The Body Clock and Its Discontents

August 3, 2015
Alfred J. Lewy, MD, PhD
Volume 32, Issue 8

Circadian components are profound in depressive disorders, such seasonal affective disorder. This article introduces a course to be given at this year’s APA Meeting in Toronto, on melatonin and light treatment.

Now is the winter of our discontent
[[{"type":"media","view_mode":"media_crop","fid":"40453","attributes":{"alt":"© Giovanni Cancemi/","class":"media-image media-image-right","id":"media_crop_651846583435","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"4148","media_crop_rotate":"0","media_crop_scale_h":"200","media_crop_scale_w":"124","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"float: right;","title":"© Giovanni Cancemi/","typeof":"foaf:Image"}}]]Made glorious summer by this
   sun of York

-W. Shakespeare, Richard III

Humans were biologically programmed to sleep at times that in the modern era are often in conflict with the demands of bosses, spouses, kids, and pets. In most advanced civilizations, there have been added discontents: there are 52 weekend-to-weekday transitions each year and, more recently, 2 equinoctial time changes-not to mention the advent of various sources of artificial light that now include electronic devices often used in the wee hours of the morning.

Remarkably, until 1980 the scientific wisdom was that humans were uniquely impervious to the non-visual effects of ocularly mediated light. All of this changed overnight when it was discovered that intensity-dependent light can suppress endogenous melatonin production in humans. Key among its many implications was that we are more responsive to sunlight than we are to indoor light. This discovery immediately led to the development of bright light therapy for seasonal affective disorder (SAD), and it continues to be generative.

Most patients with SAD become depressed every winter, some each summer. According to the phase shift hypothesis, winter depression comprises 2 circadian phase types-“owls” and “larks”-with the former predominating. The treatment of choice for both of these phase types is bright light, optimally scheduled in the morning or in the evening, respectively. Bright light in the morning causes a phase advance (a shift to an earlier time), and in the evening it causes a phase delay (a shift to a later time).

At least two-thirds of those who have winter depression have phase-delayed circadian misalignment (the owls); those in the smaller subgroup (the larks) have phase-advanced circadian misalignment (Figure 1). As days get shorter or longer, the body clock shifts in accordance with the change in dawn or dusk, but work and social schedules constrain our sleep time. Thus, there is a mismatch between these two sets of circadian rhythms, resulting in internal misalignment. In vulnerable individuals, misalignment triggers depression.

Phase-delayed misalignment occurs when the body clock is abnormally delayed (late) with respect to the sleep-wake cycle, and phase- advanced misalignment occurs when the body clock is abnormally advanced (early) with respect to the sleep-wake cycle. Sleep times can also occur earlier or later than normal, sometimes in the same direction as the body clock (Figure 2) and sometimes not (thereby further worsening internal misalignment between the set of circadian rhythms closely tied to the body clock and the set related to the sleep-wake cycle and its evoked rhythms).

Assessing body clock time

Objective phase typing of the circadian rhythms closely tied to the body clock can now be accomplished using the salivary dim light melatonin onset obtained at home in the evening; conveniently, it usually occurs before habitual sleep time. The most cumbersome constraint is that saliva samples need to be collected under dim light, because melatonin production is suppressed by light exposure. However, some strategies are in the works to liberalize this requirement, such as the use of blue-blocking amber goggles.

The dim light melatonin onset biomarker is the gold standard for assessing body clock time. Mid-sleep is the best biomarker for the timing of the sleep-wake cycle and its evoked rhythms. The time interval, or phase angle difference between dim light melatonin onset and mid-sleep, is optimally 6 hours in patients with SAD and can be used to phase type circadian misalignment objectively. Phase-delayed types have a phase angle difference shorter than 6 hours (a dim light melatonin onset that is late with respect to mid-sleep), and phase- advanced types have a phase angle difference longer than 6 hours (a dim light melatonin onset that is early with respect to mid-sleep).

In SAD, there appears to be a “sweet spot” for optimal alignment at phase angle difference 6, which turns out to be the vertex of parabolas fitted to depression ratings plotted against phase angle difference before and after treatment (see Figures 2, 3, and 4 in Lewy et al1): the more phase angle difference deviates from the sweet spot in either direction, the worse the depression. Too much phase resetting can result in a shift to the other side of the para-bolic sweet spot, which causes depression ratings to increase.

Effects of melatonin

The phase-shifting effects of melatonin have recently been shown to be equal and additive to those of bright light. Low-dose melatonin (0.3 to 0.6 mg) is preferred to provide a chemical dark signal to the body clock, overriding the effects of ambient light: when taken at wake time it causes a phase delay and in the afternoon/evening it causes a phase advance. If a patient is unusually sensitive to the soporific side effect of melatonin (which rarely occurs at these low doses), he or she should not drive immediately after taking a pill. Low-dose melatonin causes more robust phase shifts than some higher doses, because it provides a more discrete time signal.

Although some patients with SAD enjoy the experience of sitting in front of a light box, taking a pill is much more convenient and should increase adherence. As shown in Figure 2, these phase-resetting agents can be used separately or together (ie, morning melatonin plus evening light-or morning light plus afternoon/evening melatonin). The precise optimal times are tied to the dim light melatonin onset, which can vary widely among individuals. As a patient responds, follow-up dim light melatonin onsets can guide adjustment of these treatment times.

Correcting circadian misalignment

To correct phase-delayed circadian misalignment (Figure 2), the body clock needs to be shifted earlier using bright light in the morning and/or low-dose melatonin in the afternoon/evening. To correct phase-advanced circadian misalignment, the body clock needs to be shifted later using bright light in the evening and/or low-dose melatonin in the morning. The clock times in Figure 2 are averaged for people who habitually awaken at 0600, although the precise optimal times for scheduling light and melatonin are tied to the internal body clock. The gold standard biomarker for the body clock is a person’s salivary dim light melatonin onset. There is not much day-to-day variation in the dim light melatonin onset; among individuals, however, it can occur from several hours before to just before (rarely, just after) sleep onset.

Thus, while at present dim light melatonin onset/mid-sleep phase angle difference is primarily a research tool, it is perhaps the first comprehensively useful laboratory test in psychiatry that can objectively diagnose circadian rhythm disorders, determine initial treatment, monitor clinical response, and guide ongoing changes in treatment parameters. This is a good example of personalized medicine, even without the addition of genome testing, which is sure to follow. Until more commercial laboratory tests become available for processing dim light melatonin onsets, clinical features can suffice in phase typing the large majority of patients.

The circadian components are profound in persons with SAD, advanced and delayed sleep phase disorders, jet lag, shift work maladaptation, the equinoctial time transitions, and Monday morning blues, as well as in individuals with total blindness. They can be objectively assessed using dim light melatonin onset/mid-sleep phase angle difference and easily treated with one or both phase- resetting agents. The field is now poised to take on the challenge of non-seasonal psychiatric disorders, for which circadian misalignment is probably not as important as in SAD. But even if this component is small and when corrected only promotes better sleep, safe and inexpensive treatments that can be added adjunctively to just about any medication will be highly useful.

Low-dose melatonin causes phase shifts in everyone. Higher doses (1 to 10 mg) help about one-third of the population get to sleep when taken before bedtime. No serious, irreversible adverse effects have been established, and they are even less likely to occur at low, near-physiological doses. The anecdotally reported adverse effect of high-dose melatonin at bedtime of making dreams more vivid and more memorable (to be avoided in patients with PTSD) may be of special interest to the psychoanalyst. In any event, a more reliable way to help individuals fall asleep earlier (and wake up earlier) is to have them take low-dose melatonin in the afternoon to shift the circadian rhythm of sleep propensity earlier. There are few, if any, absolute contraindications for the use of bright light and low-dose melatonin. Until more clinical trials documenting their safety are undertaken, clinical exigency will determine their use in canonically high-risk populations (pregnant women, nursing mothers, younger children, the frail, and the elderly) who may one day be the greatest beneficiaries of these phase-resetting treatments.

The main challenge ahead will be to identify clinical phenotypes that represent homogeneous circadian subgroups among the non-seasonal psychiatric disorders (including MDD, bipolar disorder, anxiety disorders, autism, and ADHD) as having a clinically meaningful circadian misalignment component of one phase type or the other. It is also possible that phase typing will be optimally done on an individual basis.

An example of phase typing is illustrated in the following e-mail I received from a grateful mother. The identity of the writer has been withheld and some of the details have been changed.

I never got back to you about this, but I’ve been meaning to respond to say thank you!

You offered to do a consult with our psychiatrist (Dr S) last year about my son’s symptoms, and my suspicion that he might have reverse SAD. I had written to you several times over a few years about my concerns with his behavior/symptoms in summer, which doctors had tried to attribute to sensory issues, lack of school routine or, finally, bipolar disorder (which he does not have). We did what you recommended-morning sun, a small dose of melatonin at 4 pm-and for the first summer in years he had normal sleep patterns, baseline anxiety and agitation levels, etc. We also kept him out of the afternoon sun (usually 3 to 7-ish). I wanted to let you know what a difference it made.

As I mentioned in previous e-mails to you, the summer before, he ended up in sub-acute care with psychotic behaviors from the extreme insomnia that he experienced. I don’t know if you collect data on children with reverse SAD, but he is 12, almost 13, and we had seen an increase in his symptoms each summer until we followed your recommendations. At sub-acute care, they wanted to treat his symptoms with antipsychotics and take him off of his ADHD medications, because they were certain it was bipolar disorder initially. With your recommended treatment, which is a lot safer than antipsychotics to say the least, he does great!

I wanted to let you know how appreciative I am that you were willing to do the consult and that you were also willing to take my concerns seriously. Dr S has been great but before working with him, every doctor we talked to brushed us off about it being “impossible.”

Thank you.

Adjusting to equinoctial time changes

Here are some tips to help people adjust to the 2 equinoctial time changes. Although no one has much jet lag traveling east or west to the next time zone, the sunlight available to us on the Sunday after the time change is perversely acting to prevent the shift in our body clocks in the correct direction.

In the fall, we must delay our body clocks 1 hour later. However, we have more sunlight Sunday morning (which prevents delaying of the body clock) and less sunlight at the end of the day (which would have helped us to delay). So it is best to avoid sunlight for the first hour in the morning and to get as much sunlight before dusk (do not look directly at the sun, of course). Also, a very small dose of melatonin taken at wake time, starting on Friday (the weekend of the time change) and ending on Monday, will facilitate the 1-hour delay necessary for adjustment.

By the way, it is not necessarily true that we gain an hour of sleep right after switching to standard time, because we do not usually wake up naturally an hour later on Sunday morning and on Sunday night we go to bed an hour later and yet still wake up an hour before our alarm clock on Monday. So, many of us actually lose an hour of sleep until we adjust-which can take up to a week or more.

In the spring, everyone loses an hour of sleep because we are required to wake up 1 hour earlier on Sunday morning. The switch to daylight saving time results in more sunlight available at the end of the day, which many of us celebrate by staying outdoors Sunday evening (weather permitting).

However, exposure to outdoor sunlight before dusk works against the advance in our body clock needed for adjustment. So it is best to postpone this gratification until Monday. Getting as much outdoor sunlight as early as possible will help, as well as a small dose of melatonin taken in the afternoon, Friday through Monday (the weekend of the time change). Sleepiness is not usually associated with low-dose melatonin, but be careful driving if you are unusually sensitive to this side effect.


Much of the work done so far has been undertaken by psychiatrists and psychologists. More researchers and clinicians in other medical and scientific disciplines are expected to help continue to extend the field from human biology to mainstream medicine. Promising areas of future research include the role of circa- dian misalignment in non-restora-tive sleep and obesity/metabolic syndrome (currently associated with sleep duration). Not too long ago after a lecture, a medical student said to me, “I think this work is on the cusp of being recognized by just about everyone as important to them.” I told him that I have had a similar thought for many years.

This article was originally posted on 4/28/2015 and has since been updated.

Editor’s note: This article is based on the course “Melatonin and Light Treatment of SAD, Sleep and Other Body-Clock Disorders” given on May 16 at the 2015 Annual Meeting of the American Psychiatric Association in Toronto.


Dr Lewy is Professor Emeritus of Psychiatry and Director of the sleep and mood disorders laboratory at Oregon Health & Science University (OHSU), Portland. Dr Lewy reports that he is co-inventor on several melatonin process patents owned by OHSU and currently are not licensed to any company; he has consulted for pharmaceutical companies (such as Servier and currently, Pfizer).


There are 2 types of circadian misalignment between the circadian rhythms tightly coupled to the body clock (indicated by the beds) and those that are related to the sleep-wake cycle (indicated by the horizontal silhouettes). The best biomarker for the body clock is the dim light melatonin onset (not shown), and the best biomarker for the sleep-wake cycle is the mid-point (not specifically indicated above) of the sleep bout. Phase-delayed circadian misalignment occurs when the body clock is delayed (shifted later) with respect to the sleep-wake cycle; phase-advanced circadian misalignment occurs when the body clock is advanced (shifted earlier) with respect to the sleep-wake cycle. (Adapted with permission from Bahattacharjee Y. Science.2 Copyright © 2007 American Association for the Advancement of Science.).

There are 2 types of treatments for circadian phase disorders: low-dose melatonin (indicated by the capsules) and bright light (indicated by the sun). Although optimal timing of these treatments is best scheduled with respect to the internal body clock (as given by the endogenous dim light melatonin onset biomarker), clock times are indicated for individuals whose average habitual wake time is 0600. To provide a corrective phase advance, light should be scheduled at wake time and melatonin should be taken in the afternoon/evening; to provide a therapeutic phase delay, light should be scheduled in the evening and melatonin should be taken at wake time. These treatments are fundamentally grounded in human biology.3-7 (Adapted with permission from Lewy AJ, Sack RL. Progress in Brain Research: Hypothalamic Integration of Circadian Rhythms. Copyright © Elsevier 1996.)

1. Lewy AJ, Lefler BJ, Emens JS, Bauer VK. The circadian basis of winter depression. Proc Natl Acad Sci U S A. 2006;103:7414-7419.

2. Bahattacharjee Y. Psychiatric research: is internal timing key to mental health? Science. 2007;317: 1488-1490.

3. Lewy AJ, Wehr TA, Goodwin FK, et al. Light suppresses melatonin secretion in humans. Science. 1980;210:1267-1269.

4. Lewy AJ, Sack RL, Singer CM. Immediate and delayed effects of bright light on human melatonin production: shifting “dawn” and “dusk” shifts the dim light melatonin onset (DLMO). Ann N Y Acad Sci. 1985;453:253-259.

5. Lewy AJ, Sack RL, Miller LS, Hoban TM. Antidepressant and circadian phase-shifting effects of light. Science. 1987;235:352-354.

6. Lewy AJ, Sack RL. The dim light melatonin onset as a marker for circadian phase position. Chronobiol Int. 1989;6:93-102.

7. Lewy AJ, Bauer VK, Ahmed S, et al. The human phase response curve (PRC) to melatonin is about 12 hours out of phase with the PRC to light. Chronobiol Int. 1998;15:71-83.

8. Lewy AJ, Sack RL. The role of melatonin and light in the human circadian system. In: Buijs RM, Kalsbeek A, Romijn HJ, et al, eds. Progress in Brain Research: Hypothalamic Integration of Circadian Rhythms. Amsterdam: Elsevier; 1996:205-215.

Additional reading

• Burke TM, Markwald RR, Chinoy ED, et al. Combination of light and melatonin time cues for phase advancing the human circadian clock. Sleep. 2013;36:1617-1624.

• Emens J, Lewy A, Kinzie JM, et al. Circadian misalignment in major depressive disorder. Psychiatry Res. 2009;168:259-261.

• Keijzer H, Smits MG, Duffy JF, Curfs LM. Why the dim light melatonin onset (DLMO) should be measured before treatment of patients with circadian rhythm sleep disorders. Sleep Med Rev. 2014; 18:333-339.

• Lewy AJ. Circadian rhythms and mood disorders: a guide for the perplexed. J Clin Psychiatry. 2015; 76:e662-e664.

• Lewy AJ. Endogenous and exogenous melatonin, the sleep-wake cycle and the circadian component of affective disorder. In: Kushida C, ed. Encyclopedia of Sleep. Vol 3. Waltham, MA: Academic Press; 2013:126-137.

• Lewy AJ. Melatonin and human chronobiology. Cold Spring Harb Symp Quant Biol. 2007;72:623-636.

• Lewy AJ, Bauer VK, Cutler NL, et al. Morning vs evening light treatment of patients with winter depression. Arch Gen Psychiatry. 1998;55:890-896.

• Lewy AJ, Emens JS, Sack RL, et al. Low, but not high, doses of melatonin entrained a free-running blind person with a long circadian period. Chrono-biol Int. 2002;19:649-658.

• Lewy AJ, Tutek J, Havel L, et al. The role of circadian rhythms, light and melatonin in sad and nonseasonal affective and anxiety disorders. Curr Psychiatry Rev. 2014;10:214-222.

• Overy C, Tansey EM, eds. The Recent History of Seasonal Affective Disorder (SAD). London: University of London; 2014.