Insomnia Disorder: CBT-I First-Line, Pharmacotherapy Hierarchy, and DORA Selection

Insomnia Disorder: CBT-I First-Line, Pharmacotherapy Hierarchy, and DORA Selection

📖 16 min read🩺 Sleep Medicine👤 gomdori, MD🔬 AASM 2017 / 2024 Updates

Executive Summary

Insomnia disorder per DSM-5-TR requires dissatisfaction with sleep quantity/quality (initial, middle, or terminal insomnia), occurring ≥3 nights/week for ≥3 months, with daytime impairment and despite adequate opportunity for sleep. Cognitive behavioral therapy for insomnia (CBT-I) is the universal first-line treatment per AASM, ACP, and EAU guidelines — components include sleep restriction, stimulus control, sleep hygiene, cognitive restructuring, and relaxation. When pharmacotherapy is needed, the modern hierarchy favors dual orexin receptor antagonists (DORAs) — suvorexant, lemborexant, daridorexant — over Z-drugs (zolpidem, eszopiclone) due to lower dependence, less rebound insomnia, and physiologic sleep architecture preservation. Daridorexant 50 mg HS, lemborexant 5–10 mg HS, and suvorexant 10–20 mg HS are FDA-approved first-line pharmacologic options. Avoid chronic benzodiazepines and over-the-counter diphenhydramine in older adults. Always taper hypnotics gradually after CBT-I has consolidated.

10–15%
Adult prevalence (chronic)
70–80%
CBT-I response rate
≥3 nt/wk
DSM-5-TR threshold
2×
Risk of incident depression

1. Pathophysiology & DSM-5-TR Criteria

Pathophysiology — the hyperarousal model

Insomnia is increasingly understood as a disorder of hyperarousal — physiologic, cognitive, and emotional. The two-process model of sleep regulation describes (1) a homeostatic sleep drive (process S) that accumulates with wakefulness and dissipates during sleep, and (2) a circadian process (C) governed by the suprachiasmatic nucleus. In insomnia, the wake-promoting systems remain pathologically active during sleep periods. Implicated systems:

  • HPA axis hyperactivity — elevated evening cortisol, blunted nocturnal cortisol nadir.
  • Sympathetic dominance — increased nocturnal heart rate, reduced heart rate variability, elevated body temperature.
  • Cortical hyperarousal — increased high-frequency EEG (beta and gamma) activity during NREM sleep; the “tired but wired” phenotype.
  • Orexin/hypocretin overdrive — orexin neurons in the lateral hypothalamus stabilize wakefulness; persistent activity during the sleep period contributes to insomnia.
  • Cognitive arousal — pre-sleep worry, sleep-effort paradox (trying harder to sleep worsens it), maladaptive sleep beliefs.

DSM-5-TR diagnostic criteria for insomnia disorder

CriterionDescription
A. Predominant complaintDissatisfaction with sleep quantity or quality, with one or more: (1) difficulty initiating sleep, (2) difficulty maintaining sleep with frequent awakenings, (3) early-morning awakening with inability to return to sleep
B. Daytime impairmentCauses clinically significant distress or impairment in social, occupational, educational, academic, behavioral, or other important areas
C. FrequencySleep difficulty occurs ≥3 nights per week
D. DurationSleep difficulty present for ≥3 months
E. Adequate opportunityDifficulty occurs despite adequate opportunity for sleep
F. Not better explainedNot better explained by another sleep-wake disorder (e.g., narcolepsy, OSA, RLS, circadian rhythm disorder)
G. Not solely due to substanceNot attributable to physiologic effects of a substance (e.g., drug of abuse, medication)
H. Not coexisting conditionCoexisting mental and medical disorders do not adequately explain the predominant complaint of insomnia

Specifiers: with mental comorbidity, with medical comorbidity, with another sleep disorder; episodic (1–3 mo), persistent (≥3 mo), recurrent (≥2 episodes within 1 year). Short-term insomnia <3 months is coded separately as “Other Specified Insomnia Disorder.”

ⓘ
Clinically significant insomnia ≠ short sleep duration. A patient sleeping 5.5 hours nightly without distress or daytime impairment does not have insomnia disorder. Conversely, a patient sleeping 7 hours nightly with profound dissatisfaction, fatigue, and impairment may meet criteria. The disorder is defined by complaint and consequence, not by polysomnographically measured sleep time.
Insomnia Treatment Ladder — CBT-I to DORA Psychotropic medication escalation ladder from first-line to last-resort Insomnia Treatment Ladder — CBT-I to DORA L1 CBT-I (first-line) Sleep restriction | Stimulus control | Cognitive restructuring | Relaxation Durable benefit; no dependence; digital CBT-I if no therapist L2 DORAs (preferred hypnotic) Daridorexant 25-50mg HS | Lemborexant 5-10mg HS | Suvorexant 10-20mg HS Preserve sleep architecture; low dependence; minimal residual sedation L3 Z-drugs (second-line) Zolpidem 5-10mg HS | Eszopiclone 1-3mg HS | Zaleplon 5-20mg HS SE: Complex sleep behaviors, amnesia, dependence, next-day sedation L4 Sedating adjuncts Low-dose doxepin 3-6mg HS | Ramelteon 8mg HS | Melatonin 0.5-3mg Doxepin for sleep-maintenance; ramelteon/melatonin for circadian phase L5 Avoid (chronic) Chronic benzodiazepines | OTC diphenhydramine in older adults Beers Criteria; dependence, tolerance, falls, delirium, anticholinergic Escalation →
Insomnia Disorder — Stepwise Pharmacotherapy

2. Assessment, Sleep Diary & Differential Diagnosis

History essentials

  • Sleep latency (time from lights out to sleep onset; typically <30 min).
  • Wake after sleep onset (WASO) (cumulative awake time during the night; typically <30 min in adults).
  • Total sleep time, time in bed, sleep efficiency (TST/TIB; normal >85%).
  • Bed/rise time consistency, weekend vs weekday pattern.
  • Daytime symptoms — fatigue, sleepiness, impaired attention/memory, mood changes, motor coordination, motivation deficits.
  • Substance/medication exposure — caffeine timing/dose, alcohol (which fragments sleep architecture even at low doses), nicotine, illicit drugs, prescribed stimulants, beta-blockers, decongestants, theophylline, glucocorticoids, OCPs, SSRIs, methylphenidate.
  • Comorbid conditions — depression, anxiety, PTSD, chronic pain, GERD, BPH, congestive heart failure, COPD, RLS, OSA risk factors.

Sleep diary & objective tools

A 1–2 week sleep diary completed prospectively is the cornerstone clinical assessment tool. Validated questionnaires include the Insomnia Severity Index (ISI) (7 items, 0–28; ≥15 = clinically significant insomnia) and the Pittsburgh Sleep Quality Index (PSQI) (19 items, >5 indicates poor sleep). Actigraphy (wrist-worn motion sensors) provides 1–2 weeks of objective sleep estimates and is helpful when self-report is inconsistent with clinical impression.

Polysomnography (PSG) is NOT routinely indicated for insomnia disorder. Reserve PSG for: suspected obstructive sleep apnea, periodic limb movement disorder, REM behavior disorder, narcolepsy, parasomnias, treatment-refractory insomnia, or insomnia with excessive daytime sleepiness disproportionate to sleep restriction.

Differential diagnosis

DifferentialDistinguishing features
Obstructive sleep apnea (OSA)Snoring, witnessed apneas, morning headaches, EDS, BMI elevation, neck circumference; STOP-BANG screen — proceed to PSG
Restless legs syndrome (RLS)Urge to move legs at rest, worse evening/night, relieved by movement; iron studies (ferritin <75 ng/mL associated)
Circadian rhythm sleep-wake disordersDelayed sleep-phase, advanced sleep-phase, shift work, jet lag — sleep is normal in quality but mistimed
Mood / anxiety disorderDepression often presents with terminal insomnia and anhedonia; anxiety with initial insomnia; bidirectional with insomnia disorder
Chronic pain syndromesPain-induced fragmentation; treat the pain; analgesic-related insomnia (NSAIDs benign; opioids fragment)
Substance-inducedCaffeine, alcohol withdrawal, stimulants, glucocorticoids, beta-blockers
REM sleep behavior disorder (RBD)Dream enactment, vocalizations; α-synucleinopathy prodrome — refer to sleep medicine for PSG
Hypothyroidism, hyperthyroidismHyperthyroid causes insomnia; check TSH if clinical suspicion
Insomnia Subtypes Comparison table: Insomnia Subtypes Acute (<3 mo) Chronic (≥3 mo) Comorbid Sleep state misperception Medication-induced Description Adjustment insomnia; usually trigger-related; address stressor ≥3 nights/wk for ≥3 months; daytime impairment; CBT-I With depression, anxiety, pain, OSA; treat both Subjective vs objective discordance Stimulants, steroids, beta-agonists, alcohol withdrawal
Insomnia Categories

3. CBT-I: First-Line Treatment

The American Academy of Sleep Medicine, American College of Physicians (2016), European Sleep Research Society (2017), and 2024 international consensus all recommend CBT-I as first-line treatment for chronic insomnia disorder — superior to all pharmacotherapies in long-term effectiveness, with response rates of 70–80% and durable benefit beyond treatment cessation. CBT-I is delivered as 4–8 weekly sessions; brief (4 sessions), digital (Sleepio, SHUTi, Somryst-prescription digital therapeutic), and group formats are validated alternatives.

Five evidence-based components of CBT-I

1. Sleep restriction therapy (SRT)

The most powerful single CBT-I component. Restrict time in bed (TIB) to match average self-reported total sleep time (TST) from the baseline diary, with a floor of 5–5.5 hours. This induces mild sleep deprivation that consolidates the homeostatic drive and improves sleep efficiency (TST/TIB). Once efficiency reaches ≥85% over 5–7 days, increase TIB by 15–30 minutes weekly. Target sleep efficiency: ≥85%. Caution: contraindicated in bipolar disorder (risk of mania), seizure disorder (lowered threshold), in patients with parasomnias, and untreated OSA.

2. Stimulus control therapy

Re-establishes the bed/sleep association. Rules:

  • Use the bed only for sleep and sex.
  • Go to bed only when sleepy.
  • If unable to fall asleep within ~20 minutes, leave the bed, engage in a quiet activity in dim light, and return only when sleepy. Repeat as needed.
  • Wake at the same time every day, regardless of total sleep time.
  • No daytime naps (or ≤30 min before 3 pm only if essential).

3. Sleep hygiene education

Standalone sleep hygiene is INSUFFICIENT for insomnia disorder treatment but is a routine adjunct: limit caffeine after noon, avoid alcohol within 3–4 hours of bed, limit nicotine, exercise regularly but not within 3 hours of bedtime, dark/quiet/cool bedroom, no screens within 1 hour of bedtime, no clock watching.

4. Cognitive restructuring

Identify and challenge maladaptive sleep beliefs: “If clinicians don’t get 8 hours, clinicians will be a wreck tomorrow.” “I must sleep — sleep is critical.” Replace with realistic, flexible beliefs. Address the sleep-effort paradox — trying harder to sleep activates wake systems and worsens insomnia.

5. Relaxation techniques

Progressive muscle relaxation, diaphragmatic breathing, body scan, paradoxical intention (try to stay awake), guided imagery, mindfulness-based stress reduction. Particularly helpful for patients with prominent somatic and cognitive arousal.

✓ Pearl

Digital CBT-I (Sleepio, SHUTi, Somryst) — the FDA-cleared prescription digital therapeutic — has demonstrated efficacy comparable to in-person CBT-I in multiple RCTs. When trained therapists are unavailable, recommend a validated digital program. Insurance coverage is rapidly expanding for these tools as awareness of CBT-I as first-line grows.

✗ Pitfall

Do NOT recommend “sleep hygiene only” as standalone treatment. Multiple RCTs and meta-analyses confirm that sleep hygiene alone has minimal efficacy in chronic insomnia disorder — it underperforms not only CBT-I but most active comparators. Sleep hygiene is a foundation, not a treatment.

4. Pharmacotherapy Hierarchy

The 2024 modern hierarchy for insomnia pharmacotherapy reflects evolving evidence on safety, abuse potential, residual sedation, dependence, and sleep architecture preservation. Choose by sleep complaint pattern (sleep-onset vs sleep-maintenance vs early-morning), age, comorbidities, and avoidance of contraindicated agents.

Class / DrugStandard doseBest forNotable cautions
DORAs (preferred new first-line)
Daridorexant (Quviviq)25–50 mg HSSleep onset + maintenance; minimal next-morning residualHalf-life ~8 h; CYP3A4 substrate; avoid potent CYP3A4 inhibitors
Lemborexant (Dayvigo)5–10 mg HSSleep onset + maintenance; faster onset than suvorexantAvoid concurrent strong CYP3A4 inhibitors
Suvorexant (Belsomra)10–20 mg HSSleep maintenance; first-in-class DORAHalf-life ~12 h; next-day driving caution; narcolepsy contraindication
Z-drugs (second-line)
Zolpidem IR (Ambien)5–10 mg HS (women 5 mg)Sleep-onset insomniaComplex sleep behaviors (sleep-driving, sleep-eating); FDA black-box; women metabolize slower
Zolpidem CR (Ambien CR)6.25–12.5 mg HSSleep-onset + maintenanceSame complex behavior risk; longer next-morning impairment
Eszopiclone (Lunesta)1–3 mg HSSleep maintenance; longer half-life ~6 hBitter dysgeusia common; next-day residual
Zaleplon (Sonata)5–10 mg HS or middle of night ≥4 h before risingSleep onset; very short t½ ~1 hLimited maintenance benefit
Sedating antidepressants (off-label)
Trazodone25–100 mg HSInsomnia with depression or anxietyOrthostatic hypotension, priapism (rare); not FDA-approved for insomnia alone
Doxepin (Silenor)3–6 mg HSSleep maintenance, older adults; FDA-approved for insomniaAnticholinergic at higher doses; avoid in narrow-angle glaucoma, urinary retention
Mirtazapine7.5–15 mg HSInsomnia with depression and weight lossWeight gain, daytime sedation, restless legs
Melatonin / Melatonin agonists
Melatonin (OTC)0.5–3 mg HS (small dose better)Circadian disorders, jet lag; modest benefit in insomniaOTC quality varies dramatically; useful in older adults
Ramelteon (Rozerem)8 mg HSSleep-onset insomnia; no abuse potentialModest effect; avoid with fluvoxamine (CYP1A2)
AVOID — outdated/risky
Benzodiazepines (alprazolam, diazepam, temazepam, lorazepam)—Avoid for chronic insomniaTolerance, dependence, withdrawal; CNS depression with opioids; falls/fractures in elderly; cognitive impairment
Diphenhydramine, doxylamine (OTC)—Avoid in older adults (Beers Criteria)Anticholinergic burden, delirium risk, urinary retention, falls; tachyphylaxis within days
Quetiapine for insomnia (no psychiatric indication)—Avoid as first-line hypnoticMetabolic syndrome, weight gain, EPS, QTc; off-label use without indication is increasingly discouraged

Why DORAs are preferred

Dual orexin receptor antagonists block both OX1R and OX2R receptors of orexin (hypocretin) — the wake-promoting neuropeptide system overactive in insomnia. By dampening wake drive rather than enhancing GABA-A inhibition (the mechanism of benzodiazepines and Z-drugs), DORAs preserve sleep architecture, do not cause anterograde amnesia, do not produce complex sleep behaviors, have negligible abuse potential, and minimize next-morning residual sedation. DORAs are also less likely to suppress REM sleep or cause rebound insomnia on discontinuation. Major remaining caveats: cost (limited formulary coverage in some health systems), rare next-morning impairment with longer-half-life agents (suvorexant), and contraindication in narcolepsy.

Z-drugs — appropriate use and the FDA black box

Z-drugs (zolpidem, eszopiclone, zaleplon) bind selectively to α1-containing GABA-A receptors and remain widely used. The 2019 FDA black-box warning highlights complex sleep behaviors — sleep-driving, sleep-eating, sleep-walking, and engaging in other activities while not fully awake — particularly with zolpidem. Discontinue Z-drug if such events occur. Use shortest effective duration; reassess at 2–4 weeks; combine with CBT-I.

Sedating antidepressants — niche role

Trazodone at low dose (25–100 mg) is widely prescribed off-label and is appropriate when insomnia coexists with depression. Doxepin at very low doses (3–6 mg) is FDA-approved for sleep maintenance and is well tolerated in older adults despite its tricyclic class. Mirtazapine is useful for the depression-insomnia-anorexia triad. Avoid using sedating antidepressants alone for primary insomnia without comorbid depression.

⚠
Combination caution. Avoid combining DORAs or Z-drugs with opioids, benzodiazepines, or alcohol — synergistic CNS depression. Caution in older adults with history of falls. Document driving precaution and “do not operate heavy machinery” counseling for any new hypnotic.

5. Special Populations & Tapering

Special populations

PopulationApproach
Older adults (≥65 yr)CBT-I first-line; if pharmacotherapy needed, prefer doxepin 3–6 mg, low-dose DORAs, or melatonin; AVOID diphenhydramine (Beers), benzodiazepines, Z-drugs in fall-risk patients
PregnancyCBT-I first-line; non-pharmacologic only when possible. If essential, low-dose doxepin, very-short-course diphenhydramine (FDA pregnancy class B but anticholinergic), or sleep-positioning. Avoid Z-drugs in 1st trimester
Insomnia with depressionSSRI/SNRI for depression + CBT-I; trazodone 25–100 mg HS adjunct; mirtazapine if weight loss
Insomnia with PTSDCBT-I + imagery rehearsal therapy (IRT) for nightmares; prazosin 1–6 mg HS for nightmares; avoid benzodiazepines (worsen PTSD)
Insomnia with chronic painTreat pain; CBT-I; doxepin or trazodone HS adjunct; avoid chronic opioids for sleep
OSA with insomnia (COMISA)Treat OSA with CPAP first; many insomnia symptoms improve; remaining insomnia → CBT-I
Renal/hepatic impairmentReduce hypnotic doses; avoid Z-drugs in severe hepatic impairment; DORAs require dose reduction with strong CYP3A4 inhibitors
Shift workersStrategic napping, light therapy, melatonin timing, modafinil for excessive sleepiness; treat as circadian rhythm disorder

Tapering hypnotics — structured discontinuation

After CBT-I has been delivered and sleep has stabilized, plan a gradual taper of any hypnotic. Abrupt discontinuation produces rebound insomnia in 30–50% of patients and risks withdrawal in those on benzodiazepines or chronic Z-drugs.

  • Z-drugs and DORAs: 25% dose reduction every 1–2 weeks. Pair with continued CBT-I session focusing on cognitive restructuring around medication dependence.
  • Benzodiazepines: convert to long-acting equivalent (e.g., diazepam) and taper by 10–25% every 1–2 weeks; longer tapers (months) for high-dose or long-duration users.
  • Trazodone, doxepin: small reductions every 2–4 weeks; minimal withdrawal at low hypnotic doses.
  • Melatonin: simply discontinue; no taper required.
📍
When to refer to sleep medicine. (1) Suspected obstructive sleep apnea (snoring, witnessed apneas, EDS, BMI >30) — refer for PSG. (2) Suspected restless legs syndrome with iron deficiency. (3) Treatment-refractory insomnia despite CBT-I and adequate pharmacotherapy trial. (4) Suspected REM behavior disorder (dream enactment) — α-synucleinopathy prodrome. (5) Circadian rhythm disorder requiring chronotherapy. (6) Excessive daytime sleepiness disproportionate to sleep restriction. (7) Insomnia in patients with severe psychiatric comorbidity requiring multidisciplinary approach.

6. Frequently Asked Questions

Why is CBT-clinicians considered first-line over medication?
CBT-I produces durable improvements in sleep that persist after treatment ends, while medication benefits typically dissipate when the medication is stopped. Multiple meta-analyses show CBT-I matches or exceeds the effectiveness of pharmacotherapy at 8 weeks, with substantially greater benefit at 6 to 12 months follow-up. CBT-I has no abuse potential, no anticholinergic burden, no fall risk, and no risk of complex sleep behaviors. The only barriers are cost and access — addressed through digital CBT-I (Sleepio, SHUTi, Somryst) and brief 4-session formats. Pharmacotherapy is appropriate as adjunct or for short-term symptom relief while CBT-I is being delivered.
Should I prescribe DORAs (suvorexant, lemborexant, daridorexant) over zolpidem?
Yes, when both are clinically appropriate, DORAs are preferred. DORAs target the wake-promoting orexin system rather than enhancing GABA-A inhibition. Compared to zolpidem and other Z-drugs, DORAs preserve sleep architecture, do not cause anterograde amnesia, do not produce complex sleep behaviors (sleep-driving, sleep-eating), have negligible abuse potential, and produce minimal next-morning residual sedation. Daridorexant 25 to 50 mg HS, lemborexant 5 to 10 mg HS, and suvorexant 10 to 20 mg HS are reasonable first-line pharmacologic options when CBT-I is unavailable or inadequate. Z-drugs remain reasonable second-line in cost-constrained settings.
Is melatonin actually effective for insomnia?
Modestly. Meta-analyses show melatonin produces small reductions in sleep latency (typically 5 to 10 minutes) and small increases in total sleep time. It is most useful for circadian rhythm disorders (jet lag, shift work, delayed sleep-phase syndrome) where it acts as a phase-shifter rather than a hypnotic. In primary insomnia disorder, melatonin is far less effective than CBT-I, DORAs, or Z-drugs. It is reasonable in older adults whose endogenous melatonin secretion declines, and as adjunct therapy. Use small doses (0.5 to 3 mg, taken 30 to 60 minutes before bedtime) — higher doses do not improve effect. OTC melatonin quality is highly variable; choose USP-verified or pharmaceutical-grade.
When should clinicians avoid prescribing diphenhydramine for sleep?
Diphenhydramine is on the Beers Criteria list and should be avoided in adults aged 65 or older due to high anticholinergic burden, risk of delirium, urinary retention, falls, and constipation. It should also be avoided in patients with narrow-angle glaucoma, BPH, dementia (worsens cognition), and in pregnancy (preferred only short-term if essential). Tachyphylaxis develops within 3 to 5 nights, limiting effectiveness. There is essentially no role for chronic diphenhydramine use as a hypnotic. If a patient is using OTC diphenhydramine nightly, address it directly: replace with CBT-I and/or a more appropriate prescribed agent.
How do I taper a patient off long-term zolpidem or benzodiazepines?
First deliver CBT-I — this dramatically increases successful tapering rates. Then plan a gradual reduction. For Z-drugs: decrease by 25 percent of the current dose every 1 to 2 weeks; pause at any step where rebound insomnia is intolerable; expect total taper to take 4 to 12 weeks. For benzodiazepines: convert to a long-acting equivalent (diazepam) for smoother kinetics, then reduce by 10 to 25 percent every 1 to 2 weeks; longer tapers (3 to 6 months) for high-dose or long-duration users. Pair the taper with continued CBT-I sessions focused on cognitive restructuring around medication dependence and address rebound insomnia behaviorally rather than by reversing the taper. Document reasons for deprescribing in the chart.

S7. Practice Insights: Insomnia Management Beyond Z-Drugs and Trazodone

(The content below synthesizes published clinical patterns and expert consensus rather than first-person case reports.)Reaching for a hypnotic is the reflex when a patient reports weeks of insomnia, and it is the wrong first move — the dominant management failures are: skipping CBT-I in favor of immediate pharmacotherapy (CBT-I is first-line per AASM 2021 and has better long-term outcomes than any drug), default prescribing of benzodiazepines or Z-drugs (zolpidem, eszopiclone, zaleplon) that cause dependence, rebound insomnia, falls in elderly, and cognitive side effects, over-reliance on off-label trazodone with weak evidence and significant orthostasis/anticholinergic burden, and underutilization of dual orexin receptor antagonists (DORAs: suvorexant, lemborexant, daridorexant) which have substantially better long-term safety profile. In experienced practice, four operational principles guide successful management.

1. CBT-I as Default First-Line

In practice, every chronic insomnia patient (insomnia disorder per DSM-5-TR criteria, >3 nights/week for >3 months) starts with cognitive behavioral therapy for insomnia (CBT-I) rather than medication. CBT-I components include: sleep restriction (matching time-in-bed to actual sleep time, then gradually increasing), stimulus control (bed only for sleep and sex, get up if not asleep within 20 minutes), sleep hygiene (consistent wake time, no naps, no caffeine after noon, no screens 1 hour before bed), cognitive restructuring (challenging catastrophic thoughts about sleep loss). Access barriers (limited sleep psychologist availability) are increasingly solved by digital CBT-I platforms (Sleepio, Somryst, SHUTi) which have FDA clearance and good real-world data. Long-term outcomes from CBT-I exceed any pharmacotherapy.

2. DORAs as Preferred Pharmacotherapy

In experienced hands, when pharmacotherapy is needed, dual orexin receptor antagonists (DORAs) are the first-line: suvorexant 10-20 mg, lemborexant 5-10 mg, or daridorexant 25-50 mg. The orexin neuropeptide system promotes wakefulness; antagonizing it allows physiologic sleep. DORAs have substantially better safety profile than benzodiazepines or Z-drugs: no GABAergic dependence/tolerance, less next-day sedation, less cognitive impairment, no association with falls in elderly (vs benzo/Z-drug 2-3x fall risk), no respiratory depression. Cost is the main barrier ($300-500/month without insurance coverage), and insurance authorization often requires documented failure of CBT-I or other agents. Side effects: nightmares (less common than predicted from REM-related mechanism), morning grogginess in some patients (more with suvorexant than lemborexant or daridorexant).

3. Avoid Benzos and Z-Drugs as Default

In experienced practice, benzodiazepines (temazepam, lorazepam) and Z-drugs (zolpidem, eszopiclone, zaleplon) are NOT first-line and used only for short-term (<2-4 weeks) or as bridge therapy while CBT-I is initiated. Their harms include: dependence and rebound insomnia on discontinuation, next-day sedation and cognitive impairment, falls in elderly (~2-3x risk), motor vehicle crash risk, parasomnias (sleep-driving, sleep-eating with zolpidem), and the AGS Beers Criteria lists them as inappropriate for older adults. Long-term use is associated with cognitive decline and increased dementia risk in observational studies. When clinicians inherit a patient on chronic Z-drug or benzo therapy, experienced clinicians taper with patient buy-in over weeks-months, often transitioning to CBT-I and/or DORA bridging.

4. Trazodone Is Overprescribed

Trazodone is widely prescribed off-label for insomnia at low doses (50-100 mg), but the evidence base is weak and the side-effect profile is meaningful. In experienced practice clinicians have largely moved away from trazodone for insomnia given: orthostatic hypotension with associated fall risk in elderly, priapism risk in men (rare but irreversible if not treated promptly), anticholinergic burden adding to cognitive concerns, daytime sedation hangover. Trazodone retains a role as adjunct in patients with depression where the antidepressant effect is also wanted, but using it solely for insomnia in the absence of mood disorder is rarely the preferred approach now. Other sedating antidepressants (mirtazapine, doxepin 3-6 mg) and antipsychotics (quetiapine off-label) are similarly problematic — used for sedating side effect rather than therapeutic indication.

Illustrative Scenario: Z-Drug Taper Plus CBT-I Plus DORA Bridging

An elderly woman with 12 years of nightly zolpidem 10 mg use presented requesting “the next strongest sleeping pill” because zolpidem was no longer working. Detailed history revealed she was having nightly arousals 2-3 times, taking 60-90 minutes to fall asleep, and was experiencing morning cognitive fog and three falls in the prior 18 months. Z-drug tolerance plus rebound effect was driving the cycle.

The treating clinicians framed the path as zolpidem taper plus CBT-I plus temporary DORA bridging. She was referred to digital CBT-I (Somryst) and started lemborexant 5 mg nightly. Zolpidem was tapered by 1.25 mg every 2 weeks. The taper was uncomfortable initially with rebound insomnia in weeks 2-4 but she persisted with daily CBT-I exercises. By month 4 she was off zolpidem entirely, on lemborexant 5 mg nightly with adequate sleep, no daytime fog, no falls. By month 8 she successfully discontinued lemborexant and was managing on CBT-I behavioral changes alone. Falls did not recur in the 18 months of follow-up after completing taper.

Takeaway: Long-term Z-drug use commonly produces tolerance and rebound that paradoxically worsens sleep. The combination of digital CBT-I + DORA bridge + gradual Z-drug taper is the modern path forward, including for elderly patients on chronic hypnotics. Fall risk reduction is a major secondary benefit. Patients on chronic hypnotics deserve the conversation about transitioning rather than escalating.

Practice Pattern Refinements vs Standard Approach

  • CBT-I as first-line for chronic insomnia disorder (AASM 2021). Digital CBT-I (Sleepio, Somryst, SHUTi) solves access. Long-term outcomes exceed any pharmacotherapy.
  • DORAs (suvorexant/lemborexant/daridorexant) as preferred pharmacotherapy: no GABA dependence, less next-day sedation, no fall risk in elderly. Cost is barrier.
  • Avoid benzos and Z-drugs as default: only short-term (<2-4 wk) or bridge therapy. Beers Criteria inappropriate for elderly. Falls, cognitive impairment, parasomnia.
  • Move away from trazodone for insomnia alone: orthostasis, priapism, anticholinergic burden. Retain only as adjunct in mood disorder where antidepressant effect wanted.
Insomnia: 4-Pillar Modern Framework Four-panel framework: CBT-I first-line, DORAs preferred pharmacotherapy, avoid benzos/Z-drugs default, move away from trazodone. Insomnia: 4-Pillar Modern Framework 1. CBT-I as First-Line Four components: Sleep restriction (time-in-bed match) Stimulus control (bed = sleep + sex) Sleep hygiene Cognitive restructuring Digital CBT-I (FDA cleared): Sleepio, Somryst, SHUTi Long-term outcomes exceed any drug 2. DORAs as Preferred Pharmacotherapy Suvorexant 10-20 mg Lemborexant 5-10 mg Daridorexant 25-50 mg Orexin antagonism → physiologic sleep Advantages: No GABA dependence/tolerance No fall risk in elderly 3. Avoid Benzos and Z-Drugs Default Harms: Dependence + rebound insomnia 2-3x fall risk in elderly (Beers list) Next-day cognitive impairment Parasomnia (sleep-driving / eating) Possible dementia association Only short-term <2-4 wk or bridge 4. Move Away From Trazodone Weak evidence for insomnia alone: Orthostatic hypotension Priapism in men (rare, irreversible) Anticholinergic burden Daytime sedation hangover Retain only as adjunct in mood disorder Not first-line for insomnia in absence of MDD

Leave a Reply

Your email address will not be published. Required fields are marked *