Problematic Sleep among Adolescents

By way of illustration, Ohayon and Guilleminault (1998) reviewed all epidemiological surveys of sleep disorders published over a 20 year period. They were looking for studies on insomnia, excessive sleepiness, sleep-disordered breathing and parasomnia: not even one had adolescent sleep problems as a focus.

Adolescence is accompanied by various biological changes including a modification of sleep-wake regulation and sleep patterns (e.g. decrease in the amount of Delta sleep, reduced REM latency). According to Carskadon (1990), adolescents require more sleep than prepubertal youths but frequently get less sleep than they need. Transition to an earlier school-start time, along with sleep phase delay, significantly affects teenagers' sleep quality, sleep/wake schedule, and daytime behavior. The combination of the phase advance, late night activities or jobs, and early morning school demands can significantly constrict hours available to sleep (Wolfson, 1985). But do these changes attributable to adolescence translate into higher prevalences of sleep disorders? It is unclear. Our knowledge about normal as well as abnormal sleep patterns in school-aged children, particularly adolescents, is represented by a relatively small body of literature (Dahl, 1996; Levy et al., 1986; Morrison et al., 1985; Richman, 1987; Wolfson and Carskadon, 1998).

Other than these two latter studies, neither of which analyzed data separately for adolescents, most of the research has been school-based rather than community-based. It is difficult to interpret the prevalence rates due to variability in operational definitions, sample populations, and assessment techniques.

To our knowledge, only one community-based, epidemiological study has been carried out using definitions of sleep problems based on either DSM diagnostic criteria (APA, 1994) or the International Classification of Sleep Disorders (ASDA, 1991). Morrison and his co-authors (1992) used DSM-III and the DISC in their study of 15-year-old New Zealand adolescents. To meet criteria, youths had to report a sleep problem at least four times per week for four weeks. Using this definition, 33% of the sample overall had at least one sleep problem in the previous four weeks. The prevalence of those who had difficulty falling asleep, staying asleep, or waking too early (i.e. insomnia) was 15.2%.

RESEARCH

Our purpose here was to provide additional data on the prevalence and patterns of problematic sleep among adolescents using DSM criteria.

RESULTS

Differences between adolescents and young adults:

• 4% of the adolescents met DSM-IV criteria for insomnia. This was comparable to the young adult group.
• At least one insomnia symptom was reported by nearly 25.7% of the adolescents; this was comparable to the rate for young adults. The rate of nearly 26% is comparable to that reported in U.S. school-based samples (Dahl, 1996; Roberts et al., in press) and in a French, school-based study (Choquet et al., 1988).
• Not surprisingly, the prevalence of insomnia symptoms was quite high in those with anxiety or affective disorders.Nearly three-fourths of both adolescents and young adults with a DSM-IV diagnosis of an anxiety disorder had at least one symptom of insomnia. For those with an affective disorder, 68% of adolescents and 77% of young adults reported at least one symptom of insomnia.
• The prevalence of circadian rhythm disorders was very low among adolescents (0.4%). Contrary to other findings, we did not find a higher rate of circadian rhythm disorders in adolescents than in young adults. We observed, however, a higher occurrence of indicators of such disorders in adolescents (for example, differences between the real and wished sleep-wake schedule; extra sleep on days off; difficulties to get up in the morning). This is insufficient to establish a sleep phase delay or advance syndrome which requires at least daytime repercussions on functioning. A regular sleep-wake schedule, such as required by school or work schedules, protects against such a desynchronization by daily resetting the internal pacemaker.
• In this way, sleep/wake schedule data obtained in this study are different from that reported in previous work with American adolescents (Carskadon, 1990; Wolfson and Carskadon, 1998). In these studies, the adolescents surveyed had a later bedtime and an earlier rise time than in our European sample of adolescents and consequently, a shorter sleep time.
• However, we found a similar migration of bedtime across age. School schedules in USA high schools are different from those of European schools. For example in France, school start time is around 8:30 AM and school ends between 4:00 and 5:00 PM. Adolescents never have school on Wednesday but they go to school on Saturday. In Germany, adolescents go to school from 8 AM to 1 PM. On rare occasions, they have school in the afternoon. Later school start time appears to be more congruent with adolescent rhythm of life.
• The results of this study clearly show that sleep habits change considerably between late adolescence and young adulthood. Indeed, usual bedtime and wake up times are earlier in late adolescence than in young adulthood.
• Sleep is also longer and less disrupted in adolescents.
• Not so surprising, more adolescents than young adults would like to wake up later in the morning. The extra amount of sleep got on weekends and days off also is more important in adolescents than in young adults.

HOW DO OUR DATA COMPARE TO THOSE FROM OTHER STUDIES?

As noted earlier, only one community-based, epidemiological study has been carried out using definitions of sleep problems based on either DSM or ICSD diagnostic criteria.

• Morrison and his co-authors (1992) report 33% of the sample overall had at least one sleep problem in the previous four weeks. The prevalence of those who had difficulty falling asleep, staying asleep, or waking too early was 15.2%.
• A survey by Yang et al. (1987) of 12 to 18 year-olds in China found 14.9% of these adolescents reported difficulty falling asleep at least four nights in the past month. Beyond these three studies, the results are quite disparate, reflecting diverse samples, diverse study designs, and diverse measures of disturbed sleep.

CONCLUSIONS

Most of the studies have used measures of sleep problems idiosyncratic to their particular study. Not surprisingly, since no two studies have used the same definitions or measures, the rates of sleep problems reported have varied widely. Studies have defined the need for more sleep, wish for more sleep, daytime sleepiness, nightmares, non-restorative sleep, grinding teeth, difficulty maintaining sleep, difficulty initiating sleep, early morning waking, sleep talking, sleep walking, and more as being reflective of " having a sleep problem". Perhaps the most frequently used measures have been those directed at problems with insomnia, using some combination of difficulties falling or staying asleep or waking too early.

Prevalences from these studies range from 6% to 35% (Anders et al., 1978; Andrade et al., 1993; Coren, 1994; Kahn et al., 1989; Stoleru et al., 1997). However, most fall in the 11-15% range (Levy et al., 1986; Kirmil-Gray et al., 1984; Manni et al., 1997; Morrison et al., 1992; Strauch and Meier, 1988; Yang et al., 1987). Our prevalence for any symptom of insomnia was 30%, clearly in the upper range of reported prevalences.