Clinical Supplement

Print Friendly 

Thomas Roth, PhD
 Primary Psychiatry. 2007;14(5 suppl 3):1-11


Dr. Roth is director of research and division head of the Sleep Disorders and Research Center at Henry Ford Health System in Detroit, Michigan.

Disclosures: Dr. Roth is a consultant to Abbott, Acadia, Acoglix, Actelion, Alchemers, Alza, Ancile, Arena, AstraZeneca, Aventis, Bristol-Myers Squibb, Cephalon, Cypress, Dove, Elan, Eli Lilly, Evotec, Forest, GlaxoSmithKline, Hyperion, Jazz, Johnson and Johnson, King, Lundbeck, McNeil, MediciNova, Merck, Neurim, Neurocrine, Neurogen, Novartis, Orexo, Organon, Orginer, Prestwick, Proctor and Gamble, Pfizer, Purdue, Resteva, Roche, Sanofi, Schering-Plough, Sepracor, Servier, Shire, Somaxon, Syrex, Takeda, TransOral, Vanda, Vivometrics, Wyeth, Yamanouchi, and Xenoport; is on the speaker’s bureaus of Sanofi and Takeda; and receives grant/research support from Aventis, Cephalon, GlaxoSmithKline, Neurocrine, Pfizer, Sanofi, Schering-Plough, Sepracor, Somaxon, Syrex, Takeda, TransOral, Wyeth, and Xenoport.



Sleep disorders are highly prevalent and underdiagnosed. Approximately 40–70 million Americans suffer from some form of sleep disturbance. Chronic insomnia is typically comorbid with medical psychiatric disorders such as depression and anxiety. Chronic insomnia is prolonged and is associated with a variety of daytime consequences as well as decreased quality of life and functioning, increased risk of depression, impaired concentration, and sub-par job performance. Higher rates of insomnia are reported in women, the elderly, the unemployed, and those with comorbid medical conditions. While treatment has advanced significantly, the etiology of chronic insomnia is not well understood. Research has suggested that patients with insomnia have higher levels of arousal, as evidenced by higher basal metabolic rates. There is also evidence of a genetic component. Effective treatment, both pharmacologic and psychosocial, is available, although medication is associated with unwanted side effects. Commonly used medications include low-dose sedating antidepressants, benzodiazepines, and nonbenzodiazepines. The shorter half-life nonbenzodiazepines have been shown to be highly effective, and possess significantly preferable side-effect profiles over the benzodiazepines. Sedating antidepressants have proven effective as well, but due to adverse events are best used in patients with comorbid sleep and mood disorders. New research into melatonin agonists shows promise; these agents have an even more desirable side-effect profile than the nonbenzodiazepines. Nonpharmacologic treatments for sleep disorders include behavioral treatment, such as stimulus control, sleep restriction, and relaxation; cognitive treatment, largely focused on eliminating unrealistic sleep expectations and bolstering coping skills; and sleep hygiene, which deals with bed-times, rise-times, and use of substances. Cognitive-behavioral therapy (CBT) and relaxation therapies, have been shown to be quite effective for treating particular subsets of patients with insomnia. There is also evidence that combination CBT and medication therapy could be very effective. In the realm of psychosocial treatment, short-term treatment is needed for use in the primary care setting. More substantive research is required on which to base a treatment consensus.


To both sleep specialists and clinicians less familiar with sleep disorders, the prevalence of these disorders is astounding. Approximately 40–70 million Americans suffer from some form of sleep disturbance.1 Meta-analyses published in collaboration with the National Institute of Health (NIH) State-of-the-Science Conference on Manifestations and Management of Chronic Insomnia in Adults (CMMCIA) indicates that chronic insomnia affects 8.5% to 24.3% of the general population, 27.8% to 43% of the clinical population, and 19.8% to 53.7% of the general outpatient population (Slide 1).2 Perhaps most jolting of all, according to a National Sleep Foundation poll only 6% of patients with insomnia are diagnosed.3

The disparity between the prevalence of insomnia and treatment is a particular concern in light of the severity of impairment produced by insomnia and the wealth of proven, available management strategies. It is important to remember that chronic insomnia is a disorder of hyperarousal, not a symptom. Its disorder status was confirmed by the NIH consensus statement that accompanied CMMCIA.4 Chronic insomnia is typically comorbid with medical psychiatric disorders, and this high prevalence of comorbidity explains why clinicians have typically thought of insomnia as a symptom rather than a disorder, as well as why chronic insomnia is so prevalent in clinical populations.


Chronic insomnia is a simple, symptom-based diagnosis that can be made in the clinical setting. However, the fact that there is no consensus on the criteria for classifying chronic insomnia, has been particularly problematic in the research setting. Basic clinical criteria are: difficulty falling asleep, difficulty staying asleep, or sleep that is nonrestorative three or more times per week, lasting 1 month or more. Importantly, the difficulty with sleep must be associated with daytime impairment or daytime distress (Slide 2). The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition5 (DSM-IV) nosology does not include specific criteria for “chronic” insomnia, simply according an “insomnia” diagnosis to primary sleep disturbance lasting over a month. According to the International Classification of Sleep Disorders Manual,6 symptoms lasting >6 months are required for a diagnosis of chronic insomnia, and symptoms lasting 1–6 months ought be classified as “sub-acute”7; however, there is no clear consensus, and criteria vary across studies. Currently, self-report questionnaires are the preferred diagnostic tool. While there is no known biomarker for insomnia, researchers are examining a variety of assays of hyperarousal that may be implicated.8

Persistence and negative consequences are key in an insomnia diagnosis. Most people have difficulty sleeping on occasion, but the majority of these problems resolve on their own and seldom have any significant effect on daily functioning; chronic insomnia is prolonged and is associated with a variety of daytime consequences. Among the undesirable consequences of chronic insomnia are: increased risk of depression, impaired concentration, and sub-par job performance.8-12 A meta-study clearly linked insomnia to reduced quality of life and quality of social relationships.13 

The onus of identifying chronic insomnia is on the clinician, as patients typically do not present with insomnia as a primary complaint. There are well-defined risk factors for insomnia with which clinicians should be familiar (Slide 3). Females, the elderly, the unemployed, those with less education, those with another medical condition, and divorcees all have higher rates of insomnia.3,14,15 Insomnia is also highly comorbid with depression and anxiety. It must be emphasized that insomnia is not secondary to depression, nor depression to insomnia. Each is a risk factor for the other. Treating insomnia in patients who also have depression will result in a better response to selective serotonin reuptake inhibitors (SSRIs), and treating the depression is likely to improve the insomnia; however, insomnia is the most refractory symptom following successful treatment of depression.


The use of stimulants, working night or rotating shifts, and irregular sleep schedule can also precipitate insomnia.


While treatment has advanced significantly, the etiology of chronic insomnia is not well understood. However, some strides have been made toward understanding the underpinnings of this disorder. Specifically, research has suggested that patients with insomnia have higher levels of arousal—as evidenced by higher basal metabolic rates—as well as other assays of arousal than people without insomnia.16 There is also evidence of a genetic component.17 A 2004 study by Nofzinger and colleagues,18 using positron emission topography (PET) to assess arousal in patients with insomnia found increased cerebral glucose in metabolism during both sleeping and waking, as well as reduced decline in waking-to-sleeping metabolism. They also found reduced relative metabolism in the prefrontal cortex in awake insomnia patients. The findings suggest insomnia is produced by interactions between the hypothalamus, ascending reticular formation, the parts of the brain that regulate emotion, and the prefrontal cortex. Specific neurotransmitters involved in arousal in general, much less in insomnia, have not yet been identified. 

Chronic insomnia appears to be a disorder of hyperarousal. Clinicians have made a great deal of progress in recent years identifying risk factors, and are now working to better understand the complete etiology of insomnia. Better understanding leads to better treatment, and improving the identification and treatment of insomnia is essential. Not only can insomnia have severe negative impact on quality of life and functioning, but the evidence indicates that resolving insomnia enhances treatment of common comorbidities, such as depression and chronic pain conditions.

Pharmacologic Management of Insomnia 

Pharmacologic management of sleep disorders falls into two distinct categories: self-medication and prescription medication.

Self Medication

Self-medication comprises largely alcohol and over-the-counter sleep agents. Diphenhydramine, which is an antihistamine, is the most common substance taken for sleep disturbance. Diphenhydramine is the active ingredient in virtually all over-the-counter sleep agents and is somewhat effective. However, the substance has too long a half-life and too many anticholinergic side effects to be recommended for long-term treatment. Many people with sleep disturbances self-medicate with alcohol, but not only is alcohol’s short-term efficacy questionable, it can actually worsen insomnia over time. In short, self-medication appears very unlikely to prove effective for treating actual cases of insomnia.

Prescription Medications 

Prescription medications include low-dose sedating antidepressants, benzodiazepine receptor agonists, and nonbenzodiazepine receptor agonists (or “nonbenzodiazepines”), as well as a number of less frequently prescribed agents. The prevalence of use and the efficacy of these various treatments were the subjects of a large meta-analysis (116 studies) conducted by the 2005 NIH consensus group on insomnia.19 The studies covered benzodiazepines (51), nonbenzodiazepines (36), antidepressants (7), complementary and alternative care (L-tryptophan, melatonin, and valerian) (14), cognitive-behavioral therapy (18), alcohol (1), barbiturates (2), hormones (1), and low energy emission therapy (LEET) (1). Inclusion criteria for the meta-analysis is indicated in Slide 4.19


The most common current prescription sleep agents are all effective at inducing sleep. The differences lie in side effects and duration of action. Benzodiazepines, nonbenzodiazepines, antidepressants, and melatonin agonists have all been shown to significantly decrease sleep-onset latency (SOL). Of note, only benzodiazepines, nonbenzodiazepines, and ramelteon (a selective melatonin agonist) are Food and Drug Administration approved for the treatment of insomnia. Depending on the compound’s half life, benzodiazepines, nonbenzodiazepines, and antidepressants have also been shown to significantly decrease wakefulness after sleep onset (WASO), while melatonin has not. 

The two factors primarily distinguishing benzodiazepines are receptor-selectivity and duration of action. The nonbenzodiazepines are more selective and shorter acting than the benzodiazepines. As both substances are effective, treatment choice depends on whether the patient has trouble falling asleep or staying asleep. If the patient has trouble falling asleep, one of the nonbenzodiazepines is recommended—they have much shorter half-lives, and are less likely to negatively impact functioning the following day. On the other hand, if the patient has trouble staying asleep, certain benzodiazepines—or possibly one of the longer-lasting nonbenzodiazepines—might be best. It is very important to watch out for residual effects. The nonbenzodiazepines have half lives of 1–6 hours, while the benzodiazepines have a much wider range of activity; some last <6 hours, while others remain active for up to 50 hours (Slide 5).20-22

Benzodiazepines, nonbenzodiazepines, and sedating antidepressants—as opposed to melatonin—all demonstrate greater risk of harm than placebo, with benzodiazepines significantly more likely to produce adverse events than the nonbenzodiazepines.23-25 

The side effects of the hypnotics are dependent on dose and half-life. Amnesia, ataxia, and the parasomnias, such as somnambulism, are tied to dose—the larger the dose, the greater the risk of an adverse event. Sleep hang-over is a residual effect tied to half-life of the drug. All of the hypnotics, except ramelteon, appear to carry some risk of habit-formation in at-risk populations.26 

Both benzodiazepines and nonbenzodiazepines produce their effects by depressing the central nervous system via modulation of the g-aminobutyric acid (GABA)A receptor. The mode of action of sedating antidepressants seems to be anticholinergic and antihistaminic activity. Melatonin agonists—which may be effective in some insomnia patients, particularly those with sleep-onset problems—impact circadian regulation of sleep. As such, these agents may be particularly effective in treating circadian sleep disorders (eg, shift-work syndrome). However, while the circadian element may be most noticeable in shift work and phase delays, many individuals with insomnia appear to have an involvement of the circadian system as well.

Both L-tryptophan and valerian (neither of which is currently FDA approved for the treatment of insomnia) require further study, not only to establish possible efficacy, but to establish side-effect profile as well. L-tryptophan is a chemical precursor to melatonin, which may explain its possible sleep-inducing effects. The chemistry of valerian is complicated—as is the case for most herbal supplements—but it may target some of the same receptors as the hypnotics.

Efficacy and Side Effects Overview of Drugs Commonly Used in Insomnia19,25,27,28


Benzodiazepines have proven efficacy, but many of them have long half lives (up to 40 hours). Adverse events can include amnesic effects, drowsiness, ataxia, vertigo, impaired judgment, and confusion. The side effects of benzodiazepines are heightened by alcohol.


Nonbenzodiazepines have been shown to be similarly effective to benzodiazepines, but they are less likely to induce intolerance, and their short half lives result in fewer side effects. Adverse events are similar to those found with benzodiazepines, but generally weaker. Most notably, they do not produce next-day side effects.

Low-Dose Sedating Antidepressants

Sedating antidepressants have demonstrated efficacy in treating insomnia in the context of depression, but their use for primary insomnia—although promising—is not yet established. The most common adverse events include rapid heart rate, blurred vision, weight gain or loss, low blood pressure, constipation, nausea, mild tiredness, headache, and dry mouth. The side-effect profile of the sedating antidepressants is a significant impediment to their use as frontline treatment for insomnia.

Melatonin Agonists

Melatonin agonists show promise for the treatment of insomnia, in particular shift-work and other primarily circadian disorders. They have been demonstrated to positively impact sleep onset latency in primary insomnia, but have not been shown to affect wakefulness after sleep onset. Melatonin agonists have a relatively mild side-effect profile. Adverse events include drowsiness, fatigue, and dizziness.

Future Research

To properly assess the efficacy of low-dose sedating antidepressant treatment for insomnia, large-scale, randomized trials are needed. In addition, more research into the impact of these various treatments on quality of life is required. There are some promising alternative treatments—most notably melatonin agonists—that need further study as well.

Psychosocial Treatment of Insomnia

A Model of Insomnia

To best understand the place of cognitive and behavioral treatments for insomnia, it helps to understand the various factors driving insomnia. First, clinicians have found that there are predisposition factors that may at least in part be genetic. That is, some people are more likely to develop insomnia than others. However, many of those who are predisposed never present with the symptoms of insomnia. Many clinicians think that a predisposition in combination with certain stressors is what actually produces insomnia (Slide 6). There are also another set of factors that, once insomnia has developed, perpetuate the disorder. The cognitive and behavioral treatments seek to eliminate or minimize these perpetuating factors.


There are a wide variety of relevant stressors, ranging from life-changes to comorbid medical, sleep, and psychiatric disorders. There are numerous perpetuating factors as well, and they tend to center around the behavior and cognitions surrounding sleep. 

Nonpharmacologic treatments for sleep disorders fall into three broad categories: behavioral treatment, including stimulus control, sleep restriction, and relaxation; cognitive treatment, largely focused on eliminating unrealistic sleep expectations and bolstering coping skills; and sleep hygiene, which deals with bed times, rise times, and substances.

Sleep Hygiene

Some patients’ sleep difficulties can be improved, or even completely remedied by improving their sleep hygiene. When sleep disturbance can be dealt with through adherence to good sleep hygiene, pharmacologic and psychosocial treatments are generally unnecessary. Sleep hygiene refers to the habits and environmental factors influencing an individual’s sleep behavior (Slide 7).

Cognitive-Behavioral Treatment 

Both behavioral and cognitive-behavioral therapy (CBT) have shown to be effective for the treatment of insomnia. Behavioral treatments include sleep restriction, stimulus control, and relaxation training. Clinicians began developing behavioral treatments for insomnia in the 1970s, and, later, cognitive methods were added. The cognitive treatments focus on cognitive restructuring, which involves eradicating or controlling incorrect and anxiety-producing beliefs about sleep loss. Ironically, worry over sleep loss is often a major contributing factor in continued difficulty sleeping.

The combination of cognitive and behavioral therapies has proved equally effective to prescription medication for short-term management of insomnia, and there is further evidence suggesting that the positive effects of CBT, as opposed to those of medication, extend significantly past the period of active treatment.28-30 In addition, CBT does not have a notable side-effect profile.

As promising as these treatments appear, proper administration of CBT requires some clinician training, which can be a limiting factor on its use. Further research is needed, and clinicians need to determine whether having CBT available in the primary care environment is feasible. Currently, studies are in progress evaluating CBT in the primary care setting utilizing a truncated treatment period.     

Combined Treatment 

Combination treatment for insomnia—including combinations of CBT and prescription medication and combinations of CBT and other treatments—have also shown some promise in treating insomnia. However, there have been a paucity of studies examining the benefits of these combination treatments. Previous meta-analyses29-31 have shown more clear benefit for combination treatment than came out of the 2005 NIH Consenus study (Slide 8),19 but these previous meta-analyses had significantly looser criteria, which might have influenced the results.  

Future Research into Psychosocial Treatment

Nonpharmocolgic treatments, including CBT and relaxation therapies, have been shown to be quite effective for treating particular subsets of patients with insomnia. There is also evidence that combination therapy could be very effective. What is needed now are larger, controlled studies comparing the efficacy of cognitive-behavioral treatments with the efficacy of prescription medications, both in the short and long term (Slide 9).4          



Sleep disorders are highly prevalent and have a profound impact on quality of life. They are underdiagnosed, and often mistakenly treated as adjuncts to comorbid disorders such as depression. Effective treatment, both pharmacologic and psychosocial, is available, but the former is often accompanied by unwanted side effects, while far too few clinicians are trained in the latter to meet patient needs. Nonbenzodiazepine treatments have been shown to be effective, and they typically possess preferable side effects over benzodiazepines. Sedating antidepressants have proven effective as well, but might be best reserved to treat patients with comorbid sleep and mood disorders. New research into melatonin agonists shows promise; these agents have a generally more benign side-effect profile than either benzodiazepines or nonbenzodiazepine hypnotics. In the realm of psychosocial treatment, more substantive research is required on which to base treatment consensus. There is also a push to develop short-term treatment for use in the primary care setting. The clinician treating insomnia is in an enviable situation—evidence-based treatment is available. The next step is to determine which treatment is most effective for which patients.


1. 2003 National Sleep Disorders Research Plan. Available at: Accessed April 13, 2007.

2. Manifestations and Management of Chronic Insomnia in Adults. Available at: Accessed April 13, 2007.

3.   National Sleep Foundation. Sleep in America Poll. Data from 1997 and 2001, 2002 sleep poll. Accessed April 13, 2007.

4.   NIH State-of-the-Science Conference Statement on Manifestations and Management of Chronic Insomnia in Adults. Available at: Accessed April 13, 2007.

5. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.

6.     The international Statistical Classification of Diseases and Related Health Problems. 10th revision. Geneva, Switzerland: World Health Organization; 1992.

7. American Academy of Sleep Medicine. The International Classification of Sleep Disorders, Revised. Diagnostic and Coding Manual. Rochester, MN: American Academy of Sleep Medicine; 2001.

8.   Kryger MH, Roth T, Dement WC. Principles and Practice of Sleep Medicine. 3rd ed. Philadelphia, PA: W.B. Saunders Co; 2000.

9. Simon GE, VonKorff M. Prevalence, burden, and treatment of insomnia in primary care. Am J Psychiatry. 1997;154(10):1417-1423. 

10. Leger D, Guilleminault C, Bader G, et al. Medical and socio-professional impact of insomnia. Sleep. 2002;25(6):625-629.

11. Janson C, Lindberg E, Gislason T, et al. Insomnia in men – A 10-year prospective population based study. Sleep. 2001;24(4):425-430. 

12. Riemann D, Voderholzer U. Primary insomnia: a risk factor to develop depression? J Affect Disord. 2003;76:255-259. 

13. Shochat T, Umphress J, Israel AG, et al. Insomnia in primary care patients. Sleep. 1999;22(suppl 2):S359-S365. 

14. Zorick FJ, Walsh JK. Evaluation and management of insomnia. In: Kryger MH, Roth T, Dement WC. Principles and Practice of Sleep Medicine. 3rd Ed. Philadelphia, PA: W.B. Saunders Co; 2000. 

15. Sateia MJ, Doghramji K, Hauri P, et al. Evaluation of chronic insomnia. An American Academy of Sleep Medicine review. Sleep. 2000; 23(2):243-308. 

16. Bonnett MH, Arand D. 24 hour metabolic rate in insomniacs and matched normal sleepers. Sleep. 1995;18(7):581-588. 

17. Yves D, Morin C, Cervena K, et al. Family studies in insomnia, Sleep. 2003;26(suppl):A:304.

18. Nofzinger E, Buysse D, Germain A, et al. Functional neuroimaging evidence for hyperarousal in insomnia. Am J Psychiatry. 2004;161:2126-2128.

19. Buscemi N, Vandermeer B, Friesen C, et al. Manifestations and Management of Chronic Insomnia in Adults. Evidence Report/Technology Assessment No. 125. AHRQ Publication No. 05-E021-2. Rockville, MD: Agency for Healthcare Research and Quality; June 2005. Available at: Accessed April 5, 2007. 

20. Fuller MA, Sajatovic A. Psychotropic Drug Information Handbook. 5th ed. Hudson, OH: Lexi-Comp, Inc.; 2004. 

21. Olin BR, Hebel SK, Dombek CE, eds. Facts and Comparisons. St Louis, Mo: Facts and Comparisons; 1997. 

22. American Hospital Formulary Service. AHFS Drug Information. Bethesda, MD: American Society of Health-System Pharmacists; 2004. 

23. Panneman MJ, Goettsch WG, Kramarz P, et al. The costs of benzodiazepine-associated hospital-treated fall injuries in the EU: a pharmo study. Drugs Aging. 2003;20:833-839. 

24. Drover DR. Comparative pharmacokinetics and pharmacodynamics of short-acting hypnosedatives: zaleplon, zolpidem and zopiclone. Clin Pharmacokinet. 2004;43:227-238. 

25. Roehrs T, Roth T. Hypnotics: an update. Curr Neurol Neurosci Rep. 2003;3:181-184.

26. Physician’s Desk Reference. 59th ed. Montvale, NJ: Thomson PDR; 2005.

27. Curry DT, Eisenstein RD, Walsh JK. Pharmacologic management of insomnia: past, present, and future. Psychiatr Clin North Am. 2006;29:871-893.

28. Borja NL, Daniel KL. Ramelteon for the treatment of insomnia. Clin Ther. 2006;28:1540-1555.

29. Morin CM, Culbert JP, Schwartz SM. Nonpharmacological interventions for insomnia: A meta-analysis of treatment efficacy. Am J Psychiatry. 1994;151:1172-1180.

30. Murtagh DRR, Greenwood KM. Identifying effective psychological treatments for insomnia: a meta-analysis. J Consult Clin Psychol. 1995;63:79-89. 

31. Smith MT, Perlis ML, Park A, et al. Comparative meta-analysis of pharmacotherapy and behavior therapy for persistent insomnia. Am J Psychiatry. 2002;159:5-11.