Clinical trials said sleep paralysis affected fewer than 2% of people taking lemborexant (brand name Dayvigo), the dual orexin receptor antagonist approved for insomnia in 2019. A new study of how real patients actually experience the drug tells a different story.
Researchers from Rutgers Brain Health Institute, Virginia Commonwealth University, and the University of Sydney published a mixed-methods infoveillance study in Sleep Medicine analyzing 260 publicly posted patient reviews of lemborexant from the consumer health site Drugs.com. Using reflexive thematic analysis, the team mapped out what patients said drove them to either stick with the medication or abandon it.
A Third Benefited, Half Didn't — and 70 Quit
Of the 260 reviewers, roughly a third reported that lemborexant meaningfully improved their sleep. About half described little or no benefit. Within the full sample, 70 reviewers said they had already stopped taking the drug or planned to.
The researchers then compared that discontinuing group against everyone else, and the gap in reported side effects was stark: 40% of those who quit or planned to quit described sleep paralysis, compared with 15% of those who stayed on the medication. Nightmares and next-day grogginess followed a similar pattern, showing up disproportionately among people heading for the exits.
Duration of use mattered too. A majority of reviewers — 61% — had used lemborexant for less than one month, and longer-term users tended to report more favorable experiences overall, suggesting an early period where side effects are most likely to drive discontinuation, before benefits have had time to accumulate.
Why the Gap With Clinical Trial Data
Lemborexant's pivotal SUNRISE trials, which supported its FDA approval, reported sleep paralysis in well under 2% of participants. The discrepancy between that figure and the rates described in real-world reviews is itself informative: clinical trials typically use structured adverse-event checklists administered under controlled conditions, which may not capture symptoms patients find embarrassing, hard to articulate, or simply didn't think to mention unless specifically asked.
Online reviews, by contrast, are unprompted and emotionally candid — patients write them because something, good or bad, felt worth describing. That makes them a different kind of data source: not a substitute for randomized trial safety data, but a window into which side effects actually change behavior in everyday use.
The authors were careful to flag the limits of their method. People with unusually positive or unusually negative experiences are more likely to post reviews in the first place, which can skew a sample like this toward the extremes and away from the more neutral experience of the "silent majority" of patients. The study cannot establish how common sleep paralysis or nightmares truly are among all lemborexant users — only that, within this self-selected group, those symptoms were strongly associated with quitting the drug.
Orexin Antagonists and Dream-Related Side Effects
Lemborexant belongs to a drug class — dual orexin receptor antagonists — that works by blocking orexin, a brain signaling molecule that promotes wakefulness, rather than by broadly sedating the brain the way older sleep medications do. That mechanism is thought to produce more natural-feeling sleep architecture for many patients, but orexin signaling is also tightly linked to REM sleep regulation, which may explain why vivid dreams, nightmares, and sleep paralysis — a state in which REM-related muscle paralysis persists briefly into wakefulness — surface as a recognized, if apparently underestimated, side effect of the class.
What This Means for Patients
For patients starting lemborexant, the findings suggest the first few weeks deserve particular attention: that's when the real-world data shows side effects most often emerge and most often prompt people to stop. Patients who experience sleep paralysis, unusually vivid nightmares, or persistent next-day grogginess have a basis in this research for raising the issue directly with a prescriber rather than assuming it's a rare fluke — and clinicians counseling patients before starting the drug may want to describe these specific symptoms rather than relying on the broader, lower incidence figures from the original trials. The study doesn't argue against lemborexant's use; it argues for a more honest conversation about what a meaningful share of patients actually experience once they take it home.
