Introduction: Sleep habits change during adolescence both for biological and psychological/social reasons. Longitudinal studies on suchhabits are, however, scarce. The aim of the present study was toinvestigate the longitudinal changes of subjective sleep habits in Swedish adolescents.

Method: The present study was part of a larger investigation mainlyon food habits that originally included all (about 2500) 14 year oldfrom three representative middle-sized Swedish towns. The adolescents filled in questionnaires at 14, 15 and 18 years of age. The 638 adolescents that answered all the questions on their sleephabits at each of the three occasions were included in the present study.

Results and discussion: The adolescents developed later weekday bed- and rise times with increasing age (bedtimes from 22:35 h to23:20 h; risetimes from 06:50 h to 07:05 h). Consequently, time inbed on weekday nights decreased (from 8:15 h to 7:45 h). On weekends the adolescents went to bed later (around 2 h) and stayed in bed longer (around 1:20 h) than during the weekdays. Weekend time in bed decreased with increasing age. Differences between genders indicated that the changes observed were more obvious among boys. A dropout analysis implied that those adolescents that only participated at one occasion (at the age of 14) or at two occasions (at 14 and 15 years of age), respectively, went to bed later, woke up later and had shorter sleep compared to those who participated at all three occasions. Hence, the results might not be completely representative.

Conclusions: A delay of bedtimes and of shortening of sleep duration as a function of age was observed. Weekend changes showed a delay of bedtimes and a (presumably) compensatory increase in sleep duration. Boys seemed 'worse off'.