to 70% of patients maintained for 4 to 48 weeks after follow-up.16
have a clinical response, which is defined as a Additional digital CBT-I products (e.g., CBT-i
reduction of 7 points on the Insomnia Severity Coach, Go! To Sleep, and Sleep Reset) use simi-
Index (ISI; scores range from 0 to 28, with lar therapeutic principles but have no or limited
higher scores indicating more severe insomnia). published efficacy data.
A sample ISI form is shown in the Supplemen- Treating co-occurring conditions such as de-
tary Appendix, available with the full text of this pression and chronic pain may alleviate insom-
article at NEJM.org. Approximately 50% of per- nia symptoms but generally does not completely
sons with insomnia had remission (total ISI resolve them. Conversely, the treatment of insom-
score, <8) after 6 to 8 weeks of treatment, and 40 nia improves sleep in the context of co-occurring
to 45% had sustained remission for 12 months. conditions but has less consistent effects on the
Daytime sleepiness is a potential adverse event co-occurring conditions themselves. For instance,
in the early phase of restricting time in bed, but the treatment of insomnia alleviates depression
that effect tends to resolve as the sleep time is symptoms and reduces the incidence and recur-
increased.24 rence of depression25 but has only small effects
Digital CBT-I (eCBT-I) has gained in popularity on chronic pain.26
over the past decade and could eventually narrow Stepped-care approaches may help to address

Clinical Pr actice
Table 3. Psychological and Behavioral Therapies for Patients with Insomnia.
Therapy Description
Sleep restriction This intervention limits the amount of time spent in bed (the sleep window) to match as
closely as possible the actual sleep time and strengthens the homeostatic sleep drive
(the increase in sleep propensity that accumulates with an increased duration of wake-
fulness). After the initial restriction, the sleep window is gradually adjusted upward or
downward on a weekly basis and as a function of sleep efficiency (time asleep ÷ time
spent in bed × 100) until an appropriate sleep duration is established.
Stimulus control Go to bed only when sleepy.
Get out of bed when unable to sleep.
Use the bed and bedroom for sleep and sex only (no reading, watching television, etc.).
Arise at the same time every morning.
Avoid napping.
Relaxation training This method involves the use of clinical procedures (e.g., progressive muscle relaxation
and imagery training) aimed at reducing autonomic arousal, muscle tension, and in-
trusive thoughts that interfere with sleep. Most relaxation procedures begin with some
professional guidance and are practiced daily over a period of a few weeks. Relaxation
training is not always included in cognitive behavioral therapy for insomnia (CBT-I).
Cognitive therapy This psychological approach uses Socratic questioning and behavioral experiments to
revise common misconceptions about sleep and to reframe unhelpful beliefs about
insomnia and its daytime consequences. This method is also intended to reduce ex-
cessive worrying about sleep difficulties and their daytime consequences. Additional
cognitive strategies may also involve paradoxical intention (willingly trying to stay
awake rather than trying to fall asleep) in order to alleviate the performance anxiety
triggered by attempting to force sleep.
Sleep hygiene education The patient receives education regarding general guidelines about health practices (e.g.,
diet, exercise, and substance use) and environmental factors (e.g., light level, noise,
and excessive temperature) that may promote or interfere with sleep. This may also
include some basic information about normal sleep and changes in sleep patterns
with aging.
Acceptance and commit- ACT is a type of psychotherapy aimed at educating the patient to stay focused on the
ment therapy (ACT) present moment and accept life experiences, thoughts, and feelings (even negative
ones) without trying to change them. ACT involves the use of different methods (e.g.,
acceptance, defusion, mindfulness, and committed action) and processes in order to
increase psychological flexibility.
Mindfulness This approach is a meditation method that involves observing one’s thoughts and feel-
ings and letting go of the need to change or ruminate about things. Originally de-
signed as a method of reducing stress and anxiety, mindfulness has been adapted for
the management of insomnia and can be included as one component of ACT.
Brief behavioral treatments This abbreviated version of CBT-I emphasizes behavioral components and is typically
for insomnia implemented in fewer (one to four) sessions. It involves education about sleep regula-
tion and factors that promote or interfere with sleep, along with a tailored behavioral
prescription based on stimulus control and sleep-restriction therapy.
resource limitations with traditional psychologi- epine receptor agonists have steadily decreased
cal and behavioral therapies. One such model and prescriptions for trazodone have steadily
recommends education, monitoring, and self-help increased, notwithstanding the absence of a
approaches at the first level, digital or group- Food and Drug Administration (FDA) indication
based psychological and behavioral treatment at for the use of trazodone to treat insomnia. In
the second level, individual psychological and addition, orexin receptor antagonist drugs were
behavioral treatment at the third level, and phar-introduced in 2014 and are widely used. Hyp-
macotherapy as a short-term adjunct at each level.27
notic medications are prescribed at higher rates
for women, older adults, and non-Hispanic White
Medications patients, which reflects the epidemiologic char-
Prescribing patterns for hypnotic medications in acteristics of insomnia.29 The main classes of
the United States have changed substantially over sleep-promoting medications are summarized in
the past 20 years.28 Prescriptions for benzodiaz- Table 4. Controlled data are sparse regarding the

252
Table 4. Medications for the Treatment of Insomnia.
Medication Class and Examples and Approximate Effect Size

Types Half-Life Potential Advantages Potential Disadvantages (95% CI)*
Benzodiazepine receptor Consistent evidence of efficacy for sleep Short-term risks: sedation, anterograde Short-acting, 0.83 (0.62 to
agonists† onset and sleep maintenance for amnesia, cognitive and psychomotor 1.04); intermediate-act-
agents approved by the FDA. Range of impairment, nausea, headaches, com- ing, 0.67 (0.52 to 0.82);
half-lives can accommodate different plex sleep-related behavior (FDA warn- long-acting, 0.58 (0.42 to
The
symptom profiles. ing), rebound insomnia 0.73); eszopiclone, 0.51

Long-term risks: falls, hip fractures, physi- (0.35 to 0.68); zolpidem,
ological dependence, depression, de- 0.45 (0.36 to 0.56); zal
mentia eplon, 0.19 (0.00 to 0.37)
Benzodiazepines Triazolam (4 hr)†, temazepam
(10 hr)†, clonazepam (30
hr)†‡
n engl j med 391;3

Nonbenzodiazepines Zolpidem (2.5 hr)†, zaleplon (1
(Z-drugs) hr)†, eszopiclone (6 hr)†
Dual orexin receptor Suvorexant (12 hr), lemborex- Consistent evidence of efficacy for sleep Short-term risks: sedation, cognitive and Daridorexant, 0.23 (−0.01 to
antagonists ant (18 hr), daridorexant onset and sleep maintenance. Targeted psychomotor impairment, dizziness, 0.48); lemborexant, 0.36
nejm.org
(8 hr) mechanism of action on wake-promot- headaches, abnormal dreams, night- (0.08 to 0.63); suvorex-
n e w e ng l a n d j o u r na l
ing orexin system. Lower risk of cogni- mares, sleep paralysis, complex sleep- ant, 0.31 (0.01 to 0.62)

of
tive and psychomotor impairment than related behavior, increased depression

benzodiazepine receptor agonists; low Contraindicated in patients with narco-
potential for abuse and physiological lepsy
July 18, 2024

dependence.
Sedating antidepressants Doxepin (15 hr), trazodone Mechanisms of action involve histamine, Inconsistent efficacy evidence for insomnia Doxepin, 0.30 (−0.05 to 0.64);
m e dic i n e
(9 hr),‡, mirtazapine (30 serotonin, and adrenergic receptors. (other than doxepin 3–6 mg) trazodone, 0.52 (0.16 to
hr)†‡, amitriptyline (30 Efficacy data for maintenance, variable Short-term risks: sedation, cognitive and 0.89)
hr)†‡ evidence for sleep onset. Low potential psychomotor impairment, cardiac con-
for abuse. duction delay, anticholinergic effects,
nausea, serotonin syndrome, increased
suicidality
Long-term risks: falls, hip fractures, de-
mentia, physiological dependence
(i.e., rebound insomnia); weight gain,
metabolic effects (i.e., abnormal glu-
cose metabolism, lipid levels) with
mirtazapine
Medication Class and Examples and Approximate Effect Size
Types Half-Life Potential Advantages Potential Disadvantages (95% CI)*
Melatonin, melatonin Melatonin (1 hr)‡, ramelteon (2 Mechanism of action involves melatonin Not efficacious for sleep maintenance Melatonin, 0.13 (−0.11 to
receptor agonists hr) tasimelteon (1–4 hr)‡ receptors. Efficacy data for sleep on- Short-term risks: sedation, fatigue, dizzi- 0.38); ramelteon, 0.12
set. Efficacy evidence for insomnia in ness, nausea, abnormal dreams (−0.14 to 0.37); tasimel
children with neurodevelopmental dis- teon
orders. Generally associated with few
side effects and low potential for abuse.
Sedating antihistamines Diphenhydramine (6 hr)†, dox- Widely available over the counter and by Limited efficacy data for insomnia Insufficient data
ylamine (10 hr)†, hydroxy- prescription. Mechanism of action in- Short-term risks: sedation, cognitive and
zine (20 hr)†‡ volves antagonism of central histamine psychomotor impairment, anticholiner-
receptors. gic effects (e.g., dry mouth)
Long-term risk: dementia (anticholinergic
effect)
Sedating antipsychotics Quetiapine (6 hr)†‡, olanza Sedating in clinical trials of patients with Limited efficacy data for insomnia Insufficient data
pine (30 hr)†‡ schizophrenia or bipolar disorder. Short-term risks: sedation, dizziness, cog-
Small studies suggest efficacy on nitive and psychomotor impairment,
patient-reported and polysomno- hypotension, headache, dry mouth
graphic sleep measures in insomnia. Long-term risks: metabolic effects (e.g.,
Mechanism of action involves multiple glucose metabolism and lipid levels)
receptor types (e.g., serotonin, dopa- and weight gain
n engl j med 391;3

mine, and histamine).
Miscellaneous Gabapentin (7 hr)†‡, pregaba- Efficacy data for chronic pain (often oc- Efficacy data for insomnia sparse and in- Insufficient data
lin (6 hr)†‡ curring with insomnia). Subjectively consistent
sedating in clinical trials for other con- Short-term risks: sedation, dizziness, cog-
nejm.org
ditions. Mechanism of action involves nitive and psychomotor impairment,
Clinical Pr actice
alpha 2–delta receptors. Eliminated by edema, respiratory depression

renal excretion. Long-term risks: depression and suicidal-
ity, physiological dependence
July 18, 2024

* Effect sizes for new (use, <4 weeks) medication treatments on primary outcomes are as defined by any patient-evaluated scales, including the Insomnia Severity Index, Pittsburgh Sleep

Quality Index, Leeds Sleep Questionnaire, and sleep diaries.30 An effect size of 0.2 is considered to be small, 0.5 is considered to be moderate, and 0.8 is considered to be large.
† The Beers Criteria (a list of medications deemed to be relatively inappropriate for patients 65 years of age or older) recommends avoidance of this drug.
‡ This drug is not FDA-approved for the treatment of insomnia. All drugs included in the table are classified by the FDA as Pregnancy category C with the following exceptions: triazolam,
temazepam (category X); clonazepam (category D); diphenhydramine and doxylamine (category B).
253
long-term efficacy and side-effect profiles of hyp- Sedating Heterocyclic Drugs

notic medications, despite their frequent long- Sedating antidepressant drugs, including tricyclic
term use. drugs (e.g., amitriptyline, nortriptyline, and
doxepin) and heterocyclic drugs (e.g., mirtaza
Benzodiazepine Receptor Agonist Hypnotics pine and trazodone), are commonly prescribed to
Benzodiazepine receptor agonist hypnotics in- treat insomnia. Of these, only doxepin (at a dose
clude benzodiazepines and nonbenzodiazepines of 3 to 6 mg daily, taken at night) is FDA-approved
(also known as Z-drugs). These subclasses have for insomnia. The lower doses used in insomnia
different chemical structures, but both are allo- than in depression and the more rapid onset of
steric modulators of a common binding site on action in insomnia than in depression suggest
γ-aminobutyric acid type A (GABA A) receptors, distinct mechanisms of action for these indica-
which accounts for their similar actions and side tions. Despite their widespread use, the efficacy
effects. Some benzodiazepine receptor agonists of the sedating antidepressants in the treatment
(e.g., zolpidem) have relative specificity for sub- of insomnia is not well supported by controlled
populations of GABA A receptors that are re- trials, except in the case of doxepin. Meta-anal-
sponsible for sleep promotion relative to anxio- yses of trazodone trials have shown inconsistent
lytic, myorelaxant, and anticonvulsant effects. In effects on sleep-onset latency, wake after sleep
practice, however, pharmacodynamic differenc- onset, and total sleep time.35,36 Given these limi-
es among benzodiazepine receptor agonists are tations, current evidence suggests that sedating
less salient than differences in pharmacokinetic antidepressants in aggregate increase sleep qual-
properties, particularly elimination half-life. ity, sleep efficiency, and total sleep time, with
Clinical trials and meta-analyses have shown the little effect on sleep latency.35 Clinicians and pa-
efficacy of benzodiazepine receptor agonists for tients often prefer these medications, despite their
reducing sleep-onset latency and wakefulness lack of specific FDA indication for insomnia, be-
after sleep onset, with small increases in total cause of their mild side effects at low doses and
sleep time (Table 4).30,31 Patient-reported side ef- clinical experience of efficacy. Side effects can
fects of benzodiazepine receptor agonists include include sedation, dry mouth, cardiac conduction
anterograde amnesia (in <5%), next-day sedation delay, hypotension, and hypertension. Sedating
(in 5 to 10%), and complex behaviors during heterocyclic drugs approved for the treatment of
sleep, such as sleepwalking, eating, or driving schizophrenia and bipolar disorder, such as que-
(in 3 to 5%), a side effect that is responsible for tiapine and olanzapine, are sometimes used to
black-box warnings for zolpidem, zaleplon, and treat insomnia. However, the cardiovascular,
eszopiclone. These side effects are more likely to metabolic, and neurologic risks of these drugs
occur with higher doses, coprescription with other weigh against their use except in persons with
sedating medications, and (in the case of amnesia co-occurring psychiatric disorders.
and sedation) longer-duration agents. The devel-
opment of drug tolerance and physiological de- Orexin Receptor Antagonists
pendence marked by rebound insomnia and with- Orexin (hypocretin)–containing neurons in the
drawal syndromes occurs with repeated nightly lateral hypothalamus stimulate wake-promoting
use in 20 to 50% of patients.32 Although misuse nuclei in the brainstem and hypothalamus and
of benzodiazepine receptor agonists (i.e., use inhibit sleep-promoting nuclei in the ventrolat-
without a prescription or at larger doses or longer eral and median preoptic areas.37 Conversely,
duration than prescribed) is relatively common, inhibiting orexinergic neurotransmission inhib-
substance use disorder involving benzodiazepine its wakefulness and promotes sleep. Three dual
receptor agonists is uncommon.33 Epidemiologic orexin receptor antagonists — suvorexant, lem-
data show dose-dependent and duration-depen- borexant, and daridorexant — are FDA-approved
dent increases in the risks of hip fractures34 and for insomnia. Clinical trials support their efficacy
dementia with long-term use of benzodiazepine for sleep-onset and sleep-maintenance symp-
receptor agonists, but confounding by indication toms.30,38,39 Side effects include sedation, fatigue,
may contribute to these observed risks. and abnormal dreaming, but they produce less

Clinical Pr actice
cognitive impairment than benzodiazepine re- Complementary and Alternative Therapies

ceptor agonists.40 Because a deficiency in endog- Alternative treatments are widely used among
enous orexin causes narcolepsy with cataplexy, persons with insomnia.45 Cannabis, cannabidiol
orexin antagonists are contraindicated in patients (CBD), and delta-9-tetrahydrocannabinol (THC)
with this condition. preparations are also widely used to treat sleep
problems, but are associated with mixed find-
Melatonin and Melatonin Receptor Agonists ings. The overall quality of evidence supporting
Melatonin is a pineal hormone that is endoge- the efficacy of cannabinoids for insomnia is
nously secreted during darkness at night. Exog- low, owing to the absence of large, well-con-
enous melatonin produces supraphysiologic blood trolled clinical trials and the apparent develop-
levels for varying durations depending on the ment of tolerance to hypnotic effects that can
specific dose and formulation. The appropriate result from chronic administration. Variation
dose of melatonin for treating insomnia is not in cannabis-derived preparations is also rele-
defined. Controlled trials involving adults have vant. For instance, CBD is stimulating at low
shown a small effect on sleep onset, with little doses and sedating at high doses, and THC has
effect on wakefulness during sleep or on total the opposite effects.
sleep time.41,42 Melatonin is increasingly used to
treat sleep problems in children, although its Selection of Hypnotic Medication
efficacy and safety are not well established ex- When medication is the selected treatment, a
cept in children with neurodevelopmental dis- short-acting benzodiazepine receptor agonist,
orders.43 orexin antagonist, or low-dose heterocyclic drug
Drugs that bind to melatonin MT1 and MT2 is a reasonable first choice in most clinical sce-
receptors are approved for the treatment of narios. Benzodiazepine receptor agonists may
sleep-onset insomnia (ramelteon) and circadian- be preferred in the treatment of patients with
rhythm sleep–wake disorder (tasimelteon). Like insomnia with predominantly sleep-onset symp-
melatonin, these drugs have little effect on toms, in younger adults, and when short-term
wakefulness after sleep onset or on total sleep use is likely (e.g., in response to acute or periodic
time.42 Somnolence and fatigue are the most stressors). Low-dose heterocyclic drugs or orexin
common side effects. antagonists may be preferred in treating patients
with symptoms that are predominantly related
Other Medications to sleep maintenance or early awakening, older
Antihistamine medications obtained over the adults, and patients with substance use disorders
counter (diphenhydramine and doxylamine) and or sleep apnea. The Beers Criteria list of medica-
by prescription (hydroxyzine) are among the most tions deemed to be relatively inappropriate for
commonly used medications for the treatment of patients 65 years of age or older includes benzo-
insomnia. Data supporting their efficacy are diazepine receptor agonists and heterocyclic
weak,41 but their availability and perceived safety drugs, but does not include doxepin, trazodone,
as compared with benzodiazepine receptor ago- or orexin antagonists. Initial medication treat-
nists probably contribute to their popularity. ment often includes nightly use for 2 to 4 weeks
Sedating antihistamines can cause excessive se- followed by reevaluation of effects and side ef-
dation, anticholinergic side effects, and an in- fects. Intermittent administration (2 to 4 times
creased risk of dementia. Gabapentinoids, such as per week) is encouraged if long-term use is ap-
gabapentin and pregabalin, are commonly used propriate. Patients should be instructed to take
for the treatment of chronic pain and are also medications 15 to 30 minutes before bedtime.
first-line agents for the treatment of restless legs With prolonged medication use, drug dependence
syndrome.44 These drugs produce sedation and develops in some patients, particularly with the
increase slow-wave sleep, and are prescribed off- use of benzodiazepine receptor agonists. A sys-
label for insomnia, particularly when accompa- tematic tapering schedule (e.g., by 25% per week)
nied by pain. Fatigue, somnolence, dizziness, and can help to reduce or discontinue the use of
ataxia are the most common side effects. hypnotics after long-term use.46,47

Combination Therapy or Single Therapy A r e a s of Uncer ta in t y

Evidence from the few head-to-head comparative
studies available indicates that both CBT-I and Evidence is lacking regarding the long-term ef-
hypnotic medications (mostly Z-drugs) produce ficacy of medications and the development of
equivalent improvements in sleep continuity in tolerance to medications for insomnia. The role
the short term (4 to 8 weeks),48-51 although medi- of intermittent medication and the appropriate
cation has been shown to increase total sleep schedule for administration are still unclear. Al-
time more than CBT-I. Combined therapy pro- though network meta-analyses address the rela-
duces improvement in sleep more quickly than tive efficacy and side effects of different medica-
CBT-I alone, but this advantage decreases by tion classes, few large trials have directly compared
the fourth or fifth week of treatment,52 and different active medications. Telehealth and digi-
CBT-I used alone produces more sustained ben- tal CBT platforms offer potential solutions for
efits over time than medication or combined some patients, although more information is
therapy. Some patients may have less adherence needed to identify the patients who benefit most.
to behavioral recommendations when the easier Additional work is needed to identify reliable in-
alternative of taking a sleep medication is also somnia phenotypes55 and test whether persons
available. with those phenotypes have different responses
to more personalized therapeutic approaches.
Guidel ine s
C onclusions a nd
Current guidelines that have been endorsed by R ec om mendat ions
health care and professional organizations rec-
ommend CBT-I as the first-line treatment for The patient in the vignette spends long periods
insomnia and medications as alternative or ad- of time in bed with considerable variability in
junctive treatment, within the context of shared the time taken to fall asleep and the time sleep-
decision making.17-19,53,54 Guidelines recommend ing. She described anxiously worrying about
CBT-I with a strong level of support, and sub- falling asleep and staying asleep. We would initi-
components such as brief behavioral treatment, ate CBT-I with a focus on reducing overall time
sleep restriction, and stimulus control are rec- in bed to improve sleep consolidation, maintain-
ommended with lower levels of support. Among ing regular sleep–wake times to strengthen cir-
medications, guidelines make weak recommen- cadian sleep regulation, and performing cogni-
dations with moderate-quality evidence for the tive exercises to reduce sleep-focused rumination.
use of FDA-approved hypnotic medications (e.g., If insomnia recurred in the setting of stressful
benzodiazepine receptor agonists, doxepin, and life events, we would prescribe doxepin for inter-
orexin antagonists) and weak evidence against mittent use on those occasions.
the use of other agents, including heterocyclic The content of this article is solely the responsibility of the
authors and does not necessarily represent the views of the
drugs such as trazodone and antipsychotic agents. Patient-Centered Outcomes Research Institute or its Board of
Recommendations in this article are generally con- Governors or Methodology Committee.
sistent with these guidelines. Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
References
1. American Psychiatric Association. Di- tors, and public health burden. Sleep Med sleep duration and risk of incident cardio-
agnostic and statistical manual of mental Clin 2022;17:173-91. vascular disease and all-cause mortality:
disorders. 5th ed. Washington, DC:Ameri- 4. Baker FC, Willoughby AR, Sassoon Sleep Heart Health Study. Sleep 2018;
can Psychiatric Publishing, 2013. SA, Colrain IM, de Zambotti M. Insomnia 41(6):zsy047.
2. Ogeil RP, Chakraborty SP, Young AC, in women approaching menopause: be- 7. Shi L, Chen SJ, Ma MY, et al. Sleep
Lubman DI. Clinician and patient barriers yond perception. Psychoneuroendocrinol- disturbances increase the risk of demen-
to the recognition of insomnia in family ogy 2015;60:96-104. tia: a systematic review and meta-analy-
practice: a narrative summary of reported 5. Hertenstein E, Feige B, Gmeiner T, et sis. Sleep Med Rev 2018;40:4-16.
literature analysed using the theoretical al. Insomnia as a predictor of mental dis- 8. Schutte-Rodin S, Broch L, Buysse D,
domains framework. BMC Fam Pract 2020; orders: a systematic review and meta- Dorsey C, Sateia M. Clinical guideline for
21:1. analysis. Sleep Med Rev 2019;43:96-105. the evaluation and management of chron-
3. Morin CM, Jarrin DC. Epidemiology 6. Bertisch SM, Pollock BD, Mittleman ic insomnia in adults. J Clin Sleep Med
of insomnia: prevalence, course, risk fac- MA, et al. Insomnia with objective short 2008;4:487-504.

Clinical Pr actice
9. Carney CE, Buysse DJ, Ancoli-Israel S, 22. Benz F, Knoop T, Ballesio A, et al. The Cochrane Database Syst Rev 2018; 5:
et al. The consensus sleep diary: stan- efficacy of cognitive and behavior thera- CD010753.
dardizing prospective sleep self-monitor- pies for insomnia on daytime symptoms: 36. Yi XY, Ni SF, Ghadami MR, et al. Tra-
ing. Sleep 2012;35:287-302. a systematic review and network meta- zodone for the treatment of insomnia: a
10. Morin CM, Belleville G, Bélanger L, analysis. Clin Psychol Rev 2020;80:101873. meta-analysis of randomized placebo-con-
Ivers H. The Insomnia Severity Index: psy- 23. Alimoradi Z, Jafari E, Broström A, et trolled trials. Sleep Med 2018;45:25-32.
chometric indicators to detect insomnia al. Effects of cognitive behavioral therapy 37. Holst SC, Landolt HP. Sleep-Wake
cases and evaluate treatment response. for insomnia (CBT-I) on quality of life: a Neurochemistry. Sleep Med Clin 2022;17:
Sleep 2011;34:601-8. systematic review and meta-analysis. 151-60.
11. Buysse DJ, Reynolds CF III, Monk TH, Sleep Med Rev 2022;64:101646. 38. Khazaie H, Sadeghi M, Khazaie S,
Berman SR, Kupfer DJ. The Pittsburgh 24. Maurer LF, Ftouni S, Espie CA, Bis- Hirshkowitz M, Sharafkhaneh A. Dual
Sleep Quality Index: a new instrument for dounis L, Kyle SD. The acute effects of orexin receptor antagonists for treatment
psychiatric practice and research. Psychi- sleep restriction therapy for insomnia on of insomnia: a systematic review and me-
atry Res 1989;28:193-213. circadian timing and vigilance. J Sleep ta-analysis on randomized, double-blind,
12. Chung F, Abdullah HR, Liao P. STOP- Res 2021;30(4):e13260. placebo-controlled trials of suvorexant
Bang questionnaire: a practical approach 25. Irwin MR, Carrillo C, Sadeghi N, and lemborexant. Front Psychiatry 2022;
to screen for obstructive sleep apnea. Bjurstrom MF, Breen EC, Olmstead R. 13:1070522.
Chest 2016;149:631-8. Prevention of incident and recurrent ma- 39. Xue T, Wu X, Chen S, et al. The effi-
13. Sharon D, Allen RP, Martinez-Martin jor depression in older adults with insom- cacy and safety of dual orexin receptor
P, et al. Validation of the self-adminis- nia: a randomized clinical trial. JAMA antagonists in primary insomnia: a sys-
tered version of the international Restless Psychiatry 2022;79:33-41. tematic review and network meta-analy-
Legs Syndrome study group severity rat- 26. Selvanathan J, Pham C, Nagappa M, et sis. Sleep Med Rev 2022;61:101573.
ing scale — the sIRLS. Sleep Med 2019; al. Cognitive behavioral therapy for in- 40. Muehlan C, Roch C, Vaillant C, Ding-
54:94-100. somnia in patients with chronic pain — a emanse J. The orexin story and orexin re-
14. Buysse DJ, Germain A, Moul DE, et al. systematic review and meta-analysis of ceptor antagonists for the treatment of
Efficacy of brief behavioral treatment for randomized controlled trials. Sleep Med insomnia. J Sleep Res 2023;32(6):e13902.
chronic insomnia in older adults. Arch Rev 2021;60:101460. 41. Culpepper L, Wingertzahn MA. Over-
Intern Med 2011;171:887-95. 27. Baglioni C, Espie CA, Altena E, et al. the-counter agents for the treatment of
15. Kyle SD, Siriwardena AN, Espie CA, et Cognitive behavioural therapy for insom- occasional disturbed sleep or transient
al. Clinical and cost-effectiveness of nia disorder: extending the stepped care insomnia: a systematic review of efficacy
nurse-delivered sleep restriction therapy model. J Sleep Res 2023;32(6):e14016. and safety. Prim Care Companion CNS
for insomnia in primary care (HABIT): a 28. Bertisch SM, Herzig SJ, Winkelman Disord 2015;17(6):10.4088/PCC.15r01798.
pragmatic, superiority, open-label, ran- JW, Buettner C. National use of prescrip- 42. Low TL, Choo FN, Tan SM. The effi-
domised controlled trial. Lancet 2023; tion medications for insomnia: NHANES cacy of melatonin and melatonin agonists
402:975-87. 1999-2010. Sleep 2014;37:343-9. in insomnia — an umbrella review. J Psy-
16. Zachariae R, Lyby MS, Ritterband LM, 29. Milani SA, Raji MA, Chen L, Kuo YF. chiatr Res 2020;121:10-23.
O’Toole MS. Efficacy of internet-delivered Trends in the use of benzodiazepines, 43. Abdelgadir IS, Gordon MA, Akobeng
cognitive-behavioral therapy for insomnia z-hypnotics, and serotonergic drugs among AK. Melatonin for the management of
— a systematic review and meta-analysis US women and men before and during the sleep problems in children with neurode-
of randomized controlled trials. Sleep COVID-19 pandemic. JAMA Netw Open velopmental disorders: a systematic re-
Med Rev 2016;30:1-10. 2021;4(10):e2131012. view and meta-analysis. Arch Dis Child
17. Edinger JD, Arnedt JT, Bertisch SM, et 30. De Crescenzo F, D’Alò GL, Ostinelli 2018;103:1155-62.
al. Behavioral and psychological treat- EG, et al. Comparative effects of pharma- 44. Winkelman JW, Armstrong MJ, Allen
ments for chronic insomnia disorder in cological interventions for the acute and RP, et al. Practice guideline summary:
adults: an American Academy of Sleep long-term management of insomnia dis- treatment of restless legs syndrome in
Medicine systematic review, meta-analy- order in adults: a systematic review and adults: report of the Guideline Develop-
sis, and GRADE assessment. J Clin Sleep network meta-analysis. Lancet 2022;400: ment, Dissemination, and Implementa-
Med 2021;17:263-98. 170-84. tion Subcommittee of the American Acad-
18. Riemann D, Baglioni C, Bassetti C, et 31. Chiu HY, Lee HC, Liu JW, et al. Com- emy of Neurology. Neurology 2016; 87:
al. European guideline for the diagnosis parative efficacy and safety of hypnotics 2585-93.
and treatment of insomnia. J Sleep Res for insomnia in older adults: a systematic 45. Bertisch SM, Wells RE, Smith MT,
2017;26:675-700. review and network meta-analysis. Sleep McCarthy EP. Use of relaxation techniques
19. Qaseem A, Kansagara D, Forciea MA, 2021;44(5):zsaa260. and complementary and alternative medi-
Cooke M, Denberg TD. Management of 32. Soyka M, Wild I, Caulet B, Leontiou C, cine by American adults with insomnia
chronic insomnia disorder in adults: a Lugoboni F, Hajak G. Long-term use of symptoms: results from a national survey.
clinical practice guideline from the Amer- benzodiazepines in chronic insomnia: a J Clin Sleep Med 2012;8:681-91.
ican College of Physicians. Ann Intern European perspective. Front Psychiatry 46. Edinger JD, Wamboldt F, Johnson RL,
Med 2016;165:125-33. 2023;14:1212028. et al. Use of a blinded hypnotic tapering
20. van Straten A, van der Zweerde T, 33. Votaw VR, Geyer R, Rieselbach MM, strategy to promote hypnotic discontinu-
Kleiboer A, Cuijpers P, Morin CM, Lancee McHugh RK. The epidemiology of benzo- ation. Sleep (in press) (https://academic
J. Cognitive and behavioral therapies in diazepine misuse: a systematic review. .oup.com/sleep/advance-article/doi/10.1093/
the treatment of insomnia: a meta-analy- Drug Alcohol Depend 2019;200:95-114. sleep/zsab270/6432064).
sis. Sleep Med Rev 2018;38:3-16. 34. Donnelly K, Bracchi R, Hewitt J, Rout- 47. Morin CM, Bastien C, Guay B, Radou-
21. Brasure M, Fuchs E, MacDonald R, et ledge PA, Carter B. Benzodiazepines, co-Thomas M, Leblanc J, Vallières A. Ran-
al. Psychological and behavioral interven- Z-drugs and the risk of hip fracture: a sys- domized clinical trial of supervised taper-
tions for managing insomnia disorder: an tematic review and meta-analysis. PLoS ing and cognitive behavior therapy to
evidence report for a clinical practice One 2017;12(4):e0174730. facilitate benzodiazepine discontinuation
guideline by the American College of Phy- 35. Everitt H, Baldwin DS, Stuart B, et al. in older adults with chronic insomnia.
sicians. Ann Intern Med 2016;165:113-24. Antidepressants for insomnia in adults. Am J Psychiatry 2004;161:332-42.

Clinical Pr actice
48. Morin CM, Vallières A, Guay B, et al. R, Otto MW. Cognitive behavior therapy clinical practice guideline. J Clin Sleep
Cognitive behavioral therapy, singly and and pharmacotherapy for insomnia: a Med 2017;13:307-49.
combined with medication, for persistent randomized controlled trial and direct 54. Mysliwiec V, Martin JL, Ulmer CS, et
insomnia: a randomized controlled trial. comparison. Arch Intern Med 2004;164: al. The management of chronic insomnia
JAMA 2009;301:2005-15. 1888-96. disorder and obstructive sleep apnea: syn-
49. Morin CM, Colecchi C, Stone J, Sood 52. Morin CM, Beaulieu-Bonneau S, Ivers opsis of the 2019 U.S. Department of Vet-
R, Brink D. Behavioral and pharmacologi- H, et al. Speed and trajectory of changes erans Affairs and U.S. Department of De-
cal therapies for late-life insomnia: a ran- of insomnia symptoms during acute fense clinical practice guidelines. Ann
domized controlled trial. JAMA 1999;281: treatment with cognitive-behavioral ther- Intern Med 2020;172:325-36.
991-9. apy, singly and combined with medica- 55. Vgontzas AN, Fernandez-Mendoza J.
50. Sivertsen B, Omvik S, Pallesen S, et al. tion. Sleep Med 2014;15:701-7. Insomnia with short sleep duration: no-
Cognitive behavioral therapy vs zopiclone 53. Sateia MJ, Buysse DJ, Krystal AD, sological, diagnostic, and treatment im-
for treatment of chronic primary insom- Neubauer DN, Heald JL. Clinical practice plications. Sleep Med Clin 2013;8:309-22.
nia in older adults: a randomized con- guideline for the pharmacologic treat- Copyright © 2024 Massachusetts Medical Society.
trolled trial. JAMA 2006;295:2851-8. ment of chronic insomnia in adults: an
51. Jacobs GD, Pace-Schott EF, Stickgold American Academy of Sleep Medicine


Management of Insomnia

Uploaded by

Copyright:

Available Formats

Management of Insomnia

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Management of Insomnia

Uploaded by

Copyright:

Available Formats

The n e w e ng l a n d j o u r na l of m e dic i n e

Patrick G. O’Malley, M.D., M.P.H., Editor

n engl j med 391;3 nejm.org July 18, 2024 247

Treatment Approaches to Insomnia

248 n engl j med 391;3 nejm.org July 18, 2024

The New England Journal of Medicine

Naps and dozing

Work, school, Final

Figure 1. Use of 24-Hour History for Insomnia Assessment.

Additional psychological interventions, such as Evaluation [GRADE] method).17-19 Evidence from

n engl j med 391;3 nejm.org July 18, 2024 249

Table 2. Tools for the Clinical Assessment of Insomnia.

Domain and Measure Description

250 n engl j med 391;3 nejm.org July 18, 2024

The New England Journal of Medicine

Table 3. Psychological and Behavioral Therapies for Patients with Insomnia.

n engl j med 391;3 nejm.org July 18, 2024 251

Medication Class and Examples and Approximate Effect Size

symptom profiles. ing), rebound insomnia 0.73); eszopiclone, 0.51

n engl j med 391;3

The New England Journal of Medicine

tive and psychomotor impairment than related behavior, increased depression

July 18, 2024

n engl j med 391;3

alpha 2–delta receptors. Eliminated by edema, respiratory depression

July 18, 2024

The New England Journal of Medicine

long-term efficacy and side-effect profiles of hyp- Sedating Heterocyclic Drugs

254 n engl j med 391;3 nejm.org July 18, 2024

The New England Journal of Medicine

cognitive impairment than benzodiazepine re- Complementary and Alternative Therapies

n engl j med 391;3 nejm.org July 18, 2024 255

Combination Therapy or Single Therapy A r e a s of Uncer ta in t y

256 n engl j med 391;3 nejm.org July 18, 2024

The New England Journal of Medicine

n engl j med 391;3 nejm.org July 18, 2024 257

258 n engl j med 391;3 nejm.org July 18, 2024

The New England Journal of Medicine

You might also like