
(© digitalskillet1 - stock.adobe.com)
In A Nutshell
- When two Swiss secondary schools let students choose their own start time, 95% picked the later option, skipping the early morning slot on most school days.
- Students slept about 45 minutes longer on school nights, and standardized test scores in math and English rose significantly.
- Some health measures improved too: fewer students reported trouble falling asleep, and the share with clinically low quality-of-life scores dropped from 25% to 17%.
- No outcomes worsened, and students completed the same number of required classes under the flexible model as before.
Every parent knows how difficult it can be to get teenagers up and moving in the morning. Now, research from Switzerland suggests there may be more to this tendency than mere laziness or lack of discipline. When students were given a real, no-strings option to start school later, 95% of them took it, and their standardized test scores in math and English went up.
Researchers tracked students at two secondary schools before and after those schools introduced a flexible scheduling model. Under the new system, students could choose when to take certain required elective classes. Pick a noon or afternoon slot instead of a morning one, and the school day started later, no missed class, no make-up work, no catch. The results, published in the Journal of Adolescent Health, were clear: more sleep, some meaningful health improvements, and higher scores on standardized academic tests.
This is not entirely surprising to sleep scientists. Teenagers are not being dramatic when they say they can’t fall asleep at 9 p.m. During puberty, the brain’s internal clock shifts biologically toward later sleep and wake times. A 7:30 a.m. school bell does not reset that clock; it just shortens the night. Chronic sleep loss in adolescents is linked to lower grades, impaired attention, and higher rates of risky behavior. The question this study took on was more practical. If schools actually gave teens a workable way to sleep later, would they use it?
When Teens Get Flexible School Start Times, They Vote With Their Alarm Clocks
Before the new model launched, the typical starting time at these two Swiss schools was around 7:40 a.m. After one year of flexible scheduling, the median start time had shifted to 8:18 a.m., a nearly 38-minute delay. Of the roughly 750 students surveyed, only 5% never used the later-start option. More than 87% chose to skip the early morning slot on three or more days per week.
That near-universal uptake stood out to the researchers. Earlier studies in Germany had tried something similar, allowing students to skip a first-period self-study session, but those students only chose the later start about twice a week. The catch was that missed sessions had to be made up within the same week, and transportation logistics added another barrier. Subtract the penalties, and teenagers behave very differently. “When offered the choice to start school later, students took it,” the researchers wrote.
With start times pushed back, students woke up about 40 minutes later on school mornings. Their bedtimes barely changed, which meant the extra morning time went almost entirely toward sleep. On school days, students slept roughly 45 minutes longer after the flexible model took effect.
More Sleep, Higher Test Scores, and Nothing Got Worse
The academic results were notable. Standardized test scores in math and English rose significantly after the schedule change. Teacher-assigned grades in those subjects did not shift as clearly, but the researchers pointed out that teacher grades involve subjective factors, including participation, effort, and classroom presence, and may not capture real changes in learning as precisely as standardized tests do. In the smaller group of students tracked across both years, increases in sleep were modestly linked to improvements in math grades specifically.
Health outcomes shifted in encouraging ways, though not uniformly. Students reported fewer problems falling asleep, and the proportion with clinically low scores on a standard measure of health-related quality of life dropped from about 25% to 17%. Daytime sleepiness, depressive symptoms, and caffeine consumption held steady. So while the flexible schedule was not a cure-all, it moved several meaningful needles without pushing any in the wrong direction.
That last point deserves some weight. A common concern about giving teenagers more scheduling flexibility is that they will abuse it, sleeping through important coursework, losing structure, or falling behind. At these schools, none of that happened. The number of elective classes students completed did not decrease. Participation rates in the follow-up survey reached 98.5%.
Why This Approach Could Work Beyond Switzerland
Pushing school start times later is not a new idea, and efforts to do so have run into predictable resistance: bus routes, after-school sports, working parents, and employer schedules all get disrupted when a school shifts its entire day. The flexible model sidesteps most of that friction. Because only the elective portion of the schedule moves, the core structure of the school day stays intact. No bus company has to reroute. No coach has to reschedule practice. Students gain the option to sleep in, and those who want or need an early start can still have one.
The study was not without limitations. The new school model brought other changes alongside flexible scheduling, including more self-directed learning and greater student autonomy, so sleep alone may not explain everything. The schools were also located in a single Swiss canton, with students mostly between 13 and 15 years old, so the findings may not transfer directly to every cultural or educational context.
Still, the core takeaway is straightforward. Teenagers are not wired to thrive at 7:30 a.m., and when schools stop asking them to, the evidence suggests they sleep more, key health measures improve, and they perform better on academic tests. This study adds a practical model for how to actually make it happen.
Paper Notes
Limitations
The flexible scheduling model was introduced alongside broader changes to how these schools operated, including more individualized learning formats and closer teacher guidance. Those changes may have contributed to the improvements observed, making it difficult to attribute results to sleep gains alone. There was no control group at a school maintaining a traditional schedule, which would have allowed for cleaner comparisons. Because all students aged by one year between the two surveys, the natural decline in adolescent sleep duration over time may have partially offset gains from the later start. Sleep data were self-reported rather than measured objectively through tools such as actigraphy. The sample was limited to students aged 12 to 16 in a single region of Switzerland, and effects may differ for older teens or students in different cultural settings. The study also did not directly assess whether the flexible model was easier to implement than a straightforward schedule delay, even though that practical advantage was part of the original rationale.
Funding and Disclosures
No external funding was obtained for this study. The authors report no conflicts of interest.
Publication Details
The study was conducted by Joëlle N. Albrecht, Ph.D., Alessa Risch, M.Sc., Reto Huber, Ph.D., and Oskar G. Jenni, M.D., affiliated with the Child Development Center at the University Children’s Hospital Zurich, University of Zurich, Switzerland. Additional affiliations include the Department of Psychology at the University of Konstanz, the Children’s Research Center at the University Children’s Hospital Zurich, and the Department of Child and Adolescent Psychiatry at the University Hospital of Psychiatry, University of Zurich. | Title: “The Power of Flexible School Start Times: Longitudinal Associations with Sleep, Health, and Academic Performance” | Published in the Journal of Adolescent Health (2026), Elsevier, on behalf of the Society for Adolescent Health and Medicine. | DOI: https://doi.org/10.1016/j.jadohealth.2026.01.011







