Content uploaded by Darrel Drobnich
Author content
All content in this area was uploaded by Darrel Drobnich on Oct 19, 2017
Content may be subject to copyright.
MIND, BRAIN, AND EDUCATION
School Start Time Change: An
In-Depth Examination of
School Districts in the United
States
Judith Owens1, Darrel Drobnich1, Allison Baylor1, and Daniel Lewin1
ABSTRACT— In response to the scientific evidence doc-
umenting both profound developmental changes in sleep
and circadian biology during adolescence and the myriad
of negative health, performance, and safety outcomes risks
associated with chronic sleep loss, at least 70 public school
districts in the United States, representing approximately
1,000 schools, have successfully implemented a delay in high
school start times. However, despite the compelling evi-
dence supporting school start time change as a key strategy
in addressing the epidemic of adolescent sleep loss, many
school districts across the country with early high school
start times have not considered the option to implement
later bell schedules for adolescents. Moreover, while the cur-
rent scientific literature has clearly documented the positive
outcomes associated with delayed high school start times,
these studies contain limited information regarding the pro-
cess by which school districts consider, approve and imple-
ment bell schedule changes. us, this in-depth examina-
tion of those school districts that have been successful in
changing their bell schedules is intended to support the
efforts of other districts in various stages of contemplating
this measure. We utilized a multi-pronged approach (lit-
erature review, case studies, telephone interviews, online
survey) to summarize the experiences of school districts
across the United States in regard to challenges faced, strate-
gies employed, and lessons learned in the hope that this
1Children’s National Medical Center, George Washington School of
Medicine and Health Sciences
Address correspondence to Judith Owens, Professor of Pediatrics,
George Washington University School of Medicine and Health Sci-
ences, Director of Sleep Medicine, Children’s National Medical Cen-
ter, 111 Michigan Avenue NW, Washington DC 20010-2970; e-mail:
owensleep@gmail.com
information will be a useful tool for other school districts
looking to chart a course forward to promote the health,
safety, and academic opportunities of their students.
Many studies have documented that the average adoles-
cent in the United States is chronically sleep-deprived
and pathologically sleepy, placing adolescents at risk for
adverse consequences, including impairments in mood,
affect regulation, attention, memory, behavior control,
executive function, and impulse control. In particular, many
studies have shown an association between decreased sleep
duration and lower academic achievement at the middle
school, high school, and college levels, as well as higher rates
of absenteeism and tardiness, and decreased motivation
to learn (Curcio, Ferrara, & De Gennaro, 2006; Wolfson
& Carskadon, 2003). Other documented health-related
effects of sleep loss in adolescents include increased use
of stimulants (e.g., caffeine, prescription medications) to
counter the effects of chronic sleepiness, which in turn may
increase the risk of substance use later in adolescence and
early adulthood (Gromov & Gromov, 2009). Adolescents
are also at greater risk for drowsy driving-related crashes,
as well as athletic and other injuries, due to insufficient
sleep (Hutchens, Senserrick, Jamieson, Romer, & Winston,
2008). Chronic sleep restriction increases subsequent risk
of both cardiovascular disease and metabolic dysfunction
such as type 2 diabetes (Verhulst et al., 2008). An associa-
tion between short sleep duration and obesity in children
and adolescents has been demonstrated in numerous
cross-sectional and prospective studies, underscoring how
chronic sleep restriction can undermine the health of our
nation’s youth (Cappuccio et al., 2008). While a number
of factors, including biological changes in sleep, lifestyle,
182 © 2014 International Mind, Brain, and Education Society and Wiley Periodicals, Inc. Volume 8—Number 4
Judith Owens et al.
social, extracurricular activities, and academic demands
contribute to decreased sleep in students, the evidence
points to early school start times (SST) (i.e., before 8:00 am)
as a key contributor to sleep loss in high school students
(Carrell, Maghakian, & West, 2011; Hansen, Janssen, Schiff,
Zee, & Dubocovich, 2005; Hinrichs, 2011). Numerous stud-
ies have demonstrated that early start times significantly
impede high school students’ abilities to obtain sufficient
sleep and to maintain levels of alertness commensurate with
optimal daytime functioning, adequate physical and mental
health, and good quality of life (Epstein, Chillag, & Lavie,
1998; Spaulding, Butler, Daigle, Dandrow, & Wolfson, 2005).
e mechanisms through which early school start times
adversely impact sleep and wakefulness and consequently
health, safety, and performance involve both the resultant
inability of adolescents to obtain sufficient sleep on a chronic
basis and disruption of circadian regulation of sleep-wake
rhythms. From a biological perspective, at about the time
of the onset of puberty, adolescents begin to experience a
sleep-wake “phase delay” (later sleep onset and wake times),
as a result of well-documented changes in circadian rhythms.
is is manifested as a shift in the fall-asleep time to at least
two hours later, relative to middle childhood. At the same
time, adolescent sleep needs do not decline significantly from
pre-adolescent levels, and optimal sleep amounts remain
in the range of 8.5 to 9.5 hours per night for most teens
(Carskadon et al., 1980). On a practical level, this means
that the average adolescent cannot fall asleep much before
11 pm and has significant difficulty in waking before 8:00 am
(Carskadon, Acebo, & Jenni, 2004).
A substantial body of research has now demonstrated
that delaying school start times is an effective countermea-
sure to chronic sleep loss and has a wide range of potential
benefits for students in regard to physical and mental health,
safety, and academic achievement. Studies comparing high
schools with start times even just 30 min earlier to those
with later start times demonstrate adverse consequences
associated with earlier start times, such as shorter sleep
duration, increased sleepiness, difficulty concentrating,
behavior problems, and more school absences (Htwe, Cuz-
zone, O’Malley, & O’Malley, 2008; Owens, Belon, & Moss,
2010; Wahlstrom, 2002). Scientific literature has confirmed
that delaying high school start times results in increased
total sleep time, decreased tardiness rates and absenteeism,
improved performance on standardized tests, reduced self-
reported depression, and fewer automobile crashes (Danner
& Phillips, 2008; Vorona et al., 2011; Wahlstrom, 2014).
For a comprehensive summary of studies examining out-
comes associated with later high school and middle school
start times, as well as the impact of chronic sleep loss on
adolescents, see Appendix 1 or Owens et al. (2014a, 2014b).
However, it is an important but under appreciated fact
thatearlyhighschoolstarttimesarearelativelyrecent
phenomenon that evolved as a result of many factors which
had little to do with academics or the health and well-being
of students. e overwhelming majority of modern-day bell
schedules in U.S. public high schools are historically based
on such “adult” considerations as school budgets, trans-
portation logistics, parent work schedules, athletics, staff
commute times, and community use of fields and facili-
ties. By and large, districts have not taken into considera-
tion the evolving scientific literature on biologically based
changes in sleep patterns and circadian rhythms associated
with puberty and the evidence linking early school start
times with detriments in the health, safety, and well-being
of students. Historical and media sources suggest that school
districts in United States began advancing school start times,
especially at the high school level, first in the late 1950s
and 1960s and then increasingly so during the 1970s. e
move to earlier start times was likely in reaction to a num-
ber of increasing pressures (e.g., fiscal, political, sociological)
faced by school districts to cut costs, to close neighborhood
schools in favor of larger “feeder” schools, and basically to
“do more with less.” An historical timeline of public school
bell schedule changes, contributing factors and the relation-
ship to scientific advances in our understanding of sleep and
circadian biology is included in Appendix 2.
However, it should be noted that there are many school
districts in United States that always have maintained
healthy start times for their high school students. For
example, Loudon County, Virginia has had the same bell
schedule since 1954, with high schools starting at 9:00 am,
middle schools at 8:30 am and elementary schools at
7:50 am. Similarly, some large Texas districts, such as Dallas
and Austin, have started their high schools at 9:00 am or later
since the early 1990s. According to the U.S. Department
of Education’s National Center for Education Statistics,
a majority (60%) of the 19,000 public high schools in the
United States currently start at 8 am or later, with 45%
starting between 8 and 8:30 a.m., and 15% starting 8:30 a .m.
or later (Harpaz, 2013; National Center for Education
Statistics, 2014).
Currently, there is no comprehensive national repository
of information regarding start time change. us, a pre-
cise tally of public high schools in United States that have
prioritized and implemented a delay in school start times
nationwide is not available. To the best of our knowledge,
approximately 1,000 schools in some 70 school districts
have taken this step, yet there is little available information
regarding the process by which these school districts have
weighed the scientific evidence, maneuvered political and
financial challenges, and developed strategies to imple-
ment school start time change. is process can, in fact,
be extremely challenging, as bell changes impact not only
the obvious stakeholders in the community (e.g., parents,
students, teachers, school personnel) but also those citizens
Volume 8—Number 4 183
School Start Time Change
who may not have direct involvement in the school system
(e.g., employers of adolescents, community members using
school facilities). us, a summary of the experiences of
communities that have implemented start time change and
an integration of the most common and salient points likely
to best inform other school districts would clearly provide a
source of valuable information, guidance, and support.
e purpose of this review, therefore, was twofold: first,
to summarize what is known about the extent and nature
of high school start time change in the United States and
second, to integrate these findings into a “blueprint” for
other school districts contemplating this measure. To pro-
vide an up-to-date and comprehensive review, we used a
multi-pronged approached, including a literature and media
review, a compilation of existing databases, a national survey,
and in-depth interviews of personnel from three distinctive
districts. After reviewing all of the available information on
the process of school start time change from a wide variety of
sources as described below, we developed a summary of the
most common and salient points likely to best inform other
school districts.
METHODOLOGY
e data collection process involved in generating this anal-
ysis included three basic components: (1) the identifica-
tion and review of relevant scientific literature and existing
information from a number of diverse but complementary
sources, including online and print media articles, individual
reports from school districts that have successfully changed
start times, and case studies from other organizations (i.e.,
National Sleep Foundation, Start School Later Inc., School
Start Time.org), as well as personal notes of discussions with
sleep experts, parents, and district officials conducted over
the years; (2) conducting in-depth telephone interviews with
personnel and stakeholders in three school districts (Wilton,
CT; Arlington, VA; Milwaukee, WI) to collect qualitative
data on challenges, opportunities, and lessons learned in
more detail; and (3) administration of a brief web-based
national survey to a select number of school districts that
have successfully changed their school start times to obtain
more specific information on methods and the process used
to change schedules and gain community support.
e outcomes of this review process are presented in
the Results section as follows: (a) summary grid of selected
school districts (see inclusion criteria below) which provides
information on the process and logistics of delaying start
times along with demographic information, change strate-
gies, and comments (Table 1); (b) a summary of in-depth
school case studies that represent the scope, process, and
outcomes of individual high schools and school districts
across the country that have successfully implemented a
delay in high school start times (Appendix 3); and (c)
National School Start Times Survey results. For non-U.S.
readers, a glossary of terminology used is also included
(Table 1). is is followed by a discussion integrating the
available information and summarizing lessons learned.
To provide the best possible overview of the process and
experience of school start times change, the public school
districts listed in Table 2 met the following inclusion criteria:
(1) start time changes were successfully implemented and
maintained; (2) there were sufficient data available for the
district regarding the key details of the start time change pro-
cess (e.g., year of change, original and changed start times);
(3) they had the widest range possible of initial and final
school start times; (4) they represented a broad variety in
regards to size of the district, demographics (e.g., socio-
economic status, racial/ethnic composition), geographic dis-
tribution across the United States, and location (e.g., inner
city, suburban, rural); and (5) they had a range of motivations
for (e.g., budgetary, sleep health) and employed a variety of
strategies (e.g., transportation, curriculum, class scheduling,
community engagement) to implement change.
Importantly, while a number of districts included in
Table 2 reported anecdotal or informal outcomes following
start time change, only a small minority of districts (refer-
ences included below) conducted a systematic data-based
analysis of outcome and actually published the results
in scientific or educational peer-review journals, high-
lighting the critical need for more empirically supported
data. Finally, there have been several published studies
of school start time change outcomes in individual inde-
pendent/private schools; these are included at the end
of Table 2.
RESULTS
Selected School District Data
While extensive details regarding the “mechanics” of school
start time change are provided above in Table 1, a number of
summary points should be considered:
•e majority of school districts cited “sleep science”
and/or the evidence supporting the negative impact on
health and performance as key motivators in the decision
to delay start times.
•While many (nearly half) of the school districts listed are
small (<10,000 students), the considerable variability in
size and demographics (urban, suburban, rural) suggest
that a wide range of school district types have success-
fully implemented start time change.
•While the majority of districts moved their start times
to between 8:30 am and 9:00 am later, there were a num-
ber of schools that remained at a start time before
184 Volume 8—Number 4
Judith Owens et al.
Tabl e 1
Glossary of Terms
Bell Schedules Schedule of class periods
Block Schedule Class schedules that change across days and/or weeks
Bus Depot Central location where students are dropped off to catch a school bus
Carpool Multiple students are driven in one car
Civil Twilight Definition of morning and evening twilight onset time that changes over the course of the year. Typically, transportation
schedules are set to avoid having elementary schools students waiting for the bus or coming home in the dark
Elementary School (ES) Grades 1–6
Flexible (flex time) Student/parent choice of start time and class schedule
Flip Strategy Start time change strategy that substitutes the bell schedules for one tier with another
High School (HS) Grades 9–12 or 8–12 (can also be referred to as secondary school)
Magnet School Specialty schools typically for high achieving students
Middle School (MS) Typically Grades 5–8 or 7–8 (also referred to as “Junior High School”)
Parent Teacher Association/Organization
(PTA/PTO)
Elected or appointed group of parents and teachers who serve in advisory capacity to principals, school boards, and
district administration
Phased-in Strategy Start time change which is implemented gradually over two or more years; typically starting Year 1 in selected schools
Public Transportation State- or county-funded transportation system (not funded by school district)
Rolling Window Busing Variable start and school bus delivery times across a tier or tiers (i. e., all schools in the same tier do not start at the same
time)
School Board Governing and advisory body, typically of elected community officials who review budgets and make decisions regarding
educational policy
School District A unified group of schools that serve a county or defined region and has its own administration and governing body
School Principal Head of an individual school
School Start Time-Change (SST-C) District-wide policy regarding change of bell schedules
Secondary School (SS) Grades 7–12; combined HS and MS or synonym for either
Slide Strategy Start time strategy which delays or advances bell schedules by the same amount of time for each tier
Stakeholder Individual or groups of individuals who will be impacted by policy change
Study Hall Non-class period when student may study/complete homework
Superintendent Chief executive of a school district
Tier Groupings of school types, i.e., HS tier, MS, tier, etc.
Tiered Busing Transportation schedule in which individual buses make multiple runs in the morning and afternoon to deliver students
to each tier
Traffic Patterns School bus routes (often multiple bus stops every two blocks that slow non-school traffic and parent delivery of students)
causing congestion around schools
Transportation Fees While most districts are mandated to provide free transportation for regular bus service, sometimes fees for special
services are charged
Zero Period Non-class or class period offered before the start of the school day
Volume 8—Number 4 185
School Start Time Change
Tabl e 2
Selected School Districts that Changed Start Times
School
District
Yea r
of SST
Change
Original
Bell Times
Change
Bell Times
Absolute
SST Change
#Students
#Schools
Change
Strategies
Additional
Strategies
Outcomes/
Additional
Comments Reference
Fayette County,
KY
1998 HS: 7:30–2:20
MS: 8:00–2:50
ES: Abt. 8:30
HS: 8:25–3:15
MS: 9:05–3:55
ES: 7:45–2:35
HS: +55
MS:+55
ES:–45
40,000
66
HS/ES flip. Parents went school
board after earlier
HS times were first
proposed.
Superintendent was
looking for way to
boost attendance.
Pre and post studies
conducted. Plan was
announced 10
months ahead of
change.
Pre and post study 1 year
after showed
improved sleep and
reduced auto crashes,
increased sleep across
all grades. District
reported better
attendance and a
decrease in tardiness
in 1999.
Danner &
Phillips, 2008
Edina, MN 1996 HS: 7:25–2:10
MS: n/a
ES: n/a
HS: 8:25–3:10
MS: 7:40–2:38
ES: 8:30/9:15–3:
05/3:50
HS:+60
MS:—
ES:—
8,300
9
HS shift. Superintendent led
based on sleep
science and MN
Medical Association
actions.
One of the first districts
to change; outcomes
extensively
documented. One year
after, 92% of parents
preferred the change.
Decline in tardiness
and absenteeism
reported.
Wahlstrom, 2002
Minneapolis, MN 1997 HS: 7:15–1:45
MS: 7:05–1:35
ES: 9:40–4:10
HS: 8:10/8:35–3:00
MS: 9:40–4:10
ES: 7:30/8:05–2:
00/2:35
HS:+55/+85
MS:+155
ES:–130/–95
36,370
75
HS/MS flip. School board led to
reduce
transportation costs.
Sleep health of
students was
secondary
motivation.
Most extensively
documented
outcomes of all school
districts delaying start
times. Students
reported more sleep,
less daytime
sleepiness, fewer
depression symptoms.
Lower tardiness rates
also reported.
Wahlstrom, 2002
North
Clackamas, OR
1999 HS: 7:30–2:20
MS: n/a
ES: n/a
HS: 8:45–3:20
MS: 9:30–4:05
ES: 8:20–2:45
HS:+75
MS: n/a
ES: n/a
17,439
31
HS shift. HS principals pushed
change for a decade.
Full-year study
proceeded change.
Improved attendance
and GPA in 1st period.
Large community
acceptance.
186 Volume 8—Number 4
Judith Owens et al.
Tabl e 2
Continued
School
District
Yea r
of SST
Change
Original
Bell Times
Change
Bell Times
Absolute
SST Change
#Students
#Schools
Change
Strategies
Additional
Strategies
Outcomes/
Additional
Comments Reference
Bonneville Joint,
ID
2000 HS: 7:45–2:39
MS: n/a
ES: n/a
HS: 8:45–3:39
MS: 7:15–1:30
8:45–3:34
ES: 8:05–2:31
8:25–2:51
HS:+60
MS: n/a
ES: n/a
11,200
21
HS shift. Superintendent led
based on sleep
science.
Study showed absences
dropped 15% and
tardiness 22%.
Students got 44 min
more sleep on average.
Brevard, FL 2000 HS: 7:30–2:15
MS: 8:50–3:35
ES: 9:50–4:20
HS 8:30–3:15
MS: 9:15–4:00
ES: 8:00–2:30
HS:+60
MS:+25
ES:–115
96,000
137
HS/MS shift.
ES moves
earlier.
MS activities changed
to before school with
own transportation
required.
District reported large
reduction in tardiness
and absences.
Arlington, VA 2001 HS: 7:30–2:15
MS: 8:10–2:45
ES: 8:30–3:10
HS: 8:19–3:01
MS: 7:50–2:24
ES: 8:00–2:40
9:00–3:40
HS:+49
MS:–20
ES:–30
19,000
38
HS/MS flip. School board and
superintendent led.
Formed a large t ask
force with working
groups. Formed
district team for
implementation.
4-tier bus system.
Teachers and parents
reported students
were more alert and
focused. Teachers
raised concerns, but
none left the district.
Brunswick
County, ME
2001 HS: 7:25–1:50
MS: 7:40–2:10
ES: 8:40–2:55
HS: 7:45–2:10
MS: 8:08–2:30
ES: 9:00–3:30
HS:+20
MS:+28
ES:+20
2,645
5
MS/HS
30 min
shift.
ES 15 min
shift
Superintendent and
school board
influenced by
neighboring
Topsham district
and sleep science.
Conducted
district-wide student
survey.
Decision based on sleep
health & community
and student body
support.
Mahtomedi, MN 2002 HS: 7:30–2:10
MS: n/a
ES: n/a
HS: 8:00–2:30
MS: 8:00–2:30
ES: 9:10–3:30
HS:+30
MS: n/a
ES: n/a
3,305
4
HS/ES flip. Encouraged use of
personal
transportation.
4-period day before
and after change.
Students agreed to
shorten “passing”
time between
classes.
District saw improved
attendance, test
scores, and grades.
Decreased costs and
tardiness rates. 65%
decrease in auto
crashes.
Wahlstrom, 2014
Volume 8—Number 4 187
School Start Time Change
Tabl e 2
Continued
School
District
Yea r
of SST
Change
Original
Bell Times
Change
Bell Times
Absolute
SST Change
#Students
#Schools
Change
Strategies
Additional
Strategies
Outcomes/
Additional
Comments Reference
Wilton, CT 2003 HS: 7:35–2:10
MS: 7:35–2:10
ES: 8:15–2:45
HS: 8:15–2:50
MS: 8:15–2:50
ES: 7:40–2:10
HS:+40
MS:+40
ES:–35
4,300
5
HS (6–12) and
ES Flip
Inspired by state
senator, a
community group
formed a task force
to study the issue
and conduct
stakeholder
outreach.
Conducted survey
of students after
change.
Students reported to
be more alert and
better behaved.
More participation
in HS athletics and
ES after school
activities. HS
students reported
high satisfaction
and 35 min more
sleep.
Needham, MA 2004 HS: 7:40–2:20
MS: 7:40–1:55
ES: 8:20/9:00–2:
30/3:00
HS: 8:00–2:35
MS: 7:50–2:10
ES: 8:35–2:35
HS:+20
MS:+10
ES:+15/–25
5,523
8
HS shift. Superintendent
formed advisory
committee to
study all issues.
Change based on
concerns about
sleep health.
Jessamine
County
Schools, KY
Prior to 2005 HS: 7:30–2:15
MS: 7:40–2:25
ES: 8:30–3:15
HS: 8:40–3:25
MS: 8:50–3:35
ES: 8:00–2:45
HS:+70
MS:+70
ES:–30
7,000
11
MS/HS shift. ES
30 min
earlier.
Superintendent led.
Community and
student
engagement was
key.
Change based on
sleep health.
District reported
lower tardiness
and increased
attendance.
Denver, CO 2005 HS: 7:30–2:30
MS:7:25–2:55
ES: 8:15–3:10
9:00–3:55
HS: 7:30–4:30
MS: 7:25–2:55
ES: 8:15–3:10
9:00–3:55
HS:—
MS:—
ES:—
84,424
176
Added 2 hours
to HS day for
flexibility.
Students pick
their own
start and end
times.
District provides
public
transportation
passes to eligible
students & allows
for flexible
schedules.
Superintendent used
change to allow
students to choose
their schedule.
Saved $750,000
and 60 buses by
encouraging the
use of public
transportation.
Holyoke, MA 2005 HS: 7:15–1:52
MS: 7:15–1:52
ES: 9:05–3:05
HS: 8:15–2:52
MS: 8:15–2:52
ES: 9:05–3:05
HS:+60
MS:+60
ES:—
5,573
14
Shift HS & MS School board led
based on sleep
science.
School Board voted
with hopes to
improve sleep and
tardiness.
Topsham, ME 2005 HS: 7:30–2:00
MS: 7:20–2:00
ES: n/a
HS: 7:50–2:21
MS: 7:40–2:10
ES: n/a
HS:+20
MS:+20
ES: n/a
2,739
7
MS & HS shift HS and MS shared
buses.
Transportation cost
savings by having
HS and MS
students share
buses.
188 Volume 8—Number 4
Judith Owens et al.
Tabl e 2
Continued
School
District
Yea r
of SST
Change
Original
Bell Times
Change
Bell Times
Absolute
SST Change
#Students
#Schools
Change
Strategies
Additional
Strategies
Outcomes/
Additional
Comments Reference
Santa Rosa
County, FL
2006 HS: 8:00–2:45
MS: 8:20–2:55
ES: 7:30–1:30
HS: 9:15–3:15
MS: 8:30–2:30
ES: 7:30–1:30
HS:+75
MS:+10
ES:—
26,144
36
HS shift. Used 3-tier bus system,
MS and ED rolling
windows.
Increased graduation
rates, decreased
delinquency and
lowered bus costs.
Ithaca City, NY 2006 HS: 8:00–2:37
MS: 8:00–2:21
ES: 7:55–1:55
9:00–3:00
HS: 8:55–3:32
MS: 9:10–3:25
ES: 8:00–2:00
HS:+55
MS:+70
ES:+5
5,273
11
HS/MS shift. Superintendent led
because buses were
not arriving on time
and students
were late. Sleep
expert’s advocacy
over a few years
helped bring the
change. Discussions
held with employers,
athletic clubs and
public at board
meetings.
Change primarily
made to reduce
bus costs. District
realized about
$400–600 K.
Bentonville, AK 2007 HS: 7:45–2:45
MS: 8:00–3:00
ES: 8:00–3:00
HS: 8:45–3:45
MS: 7:40–2:40
ES: 7:30–2:30
HS:+60
MS:–20
ES:–30
11,100
14
Flip HS with
MS and ES
HS uses A/B block
schedule.
Change due to
address sleep
health and growing
traffic issues. Cost
savings.
Wes t H art f ord ,
CT
2007 HS: 7:30–2:15
MS: n/a
ES: n/a
HS: 7:30/8:15–2:15
MS: 8:00–2:50
ES: 8:35–3:20
HS:+45
MS: n/a
ES: n/a
10,222
16
Flex time for
HS.
Made first hour a study
hall and allowed
students to use “flex”
time to start 2nd
period.
District reported
positive impacts
on emotional
health, stress and
academic
performance.
South
Was hin g to n
County, MN
2009 HS: 7:35–2:05
MS: 7:30–2:00
ES: n/a
HS: 8:35–3:05
MS: 7:55–2:25
ES: 8:10–2:40
9:20–3:50
HS:+60
MS:+25
ES: n/a
17,418
26
HS/MS shifted
later.
Superintendent formed
task force to study to
optimize learning
and sleep. Initiated a
4×4 block schedule.
Part of 2014 study.
Grades in 1st and
3rdperiodclasses
rose by as much as
afullpoint.Scores
on standardized
tests improved.
58% of HS students
slept 8 hours or
more. Auto
crashesfellby6%.
Wahlstrom,
2014
Volume 8—Number 4 189
School Start Time Change
Tabl e 2
Continued
School
District
Yea r
of SST
Change
Original
Bell Times
Change
Bell Times
Absolute
SST Change
#Students
#Schools
Change
Strategies
Additional
Strategies
Outcomes/
Additional
Comments Reference
Hudson City,
OH
2010 HS: 7:30
MS: 7:20
ES: 8:30–9:15
HS: 8:00–3:00
MS: 7:56–2:50
ES: 8:55–3:40
HS:+30
MS:+36
ES:–25
4,941
6
All shifted later. Superintendent and
school board formed
advisory committee
to review health,
transportation,
outreach, and other
school districts.
Subcommittees held
separate public
meetings and
conducted surveys.
Phased in over three
years.
e whole process
took about two
years.
Superintendent
and school board
worked together to
study the issues
and gain public
support.
North Andover,
MA
2011 HS: 7:15–1:57
MS: 7:20–2:05
ES: 8:15–2:50
HS: 7:40–2:15
MS: 7:55–2:20
ES: 8:35–3:00
HS:+25
MS:+35
ES:+20
4,502
7
All shifted later. Superintendent
formed advisory
committee to collect
research, speak to
sleep experts and
conduct stakeholder
meetings.
Did not shift as much
as recommended.
Reported
outcomes included
improved grades
and attendance
and decreased
tardiness and
disciplinary action.
River Falls, WI 2011 HS: 7:25–2:35
MS: 7:30–2:35
ES: 8:35–3:40
HS: 7:45–2:50
MS: 7:45–2:50
ES: 8:45–3:50
HS:+20
MS:+15
ES:+10
3,109
8
15-min shift for
all.
Superintendent
formed committee.
Surveyed parents,
staff and others.
Added change goal
to strategic plan.
Change based on
sleep science. Bell
times did not
change as much as
leadership
recommended.
Moore County,
NC
2012 HS: 8:00–3:00
MS: n/a
ES: 8:00
HS: 9:00–4:00
MS: 8:00–3:00
ES: 7:45/7:30
HS:+60
MS: n/a
ES:–15/–30
12,491
23
HS shift. Implemented 2-tier
bus system.
Saved $700,000 in
transportation
costs.
Pulaski County,
AK
2012 HS: 7:30–2:40
MS: 7:30–2:40
ES: 7:20–2:20
HS: 8:35–3:45
MS: 8:20–8:40
ES: 7:20–2:20
7:50–2:35
HS:+65
MS:+50
ES:—/+30
17,501
38
HS & MS shift;
ES 1 tier to 2
tiers.
Announced change in
January to allow
parents to make
changes.
State’s second-largest
district. Change
done to improve
sleep health and
savings on buses.
190 Volume 8—Number 4
Judith Owens et al.
Tabl e 2
Continued
School
District
Yea r
of SST
Change
Original
Bell Times
Change
Bell Times
Absolute
SST Change
#Students
#Schools
Change
Strategies
Additional
Strategies
Outcomes/
Additional
Comments Reference
Milford County,
DE
2012 HS: n/a
MS: n/a
ES: n/a
HS: 8:16–3:20
MS: 8:00–3:00
ES: 7:45–2:15
HS: n/a
MS: n/a
ES: n/a
4,195
6
HS and ES flip. Superintendent led. HS
startat9:35onWed.
Block scheduling.
Decision based on
sleep health,
improvements in
learning and bus
cost savings.
Hattiesburg,
MS
2013 HS: 7:20–2:45
MS: 8:30–3:50
ES: n/a
HS: 8:30–3:50
MS: 7:30–2:45
ES: 8:00–3:00
HS:+70
MS:–60
ES: n/a
4,528
10
MS/HS flip Superintendent
implemented in the
last few days of the
previous school year
as an experiment.
Changed based on
sleep science.
Bedford
County, VA
2013 HS: 8:30–3:00
MS: 8:30–3:00
ES: 8:30–3:00
HS: 8:55–3:35
MS: 8:55–3:35
ES: 7:55–2:35
HS:+25
MS:+25
ES:–35
10,000
21
HS/MS flip
with ES.
Added tier to bus
system and students
shared buses. Added
online early classes
and early dismissal
for athletes.
Changes led to cost
savings.
Academy
District 20,
CO
2013 HS: 7:05–2:10
MS: 7:25–2:25
ES: n/a
HS: 7:45–2:45
MS: 7:45–2:45
ES: 8:30–3:30
ES: 8:45–3:45
HS:+40
MS:+20
ES: n/a
22,460
35
HS/MSshift. EStostartnolater
than 8:45. Reduced
tiers from 4 to 3.
Rolling MS Window.
Decision based on
sleep health and
transportation
efficiencies.
Beaufort
County, SC
2014 HS: 7:45–2:30
MS: n/a
ES: n/a
HS: 8:35–3:25
MS: n/a
ES: n/a
HS:+50
MS: n/a
ES: n/a
20,000
36
Pilot 1 HS first. Principal of pilot HS
chose to implement
based on science
provided by
superintendent.
Students’ feedback
was solicited.
Evaluation is
expected following
the 1-year pilot
period.
Albany Unified,
CA
2014 HS: 7:40–3:10
MS: 8:00–3:05
ES: 8:30–3:05
HS: 8:00–3:30
MS: 8:00–3:05
ES: 8:30–3:05
HS:+20
MS:—
ES:—
3,807
6
20-min shift as
pilot HS/MS.
Evaluate and consider
future changes.
Elimination of MS
7:00 zero hour
proposed.
HS principal very
influential in
promoting sleep
science.
Long Beach
Unified, C A
2014 HS: 7:50–2:40
MS: 8:00–2:40
ES: n/a
HS: 8:50–3:40
MS: 9:00–3:40
ES: 8:00–2:10;
9:00–3:10
HS:+60
MS:+60
ES: n/a
81,000
84
Pilot in 1 HS
and changed
1
∕
2of all MS.
ES have 40 min
preparation period
on 1 day during the
week.
ird-largest district
in state. Primary
impetus for change
was cost savings.
1
∕
2of MS already
startat9:00.
Volume 8—Number 4 191
School Start Time Change
Tabl e 2
Continued
Independent/
Private
School Location
Yea r o f
SST Change
Original
Bell Times
Change Bell
Times Absolute SST # of Students
Change
Strategies
Additional
Strategies
Outcomes/
Additional
Comments Reference
RI; boarding
school with
10% day
students
2008–2009 8:00 8.30 +30 min 201 Initiative led by
headmaster;
initial protest
from faculty;
invited sleep
experts to
present rationale
and conduct
outcome study
Changed SST in
winter term with
plan to go back to
original SST in
spring term
All outcomes
measured (e.g.,
self-reported sleep
duration, sleepiness,
depression;
attendance,
tardiness, health
center visits)
significantly
improved. Mean
bedtimes shifted
20 min earlier;high
level teacher/student
satisfaction—did not
return to earlier SST
Owens et al.,
2010
New England;
boarding
school with
20% day
students
2010–2011 8:00 8:25 +25 min 197 (all
boarding
students)
Initiative led by
headmaster;
invited sleep
experts to
present rationale
and conduct
outcome study
Implemented SST
change winter
term only
Averageincreasesleep
duration post SST
change 29 min;
students
getting ≥8 hours
sleep increased from
18 to
44%;self-reported
sleepiness,
depression, caffeine
use decreased; no
change
extracurricular
participation Results
suggest outcomes
return to baseline
when original SST
resumed; currently
at later SST
Boergers,
Gable, &
Owens, 2014
n/a =information not available.
192 Volume 8—Number 4
Judith Owens et al.
8:00 am; very few districts moved start times to after
9:00 am.
•e most common strategy was a “flip,” typically revers-
ing start times of elementary and high schools; and the
next most common strategy was a “shift” of high schools
and in some districts middle and/or elementary schools
to later start times. Flexible schedules were implemented
by a small percentage.
•Many school districts actually realized cost savings due
to increased utilization of their current bus fleet in mak-
ing multiple “runs.”
•Strategies revolving around the length of the school day
(e.g., extending the length of day, making it uniform
across tiers (ES, MS, HS)) enabled increased transporta-
tion flexibility in some districts by allowing for “stag-
gered” bus runs and thus fewer buses.
•Most districts implemented bell time changes in all
schools simultaneously, although a few utilized a “pilot”
or “phased-in” approach (Hudson City OH, Beaufort SC,
and the two independent schools).
Selected Case Studies of Start Time Change Successes
e number of published outcome studies on the impact
of delaying high school start times is still somewhat lim-
ited, and the motivations and implementation strategies of
schools and school districts deciding to adopt bell schedule
changes are not well characterized. Publicly available mate-
rials, such as those collated and presented above, examine
the challenges involved and the solutions generated but may
fail to do so in adequate detail and may not fully explore
importantnuancesoftheprocess.us,itisusefultosup-
plement the available descriptive data with more anecdotal
information from communities that have been involved in
changing start times. A more in-depth understanding of
theprocessofschoolstarttimechangeinotherdistricts
can yield important information that may guide other
school districts, suggest strategies that might not have been
considered and help to avoid both repeating mistakes and
“re-inventing the wheel.” We developed a number brief
case studies to serve as examples of the scope, process, and
outcomes of individual high schools and school districts
across the country that have successfully implemented a
delay in high school start times. e sheer variety in the
size of the student populations, geographic considerations,
initial and final bell schedules, and strategies employed are
a bit daunting; nonetheless, valuable lessons can be gleaned,
which can inform other school districts in various stages
of change contemplation. e case studies presented in
Appendix 3 were chosen based on the availability of detailed
information, instructional value, and the various types of
school districts they represent. Most involved conducting
interviews with school personnel as well as obtaining docu-
ments from other school districts, local press accounts, and
online sources provided by advocacy organizations. ey
are listed in alphabetical order by state.
National School Start Time Survey
Our approach to compiling successful strategies to delaying
high school start times also involved a survey of public
school districts across the country that had changed their
bell schedules within the past decade. We then developed
an online National School Start Time Survey and contacted
a number of identified school districts that were successful
in changing their start times. e survey consisted of three
parts: Part One—background information on the districts;
Part Two—in-depth information about their experience with
school start time change strategies and implementation; and
Part ree—actual and perceived costs and savings (financial
and otherwise) and benefits associated with the changes
(Appendix 4). is survey was intended to solicit input on a
variety of experiences with later bell schedule changes and to
supplement Table 1 data and the case studies in this report;
thus the results should neither be interpreted in isolation,
nor viewed as a “scientific” survey.
e survey was designed to be completed by a single indi-
vidual/superintendent with input as needed from ancillary
school personnel. Over the summer of 2013, our team iden-
tified 70 school districts that had previously delayed high
school start times in the past 10 years, using existing online
compilations, scientific literature, documents previously col-
lected by the team, media clips, and previous FCPS and
other school reports. In the fall of 2013, the online sur-
vey was emailed to district superintendents in identified
districts.
Results
Part One: General Information. ere were 24 initial survey
responses; however, many surveys were excluded from the
final results due to incomplete crucial data that precluded
further analysis (i.e., only surveys that reported at minimum
the pre- and post-start times for high schools were included).
A total of eight completed surveys were included in the
final analyses. All participating districts for this report were
de-identified. ere was a fair amount of variability in terms
of the districts’ sizes (Table 3) and in their final bell schedules
(Table 4). e districts’ average delay in high schools’ bell
schedules was 45 min with a range of 20–75 min. About 75%
(6/8) of the districts moved high school start times to after
8:30 am; only one district (FL) set the new start time to after
9:00 am.
Four districts moved middle school start times earlier
(range 10–70 min), while the other districts moved mid-
dle school bell times later within a much narrower range
(20–30 min). With two exceptions (VA and FL), all new mid-
dle school bell times were at 8:00 am or earlier. With regard
Volume 8—Number 4 193
School Start Time Change
Tabl e 3
Participating Districts and Population Indicators
District Population Sizes
High School
Start Time Change
(Old SST→New SST)
Middle School Start
Time Chan ge
(Old SST→New SST)
Elementary School
Start Time Change
(Old SST→New SST)
VA District 550 students (1 HS) 8:30→8:55 8:30→8:55 7:55→8:30
550 students (1 MS)
1,200 students (2 ES)
16 buses
CA District 1,200 students (1 HS) 7:40→8:00 8:00 (No change) 8:30 (No change)
900 students (1 MS)
1,800 students (3 ES)
Buses unknown
CT District 300 students (1 HS) 7:35→8:15 8:15→7:35 (Pre-K −2)
9:00→8:55;
(3–5) 8:10→7:40
1,050 students (1 MS)
1,900 students (2 ES)
33 buses
MA District 1,200 students (1 HS) 7:15→7:45 7:50→7:20 8:15→8:25
1,200 students (1 MS)
2,300 students (5 ES)
20 buses
MN District 2,900 students (1 HS) 8:00→8:40 9:15→7:50 8:00→8:40
2,300 schools (2 MS)
4,600 students (6 ES)
50 buses
AK District 3,600 students (1 HS) 8:00→8:55 7:40→8:00 7:30→8:00
4,500 students (5 MS
4,500 students (9 ES)
85 buses
MO District 5,100 students (HS) 7:45→8:55 7:30→8:00 8:20→8:50
9,100 students (6 MS)
9,100 students (19 ES)
190 buses
FL District 7,500 students (7 HS) 8:00→9:15 8:30→8:20 7:30 (No Change)
6,200 students (8 MS)
14,500 students (18 ES)
200 buses
Listed in ascending order by total (ES/MS/HS) number of students in the district.
to elementary schools, one district (CT) moved bell times
earlier (by 5–30 min), two remained the same, and the rest
delayed elementary start times (range 10–40 min). Two dis-
tricts set new start times before 8:00 am; the latest ES start
bell time was 8:50 am (MO).
Part Two: Change Strategy and Implementation (N =6).
assessed districts’ overall bell change strategies (with par-
ticular focus on transportation components) and concerns
associated with these changes. e districts surveyed
reported employing an array of change strategies: three
districts employed a “slide” (moving ES/MS/HS bell times
later), and two used a “modified slide” (moving ES/HS later
with MS earlier). None of the districts responding to the
survey employed a straight “flip” (i.e., exchanging ES and
HS bell times). Several ancillary strategies were also used;
these included a zero period (i.e., a class offered before the
start of the school day) in two districts and student choice
(i.e., flexible scheduling based upon student preference) in
one district. While only two districts (MA and MO) had
tiered busing (i.e., separate delivery of ES, MS, and HS)
prior to bell changes, all schools used tiered busing after
SST change. All districts eventually had middle and high
schools share buses, with three districts making the change
as part of a comprehensive set of bell-related transportation
changes (MA, VA, CA). Additional strategies employed in
other districts included: the consolidation of bus depots
(MO), the continuation of a flat fee for transportation (MA),
and maintaining public transit use (CA). Notably, no dis-
tricts cut transportation funding for magnet schools, and
only one school used a phased-in approach (CA), which
involved changes only for part of the district in Year 1 and
the remainder of the district in Year 2.
Part Two also assessed the top identified challenges asso-
ciated with changing school start times. When weighted to
give more credence to higher-ranked options, the top five
concerns were: (1) traffic flow at school, (2) changes in par-
ents’ work schedules, (3) after-school extracurricular pro-
gram attendance, (4) changes in teachers’ work schedules,
194 Volume 8—Number 4
Judith Owens et al.
Tabl e 4
Costs and Benefits Associated with School Start Time Change
District AK MA FL VA
Benefits
Increased daily attendance No Ye s No No
Reduced tardiness rates No Ye s No No
Improve standardized test scores No N/A Ye s Ye s
Improved grades No Ye s Yes No
Higher graduation rates No N/A Ye s No
Fewer referrals for disciplinary action No Ye s No No
Improved sports team performance No N/A No No
Fewer sports-related injuries No N/A Ye s No
Cost savings for public school system Ye s N/A Ye s Ye s
Fewer student visits to student health centers No N/A Ye s No
Lower rates of depression/suicidal thoughts No N/A No Yes
Lower rates of car accidents No N/A No No
Costs
Financial cost incurred by the school district No No No No
Loss of community support No No No No
Impact on parent work schedules No No Yes Yes
Limitations on student after-school employment No No Yes No
Financial cost incurred by families No No Yes No
Changes in traffic patterns No No No No
Reduction of student involvement in extracurricular activities/athletics No No No No
Negative impact on teacher schedules No No Yes No
Safety concerns for elementary students No No No Ye s
and (5) before-school athletics practices schedules. Of note:
when all athletic-related items (i.e., before/after school ath-
letics and games) were combined, athletics represented
the most significant concern. Finally, Part Two included
a select-all-that-apply question that solicited information
about additional strategies employed; these included sub-
stituting early morning classes with online courses, giving
athletes early dismissal, adding lighting to athletic fields, and
allowing for more flexible academic scheduling.
Part ree: Benefits/Costs (N =4). assessed benefits and
costs associated with bell changes. With more weight given
to higher ranked items, the top five identified benefits were:
(1) increased daily attendance, (2) cost savings, (3) reduced
tardiness rates, (4) improved standardized test scores, and
(5) improved grades (Table 3). Districts varied widely in
terms of perceived benefits, with one district reporting no
benefits other than cost savings. Attendance and academic
measures were the most consistently reported benefits and
all schools listed cost savings as a benefit, likely attributable
to adaptation of an increased number of busing tiers. In
regards to perceived costs, no districts identified loss of
community support, changes in traffic patterns, or a reduc-
tion in student involvement in extracurricular activities as a
consequence of SST change. e negative impact was largely
perceived to be on families (e.g., work schedules, financial
issues).
Survey Conclusions
Giventhesmallsamplesize,theconclusionsthatcanbe
drawn from this survey are very limited but nonetheless
valuable, as they add to our understanding of how and why
school districts choose to delay high school start times and
what benefits/costs they perceive as a result. While it is not
possible to generalize from this limited sample or speculate
on why particular districts did or did not respond to the
survey, a few key points can be made. First, the districts
responding were quite varied in the strategies employed in
changing bell times across all tiers. Moreover, none of the
districts changed to a bell schedule that could be termed
“biologically optimal” from a circadian/sleep perspective
for all three groups (i.e., MS and HS later and ES earlier).
Cost savings, presumably largely as a result of increasing
the number of busing tiers, was the most prominent ben-
efit (and may have been the real “driver” for the change
in some cases). e districts surveyed tended to perceive
academic-related benefits resulting from adopting later start
times for secondary schools.
DISCUSSION: A BLUEPRINT FOR CHANGE
We used a three-pronged approach to summarize the expe-
riences of a variety of school districts that have imple-
mented school start time change in order to identify both
Volume 8—Number 4 195
School Start Time Change
commonly used and novel strategies potentially applicable
to other districts across the country. e sheer variety of
(as well as lack of universally endorsed) successful imple-
mentation strategies employed strongly suggest that there is
no “one size fits all” plan and indicates that school districts
should explore a wide range of bell scenarios, transportation,
curriculum, and other strategies. Additionally, several sec-
ondary strategies merit further consideration; these include
consolidating bus depots, charging a flat-rate transporta-
tion fee, and bundling MS/HS on buses. Furthermore, it
is important to note that most of the concerns raised by
school districts prior to changing start times were not actual-
ized. In addition, many of the districts did not systematically
measure health, academic, or satisfaction outcomes, again
underscoring both the need for additional research on bell
change-related data and highlighting the unique opportunity
to contribute to public health and policy research and to pro-
vide districts considering similar changes a roadmap and a
menu of approaches.
After reviewing all of the available information on the
process of school start time change from a wide vari-
ety of sources as described above, including in-depth case
examples, we compiled a summary of the most common
and salient points to best inform school districts that are
actively contemplating a change in bell schedules. While
we acknowledge that these points are based in part on our
collective interpretation of qualitative data, we also believe
that the ten key messages listed below, categorized accord-
ing to major content themes, represent principles that have a
basic foundation in successful implementation of start time
change and are likely most applicable to the majority of
school districts, no matter their size, or complexity of issues.
e Importance of Leadership
In our sample of school districts, successful change appeared
to be associated with strong leadership on the part of the
superintendent, district staff, and the school board. Specific
areas of leadership include valuing the scientific justifica-
tion for healthy school start times, commitment to work-
ing with key community organizations to address logistical
and financial challenges, and promoting the benefits to stu-
dent health, safety, and academic and athletic performance
(e.g., Jessamine, KY; Bentonville, AR; Edina, MN; Minneapo-
lis, MN; Needham, MA; Brunswick, ME; South Washington
County, MN). Very often, the publicly stated position of the
district superintendent is the key determinant as to whether
or not a school district is successful in changing school start
times. is individual can set the tone for other staff, and can
direct communications, planning, logistics, and community
engagement. e relationship and trust (i.e., political capital)
that the superintendent has established in the community
and with the school board is also extremely important. If
the superintendent and district officials do not communicate
their strong support for the bell time change and do not keep
discussions focused on the health, safety, and academic per-
formance of students, then the process may get bogged down
with special-interest concerns.
e school board’s public support for the superintendent
and for school start time change is also critical. e school
board’s support is especially vital in communicating to the
broader public both the justification (e.g., health and safety
benefits) for changing bell schedules and the message that
any challenges can be addressed and most likely overcome.
Education of the Entire Community
Districts should seek to provide appropriate targeted
education for the entire community (e.g., students, parents,
teachers, school nurses) about justification for healthy
school start times and approaches to optimizing sleep
health. Change agents and stakeholders should have a work-
ing knowledge of the research on adolescent sleep and early
start times in order to effectively communicate the rationale
for changing bell schedules (e.g., Arlington, VA). It also
is important for school district leadership to refute mis-
conceptions (e.g., “if school starts later, teens will just stay
up later and won’t get more sleep”) while also responding
to legitimate concerns of students, parents, and teachers.
Partnering with health experts is one possible strategy
(e.g., Albany, CA; Fayette, KY). It is extremely important to
emphasize the health and safety benefits associated with
providing students the opportunity to get more sleep and
that the potential benefits go far beyond academic improve-
ments. When communicating the short and long-term
consequences of chronic sleep loss (and, by implication,
the potential dangers associated with failing to delay high
school start times), it should be emphasized that these
extend not only beyond the school grounds (e.g., drowsy
driving, depression, obesity) but very well may set stu-
dents up for debilitating (e.g., insomnia) or life-threatening
medical conditions (e.g., cardiovascular consequences such
as hypertension or metabolic dysfunction such as type 2
diabetes) in the future.
Teachers and other school personnel, especially health
and counseling professionals, should be well educated
about adolescent sleep needs and patterns, taught to rec-
ognize the signs of sleep-related difficulties among their
students, and report such symptoms to parents and school
health providers (National Sleep Foundation, 2014; Wolf-
son & Carskadon, 2005). Superintendents, school boards,
and principals should consider integrating sleep-related
education into curricula so students can learn about the
physiology of sleep, the consequences of sleep depriva-
tion, and the importance of sleep to their overall health.
is education can be provided in science, health, and
196 Volume 8—Number 4
Judith Owens et al.
athletic classes. Finally, it is particularly important that
information be provided to support families throughout
the implementation phase in culturally sensitive ways.
is includes translating basic print educational materials
into multiple languages, providing translation services at
community-wide online forums and reaching out to local
press venues that serve minority communities.
Consensus Building Among Stakeholders
It is important to inform and engage all stakeholders early
in the process to understand potential concerns and to
seek potential solutions. is includes community mem-
bers or organizations that use school district fields and
facilities on a regular basis as well as other city or county
agencies that provide programs and services to students
(i.e., libraries, parks and recreation, police, employers). e
district should notify these groups of any changes once
the decision is made in order to allow them time to adjust
their schedules and for planning, and should continue to
engage them throughout implementation in a spirit of
partnership.
Districts should consider the views of school staff and
teachers in decision-making about implementation and
develop policies that provide flexibility for teachers and
other staff to adapt to the changes (e.g., easing periods
to make transfer requests). It is also important to involve
principals from all three levels (elementary, middle, and
high schools) in internal discussions because it is likely that
all schools and students in the district will be impacted to
some degree, whether or not their own bell times change.
Engagementofthestudentsthemselvesisalsocriticalin
garnering support for the change, and often they can be the
most passionate and articulate voices in the community.
For all stakeholders, it should be noted that expressed
concerns are sometimes based on contractual or personnel
issues rather than what is good for the health, safety, and
well-being of students. In these situations, the superinten-
dent, school board and stakeholder group leaders’ public
support for start time change will be critical in overcoming
any staff or community opposition.
Early in the process of considering bell time changes, the
district should bring together key staff representing several
areas (e.g., transportation, curriculum, special or health ser-
vices, athletics) to do their own fact-finding before engag-
ing additional outside consulting groups. is allows internal
staff to identify logistical issues early and begin to develop
potential solutions before opening up the debate to the wider
community.
In regard to process, it may be prudent to engage the lead-
ership of key community groups in face-to-face meetings in
order to build trust, air mutual concerns, and establish an
open dialog prior to engaging the broader community. In
general, smaller working groups focused on unique concerns
and specific tasks may be more efficient and productive than
large groups that include many stakeholders.
Transportation as a Major Logistical and Cost Factor
In most districts, transportation logistics are a key factor in
determining start time schedules and typically represent the
largest cost. In fact, in many districts, transportation is actu-
ally the main driver for seeking changes in an attempt to
lower costs by moving from a delivery system where all stu-
dents are transported at the same time to a multi-tiered bus
schedule where the bell times are staggered (e.g., Academy
District 20, CO; Santa Rosa, CO). Districts that already
employ a multi-tiered delivery schedule may have to use
more creative strategies to find transportation savings. One
strategy that is commonly used to overcome potential trans-
portation costs is what is commonly referred to as “flipping”
secondary/high school and elementary bell schedules. is
may have the added benefit of being more “in sync” with
circadian rhythms in both groups (e.g., younger children typ-
ically fall asleep earlier and wake earlier). For some districts,
new approaches should be considered; these include encour-
aging carpools, providing incentives for using public trans-
portation (e.g., Denver, CO), creating bus depots for special
program and centers, charging parents who have sufficient
financial resources a flat-rate transportation fee for special
activities (e.g., Brevard, FL), and allowing middle school and
high school students to ride on the same buses.
e Role of Athletics and Community Use
of Recreational Facilities
Community members in districts contemplating school start
time changes frequently are concerned about the impact
on after-school programs, and athletics practices and com-
petitions; however, most of these concerns do not actually
materialize or can easily be mitigated by scheduling or pol-
icy changes (e.g., game day early dismissal, more flexible
instruction time and scheduling). We identified no districts
in which athletic programs were canceled or significantly
adversely affected by school start time change. To the con-
trary, a number of districts found that more students par-
ticipated in athletics and that sports programs grew after
high school bell times were delayed (e.g., Edina, MN), and
reported that their teams performed better following the
change (e.g., Wilton, CT; Edina, MN; Seattle, WA). us, it
is important for administration officials, coaches, and stu-
dent athletes not only appreciate the likely lack of negative
impact on athletics of delayed start times, but to also under-
stand the potential repercussions on relevant health (e.g.,
metabolic dysfunction, weight gain), performance and safety
(e.g., increased sport-related injuries) outcomes related to
Volume 8—Number 4 197
School Start Time Change
chronic sleep loss in student athletes (Milewski et al., 2014;
Spiegel, Leproult, & Van Cauter, 1999).
One Size Does Not Fit All
While there are clearly general overarching principles that
can (and should) be applied across diverse communities,
it should also be emphasized that the extent to which
changing school start times impacts a given community
and the relative importance of the different challenges
and benefits (anticipated and experienced) varies widely
across school districts. us, it follows that there is also no
“one-size-fits-all” or singular optimal approach to tackling
this issue. Among the myriad of variables that need to be
considered on a case-by-case basis include average (and
range of) student commute times, number and length of
school bus routes, availability of public transportation,
traffic patterns, community use of school recreational
facilities, the number of students enrolled in free break-
fast programs, and the impact of later dismissal times on
after-school programming both for disadvantaged students
and for high-achieving students seeking additional aca-
demic enrichment opportunities. For example, concerns
about access to personal transportation tend to be more
prevalent in less affluent school districts than in more afflu-
ent ones. Similarly, transportation plans are more complex
and may have more financial implications in larger districts,
especially those in urban or metropolitan centers with high
traffic congestion (e.g., Denver, CO).
Prioritizing Sleep Health Is an Important Corollary
to School Start Time Change
Despite the best of efforts, there is likely to be some vari-
ability in how much individual students within a district
benefit from start time change. While studies definitively
show that students overall obtain more sleep when start
times are delayed, there will be families and students who
choose not to take advantage of the additional sleep oppor-
tunity (Boergers, Gable, & Owens, 2014; Wahlstrom, 2002).
is underlines the importance of providing education about
sleep health behavior and time management to both par-
ents and students in conjunction with schedule changes.
In addition, schools may undermine the benefits of delayed
start times by rescheduling after-school programs and activ-
ities to before school (e.g., early morning sports practices).
Excessive homework, an issue frequently raised by students
and parents, may also diminish students’ abilities to obtain
optimal sleep. Districts should consider using the change
in start times as an opportunity to make other adjust-
ments that are in the best health interests of students and
which complement the benefits associated with increased
sleep.
Adjustments Take Time
It is critical to allow adequate time prior to implement-
ing changes for families and other community members to
become informed and make sufficient plans (e.g., childcare,
transportation, family time). Once finalized, district leader-
ship should communicate the details of the new schedules as
early as possible, along with information on the rationale for
making changes (e.g., Pulaski, AR). District-level organized
and comprehensive communication and outreach efforts are
absolutely key in conveying information in a timely man-
ner to the community and in addressing the misinforma-
tion and misconceptions that often circulate in the advent
of such an important (and often controversial) societal
change.
Anticipation Is Often Worse an the Reality
Similar to concerns regarding the impact of delayed start and
dismissal times on athletic practices and games, many of the
other potential problems typically raised in the community
prior to the change are often not substantiated (e.g., Arling-
ton, VA; Wilton, CT). For example, studies have shown that
participation by students in extracurricular activities does
not decline when start times are delayed (Danner & Phillips,
2008; Owens et al., 2010). Teacher retention (related to their
own childcare and commuting times) is another commonly
expressed concern that may not be realized; in Arlington VA,
for example, this district offered teachers the opportunity to
change school or tiers to aid retention, and the predicted
mass exodus of teachers in the district never occurred. Com-
munities often make adjustments to accommodate changes
in schedule; for instance, employers shift work hours for
working students and parents shift from before-school to
after-school childcare arrangements for elementary school
students. Finally, some problems dissipate over time; for
example, traffic may temporarily worsen when bus routes are
changed, until drivers in the community adjust their com-
muting patterns. To address concerns and ease the impact of
change, school districts have set up hotlines, resource guides ,
and community meetings to assist adjustment for parents,
staff, and the community
Monitoring Outcomes Is Crucial
Districts should monitor the results and outcomes following
the change to later start times, communicate positive results
to the community and seek ways to mitigate or address nega-
tive or unforeseen impacts (e.g., Arlington, VA). Ideally, dis-
tricts should work with county health professionals or local
university or medical centers to design pre- and post-surveys
and other methods to measure the impact of changing school
start times on student health, safety, and academics. School
districts that have conducted outcomes research have been
able to communicate the findings to the community to foster
198 Volume 8—Number 4
Judith Owens et al.
further acceptance of changing school start times. Addition-
ally, they have been able to provide important data for the
growing scientific literature in this area as well as invaluable
resources for other districts contemplating school start time
change.
CONCLUSIONS
Establishing healthy school start times has a clear scien-
tific rationale but can introduce considerable challenges for
communities, including school administrators, families, stu-
dents, and other stakeholders. While the potential bene-
fits to the health, safety and performance of students are
irrefutable, many school districts remain reluctant to “take
the plunge” and commit time, effort, resources, and polit-
ical capital to this effort. It is our hope that providing
this “blueprint for change” with general recommendations
regarding the process involved in changing school start times
will be an impetus for school districts across the country take
this important step.
REFERENCES
Boergers, J., Gable, C. J., & Owens, J. (2014). Later school start time
is associated with improved sleep and daytime functioning in
adolescents. Journal of Developmental and Behavioral Pedi-
atrics,35(1), 11–17. [PubMed: 24336089]
Cappuccio, F., Taggart, F. M., Kandala, N.-B., Currie, A., Ed Peile,
Stranges, S., & Miller, M. A. (2008). Meta-analysis of short
sleep duration and obesity in children and adults. Sleep,31,
619–626. [PubMed: 18517032]
Carrell, S. E., Maghakian, T., & West, J. E. (2011). A’s from Zzzz’s?
e causal effect of school start time on the academic achieve-
ment of adolescents. American Economic Journal: Economic
Policy,3, 62–81.
Carskadon, M. A., Acebo, C., & Jenni, O. G. (2004). Regulation
of adolescent sleep: Implications for behavior. Annals of the
New York Academy of Sciences,1021, 276–291. [PubMed:
15251897]
Carskadon, M. A., Harvey, K., Duke, P., Anders, T. F., Litt, I. F., &
Dement, W. C. (1980). Pubertal changes in daytime sleepiness.
Sleep,25, 453–460. [PubMed: 12224838]
Curcio, G., Ferrara, M., & De Gennaro, L. (2006). Sleep loss,
learning capacity and academic performance. Sleep Medicine
Review,10, 323–337. [PubMed: 16564189]
Danner, F., & Phillips, B. (2008). Adolescent sleep, school start
times, and teen motor vehicle crashes. Journal of Clinical Sleep
Medicine,4, 533–535. [PubMed: 19110880]
Epstein, R., Chillag, N., & Lavie, P. (1998). Starting times of school:
Effects on daytime functioning of fifth-grade children in Israel.
Sleep,21, 250–256. [PubMed: 9595603]
Gromov, I., & Gromov, D. (2009). Sleep and substance use and
abuse in adolescents. Child and Adolescent Pediatric Clinics
of North America,18, 929–946. [PubMed: 19836697]
Hansen, M., Janssen, I., Schiff, A., Zee, P. C., & Dubocovich, M.
L. (2005). e impact of school daily schedule on adolescent
sleep. Pediatrics,115, 1555–1561. [PubMed: 15930216]
Harpaz BJ. (2013). Starting high school later may help sleepy teens.
Associated Press. April 2014. <www.ap.org>
Hinrichs, P. (2011). When the bell tolls: e effects of school start-
ing times on academic achievement. Education Finance and
Policy,6, 1–22.
Htwe, Z. W., Cuzzone, D., O’Malley, M. B., & O’Malley, E. B. (2008).
Sleep patterns of high school students before and after delayed
school start time. Sleep,31(Suppl 1), A74–A75. [PubMed:
24336089]
Hutchens, L., Senserrick, T. M., Jamieson, P. E., Romer, D., & Win-
ston, F. K. (2008). Teen driver crash risk and associations with
smoking and drowsy driving. Accident Analysis and Preven-
tion,40, 869–876. [PubMed: 18460353]
Milewski, M. D., Skaggs, D. L., Bishop, G. A., Pace, J. L., Ibrahim,
D. A., Wren, T. A., & Barzdukas, A. (2014). Chronic lack
of sleep is associated with increased sports injuries in ado-
lescent athletes. Journal of Pediatic Orthopaedics,34(2),
129–133.
National Center for Education Statistics. (2014). Schools and
staffing survey. Retrieved from: https://nces.ed.gov/surveys/
sass/tables/sass1112_201381_s1n.asp
National Sleep Foundation. (2014). Eight major obstacles to delay-
ing school start times. Retrieved from: https://nces.ed.gov/sur
veys/sass/tables/sass1112_201381_s1n.asp and http://sleep
foundation.org/sleep-news/eight-major-obstacles-delaying-
school-start-times
Owens, J., & Adolescent Sleep Working Group and American
Academy of Pediatrics Committee on Adolescence. (2014a).
School start times for adolescents AAP polic y statement. Pedi-
atrics,35,1.
Owens, J. A., & Adolescent Sleep Working Group and American
Academy of Pediatrics Committee on Adolescence. (2014b).
Insufficient sleep in adolescents and young adults: An update
on causes and consequences. Pediatrics,35,1.
Owens, J. A., Belon, K ., & Moss, P. (2010). Impact of delaying school
start time on adolescent sleep, mood, and behavior. Archives of
Pediatrics and Adolescent Medicine,164, 608–614. [PubMed:
20603459]
Spaulding, N., Butler, E., Daigle, A., Dandrow, C., & Wolfson, A.
R. (2005). Sleep habits and daytime sleepiness in students
attending early versus late starting elementary schools. Sleep,
28(Suppl), A78.
Spiegel, K., Leproult, R., & Van Cauter, E. (1999). Impact of sleep
debt on metabolic and endocrine function. Lancet,354(9188),
1435–1439. [PubMed: 10543671]
Verhulst, S. L., Schrauwen, N., Haentjens, D., Rooman, R. P., Van
Gaal, L., De Backer, W. A., & Desager, K. N. (2008). Sleep dura-
tion and metabolic dysregulation in overweight children and
adolescents. Archives of Disease in Childhood,93(1), 89–90.
[PubMed: 18156481]
Vorona, R. D., Szklo-Coxe, M., Wu, A., Dubik, M., Zhao, Y., &
Ware, J. C. (2011). Dissimilar teen crash rates in two neigh-
boring southeastern Virginia cities with different high school
start times. Journal of Clinical Sleep Medicine,7, 145–151.
[PubMed: 21509328]
Wahlstrom, K. (2002). Changing times: Findings from the
first longitudinal study of later high school start times.
Volume 8—Number 4 199
School Start Time Change
National Association of Secondary School Principals Bulletin,
286, 3–21.
Wahlstrom K. (2014). Examining the impact of later high school
start times on the health and academic performance of high
school students: A multi-site study: final report.Centerfor
Applied Research and Educational Improvement. Retrieved
from: http://www.cehd.umn.edu/CAREI/sleepresources.html
Wolfson, A. R., & Carskadon, M. A. (2003). Understanding ado-
lescents’ sleep patterns and school performance: a critical
appraisal. Sleep Medicine Review,7, 491–506. [PubMed:
15018092]
Wolfson, A. R., & Carskadon, M. A. (2005). A survey of factors
affecting high school start times. National Association of Sec-
ondary School Principals Bulletin,89, 47–66.
APPENDIX 1. RECOMMENDED LITERATURE FOR
REVIEW
Boergers, J., Gable, C. J., & Owens, J. (2014). Later school
start time is associated with improved sleep and day-
time functioning in adolescents. Journal of Developmen-
tal and Behavioral Pediatrics,35(1), 11–17. [PubMed:
24336089].
Carrell, S. E., Maghakian, T., & West, J. E. (2011). A’s from
Zzzz’s? e causal effect of school start time on the aca-
demic achievement of adolescents. American Economic
Journal: Economic Policy,3, 62–81.
Danner, F., & Phillips, B. (2008). Adolescent sleep,
school start times, and teen motor vehicle crashes Jour-
nal of Clinical Sleep Medicine,4, 533–535. [PubMed:
19110880].
Dexter, D., Bijwadia, J., Schilling, D., & Applebaugh, G.
(2003). Sleep, sleepiness and school start times: A prelim-
inary study. Wisconsin Medical Journal,102(1), 44–46.
[PubMed: 12679971].
Edwards, F. (2010). Early to rise: e effect of daily start
times on academic performance.WorkingPaper,Uni-
versity of Illinois at Urbana-Champaign. Retrieved from
http://ssrn.com/abstract=1628693.
Epstein, R., Chillag, N., & Lavie, P. (1998). Start-
ing times of school: Effects on daytime functioning
of fifth-grade children in Israel. Sleep,21, 250–256.
[PubMed: 9595603].
Fredriksen, K., Rhodes, J., Reddy, R., & Way, N. (2004).
Sleepless in Chicago: Tracking the effects of adolescent
sleep loss during middle school years. Child Develop-
ment,75, 84–95.
Hansen, M., Janssen, I., Schiff, A., Zee, P. C., & Dubo-
covich, M. L. (2005). e impact of school daily schedule
on adolescents leep. Pediatrics,115, 1555–1561.
[PubMed: 15930216].
Hinrichs, P. (2011). When the bell tolls: e effects of
school starting times on academic achievement. Educa-
tion Finance and Policy,6, 1–22.
Htwe, Z. W., Cuzzone, D., O’Malley, M. B., & O’Malley,
E. B. (2008). Sleep patterns of high school students before
and after delayed school start time. Sleep,31(Suppl),
A74–A75.
Kirby, M., Maggi, S., & D’Angiulli, A. (2011). School start
times and the sleep-wake cycle of adolescents: A review
and critical evaluation of available evidence. Educational
Researcher,40(2), 56–61.
Lamberg, L. (2009). High schools find later start time
helps students’ health and performance. e Journal
of the American Medical Association,301, 2200–2201.
[PubMed:19491176].
Lufi, D., Tzischinsky, O., & Hadar, S. (2011). Delay-
ing school starting time by one hour: Some effects on
attention levels in adolescents. Journal of Clinical Sleep
Medicine,7, 137–143. [PubMed:21509327].
McGeorge, D. (2012). Sleepless in America: School start
times. Education and Health,30(3), 57–59.
Onyper,S.V.,acher,P.V.,Gilbert,J.W.,&Gradess,S.
G. (2012). Class start times, sleep, and academic perfor-
mance in college: A path analysis. Chronobiology Inter-
national,29, 318–335. [PubMed:22390245].
Owens, J. A., Belon, K., & Moss, P. (2010). Impact of
delaying school start time on adolescent sleep, mood, and
behavior. Archives of Pediatrics and Adolescent Medicine,
164, 608–614. [PubMed: 20603459].
Perkinson-Gloor, N., Lemola, S., & Grob, A. (2013). Sleep
duration, positive attitude toward life, and academic
achievement: e role of daytime tiredness, behavioral
persistence, and school start times. Journal of Adoles-
cence,36, 311–318.
Short,M.A.,Gradisar,M.,Lack,L.C.,WrightH.R.,
Dewald, J. F., Wolfson A. R., & Carskadon, M. A. (2013).
A cross-cultural comparison of sleep duration between
US and Australian adolescents: e effect of school
start time, parent-set bedtimes, and extracurricular load.
Health Education and Behavior,40, 323–330.
Wahlstrom, K. (2002). Changing times: Findings from
the first longitudinal study of later high school start
times. NASSP Bulletin, 286, 3–21.
Wahlstrom, K. (2010). School start time and sleepy teens.
Archives of Pediatrics and Adolescent Medicine,164,
676–677.
Wahlstrom, K. (2014). Examining the impact of later
high school start times on the health and academic per-
formance of high school students: A multi-site study: final
report. Center for Applied Research and Educational
Improvement. Retrieved from: http://www.cehd.umn.
edu/CAREI/sleepresources.html
Wolfson, A. R., & Carskadon, M. A. (2005) A survey
of factors influencing high school start times. NASSP
Bulletin,89, 47–66.
200 Volume 8—Number 4
Judith Owens et al.
Wolfson, A. R., Spaulding, N. L., Dandrow, C., & Baroni,
E. M. (2007). Middle school start times: e importance
of a good night’s sleep for young adolescents. Behavioral
Sleep Medicine,5, 194–209.
APPENDIX 2. HISTORY OF SCHOOL START TIMES
AND SLEEP RESEARCH ON ADOLESCENTS IN THE
UNITED STATES
Historically, public school bell times across the nation
evolved as a result of economic, social, legal, and political
pressures on school districts and municipalities—not from
sleep science pertaining to adolescents (which essentially
did not exist until the 1970s) or concerns about the health,
safety, and academic performance of students. Below is an
overview of the interplay between influential factors in the
development of public education and its transportation
systems as well as major milestones in science of sleep and
circadian biology.
1800s
Until the 1840s, the educational system is highly localized
and largely accessible only to certain privileged groups (i.e.,
the wealthy, whites, males).
Mid 1800s
Educational reform movement led by Horace Mann and
Henry Barnard leads to free public education at the elemen-
tary level for all American children.
Late 1800s
Seventeen states had operable public school transporta-
tion programs, starting with Massachusetts in 1869 (e.g.,
horseback, wagon).
At the end of the nineteenth century, 93% of the highways
in the country are dirt roads. Horse-drawn carriages and
the railroad are the leading means of transportation. Auto-
mobiles first came into use in the 1890s, and the first auto
arrived in Seattle in 1900. By the 1950s, the “Age of the Auto-
mobile” had come into its own and there is great pressure
to create an improved transportation infrastructure; i.e., to
build more and better roads that link new houses being built
outside cities and jobs that still exist mostly in urban centers.
1900s
1910—irty states had programs for transporting stu-
dents which consisted mostly of horse-drawn wagons.
1915—Navistar manufactures the first “school bus” for
Ravinia School District in South Dakota.
By 1918, all states have passed laws requiring children to
attend at least elementary school.
1919—All 48 states have laws allowing the use of public
funds for transporting children to and from school.
1920s
Dr. Nathaniel Kleitman, one of the earliest and most influ-
ential sleep researchers, begins to study the regulation of
sleep and wakefulness at the University of Chicago.
1925—Supreme Court rules in Pierce v. Society of Sisters
that states cannot compel children to attend public schools
and that children can instead attend private schools.
As the 20th century progresses, most states enact legisla-
tion extending compulsory education to age 16.
1930—e company Wayne Works produces the first
school bus with an all-steel body and windows with safety
glass. e availability of and access to affordable motor vehi-
cle transportation helps transport children from more rural
and remote areas and helps lead to the consolidation and
modernization of schools (eobald, 2004).
During the 1930s, the absorption of one-room rural
school houses and more students into geographically larger
school districts consisting of primary and secondary schools
created a need for more transportation resources, espe-
cially for high school students; many of whom could not
practically walk to school. e school bus became the
most popular form of transportation of students; requiring
significant resources in most school district budgets. e
number of students that are bused to school have risen
from 7 million students in 1950 to almost 29 million in 2013
(School Bus Fleet Magazine, 2001).
1939—Dr. Nathaniel Kleitman publishes his seminal book
Sleep and Wakefulness (Kleitman, 1963).
1940s
In the 1940s, responsibility for financing public education
becomes more regionalized at the state, district, and munic-
ipality levels. In 1940, local property taxes finance 68% of
public school expenses, while state governments contribute
30%. By 1990, local districts and states each contribute 47%
to public school revenues. e federal government provides
most of the remaining funds.
1941–1945—United States in World War II: Industrial-
ization and a postwar economic boom dramatically change
the prosperity of Americans and they buy more houses, auto-
mobiles, and start families, which will become the “baby
boomers.”
By the middle of the 20th century, most states take a more
active regulatory role in public education than in the past.
Many states consolidate school districts into larger units. In
1940, there are over 117,000 school districts in the United
States, but by 1990, the number decreases to just over 15,000.
is regionalization often results in transporting a greater
number of students over longer distances.
1950s
All in all, new highways, faster and cheaper vehicles, and
economic prosperity stimulate a tremendous urban sprawl
and the “suburbanization of America” from the 1950s to
present day.
1950s–1960s—“Baby boomers” begin to reach school
age. More than 50% of today’s schools are built during this
period.
Volume 8—Number 4 201
School Start Time Change
1953—Dr. Nathaniel Kleitman and his graduate student,
Eugene Aserinsky, make the landmark discovery of rapid
eye movement (REM) during sleep. Shortly afterwards, their
student, Dr. William Dement, describes the “cyclical” nature
of sleep and the relationship between REM and dreaming
(Dement & Kleitman, 1957).
1954—Brown v. Board of Education outlaws “separate but
equal” facilities. “White flight” from urban centers begins,
leading to the rapid development of suburban school dis-
tricts. In many regions of the country, children are bused
longer distances to assist in integrating schools.
1956—President Dwight D. Eisenhower signs the Inter-
state and Defense Highways Act and helps accelerate the
suburbanization of America. 47,000 miles of federal highway
are built.
1960s
By the early 1960s, there is a rapid increase in the
school population due to the consolidation of remaining
schools—larger and more complicated school districts are
created. Educators begin looking for new ways to deal with
the problem of overcrowding, which leads to staggered start
times being considered and implemented in some school
districts. With little or no sleep science available to guide
decision making, high school start times are typically placed
earlier than elementary schools.
1970s
During the 1970s, recession, inflation, increasing fuel
costs, and budget cuts further contribute to a “do more with
less” mentality in school systems and in state and local gov-
ernments. As a result of waning enrollment and decreasing
property tax revenues, many school districts look for ways
to cut transportation costs and adopt tiered bell schedules
so that they could move the same number of students with
fewer buses.
1970—Court-ordered busing begins to help integrate
schools, but in some regions of the country, this leads to an
even greater exodus from urban centers. e further con-
solidation of schools and the creation of still larger school
districts result in longer commutes for some students.
Dr. William Dement, “the father of sleep medicine,”
founds the first sleep research center at Stanford University.
1972—Animal studies lead to the discovery of the
suprachiasmatic nuclei in the hypothalamus is the center
of the “biologic (or circadian) clock” in the human brain
(Moore & Eichler, 1972; Stephan & Zucker, 1972).
1973–1974—Stock market crash causes inflation and
devaluation of the dollar.
U.S. experiences an energy crisis due to an oil boycott by
the Organization of Arab Petroleum Exporting Countries
(OAPEC). President Nixon asks the nation to adopt mea-
sures to conserve energy. School districts lower thermostats,
consider shorter days and many adopt tiered-busing to save
fuel, which will more than double in price by the end of the
decade.
1974—Daylight Saving Time is expanded to more states
and for a more extended time period, which leads to con-
cerns regarding younger students waiting for the bus in
the dark.
1976—Using a standardized protocol (the Multiple
Sleep Latency Test), Dr. Mary Carskadon establishes sleep
latency (time to fall asleep) as a physiologic measurement
of sleep propensity (likelihood of falling asleep). is allows
researchers to objectively measure the extent of daytime
sleepiness resulting from acute and chronic sleep loss and
begintoquantifytheimpactofsleeplossondaytimeper-
formance (Carskadon & Dement, 1975, 1977; Richardson
et al., 1978).
Late 1970s
Dr. Mary Carskadon at Brown University and others con-
duct initial research on normal biological and circadian
changes in adolescent sleep.
1979—Second energy crisis hits the nation following the
Iranian revolution; increasing pressure on school districts
and municipalities to lower transportation costs.
1980s
During the 1980s, an increase in the birth rate and new
immigration contributes to the growth of the nation’s stu-
dent population, adding pressure on many school systems.
Starting in the 1980s, important research on adoles-
cent sleep needs and sleep schedules is conducted by Dr.
Carskadon and others (e.g., Carskadon et al., 1980).
1982—Researchers develop a hypothesis about how sleep
may play a key role in learning and memory consolidation.
1986—Dr. Charles Czeisler and colleagues describe for
the first time how bright light influences the human biologi-
cal clock (Czeisler et al., 1986).
1988—U.S. Congress appoints Dr. William Dement as
Chair of the National Commission on Sleep Disorders
Research to study the prevalence of sleep deprivation
and sleep disorders and their impact on the health of all
Americans.
1990s
Sleepresearchersbegintodescribedelayedphaseprefer-
ence in teenagers and the impact of school schedules and
employment on their sleep. Researchers also begin to study
sleep disorders and the relationship between sleep loss and
depression in adolescents.
1992—e National Commission on Sleep Disorders
Research issues its report and declares, “America is seriously
sleep-deprived with disastrous consequences” (Stanford
Sleep Well Program, 1992).
1993—e Minnesota Medical Association adopts a reso-
lution calling on local school districts to eliminate early start
times for adolescents.
202 Volume 8—Number 4
Judith Owens et al.
e National Center for Sleep Disorders Research
(NCSDR) is established at the National Institutes of Health;
its mission is to coordinate research and national educational
efforts about sleep and sleep disorders.
1994—e CLOCK gene is discovered, which both
demonstrates the genetic influence on normal human
circadian functioning and emphasizes the importance of
circadian regulation on health and disease (Vitaterna et al.,
1994).
1996—Edina, Minnesota becomes the nation’s first school
district to delay start times for high school students based
on sleep research showing the impact of sleep loss on young
people.
1997—e National Institutes of Health (NIH) declares
that adolescents and young adults (ages 12–25 years) are
a population at high risk for problem sleepiness based
on “evidence that the prevalence of problem sleepi-
ness is high and increasing with particularly serious
consequences.”
1997—As a means of mitigating after-school criminal
activity by unsupervised teenagers, Congresswomen Zoe
Lofgren introduces Concurrent Resolution 227 (ZZZ’s to
As Act) expressing the “sense of Congress that secondary
schools should begin the school day no earlier than 9:00.”
1999—Congresswoman Zoe Lofgren introduces H.R.
1267, “Zs to As Act.” e bill provides grants up to $25,000
to local educational institutions that agree to begin school
for secondary students after 9:00 am. is time, the bill
focusesonthesleepneedsofadolescents.Itdoesnotpass,
but gains significant media attention and helps spur a
“national conversation” about the issue.
1999—e National Research Council holds Sleep Needs,
Patterns and Difficulties of Adolescents Workshop, which
raises awareness among federal agencies and other health
professionals about more than two decades of sleep research
on teens. Dr. William Dement declares, “Adolescence is the
time of greatest vulnerability from the standpoint of sleep.”
Dr. Eve Van Cauter and her colleagues describe the effects
of sleep debt in young adults, establishing an association
between sleep loss and metabolic and hormonal function.
e research later leads to findings linking sleep loss with an
increased risk of obesity (Spiegel, Leproult, & Van Cauter,
1999).
2000s
Sleep researchers increasingly focus on cultural and inter-
national differences in sleep habits, including public policies
related to work and school hours and their impact on sleep,
etc. ere is also emerging evidence that puberty-related
changes in sleep patterns may affect middle school students
as well, thus highlighting the importance of considering later
start times in this population as well. Scientific literature
establishing the relationship between sleep, memory, and
learning continues to grow.
2000—e National Sleep Foundation releases an Adoles-
cent Sleep Needs and Patterns: Research Report and Resource
Guide at a press conference on Capitol Hill with Con-
gresswomen Zoe Lofgren in order to draw national media
attention to the consequences of early start times on the
health and safety of adolescents (National Sleep Foundation,
2000).
2001—A state senator in Connecticut introduces the first
statewide legislation to change school start times in the
nation, but it does not pass.
2002—Dr. Kyla Wahlstrom publishes Changing times:
Findings from the first longitudinal study of high school start
times, the nation’s first major study that details the positive
impacts of later school start times (Wahlstrom, 2002).
2002—A bill is introduced in the Connecticut Senate that
bans administering state tests before 9:00, but does not pass.
2004—e State of Connecticut passes legislation that
allows districts to administer the tenth grade mastery test
as early as 8:30.
2006—e National Sleep Foundation issues its first
annual “Sleep in America” poll. e national poll is the first
of its kind, detailing the findings of telephone interviews
from a random sample of 1,602 caregivers and their ado-
lescentchildrenaboutthestudent’ssleepandsleephabits.
It finds that only 9% of high school-aged respondents get
the amount of sleep recommended by physicians on school
nights.
2009—e Centers for Disease Control and Prevention
(CDC) release data from a national survey used to assess
the prevalence of unhealthy sleep behaviors in 12 states
and declares “insufficient sleep is a public health epidemic”
(McKnight-Eily et al., 2011).
1900–2010—e percentage of teenagers who graduate
from high school increases from about 6% in 1900 to about
85% in 1996, and then declines over the next decade and a
half to 75% in 2010.
2010s
2010—e Department of Health and Human Services
releases Healthy People 2020, which for the first time gives
sleep it’s own focus area and sets the objective of increasing
“the proportion of students in grades 9 through 12 who get
sufficient sleep.”
2010—e American Medical Association (AMA)
adopts Resolution 503, “Insufficient Sleep in Adolescents”—
sponsored by the American Sleep Apnea Association—
which confirms “adolescent insufficient sleep and sleepiness
as a public health issue” and supports “education about
sleep health as a standard component of care for adolescent
patients.”
2011—Vorona et al. publish findings on adolescent auto-
mobile crash rates in Virginia Beach and Chesapeake, Vir-
ginia showing a significant increase in crashes in the district
with earlier high school start times (Vorona et al., 2011).
Volume 8—Number 4 203
School Start Time Change
e Brookings Institute issues its report, Organizing
Schools to Improve Student Achievement: Start Times, Grade
Configurations, and Teacher Assignments, identifying high
schoolstarttimedelayasoneofthethreemostimportant
strategies to improve America’s schools and projecting a
potential benefit-to-cost ratio of 9:1 (Jacob & Rockoff, 2011).
e CDC publishes an epidemiological study showing
that almost 70% of high school students are not getting
sufficient sleep and sleep loss is associated with 10 at-risk
behaviors including smoking, alcohol and marijuana use,
sexual activity, feelings of sadness, and thoughts of suicide.
2012—e Florida Chapter of the American Academy of
Pediatrics issues a position statement supporting, “consider-
ations to policy changes where students’ physical and men-
tal health is promoted. e FCAAP/FPS supports efforts to
change high school start times after 8:00 a.m.”
2013—U.S. Department of Education Secretary Arne
Duncan tweets, “Common sense to improve student
achievement that too few have implemented: let teens
sleepmore,startschoollater.”
Virginia Chapter of the American Academy of Pediatrics
issues a statement supporting later school start times for all
Virginia high schools.
2014—Dr. Kyla Wahlstrom, with funding from the CDC,
publishes the findings from a 3-year research study looking
at the impacts of later start times in 8 public high schools in
3states.estudyfindsthatlaterstarttimesimprovesleep,
academic performance, and reduce motor vehicle crashes
(Wahlstrom, 2014).
e American Academy of Pediatrics issues a Policy State-
ment in support of delaying school start times for high school
and middle school students.
References for Appendix 2
Carskadon, M. A., & Dement, W. C. (1975). Sleep studies
on a 90-minute day. Electroencephalography and Clinical
Neurophysiology,39(2), 145–155. [PubMed:50211]
Carskadon, M. A., & Dement, W. C. (1977). Sleepiness
and sleep state on a 90-min schedule. Psychophysiology,
14(2), 127–133. [PubMed: 847063]
Carskadon, M. A., Harvey, K., Duke, P., Anders, T. F., Litt,
I. F., & Dement, W. C. (1980). Pubertal changes in day-
time sleepiness. Sleep,2, 453–460. [PubMed: 7403744]
Centers for Disease Control and Prevention. (2014).
Insufficient sleep is a public health epidemic.Retrieved
from http://www.cdc.gov/features/dssleep/
Czeisler,C.A.,Allan,J.S.,Strogatz,S.H.,Ronda,J.M.
Sánchez, R., Ríos, C. D., …Kronauer, R. E. (1986). Bright
light resets the human circadian pacemaker independent
of the timing of the sleep-wake cycle. Science,233(4764),
667–671. [PubMed: 3726555]
Dement, W., & Kleitman, N. (1957). e relation of eye
movements during sleep to dream activity: An objective
method for the study of dreaming. Journal of Experimen-
tal Psychology,53, 339–346. [PubMed: 13428941]
Jacob, B., & Rockoff, J. (2011). Organizing schools
to improve student achievement: Start times, grade
configurations, and teacher assignments. e Brook-
ings Institute,8, 1–6. Retrieved from http://www.
hamiltonproject.org/files/downloads_and_links/0920
11_organize_jacob_rockoff_brief.pdf
Kleitman, N. (1963). Sleep and wakefulness. Chicago, IL:
University of Chicago Press.
McKnight-Eily,L.R.,Liu,Y.,Wheaton,A.G.,Croft,J.B.,
Perry, G. S., Okoro, C. A., & Strine, T. (2011). Unhealthy
sleep-related behaviors: 12 states, 2009. Morbidity and
Mortality Weekly Report,60, 233–238.
Moore, R. Y., & Eichler, V. B. (1972). Loss of circa-
dian adrenal corticosterone rhythm following suprachi-
asmatic nucleus lesions in the rat. Brain Research,42,
201–206.
National Sleep Foundation. (2000). Adolescent sleep
needs and patterns: Research report and resource guide.
Retrieved from: http://www.sleepinfairfax.org/docs/
sleep_and_teens_report1_NSF.pdf
Owens J., & Adolescent Sleep Working Group and Amer-
ican Academy of Pediatrics Committee on Adolescence.
(2014). School start times for adolescents AAP policy
statement. Pediatrics,35(9), 1.
Richardson, G. S., Carskadon, M. A., Flagg, W., Van de
Hoed, J., Dement, W. C., & Mitler, M. M. (1978). Exces-
sive daytime sleepiness in man: Multiple sleep latency
measurement in narcoleptic and control subjects. Elec-
troencephalography and Clinical Neurophysiology,45,
621–627. [PubMed: 81764]
School Bus Fleet Magazine. (2001). School bus fleet
2001 fact book. Torrance, CA: Bobit. Retrieved from
http://education.stateuniversity.com/pages/2512/Trans
portation-School-Busing.html
Spiegel, K., Leproult, R., & Van Cauter, E. (1999). Impact
of sleep debt on metabolic and endocrine function. e
Lancet,354, 1435–1439.
Stanford Sleep Well Program. (1992). Overview of
the findings of the national commission on sleep dis-
orders research. Retrieved from http://web.stanford.
edu/∼dement/overview-ncsdr.html
Stephan, F. K., & Zucker, I. (1972). Circadian rhythms
in drinking behavior and locomotor activity of rats are
eliminated by hypothalamic lesions. Proceedings of the
National Academy of Sciences of the United States of
America,69, 1583–1586
eobald, M. (2004). Wayne Wor k s .Retrievedfrom
http://www.coachbuilt.com/bui/w/wayne/wayne.htm
204 Volume 8—Number 4
Judith Owens et al.
Vitaterna,M.H.,King,D.P.,Chang,A.M.,Kornhauser,
J.M.,Lowery,P.L.,McDonald,J.D. …Takahashi, J.
S. (1994). Mutagenesis and mapping of a mouse gene,
clock essential for circadian behavior. Science,264(5159),
719–725. [PubMed:8171325]
Vorona, R. D., Szklo-Coxe, M., Wu, A., Dubik, M., Zhao,
Y., Ware, J. C. (2011). Dissimilar teen crash rates in
two neighboring southeastern Virginia cities with dif-
ferent high school start times. JournalofClinicalSleep
Medicine,7, 145–151. [PubMed: 21509328]
Wahlstrom, K. (2002). Changing times: Findings from
the first longitudinal study of later high school start
times. National Association of Secondary School Princi-
pals Bulletin,286, 3–21. Retrieved from http://conser
vancy.umn.edu/bitstream/139295/1/2002%20NASSP%
20Bulletin,%20Vol.%2086%20No.%20633.pdf
Wahlstrom, K. (2014). Examining the impact of later
high school start times on the health and academic per-
formance of high school students: A multi-site study: final
report. Center for Applied Research and Educational
Improvement. Retrieved from http://www.cehd.umn.
edu/CAREI/sleepresources.html
APPENDIX 3. SELECTED CASE STUDIES OF LATER
START TIME SUCCESSES
Bentonville School District, Arkansas
In 2007, the Bentonville School District, the fifth-largest in
the state (11,100 students), implemented a later start time
for its one high school, moving a one later from a 7:45 am
to an 8:45 am start time. District officials changed the start
times to better accommodate a new “A/B block” class sched-
ule that allowed students to rotate courses every other day.
e teachers reported liking the A/B block schedule and
used the extended class time to incorporate cooperative
learning groups and hands-on activities. To accomplish
the bell schedule change, the district moved all of the five
middle schools earlier (8:00 am to 7:40 am) and also moved
the nine elementary schools earlier (8:00 am to 7:30 am).
Following the changes in bell schedules, many teachers
and administrators anecdotally reported a high level of sat-
isfaction, with some teachers indicating feeling more rested
and productive throughout the day. Students who were
active in extracurricular activities or worked after school
preferred the later start time because it allowed the oppor-
tunity for more sleep. e community also found that the
bell schedule change assisted in alleviating traffic congestion.
e district found that block scheduling gave their students
more class choices and the later start time change gave the
district more flexibility in dealing with traffic and addressing
the students’ sleep needs.
Lessons: Bentonville is an example of a successful col-
laboration between district officials and the school board
to change class and transportation schedules. e changes
resulted in the promotion of sleep health and more flexibility
in students’ class schedules. Improvements in sleep and
quality of life for both students and teachers were also
reported.
(Source: http://normessasweb.uark.edu/bestpractices/
papers/Casestudies/401003.pdf)
Pulaski County Special School District, Arkansas
In 2012, the Pulaski County Special School district, with a
student population of 17,500, “flipped” the start times for
its six middle and six high schools with those of its 24 ele-
mentary schools. e middle schools moved 50 min later
from 7:30 am to 8:20 am, and high schools moved later by
65 min, from 7:30 am to 8:35 am, with all elementary schools
remaining at their start times between 7:45 and 8:00 am. e
changes were spurred by the district’s interest in reducing
transportation costs and improving the sleep the health and
safety of its students. To accomplish these changes, the dis-
trict went from a single-tier to a two-tier busing schedule.
e change was initiated by the district under the leader-
ship of the new superintendent who was charged by the State
Commissioner of the Department of Education with cutting
costs following a period of financial mismanagement by the
previous administration and the school board, which was
dissolved. e district announced the change in January with
implementation in September to allow enough time for par-
ents to plan and change their childcare arrangements and
utilized a toll-free line to field questions and comments from
the community.
Lessons: e Pulaski County Special School district
demonstrates that communities can successfully adapt
to substantial changes if given appropriate time and if
parents are provided with adequate information during
implementation.
(Source: http://www.thv11.com/news/article/189874/2/
PCCSD-proposes-changes-to-start-times and http://www.
pcssd.org/pcssd-bell-times-bus-schedules-may-change-for-
next-school-year)
Albany Unified School District, California
Albany High School has 1,800 students in a small school
district with one other high school for at-risk students,
one middle school, and three elementary schools. In 2013,
the Albany High School principal proposed moving the
current bell time of 7:40 am to 8:30 am, based on research
from economists and sleep researchers as well as feedback
from a series of community engagement sessions. A task
force, named the Challenge Success Committee, was formed
and included parents, students, teachers, counselors, and
school administrators. e committee researched the issue
Volume 8—Number 4 205
School Start Time Change
and developed a set of recommendations. After receiv-
ing the report and recommendations, the high school’s
Instructional Improvement Council (a small committee
comprised of six staff members, two parents, and two
students) agreed to move the start time to 8:00 am as a
one-year pilot and proposed to solicit feedback from the
community through a series of surveys in order to gauge
potential impact. e high school also sought input through
several governance committees and a town hall forum. After
reviewing the scientific rationale and as a result of these
community discussions, the Albany Middle School princi-
pal also recommended eliminating the 7:00 am zero period
and moving the start time to 8:00 am, which were both
adopted.
Lessons: is district is a good example of a principal
immersing himself in the sleep research and working hard to
educate both his staff and community. His commitment to
the issue and willingness to pilot the change had a positive
impact on his colleagues in the rest of the district. e
superintendent utilized local sleep experts and community
members to educate the community about the health and
safety benefits of changing start times.
(Source: http://web.ahs510.org/news/administrative-
news/iicagreestolaterstarttime and http://albany.patch.com/
groups/schools/p/later-daily-start-time-for-albany-high-
proposed)
Long Beach Unified School District, California
Long Beach Unified School District educates 81,000 students
in six high schools, 15 middle schools, and 51 elementary
schools, with four charter and five alternative schools. In
the 2013–2014 school year, the Long Beach Unified School
District adopted a plan to move their high schools from
7:50 am to 8:50 am with an end time of 3:40 pm, as part of a
pilot program. e plan, debated about and approved by the
school board early in 2013, also moved six middle schools
to a 9:00 am to 3:40 pm bell schedule to be consistent with
the district’s nine other middle schools. As part of the plan,
all affected schools were to develop action plans to provide
supervision for students in the morning. Budget reductions
were the driving factor for the realignment of bell schedules.
District officials anticipated that the changes would save the
district over $1 million, with much of the savings resulting
from bus schedule adjustments for special education stu-
dents requiring home pick-up and riding smaller buses.
Community outreach prior to schedule changes indicated
mixed levels of support and initial reservations from key
stakeholders. An online survey on school start time change,
conducted by the district, showed that the respondents
were about evenly split on the issue of changing start
times. Additionally, the local teacher’s union expressed
concern regarding contractual issues and the change’s
impact on sports and other programs. e district worked
closely with these groups to address their issues and ulti-
mately to gain their support. Additionally, the district
announced that it would convene a large, ad hoc commit-
tee (40–60 members) to evaluate the impact of the pilot
program.
Lessons: While the motivation for changing start times
in Long Beach was to cut transportation costs, this district
is taking a unique approach in forming a large committee
of stakeholders to evaluate the impact of the change. Most
school districts typically form such committees prior to
considering taking action and then disband once a decision
to delay start times is made. e ad hoc committee report is
due in the fall of 2014.
(Source: http://www.dailybreeze.com/general-news/2013
0326/long-beach-middle-schools-to-start-later-next-fall
and http://www.lbschools.net)
West Hartford Public High School, Connecticut
West Hartford Public Schools is a district with a student pop-
ulation of 10,222 in two high schools, three middle schools,
and 10 elementary schools. In the fall of 2006, the West Hart-
ford Board of Education voted 5–2 to adopt a flexible start
time schedule for juniors and seniors at its two high schools
beginning the in the fall of 2007, with implementation for
sophomores and freshmen beginning in 2008. e “flexible”
start time option gave juniors and seniors the option of start-
ing school at 8:15 am (second period), instead of at 7:30 am.
e first period was converted to a study hall and the dis-
missal bell remained at 2:15 pm for all students, regardless
of whether or not they elected the second period start. e
“flexible” start time was a strategy developed in response
to significant community resistance to a proposed “blanket”
delay of high school start times. While acknowledging that
this “flexible” option was a less-than-optimal compromise,
the Board of Education concluded that at least it offered a
delayed start time option to those students and parents who
recognized the value of obtaining sufficient sleep. According
to school administrators, the later start times had positive
impacts on academic performance, student stress levels, and
emotional health.
Lessons: West Hartford Public Schools is a good example
of a school board and administration continuing to seek
solutions to provide choices to students and to encourage
better sleep health, after deciding against full implementa-
tion of start time change due to community opposition.
(Source: Author’s notes, 2006.)
Wilton Public Schools, Connecticut
Wilton engaged in a two-year long process, which culmi-
nated in delayed start times for both its middle and high
206 Volume 8—Number 4
Judith Owens et al.
school (7:35 am to 8:15 am) in 2003. Initially, legislation
was proposed by State Senator Kevin Sullivan in 2001 to
delay start times across Connecticut, but subsequently it was
decided that this should occur on a local district-by-district
level. Wilton’s local chapter of the League of Women Voters
became involved in the debate and distributed both a review
of the literature on adolescent sleep and the results of a local
survey of students and school staff and recommended that
Wilton Board of Education consider delaying start times.
Existing community-planning teams, consisting of teach-
ers, administrators, parents, students, and citizens, were
then engaged in studying the issue. Area superintendents
met with physicians from the Connecticut oracic Soci-
ety to ask for scientific input and to request their engage-
ment in the process. It was decided by the superintendent
with input from advisory groups that any potential solu-
tions would need to meet three criteria: (1) be cost-neutral,
(2) require student bus rides no greater than 45 min, and
(3) avoid any student being picked up by a bus earlier than
7:00 am. Parents and teachers were offered the opportunity
to participate in an “advisory vote” on the superintendent’s
recommendations regarding start times, which yielded dia-
metrically opposed results (parents supported two to one,
teachers opposed two to one). While the Wilton Sports
Council published full-page ads opposing the changes on a
number of grounds, including compromising Wilton’s rep-
utation of athletic excellence, the head of the Connecticut
Interscholastic Athletic Conference provided a written state-
ment in support of delayed start times. After the start time
change,teachersreportedthatstudentswerebetterrested
and more alert during the school day. ere was no apprecia-
ble effect on athletics, and within a year, the change became
the “norm” and “part of the fabric of the community” (Robert
O’Donnell, current superintendent). While there has been
considerable interest expressed by neighboring school dis-
tricts, no other schools in the surrounding area have fol-
lowed Wilton’s lead.
Lessons: Wilton provides an illustration of several com-
mon themes related to changing school start times, includ-
ing the vital role of the superintendent, community concerns
regarding impact on athletics which were not subsequently
substantiated, and the importance of ongoing stakeholder
input. It also provides some unique perspectives (i.e., the
potential for community groups to assume a leadership role,
enlistment of local health professional societies, use of estab-
lished models such as community planning teams to provide
input), which may be applicable to other districts. In partic-
ular, the community’s identity as a self-styled “sports town”
provides a striking example of “cultural” considerations that
have the potential to derail efforts to change start times if not
adequately addressed.
(Source: Owens, J. [2013, October 30]. Telephone inter-
view.)
Denver Public Schools, Colorado
Denver Public Schools is a large district with over 84,000
students in 22 high schools, 21 middle schools and 74 ele-
mentary schools, and 59 alternative and charter schools. Due
to concerns regarding the impact of early start times on ado-
lescents and seeking a way to cut transportation costs, the
district conducted a feasibility study in 2004. In 2005, the
district took one of the more unique approaches to chang-
ing start times for all of its high school students. Under the
leadership of the superintendent, the school district adopted
a “flexible” schedule and radically changed how it provided
transportation to its students. While conducting the feasi-
bility study, the district found that 2,400 high school and
1,300 middle school students purchased bus passes from the
local public transit agency, the Regional Transportation Dis-
trict (RTD). In the “flexible” plan, students were allowed to
choose their arrival and dismissal times within a 7:30 am to
4:15 pm bell schedule, as long as they met the instructional
time requirements. e new transportation plan involved
switching more students to public transportation and pro-
viding free bus passes to all students who lived more than 3.5
miles from their base school and to students going to mag-
net schools elsewhere in the district. Other students were
still provided with bus transportation by the district.
As part of the district’s implementation plan, a series of
public outreach sessions were held including two town hall
meetings, a public hearing and a presentation by princi-
pals summarizing their discussions with parents and local
communities. ere were some initial concerns raised by
parents about students riding on public buses. e district
also conducted surveys of students and found that many of
the students were not initially open to taking advantage of
the later start time options because of their involvement in
after-school programs. However, in the 2006 school year,
about 30% students chose a start time of 8:00 am or later. As
a result, the district was able to eliminate 60 buses and save
$750,000 in its transportation budget.
Lessons: e Denver public school district was open to
unique solutions and working with the local public trans-
portation system to allow options for its students. ose
students who wished to take advantage of school schedule
choices were presented with a number of options that were
nonetheless within specific parameters. e process also
involved significant community engagement through pub-
lic meetings and online surveys. While the initiative lacked
overwhelming community and stakeholder support, district
officials decided to still pursue a course which they felt was
in the best interests of the district, both financially and for
the health and safety of its students.
(Source: National Sleep Foundation, http://www.
sleepinfairfax.org/docs/CS.Denver.pdf. High School
Transportation: Report to the Board, Department of
Volume 8—Number 4 207
School Start Time Change
Research, Planning and Special Programs, Department
of Transportation, March 18, 2004.)
Milford Public Schools, Delaware
Milford Public Schools is a small district with about 4,100
students who attend one high school, one middle school, and
four elementary schools. After many years of studying ways
to mitigate scheduling conflicts resulting from state require-
ments that mandated professional development time and
standardized testing, the Superintendent of Milford Public
Schools formed a task force to study potential solutions. e
task force included teachers and administrators who vol-
unteered to research the issue and develop recommenda-
tions. In 2012, the task force recommended the adoption
of an “A/B block” scheduling system as well as a delay in
school start times, with the high school start times mov-
ing 40 min later from 7:35 am to 8:15 am, the middle school
start times moving from 7:35 am to 8:00 am, and the elemen-
tary school start times moving earlier to 7:40 am. A unique
feature of the plan was to start school at 9:35 am for both
middle and high school students on Wednesdays. e school
accommodated parents who were concerned about leaving
their older children at home unsupervised on late start time
days by opening the libraries to students who needed to be
at school earlier. e superintendent expressed the belief
that these changes—both later start times for secondary stu-
dents and earlier start times for elementary students—would
be beneficial for all students and would increase daytime
alertness.
Lessons: Milford Public Schools illustrates a situation in
which a district initially changed its bell times to accommo-
date state requirements for standardized testing and pro-
fessional development but subsequently realized that this
schedule better accommodated the biological and learning
needs of its students. Parental opinions were mixed, with
some elementary school parents concerned about winter
civil twilight violations, while some found the earlier times
to be more accommodating for their work schedules.
(Source: https://milfordlive.com/2012/06/19/msd-to-see-
changes-next-year-2 and http://www.milfordbeacon.com/
article/20120710/NEWS/307109954)
Brevard County School District, Florida
Brevard County School District is a large system with 82
elementary schools, 16 middle schools, 21 high schools,
18 specialized centers, and a student population of 96,000
that is delivered on a three-tier bus system. Each bus has
three routes that service all three school tiers. In 2000,
the district delayed high school start times by a full hour,
from 7:30 am to 8:30 am and middle school times by 25 min
from 8:50 am to 9:15 am. e elementary schools, which
incorporate grades K-6, were moved earlier by a full hour and
fifty min, from 9:50 am to 8:00 am. e change was imple-
mented after extensive community outreach was conducted
through public meetings to allow stakeholders the oppor-
tunity to express their concerns and share their views. e
district also conducted research on other school districts
that had successfully changed their start times and provided
information to the public and school staff about improve-
ments in academic performance and attendance rates. Dis-
trict staff also worked with a local hospital to provide scien-
tific and health education presentations to the community
and the school board, using information from the National
Sleep Foundation. A proposal was developed by the super-
intendent and submitted to the school board, which subse-
quently approved the change. In order to offset some of its
transportation related costs, the district charges students for
special or out-of-boundary services.
Following the change, the school district found a sig-
nificant reduction in the rate of first period tardiness and
absences. Some of the negative outcomes reported by the
district were limited availability of school buses for field trips
and an increased need for after-school childcare.
Lessons: Brevard County demonstrates that students and
the community in a large county can adapt to significant
changes in school bell schedules. e district solicited a
tremendous amount of public input but did not let neg-
ative opinions proffered by some constituencies and indi-
viduals deter them from finding solutions. Many of the
non-academic concerns typically raised by parents regarding
potential impact on after-school employment and on athlet-
ics were not realized. Finally, Brevard’s strategy of charging
for non-academic and magnet transportation services may
help reduce costs significantly in districts that provide sub-
stantial special transportation services.
(Source: Author’s notes, 2006. http://www.neola.com/
brevardcofl/search/policies/po8600.htm)
Bonneville School District, Idaho
In the 1999–2000 school year, the Bonneville School District
adopted a new school start time for its two high schools by
shifting from a 7:45 am start time to 8:50 am. e district,
which has 21 schools (including three middle and 14 elemen-
tary schools) and a student population of 11,200, was the first
in the state to move start times later. In the year following the
change, the district studied the impacts on attendance and
found that absences dropped by 15% and tardiness decreased
by 22%. e study also found that high school students were
getting about 44 min more sleep on average after the change.
Many students reported that they were more alert, and their
teachers concurred. Some students also reported using part
of the extended period morning before school to seek help
from their teachers. For unclear reasons, the start times at
208 Volume 8—Number 4
Judith Owens et al.
the two high schools were subsequently shifted earlier to
8:25 am before 2014.
Lessons: e Bonneville School District is an example
of a district evaluating the impact of the change in school
start times and reporting the positive data back to its staff,
students, and the local community.
(Source: Author’s notes, 2006.)
Needham Public School District, Massachusetts
Needham Public School District is a small district con-
sisting of six elementary schools, one middle school, and
one high school with a student population of 5,476. Under
the direction of the superintendent, an advisory body
called the School Starting Time Advisory Committee was
formed in the fall of 2002 to determine if start times for
all schools should be changed to be more compatible with
students’ biological rhythms. e committee included one
teacher, one administrator, and a parent representative from
each level, a school committee representative, a school
nurse, a representative from the transportation depart-
ment, two high school students, and the director of the arts
department.
e committee reviewed the existing sleep and education
research available at the time and conducted surveys of high
school teachers regarding the level of alertness of students
while in class. ey also surveyed a small sample of stu-
dents regarding their sleep habits and preferences regarding
start times. e committee launched a website to provide
information about their activities and encouraged members
of the community to provide feedback. e committee also
made a concerted effort to contact school districts that had
successfully changed their start times and consulted sleep
experts, health professionals, and e National Institutes of
Health.
As a result of their research and fact-finding, the commit-
tee concluded that, “the research about the educational and
health benefits of a later high school starting time are clear
and compelling.” On the basis of a review of the potential
impact on athletics, after-school programs, religious activ-
ities, performing arts and transportation, the committee
recommended changing the start times later for both pri-
mary and secondary schools. e committee developed five
options, all of which had the high school start time moving
from 7:40 am to 8:05 am, with middle schools either staying
at their current time or also moving to 8:05 am, and all
elementary schools staying at their current time or starting
at 8:45 am. Due to redistricting in 2003, the school board
delayed their decision a full year and conducted further com-
munity engagement. In 2004, the school board moved start
times for all middle and high schools close to the time rec-
ommended by the committee with high schools starting at
8:00 am, middle schools starting at 7:50 am, and elementary
school starting 15 min later, moving from 8:20 am to
8:35 am.
Lessons: Needham Public School District exemplifies
many of the strategies used by other districts that have
successfully changed their start times. It created a small
working group composed of key department members and
parent and teacher representatives to collect information
on the scientific research (including interviews with sleep
experts), to consider the potential impact and develop pos-
sible mitigating strategies, and to conduct meetings with
stakeholders. e committee carefully studied the issues
over a protracted period of time, developed and articulated
a compelling rationale for changing the bell schedule and
presented a range of options to the community based on an
assessment of feasibility.
(Source: Needham Public Schools School Starting Time
Report and Recommendation, January 2003.)
Arlington Public Schools, Virginia
Arlington Public Schools is a large urban district just outside
of Washington, DC. Currently, the school district has more
than 19,900 students in 22 elementary schools, five middle
schools, and four high schools. In 1999, based on the growing
interest of some parents and the school board in emerging
sleep science and in order to reduce transportation costs
and to improve academic performance, the district began a
comprehensive and intricate two-year process to change its
school start times.
e school board requested that the district’s Advisory
Council for Instruction (ACI), a large body of 50 members,
form a School Start Time Steering Committee to compile
research on sleep and adolescence and to study the poten-
tial impact of changing bell schedules on transportation and
extracurricular activities. In December of 1999, the commit-
tee recommended that the school board move high school
start times 45 min later (from 7:30 am to 8:45 am) starting
in the fall of 2000. On the basis of this recommendation,
the school board voted unanimously to direct district staff
to develop plans to change start times for high schools. As
part of its direction, the school board provided four guiding
principles: (1) the change should improve student achieve-
ment, (2) no school should start before 7:50 for safety rea-
sons, (3) change in high school start times should not neg-
atively impact any group or school level, and (4) the abil-
ity of students being able to participate in extracurricular
activities should not be affected. e board also expressed
a desire that an evaluation study be conducted following
implementation.
Following the board action, the superintendent formed
a working group, comprised of both staff and community
members, to review the ACI’s findings, study all relevant
issues, and make recommendations for implementation in
Volume 8—Number 4 209
School Start Time Change
the fall of 2001. e working group consisted of a steer-
ing committee and several subcommittees to study and
make recommendations regarding such issues as transporta-
tion, public engagement, after-school activities, and sleep
research. e district also hired a transportation consultant
to study bus schedule options and to look for additional effi-
ciencies in the system. e consultant and the working group
initially developed 12 options, eventually settling on a total
of four, one of which included keeping the current bell sched-
ules (status quo).
Once the options were refined, the steering committee
and representatives from a couple of the subcommittees
engaged in significant community engagement to edu-
cate the community about sleep science and to obtain
feedback on the models. Materials were developed and
distributed in back-to-school packets, at the superinten-
dent’s public meetings, the local county fair as well as in
press releases. Information was also posted on the dis-
trict’s website and community feedback was requested via
email through the website. Letters were sent to all of the
local parent teacher associations (PTAs) and representa-
tives from the steering committee attended meetings to
provide information and answer questions. As part of its
process, the working group worked with the University
of Maryland to conduct surveys of parents, students and
teachers and held two public forums to solicit input from the
community.
On the basis of a few different periods of community
engagement and staff input, the steering committee devel-
oped additional options and made further adjustments
throughout. Finally, in October 2000, the steering commit-
tee made a recommendation to the school board, which
voted unanimously to change start times for high schools
from 7:30 am to 8:15 am, middle schools from 8:10 am to
7:50 am, and elementary schools from 8:10 am and 8:50 to
three tiers of 8:00 am, 8:25 am and 9:00 am. Following the
vote, the school board requested that district staff survey
teachers about the proposed changes on whether or not
they would leave the school district with the change taking
place. e survey found that 14% reported that they would
consider leaving their jobs.
During the implementation phase, the steering commit-
tee and school board continued to meet with the community
and key stakeholder groups in order to discuss issues and
foster support for the bell changes. A toll-free line was estab-
lished to help parents and students ask questions about the
implementation, but it was rarely used. A series of surveys
were conducted to gauge community, teacher and student
satisfaction following the
In 2005, the school district’s Office of Planning and
Evaluation did an analysis of surveys conducted after
implementation and grades collected pre (2000–2001)
and post (2001–2002) implementation. e results were
largelyinconclusiveandtheOfficestatedthatitsanaly-
sis“shouldbeinterpretedwithcaution”becauseitwas
“impossible to isolate the impact of the start time change
on academic performance.” However, the Office found a
“very slight” improvement in the first period grades for the
graduating class of 2003. Other findings were insignificant
improvements in academic performance but negligible
or largely inconclusive results regarding tardiness and
attendance. e Office did report that more high school
students reported (41% vs. 47%) participating in “class dis-
cussions” and “being prepared for class” (41% vs. 47%) all of
the time.
Lessons: Arlington is a good example of a district that con-
ducted considerable research, did comprehensive planning,
and included community members and key stakeholders
throughout the process, including after implementation.
is district staff, working group and school board con-
ducted significant community engagement efforts and
adjusted models and plans during the process but did not
let any opposition deter them from achieving what they
believedwasbestforstudents.iscasealsodemonstrates
that while people express strong anticipatory concerns
about change, such fears rarely ever come to fruition (e.g.,
no teachers left as a result of the change). As time goes
by, people adapt, traffic patterns and personal schedules
adjust.
(Source: Changing School Start Times: Arlington,
Virginia. Sleep for Teens Toolkit. National Sleep Foun-
dation, 2005. Accessed by http://www.sleepinfairfax.org/
research.htm. Impact of 2001 Adjustments of High School
and Middle School Start Times. Lewin, D. [2013, October
31] Telephone interview. Arlington Public Schools, Office
of Planning & Evaluation, June 2005. http://www.fcps.edu/
fts/taskforce07/documents/arlington605.pdf)
APPENDIX 4. NATIONAL SCHOOL START TIME
SURVEY
You are being asked to fill out this survey because your
school district has prior experience with delaying high
school start times and you participated in and/or have
knowledge about the process. ank you for your time.
General Information
1. Name and title of individual filling out survey:
2. Name and location of school district:
3a. What year did your school district begin to imple-
ment school start time change?
3b. If your district employed a phased-in approach,
over how many years was start time change imple-
mented?
210 Volume 8—Number 4
Judith Owens et al.
4. Please indicate the number of schools in your dis-
trict in the following categories:
•Elementary School
•Middle School
•High School
5. Please indicate the approximate number of students
enrolled in each category:
•Elementary School
•Middle School
•High School
6. Please indicate the total number of school buses
transporting students in your district:
7. Please indicate the school start/end times PRIOR to
the change for:
•Elementary School
•Middle School
•High School
8. Please indicate the school start/end times AFTER
the change for:
•Elementary School
•Middle School
•High School
9. Please indicate the current school start and end
times (2013–2014 academic year)
•Elementary School
•Middle School
•High School
Change Strategy and Implementation
10. Which best describes your district’s school start
time change strategy (select all that apply)?
•“Flip” (e.g., switch high school and elementary
school start times)
•All school start times (ES, MS, HS) delayed (“slide”
later)
•Addition of a “zero period” in the morning
•Students chose late or early start times
•Other (please describe)
11. Please indicate which transportation approaches
were in place before and/or after the start time change
(may select more than one):
•Tiered bus delivery schedule
•Decreased transportation budget
•Increased transportation budget
•Consolidated “bus depots”
•MS/HS ride together
•ES/MS ride together
•ES/MS/HS ride together
•No Buses provided for magnet schools
•Students pay flat-rate for buses for year
•Increased utilization of public transportation
•Other (please describe)
12. Please mark “Yes” or “No” for the following chal-
lenges that your school district faced in changing
school start times. Yes, No, N/A
1 Traffic flow at the school during drop off and pick up
2 Commuting distances
3 Before-school extracurricular program attendance
4 After-school extracurricular program attendance
5 Before-school academic enrichment program atten-
dance
6 After-school academic enrichment program atten-
dance
7 Use of school facilities by non-school community
groups (e.g., Boy Scouts)
8 Athletic game schedules
9 Before-school athletics practice schedules
10 After-school athletics practice schedules
11 Use of practice fields by non-school groups (e.g.,
Parks and Recreation)
12 Before-school child care
13 After-school child care
14 Changes in parents’ work schedules
15 Changes in teachers’ work schedules
16 Changes in staff commute times
17 Student after-school employment
18 Participation in school breakfast programs
19 “Civil twilight” violations (i.e., leaving home before
dawn or after dusk) for elementary students
20 Other (please specify):
13. Please rank order the top five issues that were
the most challenging (from 1 =most challenging to
5=least challenging)
1 Traffic flow at the school during drop off and pick up
2 Commuting distances
3 Before-school extracurricular program attendance
4 After-school extracurricular program attendance
5 Before-school academic enrichment program atten-
dance
6 After-school academic enrichment program atten-
dance
7 Use of school facilities by non-school community
groups (e.g., Boy Scouts)
Volume 8—Number 4 211
School Start Time Change
8 Athletic game schedules
9 Before-school athletics practice schedules
10 After-school athletics practice schedules
11 Use of practice fields by non-school groups (e.g.,
Parks and Recreation)
12 Before-school child care
13 After-school child care
14 Changes in parents’ work schedules
15 Changes in teachers’ work schedules
16 Changes in staff commute times
17 Student after-school employment
18 Participation in school breakfast programs
19 “Civil twilight” violations (i.e., leaving home before
dawn or after dusk) for elementary students
14. Please mark “Yes” or “No” for the following strate-
gies that your school district employed in changing
school start times. Yes, No, N/A
1 Substitute online education for early morning classes
2 Substitute summer school for early morning classes
3 Substitute Saturday classes for early morning classes
4 Increasing the length of winter break to reduce early
morning travel for elementary students (“civil twi-
light violations”) and decreasing the summer break
by a corresponding number of days
5 Flexible start and end-time scheduling (not requiring
school bus transportation)
6 Adding lighting for selected athletic playing fields
7 Increased use of public transportation (e.g., provid-
ing incentives for use)
8 Increased use of personal transportation (e.g., “kiss
and ride”)
9 Decreased use of personal transportation
10 Student athletes’ early dismissal
11 Extracurricular programs on Saturdays
12 Mid-morning “breakfast break” for students
13 Other (please specify):
15. Please rank order the top five strategies that were
the most effective (from 1 =most effective to 5 =least
effective)
1 Substitute online education for early morning classes
2 Substitute summer school for early morning classes
3 Substitute Saturday classes for early morning classes
4 Increasing the length of winter break to reduce early
morning travel for elementary students (“civil twi-
light violations”) and decreasing the summer break
by a corresponding number of days
5 Flexible start and end-time scheduling (not requiring
school bus transportation)
6 Adding lighting for selected athletic playing fields
7 Increased use of public transportation (e.g., provid-
ing incentives for use)
8 Increased use of personal transportation (e.g., “kiss
and ride”)
9 Decreased use of personal transportation
10 Student athletes’ early dismissal
11 Extracurricular programs on Saturdays
12 Mid-morning “breakfast break” for students
13 Other (please specify):
Benefits/Costs
16. Please mark “Yes” or “No” for the following ben-
eficial outcomes that your school district observed at
the high school level as a result of school start time
changes; if you did not measure an outcome, please
check “Not Applicable” N/A Yes, No, N/A
1 Increase in daily attendance
2 Reduced tardiness rates
3 Improved standardized test scores
4 Improved grades
5 Higher graduation rates
6 Fewer referrals for disciplinary action
7Improvedsportsteamperformance
8 Fewer sports-related injuries
9 Cost saving for public school system
10 Fewer student visits to school health centers
11 Lower rates of depression/suicidal thoughts
12 Lower rates of car accidents
Other (please describe):
17. Please rank order the top five beneficial outcomes
that you believe were the most important (from
1=most important to 5 =least important).
1 Increase in daily attendance
2 Reduced tardiness rates
3 Improved standardized test scores
4 Improved grades
5 Higher graduation rates
6 Fewer referrals for disciplinary action
7Improvedsportsteamperformance
8 Fewer sports-related injuries
9 Cost saving for public school system
10 Fewer student visits to school health centers
11 Lower rates of depression/suicidal thoughts
12 Lower rates of car accidents
18. Please mark “Yes” or “No” for the following neg-
ative outcomes that your school district observed as
a result of school start time changes; if you did not
212 Volume 8—Number 4
Judith Owens et al.
measure an outcome, please check “Not Applicable”
N/A Yes, No, N/A
1 Financial cost incurred by the school district
2 Loss of community support
3 Impactonparentworkschedules
4 Limitations on student after-school employment
5 Financial cost incurred by families (loss income, addi-
tional child care expenses)
6 Changes in traffic patterns
7 Reduction of student involvement in extracurricular
activities/athletics
8 Negative impact on teacher schedules
9 Safety concerns for ES students
19. Please rank order the top five negative outcomes
that you believe were the most important (from
1=most important to 5 =least important)
1 Financial cost incurred by the school district
2 Loss of community support
3 Impact on parent work schedules
4 Limitations on student after-school employment
5 Financial cost incurred by families (loss income, addi-
tional child care expenses)
6 Changes in traffic patterns
7 Reduction of student involvement in extracurricular
activities/athletics
8 Negative impact on teacher schedules
9 Safety concerns for ES students
ank you very much for completing this survey!
Volume 8—Number 4 213
