Trazodone vs Quviviq vs Belsomra

The sleep medication landscape in 2026 is in a weird place. Trazodone is still the default for most prescribers, mostly out of habit and price. The orexin antagonists (a newer class of sleep medications, Belsomra and Quviviq and Dayvigo, which work by blocking the wakefulness signal in the brain rather than just sedating you on top of it) are the first class with a mechanism that finally makes sense for chronic insomnia. The Z-drugs (Ambien, Lunesta, Sonata, the older “sleeping pills” that work on the same brain system as benzos) are getting prescribed less because of accumulated safety concerns. Doxepin sits in its own niche. Here’s how to think about picking among them, without pretending the choice is more clear-cut than it actually is.

Trazodone, the workhorse

Trazodone is an old antidepressant that almost nobody uses for depression anymore because we have better options, but at low doses it makes people sleepy and we’ve been using it for sleep for the better part of three decades. Standard dose is 50mg to 100mg at bedtime. Some patients end up at 150mg or 200mg. It’s cheap (generic, around five dollars a month), works for most people, and there’s enough familiarity with it that prescribers reach for it almost reflexively when sleep comes up.

The catch is the morning hangover. About a third of patients feel groggy until noon, and the percentage climbs at higher doses. There’s also a rare but real risk of priapism in men (an erection that won’t go down, which sounds funny on paper until it happens and lands you in the ER and possibly requires surgery to drain). The warning is real and worth mentioning before the prescription goes in, not after.

For sleep maintenance, trazodone is decent. For sleep onset, it’s slow, takes 30 to 60 minutes to land. Some patients build mild tolerance after months and need to push the dose up, which is part of how a 50mg starting prescription drifts to 200mg over a couple years if nobody’s paying attention.

The orexin antagonists (Belsomra, Quviviq, Dayvigo)

This class is the newest and the most mechanistically interesting. Orexin is a brain peptide that promotes wakefulness… it’s basically the brain’s “stay alert” signal. Block orexin receptors, the signal gets quieter, you fall asleep. It’s the first sleep medication class that targets the wakefulness side of the equation rather than just sedating you on top of an awake brain.

The practical result is sleep that subjectively feels more natural than what you get from Ambien or trazodone. Patients report actual dreams. They report feeling rested in the morning rather than drugged. The next-day cognitive effects are meaningfully smaller than with benzos or Z-drugs, which matters a lot for people whose jobs require their head to be on straight first thing in the morning.

The catch is cost. Quviviq runs around $400 to $500 a month list price. Belsomra similar. Insurance coverage is gradually improving, but most plans still require documented failure on cheaper options first. With copay cards and the right insurance, commercial out-of-pocket can drop under $30, which makes the medication realistic. Without that, it’s a budget conversation that ends in the cheaper option for most patients.

Effectiveness is decent but not dramatic… the trials show modest improvements over placebo, the patients who respond often love it, and the patients who don’t respond don’t, at which point the conversation moves on.

The Z-drugs (Ambien, Lunesta)

The reflex on these has cooled meaningfully. They work fast and hard, which is what makes them effective and also what creates the problems. Sleep behaviors (sleepwalking, sleep-eating, sleep-driving, sleep-emailing things you’d never email awake) are real and the rate is higher than the package insert language makes it sound. Next-day cognitive impairment is real. Tolerance and dependence are real, even if less severe than with benzos.

For occasional use, fine. For nightly long-term use, most prescribers worth their license would rather have you on something else, because the long-term picture on Z-drugs has gotten worse the more carefully it’s been looked at.

Doxepin, in its own lane

Low-dose doxepin (3mg or 6mg) is genuinely good for sleep maintenance, not great for sleep onset. Cheap, generic, mild profile. The very low doses we use for sleep are different from the antidepressant doses, and at the sleep doses the medication mostly just helps you stay asleep without the morning hangover trazodone leaves behind. It’s underused for what it does well.

How the choice actually gets made

The first thing that matters is what the actual sleep problem is. For sleep onset trouble, doxepin alone isn’t enough… trazodone, an orexin antagonist, or a short course of a Z-drug for occasional bad nights are the options. For sleep maintenance trouble (waking up at 3am and not getting back to sleep), doxepin is often the cleanest option, and Quviviq has the strongest data of the orexin class for this specific use.

The second thing is the patient’s cost situation. With cash pay or restrictive insurance, trazodone or doxepin go first. With commercial insurance plus a manufacturer copay card, an orexin antagonist becomes reasonable.

Third, what other medications the patient is already on. Trazodone has some serotonergic activity and we have to think about interactions if there’s an SSRI or SNRI already in the picture. Orexin antagonists have fewer drug interactions and slide into existing regimens more cleanly.

Fourth, how much next-day fog actually matters to this patient’s life. The software architect whose job depends on being clear-headed at 8am has a different answer than the retired guy whose mornings can be slow without consequence. Trazodone hangover is common, orexin antagonists have lower next-day effects, Z-drugs have the weird-behavior problem… none of which is hypothetical, all of it changes who picks what.

The pattern this is changing

The kind of guy this conversation matters most for is somebody who’s been on trazodone for a year or two for sleep, sleeps fine on it, and is foggy until 10 or 11am. He’s tried cutting the dose. At the lower dose he’s waking up at 3am again. He’s been told by primary care that the morning grogginess is just how trazodone goes and to live with it. A switch to an orexin antagonist, when cost is manageable, usually takes a few nights to settle in and then he’s sleeping through and waking up clear. It’s the kind of medication change where the patient describes it as one of the bigger quality-of-life shifts he’s made, which is what good sleep does for anyone who’s been getting bad sleep for a while.

The cost piece matters here. Without the copay card and decent insurance, the orexin antagonist isn’t doable for most people and we’re back to trazodone. With the copay card and decent insurance, it’s a different conversation entirely.

Sleep on an orexin antagonist subjectively feels more natural than what you get from Ambien or trazodone. Patients report dreams. They report feeling rested in the morning, not drugged.

What sleep meds don’t fix

If the sleep problem is actually anxiety, the anxiety still needs treating. If it’s sleep apnea, the sleep medication may mask the apnea and make the picture worse, not better, and a sleep study is the move before any prescription. If it’s depression, the depression treatment is the lead and the sleep often follows. If it’s a partner who snores like a chainsaw, that’s a completely different conversation.

CBT-I (cognitive behavioral therapy for insomnia, a structured short-term therapy specifically built for sleep) is the most effective long-term intervention for chronic insomnia and the single most underused. It runs six to eight sessions. Insurance often covers it. There are apps that deliver a version of it for people who won’t or can’t get to a therapist. For most patients with chronic insomnia, CBT-I should be part of the picture even if medication is also part of the picture. The behavior piece is what keeps things working long-term once the medication is doing its part in the short term.

Where the autonomy stance lands

The patient picks the trade. My job is the honest take on which medication does what, what the cost picture looks like, what the side effect profile is, and how the choice maps to the specific sleep problem they’re actually trying to solve. I’m a provider, not a parent. If somebody hears the trazodone-vs-orexin-antagonist conversation and decides he wants to try the cheaper option first to see if it works, fine. If somebody hears it and decides he’d rather pay the copay for the cleaner morning experience, also fine. Both are reasonable answers depending on the picture and the budget.

What’s nice to hear, because the rest of this is comparing side effect profiles, is that for the patients where any of these medications work, the improvement in how the next day feels is genuinely large. Sleep is one of the things people don’t realize how much they’ve been missing until they get it back, and the medication that finally lands tends to feel less like a treatment and more like getting your life back.

Bottom line

For most adults with chronic insomnia, the default in 2026 is doxepin if the problem is staying asleep, or an orexin antagonist if the cost works and the problem is staying asleep. Trazodone if cost is the constraint and the patient tolerates the morning hangover. Z-drugs reserved for occasional bad nights, not nightly long-term use. Whatever the medication choice, sleep hygiene and CBT-I should be part of the conversation, because the medication is the easy part and the behavior is what keeps things working long after the prescription does its job. Sleep is one of the bigger quality-of-life levers in psychiatry, and most prescribers undersell how much getting it right matters for everything else.

Sources

1. Sateia MJ, Buysse DJ, Krystal AD, et al. Clinical Practice Guideline for the Pharmacologic Treatment of Chronic Insomnia in Adults. J Clin Sleep Med. 2017;13(2):307-349. PMID 27998379.
2. Herring WJ, Connor KM, Snyder E, et al. Suvorexant in Patients With Insomnia: Pooled Analyses of Three-Month Data from Phase-3 Randomized Controlled Clinical Trials. J Clin Sleep Med. 2016;12(9):1215-1225. PMID 27397664.
3. Trauer JM, Qian MY, Doyle JS, et al. Cognitive Behavioral Therapy for Chronic Insomnia: A Systematic Review and Meta-analysis. Ann Intern Med. 2015;163(3):191-204. PMID 26054060.