A detailed clinical history is the most important diagnostic tool for the evaluation of sleep disorders. The clinical history should always elicit the type of complaint(s), chronology, and course; exacerbating and alleviating factors; and responses to previous treatments. In particular, it is important to inquire about life events, psychiatric disorders, and medical illnesses at the time of onset of the sleep problem to help establish its etiology.

Determining whether the course of the disorder has been chronic or intermittent can be crucial in the differential diagnosis of some disorders. For instance, patients with narcolepsy have persistent daytime sleepiness, whereas patients with hypersomnia related to depression have more intermittent symptoms. Insomnia complaints are classically divided into transient, short-term, and chronic types, which vary in terms of etiology and treatment (National Institute of Mental Health 1984).

Factors that exacerbate or alleviate sleep disorders may be quite different from factors for typical psychiatric problems. For instance, the patient’s response to sleep deprivation, alcohol, or even sleeping in a different environment can be quite useful diagnostically. It is also crucial to evaluate the patient’s sleep-related behaviors, including activities leading up to bedtime, and the patient’s cognitive process in bed; both behavioral and cognitive elements can strongly contribute to sleep problems. Additional history from the bed partner is often of crucial importance. For instance, patients may be unaware of their own snoring or of the presence of breathing pauses during the night, and they may underestimate or deny their degree of daytime sleepiness.

Prospective data are very often useful not only for the diagnosis but also for the treatment of sleep disorders. A sleep-wake diary or log provides information on a patient’s sleep habits, which may vary considerably from his or her global statements during an interview. In particular, sleep diaries can be very helpful for establishing a diagnosis of circadian rhythm sleep disorder and for elucidating insomnia complaints. Wrist actigraphy also has been advocated as a prospective method of data collection that may be sensitive not only to baseline sleep patterns but also to behavioral treatment interventions such as sleep restriction (Brooks et al. 1993; Hauri and Wisbey 1992). However, actigraphy does not by itself establish a particular diagnosis (Sadeh et al. 1995; Standards of Practice Committee of the American Sleep Disorders Association 1995).

For some patients with sleep disorders, overnight sleep studies or polysomnography can aid in the diagnosis and treatment planning. Nocturnal sleep studies typically include monitoring of electroencephalogram, eye movements, and muscle tension in order to establish sleep stages and wakefulness. In addition, patients may be monitored for airflow and respiratory effort, limb movements, oxyhemoglobin saturation, body position, or seizure activity.

The question of which patients should have sleep studies continues to be debated. In general, patients who present with excessive daytime sleepiness should have an overnight polysomnogram and a multiple sleep latency test, which provide an objective estimate of the degree of daytime sleepiness. The diagnoses of narcolepsy and breathing-related sleep disorder can be firmly established only with such studies, and proper treatment depends on the results of these studies. The role of polysomnography in patients with insomnia disorders is considerably less certain. Some investigators find very little evidence for specific sleep pathologies in patients with insomnia (Vgontzas et al. 1995), whereas others suggest that diagnosis is often substantiated or altered by the results of polysomnography (Edinger et al. 1989a; E. A. Jacobs et al. 1988).

Most clinicians would agree that younger patients with insomnia generally can receive an adequate diagnosis through clinical history alone. If trials of usual treatments do not result in improvement, sleep studies may be indicated. Sleep studies may also have a higher yield of diagnostically useful information in elderly patients with insomnia (Edinger et al. 1989a). Patients who present primarily with sleep disruption in the setting of circadian rhythm disturbance generally do not require polysomnography for confirmation of their diagnosis.

Finally, polysomnography is not indicated for patients with more or less typical parasomnias, such as sleepwalking or nightmares. If there is any question of seizures, rapid eye movement (REM) behavior disorder, or apnea as a cause of unusual behavior at night, however, overnight polysomnography for 1 or 2 nights should be performed. The American Academy of Sleep Medicine (formerly the American Sleep Disorders Association) recommends polysomnography for patients who present with hypersomnia or a suspicion of apnea or narcolepsy but not for most patients with insomnia (Polysomnography Task Force, American Sleep Disorders Association Standards of Practice Committee 1997; Reite et al. 1995; Standards of Practice Committee of the American Sleep Disorders Association 1995).