Insomnia therapy works for heavy drinkers, but it does not cut the drinking

Insomnia is common in people with alcohol use disorder, and the two problems feed each other: poor sleep can undermine treatment and raise the risk of relapse. Cognitive behavioural therapy for insomnia, known as CBT-I, is the recommended first-line treatment for insomnia in general, yet its track record in people who drink heavily has never been well characterised. That leaves an open question with real stakes: does the gold-standard sleep treatment still work when alcohol is in the picture, and does fixing the sleep help the drinking too? This review set out to answer both.

The researchers systematically searched four major databases up to February 2024 for randomised controlled trials of CBT-I in adults with heavy alcohol use or any level of alcohol use disorder plus comorbid insomnia. Eight trials met the bar, covering 426 adults, of whom 68.78% were men. They pooled the results to estimate how insomnia severity, measured on the Insomnia Severity Index, changed over time in CBT-I groups compared with control groups, and did the same for alcohol craving, alcohol-related psychosocial problems, and heavy-drinking or abstinent days. Each trial's risk of bias was also formally assessed.

CBT-I produced a large reduction in insomnia severity compared with control conditions: an estimated 5.51 points on the Insomnia Severity Index straight after treatment. The benefit was durable, with an estimated 4.39-point advantage at one-to-three-month follow-up across seven studies and a 4.55-point advantage at six months across four studies. The alcohol side of the story was different: alcohol-related outcomes were reported less consistently across trials, and no significant differences emerged between CBT-I and control groups. The included trials were judged to carry a low or moderate overall risk of bias for all outcomes.

This confirms that CBT-I's well-established benefit for insomnia extends to a group that has often been left out of the sleep literature: people who drink heavily or meet criteria for alcohol use disorder. The authors conclude the finding supports wide implementation of CBT-I in alcohol prevention and treatment settings. What it complicates is the hopeful idea that treating the insomnia would pull drinking down with it: on the evidence pooled here, that did not show up. Whether better sleep helps drinking outcomes over longer periods, or in trials designed to measure drinking properly, remains an open question.

If alcohol and insomnia have become entangled in your life, this is a reason to think the standard, structured insomnia therapy is still worth knowing about: it worked about as you would hope, and the improvement lasted at least six months in the trials that checked. It is equally worth knowing what it did not do: craving and drinking measures did not significantly improve, so better sleep is best seen as a goal in its own right rather than a back-door route to drinking less. The evidence base is stronger for men than women, since roughly two-thirds of participants were male.