PURPOSE: Insomnia is problematic for older adults. After behavioral modifications fail to show adequate response, pharmacologic options are used. The pharmacokinetics of agents used to treat insomnia may be altered. This review focuses on the safety and efficacy of medications used to treat insomnia. METHODS: A literature search of Medline, PubMed, and Embase was conducted (January 1966-June 2016). It included systematic reviews, randomized controlled trials, observational studies, and case series that had an emphasis on insomnia in an older population. Search terms included medications approved by the US Food and Drug Administration for insomnia: benzodiazepines (triazolam, estazolam, temazepam, flurazepam, and quazepam), nonbenzodiazepine receptor agonists (non-BzRAs; zaleplon, zolpidem, and eszopiclone), suvorexant, ramelteon, doxepin and trazodone. Off-label drugs such as other antidepressants, antihistamines, antipsychotics, gabapentin, pramipexole, tiagabine, valerian, and melatonin were also included. FINDINGS: Cognitive behavioral therapy and sleep hygiene are considered initial therapy for insomnia. Benzodiazepines are discouraged in the geriatric population, especially for long-term use. Although non-BzRAs have improved safety profiles compared with benzodiazepines, their side effects include dementia, serious injury, and fractures, which should limit their use. Ramelteon has a minimal adverse effect profile and is effective for sleep-onset latency and increased total sleep time, making it a valuable first-line option. Although the data on suvorexant are limited, this drug improves sleep maintenance and has mild adverse effects, including somnolence; residual daytime sedation has been reported, however. Sedating low-dose antidepressants should only be used for insomnia when the patient has comorbid depression. Antipsychotic agents, pramipexole, and tiagabine have all been used for insomnia, but none has been extensively studied in an older population, and all have considerable adverse effects. Gabapentin may be useful in patients with restless leg syndrome or chronic neuropathic pain and insomnia. Diphenhydramine should be avoided in the elderly. Valerian and melatonin are unregulated products that have a small impact on sleep latency and can produce residual sedation. IMPLICATIONS: An ideal treatment for insomnia should help to improve sleep latency and sleep duration with limited awakenings and be without significant adverse effects such as daytime somnolence or decreased alertness. Cognitive behavioral therapy should always be first line treatment. Clinical inertia regarding previous prominent use of benzodiazepines and non-BzRAs will be a significant challenge for patients accustomed to their issuance. The future direction of insomnia treatment should have an emphasis on nonpharmacologic interventions, treating comorbid conditions, and focusing therapy on using benzodiazepines and non-BzRAs as last resorts.