Treating Insomnia: Pharmacologic Treatment
There are many pharmacological treatment options for insomnia, including over-the-counter medications, benzodiazepines, non-benzodiazepines, and antidepressants.108 These medications are generally intended for occasional, intermittent use.
Over-the-counter (OTC) medications can be safely and effectively used to occasionally promote a good night’s sleep.109 One of the most common types of OTC sleep medications is antihistamines, such as doxylamine (Unisom) and diphenhydramine (Benadryl). Antihistamines block receptors that respond to histamine; this reduces congestion, sneezing, coughing, and allergy symptoms. Blockade of histamine receptors in the central nervous system causes sedation, and thus, antihistamines can be used as sleep aids. Side effects include daytime drowsiness, dry mouth, and constipation.
There are few rigorously designed trials to definitively determine the efficacy of OTC sleep aids.110 Diphenhydramine can remain in the body for long periods of time, resulting in sedation the following day. Note some people may develop a tolerance or dependence to these medications, and they may cause dangerous side effects when used with other medications. In addition, if taken for a long time and then stopped, they may worsen sleep problems.111
Special caution should be used when elderly people use OTC sleep aids. One review found 50‒65% of older adults misused OTC sleep medications. Drug-drug interactions and drug-age interactions were most common.112 In 2015, the Beers Criteria for Potentially Inappropriate Medication Use in Older Adults recommended individuals over age 65 not use products containing the ingredients diphenhydramine and doxylamine.113
Speak with a physician prior to using a sleep aid to ensure it does not interact with any medications you currently take or conditions you have. Avoid alcohol; do not engage in activities that require being alert after taking the medication, such as driving a vehicle; and do not use any sleep medications for more than two weeks.109
Benzodiazepines (eg, alprazolam [Xanax], clonazepam [Klonopin], and diazepam [Valium]) were the cornerstone for treatment of insomnia until the 1990s. These medications enhance effects of the neurotransmitter gamma-aminobutyric acid (GABA)—one of the main inhibitory neurotransmitters in the brain—by binding to multiple brain receptor sites.114 Studies found benzodiazepines enhance sleep onset, reduce the number of nighttime waking, and improve total sleep time and sleep quality with short-term use.108
Benzodiazepines can be classified based on their duration of action. Short-acting benzodiazepines are more likely to cause withdrawal symptoms, whereas long-acting ones are more likely to leave users feeling groggy.114 A recent longitudinal cohort study of over 200 nursing home residents in Belgium found long-term use of benzodiazepines actually decreased sleep quality over the course of one year, as compared to people who did not use these medications. This suggests chronic use of these drugs does not maintain a high sleep quality.115
Non-benzodiazepines, also called benzodiazepine-like drugs, such as zaleplon (Sonata), zolpidem, and eszopiclone, act on fewer brain receptors than benzodiazepines, and therefore are typically associated with fewer side effects.4 Zaleplon, one of the first non-benzodiazepines developed for the treatment of insomnia, has been proven effective in reducing the amount of time it takes to fall asleep. Its short half-life (1 hour) also reduces the risk of lasting effects the following morning, which may make it less useful for people who wake up during the night.116 Older adults should not take zaleplon because it is not as safe or effective as other medications available.108
Note while using zaleplon, some individuals experienced sleep disturbances such as getting out of bed and engaging in activities like driving cars, having sex, or consuming food while partially asleep. These people were typically unable to remember what happened the next day. Also, a person’s mental health may change unexpectedly while using this medication, including aggressiveness, hallucinations, memory problems, depression, confusion, and suicidal ideation. Zaleplon should only be taken under the direction and observation of a healthcare professional.116
Zolpidem's half-life (about 2.5 hours) may make it more effective at reducing the amount of time it takes to fall asleep and aid in staying asleep while reducing residual daytime sleepiness.108 A literature review found 10 mg of zolpidem in adults and 5 mg in those age 65 and over reduced sleep latency and increased sleep duration in people with insomnia. Residual daytime effects are not common, as long as the individual is in bed for at least eight hours before waking for the day.117 Zolpidem clears more slowly in women than men. Morning blood levels may be higher, which may affect psychomotor performance. In 2013, the Food and Drug Administration (FDA) required manufacturers to lower the recommended dosage from 10 mg to 5 mg for immediate-release preparations and from 12.5 mg to 6.5 mg for extended-release forms.108
Eszopiclone has also been shown to be effective at improving sleep.118,119 This medication lengthens total sleep time and helps people fall asleep more quickly. It takes longer to work than other non-benzodiazepines, but also lasts longer.4 The FDA recently lowered the recommended starting dosage of eszopiclone to 1 mg due to risk of next-day impairment to driving, memory, and concentration. A double-blind study involving 91 people between ages 25 and 40 demonstrated that 3 mg eszopiclone was associated with psychomotor and memory impairment 7.5‒11 hours after dosing. Dosages may be increased to 2‒3 mg under physician guidance, but 3 mg is associated with a state of altered mental awareness.120
Note recent research suggests hypnotic sleep aids may be associated with severe adverse events, including dependency/withdrawal, driving impairment, cognitive difficulties, and an increased risk of accidents or falls. There is also an association between use of hypnotic sleep aids and increased depression, infection, and mortality rates. These risks are most commonly observed in elderly individuals.121 Moreover, in 2012, a well-controlled study revealed an association between sleep aids, such as zolpidem, eszopiclone, and temazepam, and a more than three-fold increased risk of death.14 However, we should note those using hypnotic sleep aid drugs often have poor overall sleep quality, which could be the factor causing the sharply increased risk of death.
In a recent review, 43 of 46 epidemiological studies found that the use of hypnotics was associated with an excess mortality rate, and 45 of the studies found hypnotic use did not benefit patient survival.103 Results from two large cohort studies indicate benzodiazepine use is associated with an increase in all-cause mortality. Some researchers suggest the risks associated with hypnotic sleep aids outweigh any minimal benefit. In fact, in April 2019, the United States FDA announced that the agency was requiring boxed warnings on eszopiclone (Lunesta), zaleplon (Sonata), and zolpidem (e.g., Ambien). A boxed warning is the most prominent warning the agency requires. This move was based on safety monitoring studies that found these drugs are associated with increased risks of engaging in potentially dangerous behaviors while not fully awake, such as driving.122 These medications should be used under the direction and supervision of a physician only.
Antidepressants may be useful in treating insomnia for some people. As many people with depression also struggle with insomnia, these medications may help relieve symptoms of both conditions. Research suggests antidepressants may reduce the amount of time it takes to fall asleep and help prevent nighttime awakenings. In addition, the sedating effects of these medications allow for the relaxation necessary to fall asleep more quickly. In general, antidepressants reduce REM sleep, but seem to have little impact on deeper sleep cycles.123
Many antidepressants, such as doxepin (Silenor) (a histamine receptor antagonist with tricyclic antidepressant properties), trazodone (Desyrel) (a serotonin antagonist and reuptake inhibitor), and amitriptyline (Elavil) (a tricyclic antidepressant), are used to treat insomnia because they have sedative properties.108 Doxepin has been found to increase sleep time without causing significant adverse effects.124,125 Research indicates at doses of 1‒3 mg, doxepin improved sleep onset, duration, and quality over a 12-week period.125,126 Higher doses (3‒6 mg) led to improvements in sleep maintenance and reduction of early morning wakings.126
Some data have shown trazodone, functioning as a mild hypnotic, may temporarily help people fall asleep.127 Trazodone, a serotonin modulator used to treat major depressive disorder, may cause daytime sedation. Few well-designed studies demonstrate its effectiveness in managing insomnia.128 In a randomized, double-blind, placebo-controlled study, 16 insomniacs were given 50 mg trazodone or placebo 30 minutes before bed. The study was designed with two groups of participants, each of which took either the drug or placebo for one week, followed by a one-week washout period, and then another week of placebo or drug—whatever the participants took the first week of the trial.
Trazodone reduced night-time waking and improved sleep quality, but also resulted in small but significant impairments in short-term memory, verbal learning, and arm muscle endurance.129 Researchers concluded that while efficacious in improving sleep, its risks may outweigh its benefits, particularly in those more susceptible to these side effects, including the elderly.108,129 In a randomized, double-blind, placebo-controlled trial involving 30 Alzheimer’s patients, individuals who took 50 mg trazodone once daily for two weeks slept 42.5 minutes more per night compared with placebo. Treatment did not affect daytime sleepiness or cognitive functioning.130
Amitriptyline, a sedating antidepressant that alters brain chemistry to stabilize mood, is sometimes used off label to treat insomnia. Note this medication has been shown to increase suicidal thoughts in individuals under age 24.131 There are no controlled trials evaluating amitriptyline for insomnia in the absence of other medical conditions.128
Mirtazapine (Remeron) is a tetracyclic antidepressant used to treat insomnia. A randomized, double-blind, placebo-controlled trial involving 19 men found 7.5 mg mirtazapine and 50 mg of the neuroleptic quetiapine (Seroquel) increased total sleep time and reduced night waking, although they also increased daytime sleepiness. More research is needed to determine the long-term effects of these medications.132
A recent review of published literature on antidepressants and sleep found the best results come from low dosages of these drugs administered early at night (prior to bedtime) as part of an intervention that also involves behavioral treatments, such as cognitive behavior therapy.133
Ask the Scientist: Insomnia and Risks Associated with Hypnotic Sleep Medications
Daniel F. Kripke, MD, is Professor Emeritus of Psychiatry at UC San Diego. He has spent decades studying sleep and the side effects of sleep medications.
- Hi, Dr. Kripke. Thank you for taking time out of your day to share your thoughts with us. Would you start by telling us a little bit about your background and training?
I am a research psychiatrist who opened one of the country’s first sleep clinics to study the causes of depression. Insomnia and depression are often linked. They have common genetic predispositions, and depression seems to cause insomnia, and vice versa.
- You’ve been studying the association between hypnotic sleep aids and adverse outcomes for quite a long time. Why were you drawn to this area of study?
American Cancer Society data from one million questionnaire participants showed that self-reported long sleep duration and short sleep duration predicted early death.134 Over 40 years ago, I noticed that their data showed that reported use of sleeping pills predicted early death more so than insomnia did. I have since been trying to clarify the risks of sleeping pills.
- What are some of your most compelling findings related to the risks associated with hypnotic sleeping pills?
People who take hypnotics are at significantly greater risk of mortality than people who do not take them. Almost all of 46 studies from all over the world demonstrate this, but the amount of risk has varied greatly in the research due variations in study methodology.103 There is uncertainty surrounding how much of risk is caused by sleeping pills. Controlled clinical trials show that hypnotics can cause depression and are associated with increased risk of suicide. Also, hypnotics considerably increase the risk of infections, likely including an increased risk of potentially severe infections such as pneumonia. Finally, hypnotics impair automobile driving and may increase the risk of falls and other accidents.
- One of the limitations with the research in this area is that much of it is correlational and not necessarily reflective of a causal link between sleeping pills and negative outcomes. Are there trials underway or planned that you think will overcome this uncertainty?
The data for depression and infection include randomized studies that establish causality. Some studies of autopsied deaths, especially those linked to opioids, list hypnotics as contributing or primary causes. Hypnotics are used to put animals to sleep (permanently) and for executing prisoners, so there is no question that, at excessive doses, these drugs can kill. The question is how often they are lethal in combination with risky behavior, health problems, and other medications. There are studies planned to clarify these uncertainties, but I cannot tell you that any definitive study has already begun.
- You’ve expressed some frustration in some of your past published work that the FDA has not adequately responded to these concerns about hypnotics and adverse outcomes. Have you seen any evolution in the agency’s position recently? If not, what is keeping the FDA from taking action?
The FDA has been essentially unresponsive to the new research that demonstrates hypnotic risks. Moreover, the FDA has the authority to order manufacturers to do clinical trials of marketed drugs to establish the causality of newly recognized risks, but it has made no effort to do so. The FDA has its own animal testing facilities that could be used to help clarify the cancer risk associated with these drugs. In my opinion, the FDA position is an unscientific one.
- Do you think hypnotic sleep aids actually work? What does the evidence say?
Recent studies sponsored by the government’s Agency for Healthcare Research and Quality (AHRQ) and by the NIH have found that there is weak evidence to suggest that hypnotic sleep aids make people sleep 10 or 20 minutes more per night. It is generally agreed that hypnotic sleep aids do not make people objectively more alert and productive the day after taking them at bedtime. In fact, much of the evidence suggests sleeping pills make people sleepier and reduce performance time the next day. Moreover, there is absolutely no evidence that sleeping pills improve general health.
- Are there any newer sleep aids that you think are better options than hypnotics? If so, what are they and why would you prefer them over hypnotics?
The AHRQ report and the American College of Physicians guidelines conclude that no drug is as useful for insomnia as cognitive behavioral therapy. I think that when circadian rhythm disturbances or depression are associated with cases of insomnia, bright light treatments are superior to hypnotics. It would be nice to have more long-term studies comparing light versus hypnotics.
- What can people who don’t want to take any form of sleeping medication do to get a better night’s sleep? Are there any particular habits that you suggest?
First of all, there are many reasons people want to sleep better, and they need different approaches. In the sleep clinic, we commonly see patients who think there is something wrong with their sleep when it really is not since their sleep pattern is normal. Part of this comes from drug-company-sponsored propaganda suggesting that everyone needs eight hours of sleep or more, which is not true in all cases. Family doctors, psychologists, and sleep specialists can advise what treatments to try first. There is useful information about cognitive behavioral therapy for the treatment of insomnia on the internet or at the library. The VA now offers its patients a free website that provides cognitive behavioral therapy, and there are several commercial websites that offer similar services for less than the cost of sleeping pills.
- Do you find that people who suffer with insomnia typically make certain mistakes or engage in certain habits that contribute to their insomnia? If so, what are they, and how can they be avoided?
There are many causes of insomnia. One of the most common is spending more time in bed than is needed to sleep: spending less time in bed results in less insomnia. Another common problem is getting too little bright light exposure. People feel better and sleep better with at least an hour a day of outdoor daylight or its equivalent. Since exercise is also helpful, walking for thirty minutes to an hour a day in daylight is a simple and effective treatment. Avoiding too much caffeine and alcohol can promote healthy sleep patterns, as well as not using cell phones, tablets, and computer screens shortly before bedtime. The bright bluish light of modern display screens can be neutralized with software that decreases blue light or by wearing orange (blue-blocking) glasses for an hour or two before bedtime. An irregular sleep schedule may also contribute to insomnia. Staying up late on weekends may make it difficult to get to sleep early enough on Sunday night, which makes it hard to wake on time on Monday morning.
Your comments have been very interesting and insightful, Dr. Kripke. Thanks again for sharing your knowledge with us.