A physician specializing in sleep medicine may ask patients about their medical history; for example: neurological problems, prescription or non-prescription medications taken, alcohol use, family history, and any other sleep problems. A thorough medical and neurological exam is indicated. The patient will be asked to complete a sleep diary, recording natural sleep and wake up times, over several weeks. Sleep rating with the Epworth Sleepiness Scale may be used.

A neurological condition or another medical problem may be suspected, in which case, blood tests, a CT scan or an MRI may be used. An overnight sleep study is usually not needed to detect this disorder, but may be indicated if other sleep disorders, such as sleep apnea and periodic limb movement disorder, seem likely. The overnight sleep study is called polysomnography. It charts brain waves, heart beat, muscle activity, and breathing during sleep. It also records arm and leg movement. It will show if there are other sleep disorders that are causing or increasing the problems with ISWD.

The disorder can be considered very likely in a totally blind person with periodic insomnia and daytime sleepiness, although other causes for these common symptoms need to be ruled out. In the research setting, the diagnosis can be confirmed, and the length of the free-running circadian cycle can be ascertained, by periodic assessment of circadian marker rhythms, such as the core body temperature rhythm, the timing of melatonin secretion, or by analyzing the pattern of the sleep–wake schedule using actigraphy. Most recent research has used serial measurements of melatonin metabolites in urine or melatonin concentrations in saliva. These assays are not currently available for routine clinical use.

Since 1979, the disorder has been recognized by the American Academy of Sleep Medicine:

- "Diagnostic Classification of Sleep and Arousal Disorders" (DCSAD), 1979: Non-24-Hour Sleep–Wake Syndrome; code C.2.d

- "The International Classification of Sleep Disorders", 1st & Revised eds. (ICSD), 1990, 1997: Non-24-Hour Sleep–Wake Syndrome (or Non-24-Hour Sleep–Wake Disorder); code 780.55-2

- "The International Classification of Sleep Disorders", 2nd ed. (ICSD-2), 2005: Non-24-Hour Sleep–Wake Syndrome (alternatively, Non-24-Hour Sleep–Wake Disorder); code 780.55-2

Since 2005, the disorder has been recognized by name in the U.S. National Center for Health Statistics and the U.S. Centers for Medicare and Medicaid Services in their adaptation and extension of the WHO's "International Statistical Classification of Diseases and Related Health Problems" (ICD):

- ICD-9-CM: Circadian rhythm sleep disorder, free-running type; code 327.34 became effective in October 2005. Prior to the introduction of this code, the nonspecific code 307.45, Circadian rhythm sleep disorder of nonorganic origin, was available, and as of 2014 remains the code recommended by the DSM-5.

- ICD-10-CM: Circadian rhythm sleep disorder, free running type; code G47.24 is due to take effect October 1, 2014.

Since 2013, the disorder has been recognized by the American Psychiatric Association:

- DSM-5, 2013: Circadian rhythm sleep–wake disorders, Non-24-hour sleep–wake type; ICD-9-CM code 307.45 is recommended (no acknowledgment of 327.34 is made), and ICD-10-CM code G47.24 is recommended when it goes into effect.

An adult who is compelled to nap repeatedly during the day may have excessive daytime sleepiness; however, it is important to distinguish between occasional daytime sleepiness and excessive daytime sleepiness, which is chronic.

A number of tools for screening for EDS have been developed. One is the Epworth Sleepiness Scale which grades the results of a questionnaire. The ESS generates a numerical score from zero (0) to 24 where a score of ten [10] or higher may indicate that the person should consult a specialist in sleep medicine for further evaluation.

Another tool is the Multiple Sleep Latency Test (MSLT), which has been used since the 1970s. It is used to measure the time it takes from the start of a daytime nap period to the first signs of sleep, called sleep latency. The test is based on the idea that the sleepier people are, the faster they will fall asleep.

The Maintenance of Wakefulness Test (MWT) is also used to quantitatively assess daytime sleepiness. This test is performed in a sleep diagnostic center. The test is similar to the MSLT. However, during this test the patient is instructed to try to stay awake.

According to a May 2014 article published in NewScientist, spectral analysis may help clinicians find objective evidence for sleep state misperception:

Because a number of parasomnias may be confused with RBD, it is necessary to conduct formal sleep studies such as polysomnography (PSG) performed at sleep centers that are experienced in evaluating parasomnias in order to establish a diagnosis. In RBD, a single night of extensive monitoring of sleep, brain, and muscle activity will almost always reveal the lack of muscle paralysis during REM sleep, and it will also eliminate other causes of parasomnias.

Recently, due to the limited access to PSG, attempts have been made to identify RBD from clinical interview as well as questionnaires. Postuma et al. have validated a single-question screening tool for RBD (RBD1Q) that could be easily applied in general practice to the patient and their bed partner. A positive answer to the RBDQ1, ‘Have you ever been told or suspected yourself, that you seem to act out your dreams while asleep (for example, punching, flailing your arms in the air, making running movement etc.)?’ should encourage the medical practitioner to consider the diagnosis of RBD as it offers good sensitivity (94%) and specificity (87%). Other questionnaires, such as the Rapid Eye Movement (REM) sleep Behavior Disorder Screening Questionnaire (RBDSQ) or the REM Sleep Behavior Questionnaires – Hong-Kong are available for more detailed characterisation.

EDS can be a symptom of a number of factors and disorders. Specialists in sleep medicine are trained to diagnose them. Some are:

- Insufficient quality or quantity of night time sleep.

- Misalignments of the body's circadian pacemaker with the environment (e.g. jet lag, shift work or other circadian rhythm sleep disorders).

- Another underlying sleep disorder, such as narcolepsy, sleep apnea, idiopathic hypersomnia or restless legs syndrome.

- Disorders such as clinical depression or atypical depression.

- Tumors, head trauma, anemia, kidney failure, hypothyroidism or an injury to the central nervous system.

- Drug abuse.

- Genetic predisposition

- Vitamin deficiency, such as Biotin deficiency

- Particular classes of prescription and OTC medication

DSPD is diagnosed by a clinical interview, actigraphic monitoring, and/or a sleep diary kept by the patient for at least two weeks. When polysomnography is also used, it is primarily for the purpose of ruling out other disorders such as narcolepsy or sleep apnea. If a person can adjust to a normal daytime schedule on her/his own, with just the help of alarm clocks and will-power, the diagnosis is not given.

DSPD is frequently misdiagnosed or dismissed. It has been named as one of the sleep disorders most commonly misdiagnosed as a primary psychiatric disorder. DSPD is often confused with: psychophysiological insomnia; depression; psychiatric disorders such as schizophrenia, ADHD or ADD; other sleep disorders; or school refusal. Practitioners of sleep medicine point out the dismally low rate of accurate diagnosis of the disorder, and have often asked for better physician education on sleep disorders.

The 2001 International Classification of Sleep Disorders (ICSD) divides primary hypersomnia syndromes between narcolepsy, idiopathic hypersomnia, and the recurrent hypersomnias (like Klein-Levin syndrome); it further divides narcolepsy into that with cataplexy and that without cataplexy. This ICSD version defines narcolepsy as a disorder of unknown cause "that is characterized by excessive sleepiness that typically is associated with cataplexy and other REM-sleep phenomena, such as sleep paralysis and hypnagogic hallucinations". It also establishes baseline categorical standards for diagnosis of narcolepsy, through 2 sets of well defined criteria, as follows.

Minimal narcolepsy diagnostic criteria set #2:

- A "complaint of excessive sleepiness or sudden muscle weakness."

- Associated features that include: sleep paralysis; disrupted major sleep episode; hypnagogic hallucinations; automatic behaviors.

- Polysomnography with one or more of the following: "sleep latency less than 10 minutes;" "REM sleep latency less than 20 minutes;" an MSLT with a mean sleep latency less than 5 minutes; "two or more sleep-onset REM periods" (SOREMPs).

- "No medical or mental disorder accounts for the symptoms." (see hypersomnia differential diagnosis)

In the absence of clear cataplexy, it becomes much more difficult to make a firm diagnosis of narcolepsy. “Various terms, such as essential hypersomnia, primary hypersomnia, ambiguous narcolepsy, atypical narcolepsy, etc., have been used to classify these patients, who may be in the developing phase of narcolepsy.”

Since the 2001 ICSD, the classification of primary hypersomnias has been steadily evolving, as further research has shown more overlap between narcolepsy and idiopathic hypersomnia. The 3rd edition of the ICSD is currently being finalized, and its new classification will label narcolepsy caused by orexin deficiency as “type 1 narcolepsy,” which is almost always associated with cataplexy. The other primary hypersomnias will remain subdivided based on the presence of SOREMPs. They will be labeled: “type 2 narcolepsy,” with 2 or more SOREMPs on MSLT; and “idiopathic hypersomnia,” with less than 2 SOREMPS.

However, “there is no evidence that the pathophysiology or therapeutic response is substantially different for hypersomnia with or without SOREMPs on the MSLT.” Given this currently understood overlap of idiopathic hypersomnia and narcolepsy, the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-V) is also updating its classification of the primary hypersomnias. It reclassifies narcolepsy without cataplexy as major somnolence disorder (MSD). Additionally, MSD will encompass all syndromes of hypersomnolence not explained by low orexin concentrations, including idiopathic hypersomnia (with and without long sleep time) and long sleepers (people requiring >10 hours sleep/day).

Further complicating these updated classification schemes, overlap between narcolepsy "with" cataplexy and idiopathic hypersomnia has also been reported. A subgroup of narcoleptics with long sleep time, comprising 18% of narcoleptics in one study, had symptoms of both narcolepsy with cataplexy and idiopathic hypersomnia (long sleep time and unrefreshing naps). It is believed that this subgroup might have dysfunction in multiple arousal systems, including orexin and GABA (see idiopathic hypersomnia causes).

Diagnosis is relatively easy when all the symptoms of narcolepsy are present, but if the sleep attacks are isolated and cataplexy is mild or absent, diagnosis is more difficult. It is also possible for cataplexy to occur in isolation. Three tests that are commonly used in diagnosing narcolepsy are the polysomnogram, the multiple sleep latency test (MSLT), and administration of the Epworth Sleepiness Scale. These tests are usually performed by a sleep specialist. The polysomnogram involves continuous recording of sleep brain waves and a number of nerve and muscle functions during night time sleep. When tested, people with narcolepsy fall asleep rapidly, enter REM sleep early, and may often awaken during the night. The polysomnogram also helps to detect other possible sleep disorders that could cause daytime sleepiness.

The Epworth Sleepiness Scale is a brief questionnaire that is administered to determine the likelihood of the presence of a sleep disorder, including narcolepsy. For the multiple sleep latency test, a person is given a chance to sleep every 2 hours during normal wake times. The patient is taken in usually for an overnight sleep study. The following day the patient will have multiple tests where they will be told to nap after a full nights sleep (usually eight hours). Observations are made of the time taken to reach various stages of sleep (sleep onset latency). This test measures the degree of daytime sleepiness and also detects how soon REM sleep begins. Again, people with narcolepsy fall asleep rapidly and enter REM sleep early. Occasionally, a multiple sleep latency test can result in a false-negative for a narcoleptic.

The system which regulates sleep, arousal, and transitions between these states in humans is composed of three interconnected subsystems: the orexin projections from the lateral hypothalamus, the reticular activating system, and the ventrolateral preoptic nucleus. In narcoleptic individuals, these systems are all associated with impairments due to a greatly reduced number of hypothalamic orexin projection neurons and significantly fewer orexin neuropeptides in cerebrospinal fluid and neural tissue, compared to non-narcoleptic individuals. Those with narcolepsy generally experience the REM stage of sleep within five minutes of falling asleep, while people who do not have narcolepsy (unless they are significantly sleep deprived) do not experience REM until after a period of slow-wave sleep, which lasts for about the first hour or so of a sleep cycle.

Measuring orexin levels in a person's cerebrospinal fluid sampled in a spinal tap may help in diagnosing narcolepsy, with abnormally low levels serving as an indicator of the disorder. This test can be useful when MSLT results are inconclusive or difficult to interpret.

Idiopathic hypersomnia has historically been "difficult to diagnose at an early stage," especially because many other disorders can cause symptoms of excessive daytime sleepiness (EDS). Therefore, "at the time of presentation, most patients have had the disorder for many years."

Further complicating the diagnostic process, idiopathic hypersomnia lacks a clearly defining clinical feature. Whereas narcolepsy is associated with cataplexy and sleep-onset REM episodes, and Kleine-Levin syndrome is associated with megaphagia (compulsive food cravings) and hypersexuality, idiopathic hypersomnia has no such dramatic associated features, except perhaps sleep drunkenness. "Consequently there has been an unfortunate tendency to label all difficult-to-classify cases of excessive daytime sleepiness as idiopathic hypersomnia." For example, upper airway resistance syndrome and delayed sleep phase disorder were formerly confused with idiopathic hypersomnia, but now that they have been more clearly defined, doctors can more carefully exclude these causes of EDS in order to more correctly diagnose idiopathic hypersomnia. However, "even in the presence of other specific causes of hypersomnia, one should carefully assess the contribution of these etiological factors to the complaint of EDS and when specific treatments of these conditions fail to suppress EDS, the [additional] diagnosis of idiopathic hypersomnia should be considered."

The severity of EDS can be quantified by subjective scales, such as the Epworth sleepiness scale and the Stanford sleepiness scale (SSS), and also by objective tests, like the multiple sleep latency test (MSLT)."

In 2001, the ICSD (International Classification of Sleep Disorders) updated their criteria for the diagnosis of idiopathic hypersomnia. Essentially, EDS must be present for at least 6 months, sleep studies (polysomnography and multiple sleep latency test) must show certain characteristics, and all other known causes for long sleep time and EDS must be considered (see hypersomnia). For the patient, this diagnostic process is often tedious, expensive and time-consuming, as other than the sleep studies, it is still basically a diagnosis of exclusion.

In patients with idiopathic hypersomnia, polysomnography typically shows short sleep latency, increased mean slow wave sleep, and a high mean sleep efficiency. "Latency to REM sleep and percentages of light sleep and REM sleep were normal, compared with normal ranges." Despite this, one study has found increased sleep fragmentation in patients with idiopathic hypersomnia without long sleep time, suggesting multiple possible presentations.

It is important to note that although sleep latencies are typically short in idiopathic hypersomnia, the clinical severity may not correlate closely with the MSLT results. In fact, "latencies above 5 minutes are not uncommon in patients with clinically severe hypersomnia." When sleep latency is below 10 minutes, the presence of sleep-onset REM periods (SOREMPs) in two or more of the MSLT naps suggests a diagnosis of narcolepsy, whereas sleep periods lacking rapid eye movement (NREM sleep) in the various naps suggests a diagnosis of idiopathic hypersomnia. However, the importance of this differentiation between REM and NREM has been called into question. (see Classification)

Although the MSLT is currently the best available test to diagnose EDS in general, the MSLT protocol lacks the ability to document the extended, unrefreshing daytime naps that often occur in idiopathic hypersomnia. Complicating the matter, several groups of researchers have found normal MSLT results in patients who otherwise seem to have idiopathic hypersomnia. Therefore, when idiopathic hypersomnia is suspected, researchers suggest appending a 24-hour continuous polysomnography to the standard overnight/MSLT study in order to record total sleep time. Alternatively, an assay of the patient's cerebrospinal fluid (CSF) can be performed in order to test for an adequate level of hypocretin (to exclude narcolepsy with cataplexy) and to determine whether the patient’s CSF abnormally boosts GABA receptor sensitivity (thought to underlie many cases of idiopathic hypersomnia and narcolepsy without cataplexy). Globally, there are very few labs capable of performing the CSF assays referenced above.

It is also important to note that whereas narcolepsy is strongly associated with the HLA-DQB1*0602 genotype, "HLA typing is of no help in the positive diagnosis of idiopathic hypersomnia." This is "despite some reports that suggest an increase frequency of HLA Cw2 and DRS in idiopathic hypersomnia subjects."

Polysomnography is also used to aid in the diagnosis of other sleep disorders such as obstructive sleep apnea (OSA), narcolepsy, and restless leg syndrome (RLS). Normal test results show little to no episodes of sleep apnea and normal electrical activity in the individual's brain and muscles during sleep.

Oximetry, which may be performed over one or several nights in a person's home, is a simpler, but less reliable alternative to a polysomnography. The test is only recommended when requested by a physician and should not be used to test those without symptoms. Home oximetry may be effective in guiding prescription for automatically self-adjusting continuous positive airway pressure.

Polysomnography is a study conducted while the individual being observed is asleep. A polysomnograph (PSG) is a recording of an individual's body functions as they sleep. Complete sleep studies are most commonly facilitated at a designated sleep center. Specialized electrodes and monitors are connected to the individual and remain in place throughout study. Video cameras can be used in certain cases to record physical behaviors occurring while the individual is asleep. Typically the unwanted sexual behaviors do not present on film, therefore the majority of information is taken from a sleep study.

In general, there are two broad classes of treatment, and the two may be combined: psychological (cognitive-behavioral) and pharmacological. In situations of acute distress such as a grief reaction, pharmacologic measures may be most appropriate. With primary insomnia, however, initial efforts should be psychologically based, including discussion of good sleep hygiene. Other specific treatments are appropriate for some of the disorders, such as ingestion of the hormone melatonin, correctly timed bright light therapy and correctly timed dark therapy or light restriction for the circadian rhythm sleep disorders. Specialists in sleep medicine are trained to diagnose and treat these disorders, though many specialize in just some of them.

In patients who are at high likelihood of having OSA, a randomized controlled trial found that home oximetry (a non-invasive method of monitoring blood oxygenation) may be adequate and easier to obtain than formal polysomnography. High probability patients were identified by an Epworth Sleepiness Scale (ESS) score of 10 or greater and a Sleep Apnea Clinical Score (SACS) of 15 or greater. Home oximetry, however, does not measure apneic events or respiratory event-related arousals and thus does not produce an AHI value.

Several circumstances have been identified that are associated with an increased risk of sleep paralysis. These include insomnia, sleep deprivation, an erratic sleep schedule, stress, and physical fatigue. It is also believed that there may be a genetic component in the development of RISP, because there is a high concurrent incidence of sleep paralysis in monozygotic twins. Sleeping in the supine position has been found an especially prominent instigator of sleep paralysis.

Sleeping in the supine position is believed to make the sleeper more vulnerable to episodes of sleep paralysis because in this sleeping position it is possible for the soft palate to collapse and obstruct the airway. This is a possibility regardless of whether the individual has been diagnosed with sleep apnea or not. There may also be a greater rate of microarousals while sleeping in the supine position because there is a greater amount of pressure being exerted on the lungs by gravity.

While many factors can increase risk for ISP or RISP, they can be avoided with minor lifestyle changes. By maintaining a regular sleep schedule and observing good sleep hygiene, one can reduce chances of sleep paralysis. It helps subjects to reduce the intake of stimulants and stress in daily life by taking up a hobby or seeing a trained psychologist who can suggest coping mechanisms for stress. However, some cases of ISP and RISP involve a genetic factor—which means some people may find sleep paralysis unavoidable. Practicing meditation regularly might also be helpful in preventing fragmented sleep, and thus the occurrence of sleep paralysis. Research has shown that long-term meditation practitioners spend more time in slow wave sleep, and as such regular meditation practice could reduce nocturnal arousal and thus possibly sleep paralysis.

Episodes of sleep paralysis can occur in the context of several medical conditions (e.g., narcolepsy, hypokalemia). When episodes occur independent of these conditions or substance use, it is termed "isolated sleep paralysis" (ISP). When ISP episodes are more frequent and cause clinically-significant distress and/or interference, it is classified as "recurrent isolated sleep paralysis"(RISP). Episodes of sleep paralysis, regardless of classification, are generally short (1–6 minutes), but longer episodes have been documented. With RISP the individual can also suffer back-to-back episodes of sleep paralysis in the same night, which is unlikely in individuals who suffer from ISP.

It can be difficult to differentiate between cataplexy brought on by narcolepsy and true sleep paralysis, because the two phenomena are physically indistinguishable. The best way to differentiate between the two is to note when the attacks occur most often. Narcolepsy attacks are more common when the individual is falling asleep; ISP and RISP attacks are more common upon awakening.

The light-dark cycle is the most important environmental time cue for entraining circadian rhythms of most species, including humans, and bright artificial light exposure has been developed as a method to improve circadian adaptation in night workers. The timing of bright light exposure is critical for its phase shifting effects. To maximize a delay of the body clock, bright light exposure should occur in the evening or first part of the night, and bright light should be avoided in the morning. Wearing dark goggles (avoiding bright light) or blue-blocking goggles during the morning commute home from work can improve circadian adaptation. For workers who want to use bright light therapy, appropriate fixtures of the type used to treat winter depression are readily available but patients need to be educated regarding their appropriate use, especially the issue of timing. Bright light treatment is not recommended for patients with light sensitivity or ocular disease.

The "DSM-IV-TR" diagnostic criteria for sleep terror disorder requires:

- recurrent periods where the individual abruptly wakes from sleeping with a scream

- the individual experiences intense fear and symptoms of autonomic arousal, such as increased heart rate, heavy breathing, and increased perspiration

- the individual cannot be soothed or comforted during the episode

- the individual is unable to remember details of the dream or details of the episode

- the occurrence of the sleep terror episode causes "clinically significant" distress or impairment in the individual's functioning

- the disturbance is not due to the effects of a substance or general medical condition

In most children, night terrors eventually subside and do not need to be treated. It may be helpful to reassure the child and their family that they will outgrow this disorder.

Psychotherapy or counseling can be helpful in many cases. There is some evidence to suggest that night terrors can result from lack of sleep or poor sleeping habits. In these cases, it can be helpful to improve the amount and quality of sleep which the child is getting. If this is not enough, benzodiazepines (such as diazepam) or tricyclic antidepressants may be used; however, medication is only recommended in extreme cases.

Possible treatments for circadian rhythm sleep disorders include:

- Behavior therapy or advice about sleep hygiene where the patient is told to avoid naps, caffeine, and other stimulants. They are also told to not be in bed for anything besides sleep and sex.

- Dark therapy, for example the use of blue-blocking goggles, is used to block blue- and bluegreen wavelength light from reaching the eye during evening hours so that the production of melatonin is not decreased or eliminated.

- Medications such as melatonin and modafinil (Provigil), or other short term sleep aids or wake-promoting agents can be beneficial; the former is a natural neurohormone responsible partly and in tiny amounts for the human body clock. The melatonin agonist Tasimelteon, trade name Hetlioz, has been approved in the USA solely for the treatment of non-24-hour sleep–wake disorder in totally blind people.

- Sleep phase chronotherapy may progressively advance or delay sleep time.

Experts agree that there is no such thing as an "ideal" night work schedule, but some schedules may be better than others. For example, rotating shifts every two weeks in a forward (delaying) direction was found to be easier than rotation in a backward (advancing) direction. Gradual delays (“nudging” the circadian system about an hour per day) has been shown in a laboratory setting to maintain synchrony between sleep and the endogenous circadian rhythms, but this schedule is impractical for most real world settings. Some experts have advocated short runs (1 to 2 days) of night work with time for recovery; however, in the traditional heavy industries, longer (5 to 7 day) runs remain the rule. In the end, scheduling decisions usually involve maximizing leisure time, fairness in labor relations, etc. rather than chronobiological considerations. Shift workers can benefit from adhering to sleep hygiene practices related to sleep/wake scheduling. Symptoms typically only fully resolve once a normal sleep schedule is resumed.

Many night workers take naps during their breaks, and in some industries, planned napping at work (with facilities provided) is beginning to be accepted. A nap before starting a night shift is a logical prophylactic measure. However, naps that are too long (over 20–30 minutes) may generate sleep inertia, a groggy feeling after awakening that can impair performance. Therefore, brief naps (10 to 30 minutes) are preferred to longer naps (over 30 minutes). Also, long naps may also interfere with the main sleep bout.

In the transportation industry, safety is a major concern, and mandated hours of service rules attempt to enforce rest times.

There are three types of sleep apnea. OSA accounts for 84%, CSA for 0.4%, and 15% of cases are mixed.