Insomnia is defined as difficulty in initiating or maintaining sleep or having sleep that is non-restorative. This definition gives rise to other terms such as initial insomnia, where there is increased sleep latency (the interval time between when patients lie down to sleep and the time they actually fall asleep) and middle insomnia, where there is difficulty getting back to sleep after waking up. Other individuals experience early morning wakening in the later stages of sleep, with awakening accompanied by a feeling of being unrefreshed by sleep. The main categories of insomnia are primary insomnia and comorbid insomnia:
- Primary insomnia is defined in DSM by insomnia symptoms lasting at least one month, causing fatigue or impairment, not caused by a mental or physical disorder, a primary sleep disorder, or the effects of a medication
- Comorbid insomnia is defined as insomia appearing in the presence of at least one other disorder such as major depression, generalised anxiety, substance abuse, attention deficit and hyperactivity disorder in children, dementia and a variety of physical problems.
Insomnia is prevalent in the general adult population, and is more common in medical and psychiatric patients. Studies of general population samples have found a prevalence of between 10-48%. Age, gender and social class has been shown to be influential. Most studies investigating insomnia have shown increased prevalence associated with increased age, with a prevalence rate that is stable between 15-44 years of age but increasing after 45 years of age and continuing to increase up to 65 years of age. Despite the widely held belief that ‘people don’t sleep well as they age’ and evidence of changes in sleep physiology associated with aging such as reductions of slow wave sleep and tendencies to greater arousability, it has now been shown that comorbid medical problems play the primary role in the increased rates of insomnia in this population.
Across all age groups, it has also been noted that women are more likely to report problems with disturbed sleep than men. Menopausal and perimenopausal symptoms may play a clear role in the development of insomnia in women.
Social class and marital status have also been shown to have an impact on sleep, with most studies showing an increased risk for insomnia in people who are divorced, separated and widowed, as well as other factors such as limited years of education, low income and unemployment.
Stress has been demonstrated to play a major role in the development and perpetuation of insomnia. Epidemiological studies have shown that increased stress levels in the workplace are associated with increased rates of sleep complaints, and patients with chronic insomnia have consistently reported their insomnia as being precipitated by increased stress levels. Lastly, disturbances of circadian rhythm has been linked to insomnia, with higher rates of complaints among shift workers.
History, clinical examination and investigation
Evaluation of insomnia should include a detailed history, as well as clinical examination and relevant investigations.
Clinicians need to first inquire about the duration of the problem with sleep, to differentiate whether is it a transient insomnia (limited to a few nights), short-term insomnia (less than three weeks) or long-term insomnia (more than three weeks). Generally, the former two have different sets of aetiological factors compared to long-term insomnia.
Both transient and short-term insomnia can occur following periods of emotional distress or bereavement, initiation or discontinuation of medication, use of illegal substances or alcohol, recent onset of a physical or painful illness, work-shift changes and jetlag. Causes for long-term insomnia may be similar but are more likely to be associated with significant medical diseases, major psychiatric disorders or substance abuse (see below).
The clinician should try to identify the pattern of the sleeping difficulty, whether is it initial insomnia, middle insomnia, broken pattern of sleep, non-restorative sleep or early morning wakening. This may have an impact on the choice of short term medication to be recommended.
It is also important to enquire about recent stressors and lifestyle changes, as patients may not be aware that recent events such as jetlag, changes in work shift or a personal crisis can affect their sleep.
Clinicians should then proceed to identify clinical entities that can be associated with insomnia, including medical diseases such as hypertension, hyperthyroidism, respiratory insufficiency, cardiovascular insufficiency and chronic pain, and psychiatric disorders such as depression, bipolar affective disorder, psychotic illness and anxiety disorders.
Prescription medication should also be addressed, since medications that include antihypertensives, diuretics, steroids, bronchodilators, decongestants, histamine antagonists, antidepressants and antipsychotics can disturb sleep.
A crucial aspect of the history would be the identification of substance abuse and alcohol-related difficulties. Studies have shown that 10-15% of patients with insomnia have substance abuse problems. On the other hand, alcohol has been known to initially cause a sedative effect, followed by rebounds and awakening or disrupted, non-restful sleep. Furthermore, daily intake of caffeine should also be assessed, particularly any intake after 3pm.
Other useful screening questions could focus on eliciting symptoms of a primary sleep disorder, which include restless leg syndrome, sleep apnoea, narcolepsy and insufficient sleep syndrome. Consequently, collateral history from the patient’s sleeping partner is invaluable, such as witnessed symptoms of apnoeic episodes or snoring consistent with sleep apnoea.
It is important to clarify the impact of insomnia on the individual, be it impaired cognitive functioning, decreased occupational performance and association with psychiatric disorders including new depression.
Management of insomnia
The initial management of insomnia is focused on treating any underlying medical conditions which may be contributing to the condition, as well as reversing any reversible causes. If these interventions have been performed but the patient continues to suffer from insomnia then further management is required. Management can be divided into non-pharmacological and pharmacological options.
Non-pharmacological options focus mainly on behavioural treatments1 and options are both many and varied. The main options employed include:
This involves limitation of the amount of time a patient spends in bed to the actual amount of time spent asleep, creating mild sleep deprivation: sleep time is then increased as a patients sleeping improves.
Stimulus control therapy
This method is based on the idea that some people with insomnia have learned to associate the bedroom with staying awake rather than sleeping. Therefore, these patients are given strict instructions to change this learned behaviour, eg. spending no more than 20 minutes lying in bed trying to fall asleep. If the person cannot fall asleep within 20 minutes, the person is advised to get up, go to another room and read or find another relaxing activity until they feel sleepy again. Activities such as eating, doing housework, watching TV, or studying for a test, which ‘reward’ you for staying awake, should be avoided. When the patient starts to feel sleepy, they can return to bed but if they cannot fall asleep in another 20 minutes, the process is repeated. It is crucially important to set an alarm clock and get up at the same time every day, including weekends. Do not take a nap during the day.
Cognitive behaviour therapy
This method aims to address a patient’s maladaptive behaviours, eg. frequent daytime napping, clock-watching and going to bed too early. Treatment targets also include unrealistic expectations (‘I must get eight hours of sleep or I will be unable to function’), faulty causal attributions (‘my insomnia is something I was born with and is out of my control’), and pathologic anxiety about consequences of insomnia (‘my insomnia is causing my poor health’). Strategies include developing realistic expectations, minimising the importance of sleep and teaching a patient not to catastrophise during, or after, a poor night’s sleep.
This method focuses on correcting behaviours that cause a patient to remain awake despite patients often realising the link between this behaviour and their insomnia. For example, patients are advised to avoid caffeine late in the day, to avoid clock watching, to ensure there is no excessive light or noises in the bedroom. Patients are encouraged to relax in the evening before bed.
The advice is to maintain a regular schedule with approximately the same bedtime and rising time daily. Daytime naps should be avoided. Avoid heavy meals prior to sleep. Encourage regular exercise three to four times per week (advise patient that regimes may take up to six weeks to take effect). Control bedroom temperature and surroundings. Encourage ritual wind-down in the hour or two before going to bed. Avoid alcohol after dinner. Use bed only for sleeping and sexual activity – it should not to be used for watching TV, eating or other activities.
When sleeping medications are prescribed, the least amount of medication that permits restorative sleep should be used for the shortest period of time until the patient’s normal sleep pattern re-emerges. For responsible patients at low risk of substance abuse, sleeping medications can generally be prescribed for a period of several days to several weeks. There is a minimal risk of rebound insomnia over this time period (increased difficulty sleeping after a patient who has been taking a sleep medication on a chronic basis stops the medication suddenly). When prescribing, be sure to give appropriate warnings about possible excess sedation, potential interactions and increased sensitivity to alcohol.
Over the counter preparations include antihistamines, L-tryptophan, melatonin or valerian root.
Antihistamines are believed to be the most common OTC treatment in chronic insomnia.2 Limited evidence exists for its use but there are many concerns regarding adverse effects. These include residual daytime sleepiness, diminished cognitive function and delirium, especially in the elderly. Other adverse effects include a risk of dry mouth, blurred vision, urinary retention and constipation.
Many drugs are used off-licence in the treatment of insomnia. One of the biggest groups is antidepressants. Of this group, trazadone is the antidepressant most commonly used for this purpose in the US. One particular study compared trazadone 50mg, zolpidem 10mg and a placebo during two weeks duration in well-defined primary insomnia patients and found increased total sleep duration in zolpidem and trazadone. However, zolpidem was found to be more effective than trazadone overall.3
Other antidepressants that are used include amitriptyline and doxepin. Gabapentin and clonidine may also be used off-licence. Despite being used off-label, these medications are often favoured due to their sedative and non-habituating properties.
It is worth mentioning that currently, olanzapine and quetiapine are used off-label to promote sleep in patients with or without major psychiatric problems. Due to a lack of appropriate data, dosage of these drugs can prove problematic, however. There is a well-recognised risk of hyperglycaemia with atypical anti-psychotics, as well as a risk of tardive dyskinesia and neuroleptic malignant syndrome with older agents such as chlorpromazine.
Medications specifically designed for sleep are generally divided into benzodiazepine compounds and non-benzodiazepine compounds.
As a class these have been in common use as a sedative/hypnotic for several decades. Benzodiazepine use has been a recognised problem within Ireland for a number of years. Research undertaken by the Health Research Board prompted increased awareness of the problem by identifying a large population of problem benzodiazepine users. This prompted ongoing efforts by the HSE to curb prescribing, with every practice in Ireland being sent a breakdown of their benzodiazepine subscription rates.
The high degree of benzodiazepine usage reflects a predictable anti-anxiety effect at low doses and sleep-inducing effects at moderate doses with relatively few serious side-effects.4 Abuse, dependence, discontinuation syndrome and over sedation are well-documented side-effects. However, these compounds are remarkably useful and safe when prescribed intelligently and used in moderation.
There is no known therapeutic advantage of one benzodiazepine over the other, other than the issues created by longer half-lives, which can cause possible ‘hangover effects’ or amnesia in the morning.
When choosing medication, one must aim to match the benzodiazepine or other sedative hypnotic being used to the type of difficulty the patient is having. If the difficulty is with initiating sleep and not with maintaining sleep then a short-acting hypnotic would be preferable. Such a compound would be avoided in patients with sleep continuity disturbances as the compound would be fully excreted when the patient needed it most. These patients would benefit from intermediate acting compounds. Patients with early morning wakening would also benefit from intermediate-acting compounds.
Particular caution should be exercised when prescribing for older patients as they are at increased risk of drug accumulation and drug hangover, leading to increased risk of falls or accidents.
Over time, some patients will lose the hypnotic effect of benzodiazepines. If taken for more than seven months patients can experience some rebound insomnia, lasting one to seven days, when the medication is withdrawn. With very high doses taken over extended periods of time, frank physical withdrawal and seizures may result from sudden discontinuation.
A less common side-effect associated with triazolam use is retrograde amnesia for several hours before the medication was taken. Given the fact that benzodiazepines have a street value in Ireland that is grossly above their retail price, one should always be wary of drug-seeking behaviours.
The four main types are zolpidem, zopiclone, zaleplon, and eszopiclone. All four are believed to exert their mechanism of action at the benzodiazepine receptor site in the brain. They are preferable to benzodiazepines as the habit-forming potential is less and is relatively small. Some measure of physical dependence has however, been reported with long-term use of very high doses. Possible disruption to normal sleep architecture, medication interactions and rebound insomnia are minimal compared to benzodiazepines.5,6,7,8,9,10 The primary downside to these medications is their cost, as they are more expensive than benzodiazepines. They are also associated with confusion and memory impairment when used in large doses or sensitive individuals. They have not shown the muscle relaxation and anxiolytic effects of typical benzodiazepines which lowers the potential for abuse.
Follow-up is important irrespective of what medication the patient has been prescribed. One should focus on daytime sleepiness, lack of attention and co-ordination as well as other side-effects of over dosage/drug accumulation.
A patient should never be prescribed chronic sleep medications without several trials of medication at varying intervals to assess continued need. Patients in institutional care settings or nursing homes are often prescribed sleep medication with minimal follow-up, so one should be extra vigilant in these settings. In such patients any attempted trial off medication should continue for a long enough period to allow any rebound insomnia to resolve. While every effort should be made to stop sleep medication, some patients will require long-term chronic prescription.11
Individuals with a history of substance abuse are generally poor candidates for benzodiazepine and other sedative hypnotics. Trials of OTC medication as well as gabapentin, trazadone, doxepin or trimipramine would be preferred. Patients undergoing withdrawal from alcohol can expect disrupted sleep for several days to several weeks. It is not useful to routinely prescribe sedative hypnotics to such patients except as part of a detoxification profile. If the above measures have been trialled unsuccessfully but the patient, despite a history of substance abuse, has been substance-free for a period of time, then we suggest they be carefully tried on sedative hypnotics. If a prescription is given, then encourage intermittent use and for the shortest possible period of time.
Patients with liver impairment have limited options, as most of the medications mentioned above are metabolised in the liver. Gabapentin is the main exception as it is excreted via the kidneys and so can be a worthwhile option.
Elderly patients are at increased risk for insomnia and are often at greater risk for side-effects from medications.12,13 The general principles used for prescribing include opting for shorter-acting medications and intermittent dosing. This helps to prevent drug accumulation.
We recommend giving special attention to the non-pharmacological options for elderly patients. More specifically, one should encourage increased exercise, avoidance of daytime naps and removal of the clock from their bedroom.
Referral to a sleep clinic or a specialist with expertise in sleep medicine may be required if a primary sleep disorder is suspected, there is doubt regarding the diagnosis, or long-term insomnia has not responded to management in primary care.
- Morin C, Espie C. Insomnia: a Clinical Guide to Assessment and Treatment. New York, NY: Springer: 2004.
- Yoshihawa N et al (1998) A case of insufficient sleep syndrome. Psychiatry Clin Neuro 52(2): 200-201.
- Walsh JK, Erman M, Erwin CW, et al. Subjective hypnotic efficacy of trazodone in DSM3-R primary insomnia. Human Psychopharmacol 1998; 13: 191-198
- McGee M and Pres R (2002)Benzodiazepines in primary practice; risks and benefits. Resident Staff Physician 48(4): 42-49.
- Lippman S et al.  Insomnia: therapeutic approach. South Med J 94(9): 870-872.
- Stimmed GL (1999) Future directions in drug treatment of insomnia. Psych Times (Suppl.): 1-8
- Darcourt G (1999) Safety and tolerability of zolpidem: an update. J Psychopharmacol 13(1): 81-93
- Bowes M (1999) Sedative hypnotic medications for insomnia. Psych Times (Suppl.): 9-26
- Doghranji K (1999) Treatment of insomnia in ageing patients. Sleep Disorders July: 5-6
- Richardson GS et al (2002) Management of insomnia: the role of Zaleplon. General Medicine 4(1). Available at www.medscape.com.
- Schenck C and Mahowald MW (1996) Long-term benzodiazepine treatment of injurious parasomnias and other disorders of disrupted nocturnal sleep in 170 adults. Am J Med 100: 333
- Mosier WA et al. (1998) Wanted: a good night’s sleep. Adv Nurse Pract 6: 30-35
- Nakra BRS et al. (1991) Insomnia in the elderly. Am Fam Physician 43: 477-483