Does CBT-I meet formal criteria for an empirically supported treatment across ages and conditions?

What it actually means to call a treatment "empirically supported"

"Empirically supported treatment" is a formal label, not just a compliment. It comes from a structured checklist—often called Tolin's criteria—that asks whether a therapy has been tested in controlled studies of sufficient quality and quantity to justify calling it well established. A 2020 systematic review took that checklist and applied it directly to cognitive behavioral therapy for insomnia, following the step-by-step procedures the framework specifies [1]. In other words, someone did the bookkeeping: they gathered the treatment-outcome research on CBT-I and judged it against the formal bar rather than simply asserting that it works.

That matters because the question "does CBT-I meet the criteria?" is different from "is CBT-I effective?" A therapy can look promising in individual trials and still fall short of a formal designation, or it can clear the bar while leaving open how broadly it applies. The 2020 review addresses the first question—the formal designation—but, notably, it does not examine whether CBT-I holds up across different age groups or medical and psychiatric conditions [1]. So the formal stamp and the breadth-of-application question have to be answered from separate evidence.

Does it work for adolescents as well as adults?

The age picture is encouraging but comes from clinical and review sources rather than from the formal-criteria review. A 2024 workshop paper describes CBT-I as consisting of teachable, portable, time- and cost-effective skills, and it presents the approach as applicable to both adolescents and adults, with guidance on designing treatment plans for adolescent and young adult patients diagnosed with insomnia [3]. The same source notes that many studies find CBT-I works as well as medication, is extremely cost-effective, and produces longer-lasting benefits—while also pointing out that most child and adolescent psychiatrists are not trained in it [3]. That last point is a practical gap in delivery, not evidence that the treatment fails in younger patients.

It helps to see what CBT-I actually involves, because the components are the same skills regardless of age. The treatment combines behavioral training (stimulus control, sleep restriction, sleep hygiene), cognitive training for unhelpful sleep-related thoughts and behaviors, and attention to physiological factors such as hyperarousal and tension [3]. Because these are structured, learnable skills rather than a drug, the same core package can be adapted to different patients—which is part of why it is described as portable across adolescent and adult settings [3].

What about people with other psychiatric conditions?

Insomnia often travels with other mental health problems, so a natural question is whether CBT-I still helps when a psychiatric disorder is present. A 2014 systematic review looked specifically at CBT-I in psychiatric populations and examined its effectiveness in patients with co-morbid psychiatric disorders [4]. The review treats CBT-I as a collection of separate interventions and notes that some components have stronger empirical support than others—so "CBT-I" is not a single uniform ingredient but a bundle whose parts differ in how well they are backed [4]. Importantly, this review assessed applicability across clinical conditions but did not evaluate whether CBT-I meets the formal empirically supported treatment criteria [4].

Newer work extends the reach further, though in a different format. A 2026 review of digital CBT-I reports evidence of efficacy across diverse populations and clinical conditions, and discusses how the intervention can be adapted and scaled up [2]. Digital delivery matters because access to CBT-I is limited even though it is the recommended treatment for insomnia [2]. Again, though, this review does not explicitly apply the formal criteria—it speaks to breadth and delivery, not to the formal designation [2].

If it is recommended, why do so few people actually get it?

Evidence and formal status do not automatically translate into real-world use. A 2023 study of the Veterans Health Administration found that among more than 5.5 million patients, only about 0.2% received CBT-I as a first-line treatment, while about 4.1% were newly prescribed a sleep medication without prior CBT-I [5]. Patients over 60 and those with substance use disorders were less likely to receive first-line CBT-I than other patients [5]. The authors conclude that adherence to guidelines recommending CBT-I as first-line treatment remains a challenge and call for better implementation and training strategies, with targeted efforts for older patients and those with substance use disorders [5].

This is a useful reality check on the original question. The formal criteria and the population evidence describe what the treatment can do under study conditions; the implementation data describe how rarely it is actually delivered as intended [5]. So the honest answer has two layers: CBT-I has been formally evaluated against empirically supported treatment criteria [1], and it has evidence of working across ages and conditions [2][3][4]—but whether a given patient receives it depends heavily on training, access, and practice patterns rather than on the strength of the evidence alone [3][5].

Sources used in this answer

Related research questions