The research, published in the Journal of Emergency and Disaster Medicine, examined 96 adult encounters for acute peripheral vertigo at Rasool-e-Akram Hospital in Tehran between January and June 2024. Peripheral vertigo, in which the sensation of spinning originates in the inner ear rather than the brain, is most often caused by benign paroxysmal positional vertigo, vestibular neuritis, or Ménière’s disease. Yet because these same symptoms can occasionally signal a stroke, emergency clinicians face a diagnostic tightrope: they must rule out dangerous central causes while relieving distressing symptoms and keeping patients moving through a crowded department.

The research team, led by emergency medicine physicians and researchers affiliated with Iran University of Medical Sciences, conducted a cross-sectional review of patient charts, excluding anyone with a documented central cause of vertigo, incomplete management records, or a history of chronic vertigo. The cohort, with a mean age of 60.3 years and a striking 68.8 percent female majority, was analyzed for five vertigo-specific drugs commonly used in emergency settings: ondansetron, betahistine, promethazine, metoclopramide, and benzodiazepines such as diazepam. The primary outcome was length of stay, measured in minutes from documented arrival to physical departure from the department.

The headline result was stark. Patients who received three or more distinct vertigo medications stayed a median of 652 minutes, compared with 467.5 minutes for those given two or fewer drugs, a difference that held up under non-parametric statistical testing. In a multivariable linear regression adjusted for age, sex, consultation patterns, and self-discharge status, complex regimens were independently associated with an additional 180.9 minutes of department time, with a 95 percent confidence interval of 76.0 to 285.7 minutes and a p-value of 0.0009. The mean stay in the simple-regimen group was 528 minutes, while the complex-regimen group averaged 752.3 minutes, a gap of nearly four hours.

Drug-specific analyses sharpened the picture further. Benzodiazepines, the sedating class often deployed to dampen the anxiety and vestibular storm of acute vertigo, were strongly linked to longer stays, adding an estimated 356.2 minutes even after adjustment for regimen complexity and consultations. The authors suggest this may reflect the sedation burden such drugs impose, which typically mandates prolonged observation before patients can be safely discharged. Promethazine, by contrast, was independently associated with a shorter stay, reducing department time by an estimated 161.3 minutes, possibly because it achieves faster symptom control or because clinicians reserve it for more straightforward presentations. Ondansetron, betahistine, and metoclopramide showed no clear independent association, though the authors caution that small exposed groups and confounding by indication, in which sicker patients receive certain drugs, make these estimates imprecise.

Consultations emerged as the other major lever on throughput. Neurology was the most frequently requested specialty, consulted in 28.1 percent of encounters, followed by cardiology at 15.6 percent. More than half of visits required no consultation at all, but 40.6 percent involved one and 7.3 percent involved two. In the adjusted model, neurology consultation added an estimated 163.9 minutes, while internal medicine and ear, nose, and throat consultations were associated with even larger increases of 501.3 and 476.0 minutes respectively, all statistically significant. The authors interpret these large effects not as evidence that the consultations themselves are wasteful, but as a signal that their involvement flags clinically challenging presentations, such as cardiovascular comorbidity or overlapping otologic disease, that demand additional diagnostic workup.

The prescribing patterns documented in the study reveal a clear hierarchy of practice. Ondansetron, a serotonin-receptor antagonist used to control vertigo-related nausea, was the first drug ordered for nearly half the cohort and was administered at some point to more than 70 percent of patients, consistent with international guideline support for this class. When a second-line agent was needed, betahistine topped the list, reflecting its long-standing track record in peripheral vestibular disease. Diazepam appeared in a smaller number of regimens, hinting at clinicians’ awareness of the anxiety that frequently accompanies acute vertigo, a dimension emphasized in the psychosocial literature on vestibular disorders.

The authors are careful to frame their findings as hypothesis-generating rather than prescriptive. Because the study lacked standardized measures of vertigo severity or clinical acuity at arrival, it is entirely possible that patients who appeared more unwell both received more medications and required longer observation, meaning residual confounding by illness severity may partly explain the association between regimen complexity and longer stays. The single-center design, the absence of data on time to symptom relief, functional outcomes such as safe ambulation at discharge, and unplanned return visits, and the inability to analyze non-pharmacologic interventions like canalith repositioning maneuvers all constrain the conclusions. Rare consultation types produced wide confidence intervals that larger samples would need to tighten.

Even so, the operational message resonates with a broader body of evidence linking polypharmacy and regimen complexity to slower hospital throughput in other acute conditions, from bronchiectasis to sedation-heavy critical care. The findings align with expert recommendations favoring stepwise pharmacotherapy, in which drugs are added sequentially as needed, rather than simultaneous multi-drug administration from the outset. They also echo prior reports that specialty input, while essential when central causes are suspected, can inadvertently lengthen emergency department stays when requested routinely rather than selectively for peripheral vertigo.

The study’s authors propose that emphasizing focused bedside assessment, tools such as the head-impulse, nystagmus, test-of-skew examination, stepwise rather than simultaneous drug therapy, and judicious specialty consultation could improve both the quality and the pace of vertigo care. Prospective, multi-center interventional studies will be needed to determine whether pathways built on these principles can safely shorten emergency department stays without compromising diagnostic safety, particularly given the stakes of missing a stroke masquerading as benign dizziness. For now, the study offers emergency departments a concrete, testable target: fewer drugs, chosen deliberately, may be the fastest route to getting dizzy patients back on their feet and out the door.

Subject of Research: The association between pharmacological treatment complexity, specialty consultations, and emergency department length of stay in adults with acute peripheral vertigo.

Article Title: Therapeutic approaches and their association with hospitalization duration in patients with peripheral vertigo presenting to the emergency department

Article References: Therapeutic approaches and their association with hospitalization duration in patients with peripheral vertigo presenting to the emergency department. (n.d.). https://doi.org/10.1007/s44467-025-00007-4

Image Credits: AI Generated

DOI: 10.1007/s44467-025-00007-4

Keywords: peripheral vertigo, emergency department, length of stay, polypharmacy, benzodiazepines, promethazine, betahistine, ondansetron, neurology consultation, vestibular disorders, regimen complexity, emergency medicine

The research, led by Kentaro Matsui of the National Center of Neurology and Psychiatry in Tokyo and the Japan Somnology Center, together with colleagues including senior author Yuichi Inoue, took advantage of a uniquely valuable resource: the Dokenpo health insurance claims database, which covers a large, anonymous Japanese population from 2019 through 2024. The investigators identified 605 patients who had been continuously using benzodiazepines or benzodiazepine receptor agonists, which include the so-called Z-drugs such as zolpidem, for at least six months before starting either suvorexant or lemborexant, the two dual orexin receptor antagonists available in Japan. To qualify, patients had to continue the orexin antagonist for at least six months, ensuring that the analysis captured sustained treatment rather than brief trials. All bedtime sedative doses were converted into diazepam equivalents, a standard metric that allows the potency of very different hypnotics to be compared on a single scale.

The study’s design is its methodological strength. Rather than comparing one group of patients against another, the researchers used a mirror-image, self-controlled approach in which each patient served as his or her own control. Dose trajectories in the six months and more before orexin antagonist initiation were compared directly with dose trajectories afterward. This design elegantly neutralizes a major source of bias in observational drug research: the stable confounders that differ between people who switch medications and those who do not. Genetics, personality, severity of insomnia, socioeconomic circumstances, and dozens of other variables that could otherwise distort results are held perfectly constant, because every patient contributes data to both sides of the comparison. What remains vulnerable to bias is primarily time-related change, such as natural fluctuation in insomnia severity or secular trends in prescribing practice, which the authors acknowledge as a limitation of the method.

The findings are striking. Before starting the orexin antagonists, the 605 patients were taking an average of 1.9 different sedative medications simultaneously, at a mean total bedtime dose of 8.2 milligrams of diazepam equivalent. After at least six months of adjunctive treatment with suvorexant or lemborexant, 42.1 percent of patients had cut their bedtime benzodiazepine or benzodiazepine receptor agonist dose by at least half, and 21.3 percent had discontinued these drugs entirely. The mean bedtime dose fell from 7.2 to 5.2 milligrams of diazepam equivalent in the suvorexant group and from 8.3 to 5.2 milligrams in the lemborexant group. For a population in which dose reduction has historically proven so difficult that many patients remain on stable hypnotic doses for a decade or more, these numbers represent a clinically meaningful shift rather than a marginal one.

The statistical modeling behind the study adds important nuance. Using logistic regression with adjustment for potential confounders, the team identified a constellation of factors that predicted success. The most powerful was age: patients aged 65 years or older had roughly two and a half times the odds of achieving a 50 percent or greater dose reduction compared with younger patients, with an odds ratio of 2.485 and a 95 percent confidence interval of 1.190 to 5.192. The same group showed similarly elevated odds of full discontinuation, at an odds ratio of 2.636. This result carries particular public health weight, because older adults are precisely the population in which benzodiazepine-related harms, from hip fractures to accelerated cognitive decline, are most devastating. A pharmacological strategy that works best in the most vulnerable patients inverts the usual risk calculus of sedative prescribing.

Two additional predictors emerged from the adjusted analyses. Patients who began with lower total benzodiazepine doses were more likely to succeed in tapering, a pattern consistent with pharmacological intuition, since lighter users face weaker physical dependence and less severe withdrawal when doses fall. And lemborexant use was independently associated with both dose reduction and discontinuation compared with suvorexant, hinting that the choice of orexin antagonist may matter. The authors caution, however, that this observational finding cannot establish that one drug is intrinsically superior; prescribing patterns, patient characteristics, and physician preferences may all have influenced which patients received which agent. Still, the signal aligns with emerging trial data, including the SUNRISE 2 study, which demonstrated the long-term efficacy and tolerability of lemborexant, and with network meta-analyses that have compared the orexin antagonists head to head.

Understanding why the strategy works requires a brief tour of sleep neurobiology. Orexin, also called hypocretin, is a neuropeptide produced by a small population of neurons in the lateral hypothalamus that stabilizes wakefulness. The loss of orexin neurons causes narcolepsy, and conversely, blocking orexin receptors promotes sleep by damping down the brain’s arousal system rather than by globally suppressing neural activity. Benzodiazepines and Z-drugs, by contrast, work through the GABA-A receptor, the same broad braking mechanism recruited by alcohol and barbiturates, which explains their sedative potency but also their liability for tolerance, dependence, respiratory effects, and next-morning impairment. Because dual orexin receptor antagonists act on an entirely separate pathway, they can maintain sleep quality while the GABAergic sedative is tapered, theoretically smoothing the transition and preventing the rebound insomnia that so often defeats deprescribing attempts. Earlier work, including the SLIM study of lemborexant switching and clinical case reports, had suggested this feasibility in smaller and more selected samples; the new claims-based analysis extends the evidence to routine clinical practice at scale.

The broader context makes the findings urgent. Prevalence studies have documented widespread hypnotic use in Japan and globally, and cohort analyses have shown that a substantial fraction of users continue these drugs for years, often at escalating doses and in risky combinations. Deprescribing guidelines recommend gradual tapering combined with cognitive behavioral therapy for insomnia, the gold standard non-pharmacological treatment, but access to behavioral therapy remains limited in many health systems, and tapering without an adequate sleep substitute frequently fails. Previous pharmacological substitution attempts, using melatonin or trazodone, produced mixed results. What the mirror-image analysis adds is real-world evidence that a systematic switching strategy with orexin antagonists can move the needle in ordinary insurance claims data, not just in carefully curated clinical trials.

Caveats remain, and the authors are careful to enumerate them. The mirror-image design cannot fully exclude secular trends or regression to the mean; patients who initiate and sustain a new medication for six months are, by definition, a selected and possibly more motivated group; and claims data cannot capture insomnia severity, cognitive behavioral therapy receipt, or the clinical reasoning behind dose changes. The database also cannot distinguish whether dose reductions were driven primarily by physician advice or patient initiative. Randomized controlled trials of DORA-assisted tapering remain the logical next step, and the odds ratios and effect sizes reported here provide exactly the kind of signal that justifies such trials. Funding for the study came from the Japan Society for the Promotion of Science KAKENHI program, and the data provider’s restrictions mean the underlying claims cannot be shared publicly.

Even with these qualifications, the study’s practical implication is hard to overstate. For clinicians confronting the entrenched problem of long-term benzodiazepine use, particularly in older patients, the results support a concrete, testable protocol: initiate suvorexant or lemborexant, stabilize sleep for several weeks, and then begin a structured taper of the GABAergic hypnotic, monitoring for withdrawal and rebound. More than four in ten patients may halve their dose, and one in five may stop entirely. Given that the alternative is often indefinite exposure to medications associated with falls, fractures, cognitive decline, and dependence, those odds transform a discouraging clinical stalemate into a genuine treatment opportunity, and they position orexin receptor antagonism as a central tool in the global effort to make sleep medicine safer.

Subject of Research: Dual orexin receptor antagonist-assisted dose reduction of benzodiazepines and benzodiazepine receptor agonists in long-term users with chronic insomnia

Article Title: Dual orexin receptor antagonist-assisted dose reduction of benzodiazepines and benzodiazepine receptor agonists: a mirror-image analysis of insurance claims data

Article References: Matsui, K., Sugiura, K., Shimura, A., Takagi, S., Kurihara, K., Takaesu, Y., & Inoue, Y. (2026). Dual orexin receptor antagonist-assisted dose reduction of benzodiazepines and benzodiazepine receptor agonists: a mirror-image analysis of insurance claims data. Journal of Clinical Sleep Medicine, 22(1), Article 163. https://doi.org/10.1007/s44470-026-00164-x

Image Credits: AI Generated

DOI: 10.1007/s44470-026-00164-x

Keywords: insomnia, benzodiazepines, dual orexin receptor antagonists, lemborexant, suvorexant, deprescribing, hypnotics, sleep medicine, mirror-image study, claims data, older adults, drug tapering