Lack of short-wavelength light during the school day delays dim light melatonin onset (DLMO) in middle school students

Abstract

Circadian timing affects sleep onset. Delayed sleep onset can reduce sleep duration in adolescents required to awake early for a fixed school schedule. The absence of short-wavelength ("blue") morning light, which helps entrain the circadian system, can hypothetically delay sleep onset and decrease sleep duration in adolescents. The goal of this study was to investigate whether removal of short-wavelength light during the morning hours delayed the onset of melatonin in young adults.
Dim light melatonin onset (DLMO) was measured in eleven 8th-grade students before and after wearing orange glasses, which removed short-wavelength light, for a five-day school week.
DLMO was significantly delayed (30 minutes) after the five-day intervention, demonstrating that short-wavelength light exposure during the day can be important for advancing circadian rhythms in students.
Lack of short-wavelength light in the morning has been shown to delay the circadian clock in controlled laboratory conditions. The results presented here are the first to show, outside laboratory conditions, that removal of short-wavelength light in the morning hours can delay DLMO in 8th-grade students. These field data, consistent with results from controlled laboratory studies, are directly relevant to lighting practice in schools.

Full text

To cite this article:
Neuroendocrinol Lett 2010; 31(1):92–96
ORIGINAL ARTICLE
Neuroendocrinology Letters Volume 31 No. 1 2010
Lack of short-wavelength light during the
school day delays dim light melatonin
onset (DLMO) in middle school students
Mariana G. F and Mark S. R
Lighting Research Center, Rensselaer Polytechnic Institute, Troy, New York, USA
Correspondence to: Mariana Figueiro
Lighting Research Center
21 Union St., Troy, NY 12180, USA.
: +1 (518) 687-7100; : +1 (518) 687-7120; -: figuem@rpi.edu
Submitted: 2009-10-21 Accepted: 2009-12-15 Published online: 2010-02-16
Key words: daylight; melatonin; circadian system; adolescents; sleep
Neuroendocrinol Lett 2010; 31(1):92–96 PMID: 20150866 NEL310110A04 © 2010 Neuroendocrinology Letters • ww w.nel.edu
Abstract
OBJECT IVE: Circadian timing affects sleep onset. Delayed sleep onset can reduce
sleep duration in adolescents required to awake early for a fixed school schedule.
The absence of short-wavelength (“blue”) morning light, which helps entrain the
circadian system, can hypothetically delay sleep onset and decrease sleep dura-
tion in adolescents. The goal of this study was to investigate whether removal of
short-wavelength light during the morning hours delayed the onset of melatonin
in young adults.
METHOD S: Dim light melatonin onset (DLMO) was measured in eleven 8th-grade
students before and after wearing orange glasses, which removed short-wavelength
light, for a five-day school week.
RESULTS : DLMO was significantly delayed (30 minutes) after the five-day inter-
vention, demonstrating that short-wavelength light exposure during the day can
be important for advancing circadian rhythms in students.
CONCLUS IONS: Lack of short-wavelength light in the morning has been shown
to delay the circadian clock in controlled laboratory conditions. The results
presented here are the first to show, outside laboratory conditions, that removal
of short-wavelength light in the morning hours can delay DLMO in 8th-grade
students. These field data, consistent with results from controlled laboratory stud-
ies, are directly relevant to lighting practice in schools.
INTRODUCTION
In terrestrial mammals, circadian rhythms are
regulated by the interaction of the internal bio-
logical clock located in the suprachiasmatic nuclei
(SCN) of the hypothalamus with the earth’s natu-
ral 24-hour light-dark pattern (Refinetti 2006).
The SCN are self-sustaining oscillators with an
intrinsic period that is typically slightly longer or
shorter than 24 hours. The timing of the SCN is
set by the local light-dark pattern, usually ensur-
ing that the organism’s behavioral and physiologi-
cal rhythms are synchronized with its photic niche
(nocturnal, diurnal, or crepuscular).
Light incident on human retinas will entrain or
phase shift SCN timing, depending upon the time,
duration, spectrum and intensity of the stimulus
(Stevens & Rea 2001). These fundamental light

93
Neuroendocrinology Letters Vol. 31 No. 1 2010 • Article available online: http://node.nel.edu
Lack of short-wavelength light during the school day delays dim light melatonin onset
characteristics affect the circadian system differently
than they affect the visual system. Although we now
know the human circadian system is more sensitive to
light than was originally thought (Lewy et al. 1980), it is
much less sensitive to light than the visual system (Rea
et al. 2002). It is also well established that the human
circadian system is maximally sensitive to short-wave-
length (450 nm to 480 nm) light (Brainard et al. 2001,
Thapan et al. 2001, and Rea et al. 2005). Most electric
light sources illuminating our indoor environments
are designed to support the visual system by providing
relatively low levels of light dominated by wavelengths
near 555 nm, the peak of the photopic luminous effi-
ciency function (CIE 1978). Moreover, for conve-
nience, electric light sources are available night or day
and for variable durations. More and more then, people
throughout the world are living under a roof illumi-
nated by electric light sources, exposing them to dim
days and extended dim light at night, separating them
from the robust, natural light-dark cycle.
Studies have shown that adolescents report going to
bed later as they get older (Crowley et al. 2007). These
age-related changes in bedtimes have been associated
with reduced parental influence on bedtimes, increased
homework and extra-curricular activities, and other
activities such as playing on computers and watching
television. Evidence to date supports the hypothesis
that adolescents have a late circadian phase, contribut-
ing to these late bed times. With a highly structured
school schedule requiring early rising, these adoles-
cents typically experience reduced sleep durations.
Indeed, on unrestricted weekends, adolescents rise 1.5
to 3 hours later than they do on weekdays (Crowley et
al. 2007).
Light during the day is important for entrainment;
that is, for aligning circadian phase to the rest-activ-
ity cycle required by attending school. For reasons
described above, however, electric lighting, including
that common in schools, may not provide adequate
light for circadian entrainment. Without a robust light
stimulus during the day then, adolescents would logi-
cally be expected to exhibit late circadian phase and
therefore go to bed late and experience restricted sleep.
Daily morning short-wavelength light exposures
(after minimum core body temperature) are expected
to slightly advance the clock every day and thereby
maintain entrainment to the solar day (Jewett et al.
1997). The impact of reduced daily short-wavelength
light exposure on the circadian system of young adults,
as might be experienced by students without adequate
daylight (or electric light) exposure, has never been for-
mally investigated. A simple before-and-after, within-
subjects field experiment was conducted in a school
with documented good daylight design to determine
whether removal of short-wavelength light on five con-
secutive school days would delay circadian phase rela-
tive to a baseline measurement obtained prior to the
intervention.
METHODS
Site
The study was conducted at Smith Middle School,
Chapel Hill, North Carolina in May 2009. Smith Middle
School is unusual with respect to current architectural
practice in terms of the amount of daylight provided to
interior spaces (LRC 2004). The building uses south-
facing roof monitor skylights in most spaces to deliver
daylight to the interior spaces. Diffuse toplight ing pre-
vents occlusions due to blinds or wall displays typi cal
of sidelighting. To minimize glare from direct sunlight
entering the spaces, light entering the roof monitor is
baffled with cloth partitions; only diffuse light enters
the conditioned room. The electric lighting system is
controlled with motion sensors and photosensors that
modulate the fluores cent lamp output with dimming
ballasts. This strategy allows electric lights to be off
most of the day for electric energy savings.
The daylighting conditions were evaluated as part
of an extensive case study in 2004 (LRC 2004). On a
sunny afternoon in March 2004, researchers measured
light levels on several surfaces in a classroom with a
calibrated illuminance meter having a photopic spectral
response (CIE 1978). At the time of the site measure-
ments, all illu mination was provided by daylight. Hori-
zontal light level measurements were made by placing
the illuminance meter on desks; these ranged from
1330–2150 lux. Vertical illu minances on the chalkboard
ranged from 996–1 265 lux. The vertical light measure-
ments were made by placing the illuminance meter on
the chalkboard at eye level. Typical levels found in spaces
illuminated only by electric light sources are approxi-
mately 80% lower. That is, these illuminance levels
were approximately 5× higher than commonly found in
buildings only illuminated by electric lights. Based on
calculations using the model of human circadian pho-
totransduction developed by Rea and colleagues (2005),
these vertical illuminance levels would result in at least
60% melatonin suppression (at night), suggesting that
the light stimulus students receive in Smith’s classrooms
is strong enough to activate the circadian system. Based
upon results by Zeitzer et al. (2000) who showed that
the half maximum saturation for phase shifting was
80–160 lux from cool white fluorescent light sources for
a 6.5 hr exposure, the light levels measured in Smith
Middle School would also be highly effective for phase
shifting and, therefore, entrainment. Battery-powered
monitoring devices also recorded illumin ances on the
teacher’s desk over a long period of time. The desk was
located near the perimeter of the room rather than
directly under the roof monitor. These illuminance
levels aver aged 550 lux on sunny days and 320 lux on
partly cloudy days. Based upon the previous measure-
ments, it was expected that students at Smith Middle
School would be exposed to some of the highest illu-
minances typically found in an in door classroom envi-
ronment, making this an appropri ate site for the study.

94
Copyright © 2010 Neuroendocrinology Letters ISSN 0172–780X • www.nel.edu
Mariana G. Figueiro and Mark S. Rea
Procedures
The within-subjects study began at the school on a
Friday morning. Eleven subjects (nine males and two
females, ages 13–14 years) were given an explanation
about the study and were asked to wear orange glasses
that attenuated all light of wavelengths shorter than
about 525 nm from reaching the eyes (Figure 1) during
the study period, Monday to Friday on the following
week, from the time they awoke until they returned
home after school (approximately 15:00); thus, subjects
were required to wear the orange glasses during school
and on the commute to school in the morning when
they are likely to be exposed to daylight. Participants
were then asked to refrain from consuming caffeinated
products for the remainder of the day because saliva
samples would be gathered from them in the evening.
Finally, subjects were instructed to return to the school
at 19:00 for saliva sample collection. The participants
stayed in the dimly illuminated school library during
sample collection. All electric lighting was kept off and
all blinds were pulled down to avoid daylighting in the
space. The room was lit with a dim red light (less than
5 lux at the cornea), dur ing which time the participants
were allowed to watch movies, play games, read or
study. Serial saliva samples (Salivette, Sarstedt, Newton,
NC, USA) were collected every 30 mi nutes (from 19:30
to 23:00) to determine DLMO. The subjects chewed on
a plain cotton cylinder until saturated. These samples
were then, in turn, centrifuged and refrigerated by
the researcher. To pre vent contamination of the saliva
samples, the subjects were not allowed to eat or drink
between sample times. The re frigerated samples were
later sent to Pharmascan, Os ceola, WI, for melatonin
assay. On the next Friday evening, participants returned
to school at 19:00 to repeat the saliva sample data col-
lection in the dimly illuminated library. The study was
approved by Rensselaer Polytechnic Institute’s Institu-
tional Review Board and meets the international ethical
standards of this journal (Portaluppi et al. 2008).
RESULTS
DLMO, in decimal hours, was calculated for each subject
using a threshold of 4.0 pg/ml. DLMO was determined
by using linear interpolation between the melatonin
values that fell above and below threshold. DLMO for
the eleven subjects on the Friday prior to wearing the
orange glasses averaged 21.15±0.61 and DLMO for the
same subjects averaged 21.66±0.81 after five consecu-
tive days of wearing the orange glasses. One subject had
not achieved the threshold value (4 pg/ml) at 23:00 on
the second Friday, suggesting that his DLMO occurred
later than 23:00, but as a conservative estimate, we
used 23:00 as his DLMO time. Using a two-tail paired
students’ t-test, this difference was significant with a
probability of 0.006 of a Type 1 error. Figure 2 shows
the cumulative frequencies of DLMO for the partici-
pants before and after the orange glasses intervention.
DISCUSSION
In one respect the results of this field study are trivial
because they simply confirm what has been shown
before (Warman et al. 2003) namely that removing
short-wavelength light exposure in the morning delays
circadian phase. In another respect, however, the
results of this study are quite important because they
Figure 1. The spectral irradiance distribution (SIR) of daylight varies
continuously throughout the day at every location on the earth.
Shown here is the relative SIR of one, standard phase of daylight
(left ordinate), defined by CIE as illuminant D65 (Wyszecki &
Stiles 1982) to represent natural daylight at 6500 K. Also shown
is the spectral transmittance of the orange glasses (amber/
orange, UV Process Supply), in percent (right ordinate) used in
the study. Irrespective of the actual and highly variable SIR in
daylight present in Smith Middle School during the experiment,
the orange glasses would have attenuated all short-wavelength
light from both the natural and the electric sources seen by the
students.
Figure 2. Cumulative frequencies of DLMO for the students before
and after the orange glasses intervention.

95
Neuroendocrinology Letters Vol. 31 No. 1 2010 • Article available online: http://node.nel.edu
Lack of short-wavelength light during the school day delays dim light melatonin onset
validate controlled laboratory findings with actual field
measurements. Specifically, these data are consistent
with the inference that removing short-wavelength light
during school days delays circadian phase in 8th-grade
students. After five consecutive school days of wearing
orange glasses, DLMO was delayed by about 30 min-
utes. Although it is known that individual SCN clocks
can have different periods, the phase delay of about 6
minutes per day observed here is consistent with the
typical free-running period in humans of 24.18±0.04 hrs
(Czeisler et al. 1999). Therefore, both the direction and
the magnitude of the predicted effects from laboratory
studies were obtained in this field study.
It has been estimated that bedtime occurs approxi-
mately two to three hours after DLMO (Burgess et al.
2003; Burgess & Fogg 2008). Since the present results
showed that removing short-wavelength light during
the school day will delay DLMO, sleep times are likely
to be delayed as well. Wake-up times are fixed for most
students, so those who do not receive short-wavelength
light during the day will probably have reduced sleep on
school nights. One study showed that students who had
poorer performance in school were those who obtained
about 25 minutes less sleep per night and went to bed
on average 40 minutes later on school nights than those
who were good performers (Wolfson and Carskadon
1998). By extension then, those who do not get enough
short-wavelength light during the school day would
exhibit reduced scholastic performance.
These findings, bridging controlled laboratory
results to a real school environment, should have
important, and practical, implications for school design
because it seems necessary to expose students to short-
wavelength light during the early part of the day to
maintain entrainment. Conscious delivery of short-
wavelength light in schools may be a simple, effective,
non-pharmacological treatment for students to help
them increase sleep duration and, perhaps, scholastic
performance. Daylight in a school like that provided in
Smith Middle School appears to be an ideal source to
accomplish this goal because it can deliver the proper
quantity and spectrum as well as the proper timing
and duration of light exposure. Electric lighting could
also serve this purpose, but current electric lighting is
manufactured, designed and specified to meet visual
requirements. Electric lighting could have an advantage
over daylight for the purpose of circadian entrainment,
because electric lighting can be precisely controlled, not
only during the day, but during the night when expo-
sure to light emulating daylight would be counterpro-
ductive for entrainment. Indeed, electric lighting can
provide a complete 24-hour light exposure pattern to
help ensure entrainment, but these deliberations repre-
sent an entirely new framework for architectural light-
ing design and practice (Figueiro 2008).
ACKNOWLEDGMENTS
The authors would like to acknowledge the U.S.
Green Building Council (USGBC) for sponsoring this
research. The Trans-NIH Genes, Environment and
Health Initiative Grant U01 DA023822 also provided
support this project. The authors would also like to
thank Dr. Mary Carskadon and Dr. Stephanie Crowley
of Brown University and Bradley Hospital for helping
with the DLMO calculations. A. Bierman, D. Guyon, J.
Brons, B. Plitnick, R. Leslie, and J. Taylor of the Lighting
Research Center, Rensselaer Polytechnic Institute, are
thanked for their assistance with this project and manu-
script. We would also like to thank the staff, teachers,
and principal of Smith Middle School for making this
project possible. And, finally, we would like to thank the
students and parents who participated in this research
project.
REFERENCES
1 Brainard G, Hanifin J, Greeson J, Byrne B, Glickman G, Gerner E, et
al. (2001). Action spectrum for melatonin regulation in humans:
Evidence for a novel circadian photoreceptor. J Neurosci. 21:
6405–6412.
2 Burgess HJ, Savic N, Sletten T, Roach G, Gilbert SS, Dawson D.
(2003). The relationship between the dim light melatonin onset
and sleep on a regular schedule in young healthy adults. Behav.
Sleep Med. 1: 102–114.
3 Burgess HJ, Fogg LF. (2008). Individual differences in the amount
and timing of salivary melatonin secretion. PLoS One. 3(8):
e3055.
4 Commission International de l’Eclairage (CIE). (1978). Light as
a true visual quantity: Principles of measurement. Commission
Internationale de l’Eclairage, Paris.
5 Crowley SJ, Acebo C, Carskadon MA. (2007). Sleep, circadian
rhythms, and delayed phase in adolescence. Sleep Med. 8:
602–612.
6 Czeisler C, Duffy JF, Shanahan TL, Brown EN, Mitchell JF, Rimmer
DW, et al. (1999). Stability, precision, and near-24-hour period of
the human circadian pacemaker. Science 284: 2177–2181.
7 Figueiro MG. (2008). A proposed 24 hour lighting scheme for
older adults. Light. Res. Technol. 40: 153–160.
8 Jewett M, Rimmer DW, Duffy JF, Klerman EB, Kronauer RE,
Czeisler CA. (1997). Human circadian pacemaker is sensitive to
light throughout subjective day without evidence of transients.
Am. J. Physiol. Regul. Integr. Comp. Physiol. 273: R1800–R1809.
9 Lewy AJ, Werh TA, Goodwin FK, Newsome DA, Markey SP. (1980).
Light suppresses melatonin secretion in humans. Science 210:
1267–1269.
10 Lighting Research Center. (2004). Daylight Dividends Case Study:
Smith Middle School, NC. Rensselaer Polytechnic Institute, Troy,
N.Y., 12 pp.
11 Portaluppi F, Touitou Y, Smolensky MH. (2008). Ethical and
methodological standards for laboratory and medical biological
rhythm research. Chronobiol. Int. 25: 999–1016.
12 Rea M, Figueiro MG, Bullough J. (2002). Circadian photobiology:
An emerging framework for lighting practice and research. Light.
Res. Technol. 34: 177–190.
13 Rea M, Figueiro MG, Bullough J, Bierman A. (2005). A model of
phototransduction by the human circadian system. Brain Res.
Rev. 50: 213–228.

96
Copyright © 2010 Neuroendocrinology Letters ISSN 0172–780X • www.nel.edu
Mariana G. Figueiro and Mark S. Rea
14 Refinetti R. (2006). Circadian physiology. 2nd ed. CRC Taylor &
Francis, Boca Raton, London, New York.
15 Stevens R, Rea M. (2001). Light in the built environment: Poten-
tial role of circadian disruption in endocrine disruption and
breast cancer. Cancer Causes Control 12: 279–287.
16 Thapan K, Arendt J, Skene DJ. (2001). An action spectrum for
melatonin suppression: evidence for a novel non-rod, non-cone
photoreceptor system in humans. J. Physiol. 535(Pt 1): 261–267.
17 Warman VL, Dijk DJ, Warman GR, Arendt J, Skene DJ. (2003)
Phase advancing human circadian rhythms with short wave-
length light. Neurosci Lett. 342(1–2): 37–40.
18 Wolfson AR, Carskadon MA. (1998). Sleep schedules and daytime
functioning in adolescents. Child Dev. 69: 875–887.
19 Wyszecki G, Stiles W. (1982). Color science: Concepts and meth-
ods, quantitative data and formulae. 2nd ed. Reprinted in 2000.
John Wiley and Sons, New York.
20 Zeitzer J, Dijk D, Kronauer R, Brown E, Czeisler C. (2000). Sensitiv-
ity of the human circadian pacemaker to nocturnal light: Mela-
tonin phase resetting and suppression. J. Physiol. 526: 695–702.