Trazodone is a favourite off-label sleeping pill. The guideline suggests against it

Trazodone, an old antidepressant now prescribed at low doses far more often as a sleep aid than as a treatment for depression, sits on much thinner evidence than its popularity implies — and the main sleep-medicine guideline explicitly suggests clinicians do not use it for insomnia [s1]. The best synthesis finds a modest, low-certainty improvement in how people rate their sleep, no measurable improvement in objective sleep efficiency, and more next-day side effects than placebo [s2].

Why a depression drug became a sleep drug

Trazodone was approved as an antidepressant, but at the antidepressant doses it is sedating, and clinicians began using much smaller doses to make patients sleepy. It became attractive for reasons that have little to do with proven efficacy: it is cheap, it is not a controlled substance, and it is widely assumed to carry less dependence risk than the hypnotics licensed for insomnia. That combination made it one of the most commonly prescribed drugs for sleep in several countries, most of it off-label — used for a purpose the regulator never evaluated it for.

What the guideline says

The American Academy of Sleep Medicine's clinical practice guideline for the pharmacologic treatment of chronic insomnia in adults graded individual drugs, rather than broad classes, using the GRADE method [s1]. Its weak recommendations for treatment cover a specific list of agents — including suvorexant, eszopiclone, zaleplon, zolpidem and triazolam, each "versus no treatment" and each only weak [s1]. Trazodone is on the other list: the task force suggests that clinicians not use trazodone as a treatment for sleep-onset or sleep-maintenance insomnia, alongside the same recommendation against tiagabine, diphenhydramine, melatonin, tryptophan and valerian [s1].

A weak recommendation against is not a ban, and the guideline is careful that "weak" refers to the strength of the underlying evidence, not the size of any effect in a given patient [s1]. But the direction is unambiguous: on the evidence the task force reviewed, the balance of benefits and harms did not favour using trazodone for insomnia [s1].

What the trials actually show

The most complete synthesis is a Cochrane review of antidepressants for insomnia, which identified 23 randomised trials covering 2,806 participants across the whole drug class [s2]. For trazodone specifically, three studies with 370 participants could be pooled, and they indicated a moderate improvement in subjective sleep outcomes over placebo (standardised mean difference −0.34, 95% CI −0.66 to −0.02) [s2]. That confidence interval nearly touches zero, and the certainty of the evidence was low [s2].

The more revealing result came from the objective measurements. Two trazodone studies used polysomnography — the sleep-laboratory recording that measures sleep directly — and found little or no difference in sleep efficiency against placebo (mean difference 1.38 percentage points, 95% CI −2.87 to 5.63; 169 participants), again low-quality evidence [s2]. The gap between the two is the story: people taking trazodone tend to say they slept a little better, while the instruments record essentially no change. And two studies found more adverse effects with trazodone than placebo — morning grogginess, dry mouth and thirst — though the review judged even this low-quality, because the trials reported harms so sparsely that the drug's tolerability and safety remain genuinely uncertain [s2].

The review's authors were measured about the class as a whole: there may be a small improvement in sleep quality with short-term, low-dose doxepin and trazodone, but the safety data are thin, there was no usable evidence for long-term antidepressant use in insomnia, and none at all for amitriptyline despite its common use in practice [s2]. Low-dose doxepin is the one antidepressant the AASM guideline does treat differently — but trazodone is not in that company [s2].

What it means for a reader

The honest summary is that trazodone for sleep rests on a handful of small, short trials, a subjective benefit that barely clears the threshold of significance, no objective effect on sleep efficiency, and adverse-event reporting too sparse to be reassuring [s2]. That is a weaker evidence base than most people prescribed it would assume, and it is why the guideline steers away from it [s1]. None of this is a reason for anyone to stop a prescribed medicine on their own; dosing, tapering and the choice of drug belong to a clinician, and trazodone's sedative effects and interactions make abrupt changes something to discuss rather than improvise.

For chronic insomnia, the guidelines that matter still put a behavioural treatment first: Health Newspapers has covered why cognitive behavioural therapy for insomnia carries a stronger recommendation than any drug, and the same weak-evidence problem runs through melatonin sold as a sleep fix and the wider pattern of reaching for sedatives to sleep. This article is informational and is not medical advice.

What to watch

Whether the high-quality trials the Cochrane authors called for ever materialise — a drug this widely used for sleep has, remarkably, never been tested at the scale its prescribing volume would justify [s2]. Until it is, trazodone for insomnia remains a habit of practice running well ahead of its evidence.

Sources

• Clinical Practice Guideline for the Pharmacologic Treatment of Chronic Insomnia in Adults — Journal of Clinical Sleep Medicine, 2017-02-15
• Antidepressants for insomnia in adults — Cochrane Database of Systematic Reviews, 2018-05-15