To Name :
To Email :
From Name :
From Email :
Comments :

Evidence-Based Reviews

‘Night owls’: Reset the physiologic clock in delayed sleep phase disorder

Therapies can help adjust inherent ‘eveningness’ in drowsy young patients

Vol. 8, No. 3 / March 2009

Comment on this article

Jason, age 16, has had difficulty with sleep initiation for 2 years. He describes going to bed at 10:30 PM on school nights but falling asleep no sooner than midnight and typically after 1:30 AM. He denies contributions from an “active mind” or environmental disturbances, and his bedroom contains no TV, computer, or other media devices. He does not sleep better with a change in environment. He denies pervasive low mood symptoms and believes his mood hinges predominantly on his ability to achieve sufficient sleep.

Once asleep, Jason generally enjoys good sleep consolidation until he needs to arise at 6:30 AM. His mother awakens him with difficulty, as he often sleeps through his alarm. He sleeps approximately 5 hours nightly during the school week, endorses impaired concentration, and often dozes during his first several classes. When he returns home from school, he finds it very difficult to resist napping.

On weekends he retires at 1 AM or later and typically falls asleep within 30 minutes. He usually awakens at noon but can sleep as late as 4:30 PM. He feels slightly more refreshed on weekends and describes his mood then as improved. During a recent spring break, he felt much better when allowed to sleep as much as he wanted.

Delayed sleep phase disorder (DSPD)—characterized by a pathological “night owl” circadian preference—is seen most commonly in adolescents and is associated with psychiatric morbidity, psychosocial impairment, and poor academic performance. Proper identification of the condition can be enhanced with a variety of assessment tools, and successful treatment requires an awareness of potential endogenous and exogenous contributors.

This article describes what is known about DSPD and uses the case example to illustrate diagnostic assessment and treatment choices. Intriguing data support various pathophysiologic explanations for DSPD (Box 1).1-6 Facilitating the adjustment of patients’ physiologic clocks is the overall goal in managing DSPD.

Box 1

What causes delayed sleep phase disorder?

In individuals normally entrained to the light/dark cycle, circadian rhythms are:

  • delayed by evening exposure to bright light (≥ 2,500 lux) prior to the core body temperature minimum (Tmin)
  • advanced by morning light exposure after the Tmin.1

These opposing effects attune most people to the light/dark cycle, with sleep and wakefulness occurring on a conventional schedule. Persons with delayed sleep phase disorder (DSPD) live at a delayed phase that resists advancement and is incompatible with their personal and social obligations.

Theories have been proposed, but DSPD’s etiology has not been fully explained. Affected adolescents may exhibit an extreme in circadian preference. Case reports also describe DSPD emerging after traumatic brain injury.2

Intriguing evidence supports various pathophysiologic explanations for DSPD. An abnormally long intrinsic circadian period (>25 hours) was recently demonstrated during temporal isolation in 1 individual with DSPD.3 Both this case report and controlled studies describe deviations from expected relationships between the sleep/wake cycle and physiologic circadian markers. Most consistently described are longer intervals from Tmin4 to sleep offset (final rise time) in DSPD patients compared with controls.

Other research suggests:

  • hypersensitivity to nocturnal photic stimulation in select DSPD patients5
  • impaired homeostatic sleep processes, as DSPD patients show a diminished ability to initiate sleep after sleep restriction, compared with controls.6

Extreme ‘eveningness’

Because of their extreme seemingly innate preference to retire and arise at relatively late clock hours (an “eveningness” trait), school-aged patients with DSPD represent a high-risk population for problematic sleepiness. In a survey of 612 high school students, the 63% who felt they needed more sleep on school nights showed a strong eveningness preference (as assessed by questionnaire), compared with students who described getting sufficient sleep.7 Other studies have revealed psychiatric morbidity (including affective and personality disorders), psychosocial impairment, and poor academic performance associated with the condition.8-10

DSPD may affect 7% to 16% of patients presenting with insomnia complaints in sleep medicine clinics.11 The condition appears most common among young cohorts and has been reported to affect up to 7% of adolescents in the United States.12 Its high frequency in this age group may be a pathologic exaggeration of the normal tendency toward delayed timing of sleep and wakefulness linked with pubertal development.13

Sleep and wakefulness regulation

Conceptually, 2 processes govern sleep and wakefulness:

  • The homeostatic drive to sleep (process S) is proportional to the duration of sleep restriction and becomes maximal at about 40 hours.
  • Circadian regulation (process C) creates a drive for wakefulness that variably opposes process S and depends upon intrinsic rhythms.14

Neurons of the suprachiasmatic nucleus in the hypothalamus exert master coordination of this sleep/wake rhythm, along with other behavioral and physiologic variables.15 Because the typical intrinsic period is slightly longer than 24 hours, synchronization to the 24-hour day (entrainment) is accomplished by environmental inputs (zeitgebers, or “time givers”), the most important of which is exposure to light.16

Misalignment between endogenous circadian rhythms and the light/dark cycle can result in circadian rhythm sleep disorders, such as:

  • delayed sleep timing (DSPD)
  • advanced sleep timing (advanced sleep phase disorder)
  • erratic sleep timing (irregular sleep/wake rhythm)
  • complete dissociation from the light/dark cycle (circadian rhythm sleep disorder, free-running type).

These 4 conditions are thought to involve predominantly intrinsic mechanisms, but circadian dysrhythmias also can be induced by exogenous factors. Extreme work schedules or rapid travel across time zones can challenge the circadian system’s ability to acclimate and the individual’s ability to achieve a desired sleep schedule.17

Differential diagnosis

Because DSPD relates primarily to an aberration in timing of sleep, it is characterized as a disorder only if the individual’s preferred schedule interferes substantially with social or occupational functioning. The International Classification of Sleep Disorders (ICSD) provides detailed diagnostic criteria (Table).17


Diagnostic criteria for delayed sleep phase disorder

A. Delay exists in the phase of the major sleep period in relation to desired sleep time and wake-up time, as evidenced by:

  • a chronic or recurrent complaint of inability to fall asleep at a desired conventional clock time AND
  • inability to awaken at a desired and socially acceptable time.

B. When allowed to choose a preferred schedule, patients exhibit normal sleep quality and duration for age and maintain a delayed but stable phase of entrainment to the 24-hour sleep/wake pattern.

C. Monitoring with a sleep log or actigraphy (including sleep diary) for at least 7 days demonstrates a stable delay in the timing of the habitual sleep period.

D. The sleep disturbance is not better explained by another sleep disorder, medical or neurologic disorder, mental disorder, medication use, or substance use disorder.

Source: Adapted and reprinted with permission from International classification of sleep disorders. Diagnostic and coding manual. 2nd ed17

Depression and anxiety often manifest with sleep difficulties, as do inadequate sleep hygiene and other conditions associated with prolonged sleep initiation. According to ICSD criteria, primary insomnia can be differentiated from DSPD if the patient readily initiates and maintains sleep when allowed to sleep on his/her desired sleep/wake schedule. Accumulated evidence has largely debunked this notion, however, as polysomnographic studies have demonstrated both prolonged sleep latency and impaired sleep efficiency in DSPD patients versus matched controls.3

Assessment tools can complement the clinical history in diagnosing DSPD. Either a sleep log or actigraphy is required to demonstrate a stable phase delay, but actigraphy typically generates more reliable data.18 Actigraphs are compact “motion detectors” whose output while being worn by patients allows longitudinal assessment of sleep/wake parameters.

Eveningness tendencies of presumptive DSPD patients can be further verified with the Morningness-Eveningness Questionnaire (MEQ) (Box 2).19 Low scores are associated with evening types—felt to correspond to the endogenous circadian period—and can help narrow the differential diagnosis of sleep-initiation complaints.20

Box 2

A morning or evening person? A self-assessment questionnaire

The Morningness-Eveningness Questionnaire (MEQ) developed by Horne and Ostberg19 can be used to verify eveningness tendencies of patients with presumptive delayed sleep phase disorder. The MEQ is a 19-item self-assessment tool with responses that are assigned values totaling up to 86 points. Examples of the questions include:

  • Considering only your own ‘feeling best’ rhythm, at what time would you get up if you were entirely free to plan your day?
  • Considering only your own ‘feeling best’ rhythm, at what time would you go to bed if you were entirely free to plan your day?
  • How easy do you find it to get up each day?
  • When you have no commitments the next day, how much later do you go to bed compared to your usual bedtime?
  • One hears about ‘morning’ and ‘evening’ types of people. Which ONE of these types do you consider yourself to be?

Lower scores are associated with evening types—felt to correspond to the endogenous circadian period—and can help in narrowing the differential diagnosis of sleep-initiation complaints.20 Scores on the MEQ are interpreted as:

  • 70 to 86: definite morning type
  • 59 to 69: moderately morning type
  • 42 to 58: neither type
  • 31 to 41: moderately evening type
  • 16 to 30: definite evening type

CASE CONTINUED: ‘Definite evening type’

Jason scores 28 on the MEQ, consistent with a “definite evening type.” Actigraphic monitoring is scheduled during a school holiday, when he is instructed to sleep according to his preferred schedule with the least possible restriction.

A clearly delayed sleep phase is evident, with the habitual sleep period occurring between 5 AM and 1 PM. Even on days when he was quite sleep-restricted because of an enforced wake time, sleep onset on the ensuing evening was substantially delayed, suggesting an obligate nature for the delayed sleep/wake schedule. Overall, Jason had few complaints with respect to impaired alertness while on this unrestricted schedule and experienced a much more stable mood.


Without physiologic assessments, understanding the patient’s “natural” sleep schedule can allow for rational recommendations about using phototherapy and oral melatonin (Figure21). However, referral to a sleep specialist is required unless the general psychiatrist has experience in treating circadian rhythm sleep disorders.

Morning phototherapy. Properly timed morning bright light therapy (≥2,500 lux) has been shown to help DSPD patients achieve physiologically measured sleep phase advances, objective improvements in daytime alertness, and earlier reported bedtimes compared with controls.22 Unfortunately, the described 2-hour treatment duration make this research protocol clinically impractical, and most clinicians commence with a 30-minute duration of therapy, as described in the seasonal affective disorder literature.

Relatively new and widely available blue light boxes have been reported to exhibit at least equivalent efficacy to bright light devices (as reported in the literature pertaining to seasonal affective disorder), but with markedly decreased light intensity and fewer associated adverse effects.23 As the research addressing their use in the treatment of circadian rhythm sleep disorders is still emerging, their future role remains uncertain.

Precautions. Most psychiatrists would not perform a physiologic determination of a patient’s circadian phase, and further undesired phase delays can occur if phototherapy is administered before the core body temperature minimum (Tmin).24 Also, use caution if prescribing phototherapy to patients taking photosensitizing drugs and/or those with ocular or retinal pathology.20

Evening light avoidance. Whether or not you prescribe morning phototherapy, recommending that DSPD patients avoid evening light is essential to avoid further induction or exacerbation of phase delays. Protective eyewear is warranted in instances where these advisory precautions are insufficient (see Related Resources). Such an intervention has been shown effective in decreasing light exposure and undesired phase advances in studies involving subjects exposed to simulated shift work.25

Oral melatonin. Abundant evidence supports melatonin use in achieving phase advances in individuals with DSPD.26,27 A synergistic effect can be obtained when melatonin is combined with phototherapy.28

Proper timing of melatonin to achieve a maximal phase advance can be estimated based on the individual’s dim light melatonin onset (DLMO), which occurs approximately 14 hours after the habitual (unrestricted) wake time.29 Maximal phase advances appear to occur when melatonin is given approximately 6 hours before the DLMO.26 Thus, a rational practice is to recommend that patients take melatonin 8 hours after their natural wake time. Doses of ≤0.5 mg appear to achieve the maximal chronobiotic effect while avoiding an undesired hypnotic effect.30

Did you miss this content?
Botulinum toxin for depression? An idea that’s raising some eyebrows