Importance: Although insomnia guidelines recommend the use of several individual hypnotics, the most useful hypnotic for treating insomnia in a clinical setting remains unclear.
Objective: To determine which guideline-recommended hypnotics have lower risks of monotherapy failure and which hypnotics have a higher risk of long-term prescription for insomnia treatment.
Design, setting, and participants: This retrospective observational cohort study used data from the Japan Medical Data Center Claims Database from April 1, 2005, to March 31, 2021. Participants included adults whose first prescribed pharmaceutical treatment for insomnia was guideline-recommended hypnotic monotherapy. Data were analyzed from December 24, 2022, to September 26, 2023.
Exposures: Suvorexant, ramelteon, eszopiclone, zolpidem, and triazolam monotherapy.
Main outcomes and measures: The primary outcome was monotherapy failure, defined as a change in hypnotic or having an additional hypnotic prescribed for insomnia within 6 months of the first prescription of a guideline-recommended hypnotic monotherapy. The secondary outcome was monotherapy discontinuation, defined as no prescription of any hypnotic for 2 consecutive months within 6 months after prescribing a guideline-recommended hypnotic in patients for whom monotherapy did not fail. Monotherapy failure and discontinuation were compared using Cox proportional hazards and logistic regression models, respectively.
Results: The study included 239 568 adults (median age, 45 [IQR, 34-55] years; 50.2% women) whose first prescription for insomnia was guideline-recommended hypnotic monotherapy. During the 6-month follow-up period, 24 778 patients (10.3%) experienced failure of monotherapy with a guideline-recommended hypnotic. In comparison with eszopiclone, there were more cases of monotherapy failure for ramelteon (adjusted hazard ratio [AHR], 1.23 [95% CI], 1.17-1.30; P < .001), fewer cases for zolpidem (AHR, 0.84 [95% CI, 0.81-0.87]; P < .001) and triazolam (AHR, 0.82 [95% CI, 0.78-0.87]; P < .001), and no significant difference between suvorexant and eszopiclone. Among those without monotherapy failure, monotherapy was discontinued in 84.6% of patients, with more discontinuations for ramelteon (adjusted odds ratio [AOR], 1.31 [95% CI, 1.24-1.40]; P < .001) and suvorexant (AOR, 1.20 [95% CI, 1.15-1.26]; P < .001) than for eszopiclone and no significant difference between zolpidem or triazolam and eszopiclone.
Conclusions and relevance: Due to uncontrolled confounding factors in this cohort study, no conclusions regarding the pharmacologic properties of guideline-recommended hypnotics can be drawn based on these results. Further studies accounting for confounding factors, including diagnoses of chronic vs acute insomnia disorder, insomnia and psychiatric symptom severity, and physician attitudes toward hypnotic prescription, are needed.