Should Teens Be Able to Sleep Later?

From Medscape Pediatrics > Viewpoints
William T. Basco, Jr., MD

Impact of Delaying School Start Time on Adolescent Sleep, Mood, and Behavior
Owens JA, Belon K, Moss P
Arch Pediatr Adolesc Med. 2010;164:608-614.

Study Summary
Owens and colleagues review evidence suggesting that biological changes during puberty result in adolescents shifting their sleep periods (ie, falling asleep later in the evening), but they still require the same amount of sleep (approximately 9 hours).
Many studies have shown that adolescents are generally sleep-deprived, and sleep deprivation is associated with worse academic performance.
This study was conducted in a single school that had both boarding students and day students in 9th-12th grades (total students, 357; 81.5% boarding students).
The school had historically begun morning classes at 8:00 am, but during the trial period (January and February 2009) the start time was moved to 8:30.
The evening routines for boarding students did not change during the trial period, with "lights-out" at graduated times depending on grade level.

The investigators collected self-completed survey data pre- and postintervention, including a survey of sleep habits, a survey that measured sleepiness, and another survey that assessed whether the student felt that he or she was a "morning" or "evening" person. The investigators also collected objective data on visits to the school health center, missed appointments, tardiness, and breakfast consumption.

More than 75% of the student body completed the surveys (278 students). Fifty-four percent of the students were girls, the mean age was 16.4 years, and > 80% had grades that were "B" or better.
The change in classroom start time was associated with an average increase in sleep duration of 45 minutes (95% confidence interval, 27-49). Most of the difference was accounted for by a wake time that was an average of 30 minutes later.
In addition, students went to bed an average of 18 minutes earlier. On average, even with the start-time change, 12th graders slept 40 minutes less than 9th graders.
The sleep duration of boarding students on school nights did not different significantly from that of day students.
Overall, students were much less likely to report rarely/never getting enough sleep or dissatisfaction with the amount of sleep after the time change.
The students' assessments of their own sleepiness declined after the change. Health outcomes also improved, with a reduction in the percentage of students who visited the health clinic (15.3% at first survey compared with 4.6% at second survey, P = .03).
Fewer missed morning appointments, fewer requests for late passes, and more than a 50% reduction in the percentage of students who used the health center for afternoon respite were also observed after the change in school start time.
The investigators concluded that even a modest delay in the start of the school day is associated with improved mood, health, and alertness in adolescents.

Viewpoint
In quasi-experimental designs (ie, nonrandomized, as in this case), 1 way that investigators can make the case that the intervention was the cause of the outcome (and not just associated with it) is by finding associations between the change and multiple outcome measures.
In this case, credit goes to the investigators for adding objective outcomes, such as health center utilization, to self-assessment surveys. Looking at objective changes in grades would have been another objective measure, but this was probably precluded by the short duration of the experiment.
It is worth noting that this is an independent school population, and > 80% of the students had a structured evening routine.
It is difficult to know whether the same benefits would be accrued by students who have more varied routines or by students in other settings with more disparate economic and educational backgrounds.