AASM 2025: when CBT-I meets a sleeping pill, the pill is the one that loses

If you treat patients with comorbid insomnia and anxiety or depression — that is, most of your caseload — this guideline tells you something the field has been hedging on for fifteen years. CBT-I plus a hypnotic is not better than CBT-I alone for the outcomes patients actually care about. Sleep duration may inch up early in treatment; daytime functioning, mood symptoms, anxiety scores, and global insomnia severity do not.

The clinical message is uncomfortable for prescribers. It is also liberating for psychotherapists.

Why this matters for non-sleep clinicians

The 2021 AASM behavioral CPG already named CBT-I a strong first-line treatment, but in real-world practice patients arrive in the consulting room already on zolpidem or trazodone, often for years. Until now there was no formal guidance on whether to add CBT-I to existing pharmacotherapy or to consider tapering. The 2025 guideline reframes the conversation: combination therapy is justifiable mainly when the patient explicitly prioritises rapid increase in total sleep time and is willing to tolerate residual daytime sedation. For everyone else — and especially for patients with mood or anxiety comorbidity, where daytime functioning is the treatment target — CBT-I monotherapy is the cleaner option.

The endorsement by ADAA matters more than it looks. It signals that anxiety-disorder specialists no longer view CBT-I as adjunctive sleep hygiene; they see it as a primary intervention for the insomnia component of mood and anxiety disorders. Insomnia is no longer a symptom to be managed with bedtime medication while you treat the "real" disorder. It is a treatment target with its own evidence-based protocol.

What changes in your practice

For psychotherapists already trained in CBT-I: your scope just widened. You can now refer patients into your protocol with cleaner backing — not as a complement to a sleeping pill, but as an alternative. Document the conversation: the AASM has formally said combination therapy does not outperform CBT-I alone on patient-relevant endpoints. That is a defensible clinical decision.

For psychotherapists not yet trained in CBT-I: the workforce gap is the bottleneck. The VA's national dissemination programme — the largest CBT-I rollout in any health system — has trained licensed mental-health clinicians in 4-8 session protocols and reported pre-post effect sizes of d = 2.2-2.3 in routine care, comparable to RCT outcomes. The training pathway is not specialty sleep medicine; it is brief, manualised, and within the technical reach of any CBT-trained therapist. The IOCDF / Beck Institute / Society of Behavioral Sleep Medicine training landscape is active. The clinical demand is enormous. The CPT code is established (90834).

For prescribers in your referral network: the guideline gives you a script for tapering conversations. "AASM 2025 says combination is not better than CBT-I alone for daytime symptoms" is a sentence that lands.