Pharmacologic Treatment of Insomnia Disorder: An Evidence Report for a Clinical Practice Guideline by the American College of Physicians

Background: Pharmacologic interventions are often prescribed for insomnia disorder. Purpose: To assess the benefits, harms, and comparative effectiveness of pharmacologic treatments for adults with insomnia disorder. Data Sources: Several electronic databases (2004–September 2015), reference lists, and U.S. Food and Drug Administration (FDA) documents. Study Selection: 35 randomized, controlled trials of at least 4 weeks' duration that evaluated pharmacotherapies available in the United States and that reported global or sleep outcomes; 11 long-term observational studies that reported harm information; FDA review data for nonbenzodiazepine hypnotics and orexin receptor antagonists; and product labels for all agents. Data Extraction: Data extraction by single investigator confirmed by a second reviewer; dual-investigator assessment of risk of bias; consensus determination of strength of evidence. Data Synthesis: Eszopiclone, zolpidem, and suvorexant improved short-term global and sleep outcomes compared with placebo, although absolute effect sizes were small (low- to moderate-strength evidence). Evidence for benzodiazepine hypnotics, melatonin agonists, and antidepressants, and for most pharmacologic interventions in older adults, was insufficient or low strength. Evidence was also insufficient to compare efficacy within or across pharmacotherapy classes or versus behavioral therapy. Harms evidence reported in trials was judged insufficient or low strength; observational studies suggested that use of hypnotics for insomnia was associated with increased risk for dementia, fractures, and major injury. The FDA documents reported that most pharmacotherapies had risks for cognitive and behavioral changes, including driving impairment, and other adverse effects, and they advised dose reduction in women and in older adults. Limitations: Most trials were small and short term and enrolled individuals meeting stringent criteria. Minimum important differences in outcomes were often not established or reported. Data were scant for many treatments. Conclusion: Eszopiclone, zolpidem, and suvorexant may improve short-term global and sleep outcomes for adults with insomnia disorder, but the comparative effectiveness and long-term efficacy of pharmacotherapies for insomnia are not known. Pharmacotherapies for insomnia may cause cognitive and behavioral changes and may be associated with infrequent but serious harms. Primary Funding Source: Agency for Healthcare Research and Quality. (PROSPERO: CRD42014009908)