Introduction
It is imperative to understand the degree to which shifts to later school start times positively influence the psychological well-being of adolescents as questioned in Duraccio (Reference Duraccio2023). Adolescents are middle- to high-school matriculated children aged 13–18 years old. Although much of the supporting research is correlational, later school start times are associated with improved students’ cognitive functioning, decision-making, mental health and academic outcomes and accompany reduced tardiness and absences (Carrell et al., Reference Carrell, Maghakian and West2011; Edwards, Reference Edwards2012; Wolfson et al., Reference Wolfson, Spaulding, Dandrow and Baroni2007; Kim, Reference Kim2022). Performance indicators such as test scores and grades can also be mediated by affective factors such as mood and well-being (Carrell et al., Reference Carrell, Maghakian and West2011; Edwards, Reference Edwards2012). Psychological well-being is defined as the presence of positive feelings and the absence of negative feelings (ScienceDirect, 2024). Along this affective continuum, positive feelings include happiness, self-esteem, joy and interest, while negative feelings include sadness, anxiety and anger. Importantly, insufficient sleep and disruptions to daily rhythms can tip the balance of this continuum towards negative affect.
Figure 1. Barriers to Mitigating Adolescent Health and Performance Consequences from Early School Start Times (Graphics adapted from Microsoft 365 PowerPoint stock images/icons).
Getting the appropriate amounts of sleep is critical to both academic performance and mood stability in adolescents. The American Academy of Sleep Medicine examined 864 scientific papers and published a consensus statement recommending that adolescents get 8–10 hours of sleep per night (Paruthi et al., Reference Paruthi, Brooks and D’Ambrosio2016). Despite the Centers for Disease Control and Prevention (CDC) supporting the recommendation for 8–10 hours, 72.7% of adolescents in the US report inadequate sleep on school nights (CDC, 2024). A 2015 cohort of more than 250,000 participants showed that from 1991 to 2012, there was a global trend of insufficient sleep during adolescence (Keyes et al., Reference Keyes, Maslowsky, Hamilton and Schulenberg2015). Furthermore, a randomized control study by Carrell et al. found that a delay in school start times by 50 minutes led to a substantial, positive effect on academic achievement among freshman students (Carrell et al., Reference Carrell, Maghakian and West2011). Academic achievement was assessed by grades in core courses: calculus, chemistry and English (Carrell et al., Reference Carrell, Maghakian and West2011). In a quasi-experimental study by Edwards, middle school students grades 6–8 had significant standardized reading and math scores after a delayed school start times by 60 minutes was implemented (Edwards, Reference Edwards2012). Improved academic achievement was more pronounced among lower-performing students as students in the 30th percentile gained around 3 points for math, while those in the 90th percentile gained 1 point (Edwards, Reference Edwards2012). These impacts were long-term, as 8th grade students showed sustained improvement into 10th grade (Edwards, Reference Edwards2012).
Circadian rhythms are offset in adolescents (Owens, Reference Owens2010) and falling asleep after 11 p.m. and waking up after 8 a.m. aligns better with their phase-delayed cycle (Council on School Health, 2014). It is difficult to determine mechanisms that underlie the phase delay versus those that are derivative, but many have been investigated including melatonin secretion, puberty, altered endogenous pacemakers and increased sleep drive (Carskadon et al., Reference Carskadon, Acebo and Jenni2004). Furthermore, the interrelationships between, melatonin, hormones, sleep drive (Council on School Health, 2014), evening circadian preference and extended intrinsic periods add to the explanatory complexity of circadian changes in adolescents, later circadian phase are other hypotheses that are supported (Crowley et al., Reference Crowley, Acebo and Carskadon2007). Inherent delayed circadian rhythms support the push for later school start times, as adolescents are predisposed to perform better later in the day (Wolfson et al., Reference Wolfson, Spaulding, Dandrow and Baroni2007).
Earlier school start times are associated with worse outcomes, such as daytime sleepiness and insufficient sleep (Carskadon et al., Reference Carskadon, Wolfson, Acebo, Tzischinsky and Seifer1998) that manifest in increased dropout rates, tardiness (Wolfson et al., Reference Wolfson, Spaulding, Dandrow and Baroni2007), decreased concentration and poorer school and standardized test performance (Owens, Reference Owens2010). Additionally, later school start times correlate with less fatigue and daytime sleepiness (Owens, Reference Owens2010) and decreased incidence of depressed mood (Owens et al., Reference Owens, Belon and Moss2010). A mere 30-minute delay in school start time affects emotional valence as evidenced by decreased depressed mood scores among adolescents (Boergers et al., Reference Boergers, Gable and Owens2014). Likewise, a 1-hour delay in start times increased total sleep time and coincided with decreased depression, anxiety and stress (Chan et al., Reference Chan, Poon, Leung, Lau and Lau2018). A natural experiment and quasi-experimental study among South Korean 10th–12th grade students found that a school start time delay in approximately 45 minutes caused an increase of sleep of 10–16 minutes with a significant improvement of sleep satisfaction and stress (Kim, Reference Kim2022).
In addition, sleep loss in adolescents is associated with a litany of untoward health and safety consequences including, but not limited to, risky behaviors, motor vehicle accidents, obesity, type II diabetes and mood disorders (Wolfson et al., Reference Wolfson, Spaulding, Dandrow and Baroni2007; Owens et al., Reference Owens, Belon and Moss2010). Jung and Fenelon conducted a quasi-analysis of 1,133 South Korean adolescents spanning grades 7–12 that showed that later school start times between 30 and 90 minutes caused increased sleep of 19 minutes and decreased frequency of smoking and drinking (Jung and Fenelon, Reference Jung and Fenelon2024). This impact was more pronounced as one-year post-implementation, the increase in sleep was only 7 minutes and substance abuse was less pronounced (Jung and Fenelon, Reference Jung and Fenelon2024).
The U.S. Department of Education states that out of all public high schools in the U.S., more than 42.5% started before 8 a.m. in 2011–2012 (National Center for Education Statistics, 2024a) and 31.8% in 2017–2018 (National Center for Education Statistics, 2024b). The average school start time for high school in 2020–2021 was 8:07 a.m. (Taie and Lewis, Reference Taie and Lewis2022). Between 2011–12 and 2017–18, 10.7% of US public schools adopted delayed start times (National Center for Education Statistics, 2024b). Herein we explore the effects of sleep loss related to the development of depression and anxiety during adolescence and obstacles to implementing later school starts.
The role of sleep and circadian rhythms in the development of depression during adolescence
Mental disorders are increasingly common in today’s society, especially among adolescents. Adolescence is an at-risk developmental period during which individuals are particularly vulnerable to developing mental health conditions and mood disorders (Kessler et al., Reference Kessler, Avenevoli and Ries Merikangas2001). During the COVID pandemic, UNICEF estimated that 1 in 7 adolescents struggled with mental wellness (UNICEF, 2024). Importantly, the societal implications of mental health conditions are greater than those of other chronic illnesses for several reasons including prevalence and age of onset (Kessler et al., Reference Kessler, Avenevoli and Ries Merikangas2001).
Adolescents being evaluated for insomnia had higher incidences of mental health diagnoses, specifically anxiety, Attention deficit hyperactivity disorder (ADHD) and affective disorders (Van et al., Reference Van, Becker and Byars2019). Inadequate sleep is also coupled with a higher incidence of depressed moods compared to well-rested (Carskadon et al., Reference Carskadon, Acebo and Jenni2004). A meta-analysis examining the directional relationship between sleep disturbance and depression among adolescents found that sleep disturbance is indeed a precursor of depression (Lovato and Gradisar, Reference Lovato and Gradisar2014). Moreover, adolescents with depression had higher instances of sleep disturbances as well (Lovato and Gradisar, Reference Lovato and Gradisar2014). Conversely, when school as delayed 30 minutes, students were able to get more sleep (Owens et al., Reference Owens, Belon and Moss2010). This increase in sleep correlated to improved scores on the Depressed Mood Scale, specifically less feelings of irritation and annoyance (Owens et al., Reference Owens, Belon and Moss2010). There were higher Depressed Mood Scale scores among adolescents who reported getting less sleep (Owens et al., Reference Owens, Belon and Moss2010). Less depressed moods were also reported among high-school students who got 27 more minutes of sleep on a school night (Bonnar et al., Reference Bonnar, Gradisar, Moseley, Coughlin, Cain and Short2015). Students were able to get more sleep from sleep education programs and parental involvement without a school start time change (Bonnar et al., Reference Bonnar, Gradisar, Moseley, Coughlin, Cain and Short2015).
The impact of shorter sleep duration is more profound on positive affect versus negative affect (Shen et al., Reference Shen, van Schie, Ditchburn, Brook and Bei2018). A meta-analysis by Short and colleagues (Short et al., Reference Short, Booth, Omar, Ostlundh and Arora2020), found that shorter sleep durations conferred a 2-fold reduction in positive affect, but also increased odds of anger, depressed mood, negative affect and anxiety. Suzuki et al. found that when students aged 12–18 assessed their sleep as “very bad” versus “very good,” they had a significantly greater loss of positive emotion (Suzuki et al., Reference Suzuki, Kaneita and Osaki2011). Fuligni et al. found that among students aged 12–20, higher levels of anxiety was associated with both “insufficient and excessive sleep” (Fuligni et al., Reference Fuligni, Bai, Krull and Gonzales2019). It is important to note that “excessive” sleep was measured through an individualistic approach, depending on the individual’s optimal sleep duration recorded (Fuligni et al., Reference Fuligni, Bai, Krull and Gonzales2019). When it comes to sleep duration and mood, however, it is not just a lack of sleep that is troublesome; sleeping a duration of longer than an adolescent’s optimal sleep duration is associated with increases daily levels of distress and loss of positive emotion (Suzuki et al., Reference Suzuki, Kaneita and Osaki2011; Fuligni et al., Reference Fuligni, Bai, Krull and Gonzales2019).
The role of sleep and circadian rhythms in the development of anxiety during adolescence
The effects of anxiety appear to be stronger than depression among mental health conditions prevalent in adolescent populations (Kessler et al., Reference Kessler, Avenevoli and Ries Merikangas2001). Reduced sleep duration is associated with increased anxiety and suicidal ideation (Sarchiapone et al., Reference Sarchiapone, Mandelli and Carli2014). There is a strong negative relationship between sleep duration and depression and anxiety (Short et al., Reference Short, Gradisar, Lack, Wright and Dohnt2013).
Globally, numerous studies support this conclusion. Compared to 10 hours in bed over five days, adolescents with only 6.5 hours had more feelings of anger and anxiety, but not depression (Baum et al., Reference Baum, Desai, Field, Miller, Rausch and Beebe2014), and there was a negative relationship between time in bed and anxiety on school nights (Díaz-Morales, Reference Díaz-Morales2016). The effects of sleep time and mental health are cross-cultural. Greek adolescents showed a negative correlation between short and long-term anxiety and sleep duration (Lazaratou et al., Reference Lazaratou, Anagnostopoulos and Vlassopoulos2013). Shorter sleep durations (< 7 hours) were associated with anxiety and depression in Chinese and Japanese youth (Suzuki et al., Reference Suzuki, Kaneita and Osaki2011; Liu and Zhou, Reference Liu and Zhou2002). A cross-sectional observational study by Oginska and Pokorski found that shorter sleep durations were correlated to higher scores of anxiety and depression; it is important to note that this correlation was stronger among adolescents than adults (Oginska and Pokorski, Reference Oginska and Pokorski2006). The exact amount of time of sleep was not recorded (Oginska and Pokorski, Reference Oginska and Pokorski2006).
However, the benefits of starting school later are not guaranteed. A large-scale quasi-experimental study by Hinrichs found that between 1993 and 2003, there was no significant effects of delaying school start times and standardized test scores results among 11th and 12th graders (Hinrichs, Reference Hinrichs2011). In a cohort study, total sleep time was not associated with anxiety measured by the Spence Children’s Anxiety Scale in 145 healthy adolescents; however, sleep quality measured by the Pittsburg Sleep Quality Index was significantly associated with higher levels of anxiety in both healthy adolescents and adolescents with chronic fatigue syndrome/myalgic encephalomyelitis (Josev et al., Reference Josev, Jackson and Bei2017). It is important to note, however, that this study is among few that support this claim. Furthermore, sleep timing in adolescents vary significantly as it is a complex relationship between environmental factors, culture and geographic location. Sleep quality is a subjective measure so the perception of sleep quality may differ among individuals.
Obstacles to implementing later school starts
Legislation supporting delayed school start times is sparse. To date, California and Florida are the only states to hpass a directive with prescribed school start times. California’s Education Code 46148 on student attendance states that high schools are required to start no earlier than 8:30 a.m. and middle school no earlier than 8 a.m. (State Policy Database, 2024a). Florida’s HB733 states that middle school will begin no earlier than 8 a.m. and high school will begin no earlier than 8:30 a.m. Maryland’s Code 7–122 Orange Ribbon for Healthy School Hours Certification encourages schools to adopt later school start times by incentivizing districts with an Orange Ribbon certificate if certain requirements are achieved. The certificate signifies public acknowledgement of the school’s commitment to student health and well-being (State Policy Database, 2024b). Other states, such as Connecticut, New Jersey and Washington State have implemented pilot programs, proposed legislation and/or made recommendations that advocate for later start times – none of which have become state law yet. Many states have attempted to pass bills that mandate later school start times but have not been passed – including Alabama, Rhode Island, Massachusetts and Texas. Concerted efforts to delay school start times may be more productive if instituted at the local and district levels as implementation at the state level have proven difficult to enact.
Delaying school start times also presents significant operational, logistical and financial challenges, in addition to opposition from various stakeholders (Walker, Reference Walker2022). These challenges have the potential to impact families unequally, particularly in lower-income households, for whom workplace flexibility is more limited. A delayed start time may force parents in these families to adjust their work schedules or find alternative childcare options, which will add strain to their daily routines. From an operational perspective, changing school start times requires adjustments to bus routes, after-school programs and other extracurricular activities. Additionally, school staff contracts may need to be renegotiated to accommodate new hours, which further complicates this process. Implementing these changes would involve a substantial financial burden on the city and/or state. A more feasible plan to help adolescents obtain more sleep was demonstrated in a randomized control trial by Bonnar et al., which implemented a motivation Sleep Education Program among 192 high-school students. This program helped students obtain a 27-minute increase in sleep duration on school nights without a change in school start times (Bonnar et al., Reference Bonnar, Gradisar, Moseley, Coughlin, Cain and Short2015).
Another concern is the potential safety risk for students who arrive at school early when later school start times are implemented due to parents who have conflicting schedules with the new school start times. This would pose a concern for students’ safety, especially for younger students.
Resistance to change exists even when unfounded. For example, a voiced concern of parents is that adolescents will just go to sleep later when accommodated with later school start times. However, students with later school start times were found to get more sleep as the time they went to sleep at night remained relatively the same, therefore they had an increased time in bed (Wolfson et al., Reference Wolfson, Spaulding, Dandrow and Baroni2007; Council on School Health, 2014). While there are clear benefits for later school start times, practicality and safety concerns must be carefully considered when implementing and advocating for these changes.
Conclusions and impact statement
Later school start times are associated with more sleep and fewer negative health outcomes (Yip et al., Reference Yip, Wang, Xie, Ip, Fowle and Buckhalt2022; Ke and Ns, Reference Ke and Ns2016). Despite much literature supporting later school start times and its positive impact on adolescents, there is a need for more research and legislature to implement policies into a vast majority of the U.S. While the path to statewide and nationwide change presents challenges including operational and financial concerns, a continued focus on educating families, teachers and policymakers with ongoing research and scientific investigations are vital to support this initiative (Figure 1). These changes would better the health of adolescents and expand to broader social implications – benefit a more productive and healthy future generation.
Data availability statement
Data availability is not applicable to this article as no new data were created or analyzed in this study.
Acknowledgements
We thank Suzanne Fricke for her guidance in the research process – to better navigate academic databases, refine the literature search and improve the quality of sources used.
Author contributions
L.Y. Lee: Conceptualization, Investigation, Writing – Original Draft/Review and Editing, Funding acquisition (if from ESFCOM). A.J. Jolliffe: Writing – Review and Editing C.J. Davis: Supervision, Writing – Review and Editing.
Financial support
Funding for this work was provided by (ESFCOM or The Journal).
Competing interests
The authors have no conflicts of interest to declare in the writing or content of this publication.
Ethical standards
This manuscript constitutes original writing that has not been published or submitted in part or in full elsewhere.
Open access
