Abstract
Objective:
Recent attention to pervasive sleep deficits in U.S. adolescents has focused on sleep patterns and insomnia, but there are limited data on the prevalence and correlates of hypersomnolence symptoms.
Method:
The sample included 6,483 adolescents ages 13–18 years who were interviewed directly and had parent reports in the National Comorbidity Survey Replication Adolescent Supplement (NCS-A), a nationally representative sample of US youth. Information on sleep patterns/symptoms that were collected in the interview was used to determine the population prevalence of DSM-5 criterion-A defined hypersomnolence and component symptoms. Logistic regression analyses were used to examine associations between sleepiness and sub-symptoms of hypersomnolence with weekday/weekend bedtime, sleep duration, mental disorders and psychotropic medication use.
Results:
41.5% of US adolescents reported feeling sleepy during the daytime and 11.7% met criteria for hypersomnolence. The prevalence of hypersomnolence varied depending on age (p<0.001) and was more common in adolescent girls (OR=1.40; 95% CI=1.09–1.78). Excessive sleepiness and hypersomnolence symptoms were associated with reduced sleep duration and delayed bedtimes on weekdays and weekends Hypersomnolence was significantly associated with insomnia (OR=2.45; 95% CI=1.87–3.21) and mental disorders (OR=1.99; 95% CI=1.42–2.77). After accounting for insomnia, hypersomnolence was no longer associated with use of psychotropic medication (OR=1.61; CI=0.97–2.66).
Conclusion:
Among adolescents with adequate sleep duration, we found that 11.7% still report symptoms of hypersomnolence. The strong association between hypersomnolence with insomnia suggests that sleep disorders in adolescents may fluctuate between over- and under-sleeping. Potential mechanisms underpinning the strong associations between sleep disturbances and mental disorders should be further pursued and may provide insight into prevention efforts.
Keywords: Hypersomnia, sleepiness, adolescents, sleep disturbance, insomnia
Introduction
Sleepiness and inadequate sleep in U.S. teens are associated with deterioration in school performance, health and safety1,2 and are significant public health concerns. Sleepiness in adolescents and young adults contributes to motor vehicle accidents, the leading cause of mortality in these age groups3. Improving sleep health, reducing sleepiness and increasing sleep opportunity are important areas for intervention to improve overall adolescent health and wellbeing 1. However, most community-based research on sleep in adolescents has focused on inadequate sleep time, and there are very limited data on the prevalence and correlates of sleepiness and hypersomnolence in U.S. adolescents.
The Diagnostic Statistical Manual- 5th Edition (DSM-5) defines hypersomnolence as symptoms of excessive sleepiness associated with lapses into sleep, feeling unrefreshed despite adequate sleep time and difficulty waking in the morning which occur despite adequate nighttime sleep4. A previous study that examined somnolence in a single school district in the U.S. revealed that as many as 45% of teenagers complained of sleepiness during the day and 16% noted falling asleep in class5. Similarly, 51% of U.S. teens reported sleepiness or tiredness during the daytime in an online survey conducted by the National Sleep Foundation6. However, these studies did not examine the influence of sleep duration, psychiatric comorbidity and psychotropic medication use. Additionally, the impact of inadequate nighttime sleep duration on excessive sleepiness and other symptoms of hypersomnolence is under explored in samples of community dwelling adolescents.
Previous analyses of the National Comorbidity Survey Replication Adolescent Supplement (NCS-A), a nationally representative population based sample showed that the average sleep duration of U.S. adolescents is less than even the minimum recommended 8 hours of sleep each night7, primarily due to later weekday bedtimes8. In turn, later bedtimes and shorter sleep durations are associated with significant psychiatric comorbidity8. However, the prevalence of sleepiness, hypersomnolence and its component sub-symptoms and their association with mental disorders, psychotropic medication use, sleep duration and insomnia in the NCS-A has not been previously explored.
In the current study, we utilized data from the NCS-A to examine 1) the association between excessive sleepiness and other component symptoms of hypersomnolence defined in accordance with DSM-5 criterion A with weekday and weekend bedtime/sleep duration 2) determine the population prevalence and associated socio-demographic characteristics of hypersomnolence among those with adequate nighttime sleep duration in a representative sample of adolescents; and 3) examine the association of hypersomnolence with insomnia, mental disorders and psychotropic medication use.
Methods
Sample:
The NCS-A was a face-to-face computer assisted interview of adolescents that was supplemented by a self-administered questionnaire (SAQ) mailed to a biological parent/guardian9. The NCS-A included a nationally representative sample of 10,123 adolescents aged 13 to 18 years who were living in the continental U.S. and was carried out between February 2001 and January 2004. It included dual household and school subsamples with a combined response rate of 82.9%. The response to the parent/guardian SAQ was 83.3%. The analytical sample of this study included information from 6,483 adolescents who had data from both the adolescent interview and SAQ.
Measures:
Self-reported information on sleep related questions including sleep duration and sleep disturbance was collected in the interview section of the Chronic Conditions in the adolescent in person interview. The possible contribution to hypersomnolence from inadequate sleep time was assessed utilizing the data from self-reported total sleep duration and bedtime on weekdays and weekends. The NCS-A did not include any objective measures of sleep time.
While the DSM-5 requires a main sleep period lasting at least 7 hours before hypersomnolence can be diagnosed, the American Academy of Sleep Medicine consensus guidelines recommend that teenagers obtain at least 8 hours of sleep each night10. The National Sleep Foundation expert panel recommendations suggest that 13-year-old obtain at least 9 hours of sleep11. In the current study, 13 and 14 year-olds obtaining less than 9 hours of sleep and 15–17 year-olds obtaining less than 8 hours of sleep per weekday were considered to be obtaining inadequate sleep10.
In this study hypersomnolence was defined in accordance with the DSM-5 Criterion A, with modifications. Subjects who endorsed problems with feeling sleepy during the daytime for at least 2 weeks were considered to have “daytime sleepiness.” Respondents with adequate nighttime sleep duration who had one or more of the following sub-symptoms, at least sometimes or often, in conjunction with a complaint of daytime sleepiness, a) a tendency to fall asleep during the daytime; b) not feeling rested upon awakening; or c) difficulty waking in the morning, were considered to have “hypersomnolence”.
“Insomnia” was defined as 1) difficulty getting to sleep, 2) staying asleep or 3) waking too soon. DSM-5 criteria B through F which include frequency specifiers that the complaints occur at least three times per week for three months, result in significant impairment or distress and are not better explained by other sleep or medical disorders were not part of the hypersomnolence and insomnia definitions used in this study.
Socio-demographic information on age, sex, race/ethnicity, parent education, and urbanicity was obtained during adolescent interview. The parent/guardian indicated the marital status of parents and family income which was normed against federal poverty levels. Self-reported height and weight were used to calculate sex-specific body mass index (BMI) for age according to Center for Disease Control (CDC) guidelines. Duration of symptoms was ascertained based on self-report of length of problems with daytime sleepiness and/or insomnia. Adolescents were administered the modified World Health Organization Composite International Diagnostic Interview (WHO-CIDI) Version 3.0. This instrument collected information required by the Diagnostic and Statistical Manual of Mental Disorders- IV Edition (DSM-IV) criteria for the following five major classes: mood disorder (major depressive disorder and dysthymia, bipolar I/II), anxiety disorders (panic disorder, agoraphobia, social phobia, specific phobia, generalized anxiety disorder, posttraumatic stress disorder, and separation anxiety disorder); attention-deficit/hyperactivity disorder (ADHD) and other behavior disorders (oppositional defiant disorder and conduct disorder); eating disorders (anorexia, bulimia, binge eating), and substance use disorders (alcohol or drug abuse/dependence, nicotine dependence). The diagnostic information was based on adolescent reports for anxiety, mood, eating, and substance use disorders, parent reports for ADHD, and both adolescent and parent reports combined at symptom level using the ‘or’ rule for other behavior disorders.
Psychotropic medication use was ascertained from the Pharmacoepidemiology section of the adolescent interview. We examined antidepressants, antipsychotics and stimulant medications (methylphenidate, amphetamines and mixed amphetamine salts) that had sufficient frequency for meaningful analyses.
Statistical Analysis:
The rates of sleepiness and sub-symptoms of hypersomnolence (symptoms a, b and c described above) in adolescents with and without adequate sleep duration were estimated. The prevalence of hypersomnolence as defined above was calculated. The association between daytime sleepiness and the number of sub-symptoms of hypersomnolence with insomnia, sleep duration and bedtime on weekdays and weekends were examined. For these analyses sleep duration and bedtime were ordinally divided into four groups: sleep durations < 7 hours, 7–8 hours, 8–9 hours, > 9 hours and bedtimes before 10 p.m., 10–11 p.m., 11 p.m.−12 a.m. and after 12 a.m. Logistic regression analyses controlling for age and sex, where sleep patterns were treated as independent variables, were performed. The associations between past year hypersomnolence and various demographic characteristics were calculated utilizing odds ratios (ORs) where demographic characteristics were treated as exposures of interest and hypersomnolence was considered a dichotomous outcome. Separate logistic regression analyses were used to examine the associations between hypersomnolence and DSM-IV mental disorders and psychotropic medication use, in which hypersomnolence status was treated as the main exposure of interest. These models were analyzed progressively to adjust for 1) demographic characteristics 2) other 12-month DSM-IV mental disorders and 3) insomnia.
All statistical analyses were completed with the SAS version 9.3 and SUDAAN version 11 using the Taylor series linearization method to take into account the complex survey design and sample weighting. The Taylor series design-based variance-covariance matrices were used for variance estimates and 95% confidence intervals (CIs). The design-adjusted Wald χ2 tests were used to examine differences by hypersomnolence status. All statistical significance was based on two-sided tests evaluated at the 0.05 level of significance.
Results
Prevalence of hypersomnolence symptoms and sleep patterns
A significant proportion of adolescents reported feeling excessively sleepy during the daytime (41.5%; SE=1.1). Sleepiness was much more common in adolescents who were not obtaining an adequate duration of sleep at night time, compared to those had an adequate duration of sleep (45.5% (SE=1.3) versus. 36.2% (SE=1.4); p<.001) (See figure 1).
Figure 1.
Distribution of Adequate Sleep Duration and Hypersomnolence Symptoms in the National Co-morbidity Survey Adolescent Supplement (n=6,483) Note:a Adequate sleep duration was defined as self-reported weekday sleep duration of > 9 hours if aged 13– 14 years or >8 hours if aged 15 years or older b Three sub-symptoms included 1) falling asleep 2) non-restorative or unrefreshing sleep and 3) difficulty awakening c Hypersomnolence was defined as daytime sleepiness with at least one of the three subsymptoms in adolescents with adequate sleep
Daytime sleepiness (OR= 6.07; 95% CI=5.31 – 6.93) and sub-symptoms of hypersomnolence (OR= 5.46; 95% CI=4.60 – 6.48) were much more common in teens who reported symptoms of insomnia. A total of 69% of subjects reporting sleepiness also reported insomnia. Daytime sleepiness and sub-symptoms of hypersomnolence were more frequent among teens reporting sleep durations of ≤ 8 hours and bedtimes later than 11 p.m. on weekdays as well as in those reporting sleep durations of ≤ 7 hours or less and bedtimes later than 12 a.m. on weekends (Table 1).
Table 1.
Past-Year Hypersomnolence Symptoms and Sleep Patterns, National Comorbidity Survey Replication Adolescent Supplement (NCS-A) (n=6,483)
| Sleep patterns | n | The leading symptom | Number of sub-symptomsa endorsed, besides the leading symptom | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Daytime sleepiness | + none | + any one | + any two | + all three | OR (95% CI)c | ||||
| % (SE) | OR (95% CI)b | % (SE) | % (SE) | % (SE) | % (SE) | ||||
| Total | 6,483 | 41.5 (1.1) | - | 67.5 (1.0) | 14.8 (0.7) | 11.2 (0.6) | 6.6 (0.4) | - | |
| Insomnia | Yes | 2,124 | 69.4 (1.3) | 6.07 (5.31 – 6.93) | 41.7 (1.5) | 24.5 (1.3) | 20.6 (1.3) | 13.3 (1.0) | 5.46 (4.60 – 6.48) |
| No | 4,359 | 27.4 (1.2) | Reference | 80.6 (1.0) | 9.8(0.7) | 6.5(0.7) | 3.1 (0.5) | Reference | |
| Weekday sleep duration | < 7hrs | 1,008 | 61.0 (1.9) | 3.21 (2.40 – 4.28) | 47.1 (2.3) | 18.6 (1.7) | 21.9 (2.7) | 12.4 (1.0) | 3.09 (2.18 – 4.40) |
| 7 – 8 hrs | 1,636 | 43.0 (2.0) | 1.51 (1.19 – 1.92) | 66.1 (2.1) | 15.3 (1.7) | 10.7 (1.1) | 8.0(0.9) | 1.38 (1.07 – 1.78) | |
| 8 – 9 hrs | 3,207 | 36.1 (1.5) | 1.21 (0.95 – 1.54) | 73.4 (1.2) | 14.1(1.0) | 8.4(0.8) | 4.1 (0.6) | 1.02 (0.78 – 1.33) | |
| > 9 hrs | 525 | 30.8 (2.3) | Reference | 76.0 (2.3) | 9.2(1.9) | 8.9(1.9) | 5.9 (1.7) | Reference | |
| Weekend sleep duration | < 7hrs | 771 | 50.4 (2.9) | 1.53 (1.17 – 2.00) | 60.0 (2.7) | 17.6 (2.3) | 15.5 (2.1) | 6.9 (1.2) | 1.32 (1.08 – 1.62) |
| 7 – 8 hrs | 555 | 42.2 (3.5) | 1.06 (0.78 – 1.43) | 66.6 (3.3) | 15.5 (2.2) | 10.3 (2.2) | 7.6 (1.5) | 0.99 (0.73 – 1.35) | |
| 8 – 9 hrs | 2,303 | 39.7 (2.1) | 0.95 (0.76 – 1.18) | 70.4 (2.2) | 14.5 (1.7) | 8.4(1.0) | 6.6(0.7) | 0.82 (0.65 – 1.04) | |
| > 9 hrs | 2,739 | 40.1(1.4) | Reference | 67.5 (1.2) | 14.0 (1.2) | 12.3 (1.3) | 6.2(0.7) | Reference | |
| Weekday bedtime | Before 10pm | 1,370 | 34.6 (2.2) | Reference | 73.5 (2.3) | 13.3 (1.8) | 7.5(1.0) | 5.7 (1.1) | Reference |
| 10pm - 11pm | 2,578 | 39.6 (1.7) | 1.18 (0.93 – 1.50) | 70.5 (1.3) | 13.3 (0.7) | 11.1 (0.9) | 5.1 (0.7) | 1.09 (0.84 – 1.42) | |
| 11pm - 12am | 1,937 | 45.0 (1.7) | 1.48 (1.13 – 1.94) | 63.1 (1.7) | 17.2 (1.3) | 11.8 (1.3) | 8.0(0.9) | 1.53 (1.14 – 2.06) | |
| After 12am | 444 | 55.6 (4.5) | 2.43 (1.53 – 3.84) | 54.3 (4.5) | 14.4 (2.4) | 20.1(2.7) | 11.3 (2.1) | 2.57 (1.72 – 3.83) | |
| Weekend bedtime | Before 10pm | 164 | 31.2 (6.7) | Reference | 73.9 (6.2) | 11.8 (3.2) | 8.0 (3.4) | 6.3(3.8) | Reference |
| 10pm - 11pm | 653 | 30.3 (3.2) | 0.91 (0.42 – 1.97) | 78.6 (2.8) | 10.5 (1.7) | 6.8 (1.6) | 4.1 (0.9) | 0.73 (0.32 – 1.63) | |
| 11pm - 12am | 2,536 | 35.9 (1.6) | 1.12 (0.61 – 2.05) | 72.9 (1.3) | 12.2 (0.7) | 9.4(0.7) | 5.5(0.8) | 0.95 (0.51 – 1.77) | |
| After 12am | 2,806 | 50.1(1.7) | 2.01 (1.14 – 3.55) | 59.5 (1.6) | 18.3 (1.3) | 13.8 (1.0) | 8.4(0.8) | 1.68 (0.91 – 3.12) | |
Note: % standard error indicates prevalence. Odds ratio (OR) with 95% CI was based on logistic regression models using sleep pattern variables as independent variables (one at a time) controlling for age and sex
the leading symptom of hypersomnolence is daytime sleepiness; three sub-symptoms included falling asleep daytime, non-restorative or unrefreshed sleep, and difficulty awaking
PROC RLOGIST daytime sleepiness as dependent variable, treated as dichotomous variable
PROC MULTILOG number of sub-symptoms as dependent variable, treated as ordinal variable (0,1,2,3), using CUMLOGIT as link function
Prevalence of hypersomnolence and its socio-demographic correlates
The prevalence of hypersomnolence and its socio-demographic correlates are outlined in table 2. The overall prevalence of hypersomnolence defined in accordance with DSM-5 criterion A i.e. in those with adequate sleep duration was 11.7% (SE=0.7). The prevalence was 5.0% (SE=0.5) after excluding those with insomnia. Among adolescents with hypersomnolence, symptoms lasted for a median of 4.6 weeks (SE=1.1).
Table 2.
Past-Year Hypersomnolence by Demographic Characteristics, National Comorbidity Survey Replication Adolescent Supplement (NCS-A) (n=6,483)
| Characteristics | Hypersomnolence | |||
|---|---|---|---|---|
| n | %(SE) | OR (95% CI) | ||
| Total | 6,483 | 11.7 (0.7) | - | |
| Sex | Female | 3,333 | 13.5 (1.0) | 1.40 (1.09 – 1.78) |
| Male | 3,150 | 10.1 (0.8) | Reference | |
| Age Group | 13–14 yrs | 2,611 | 5.7 (0.9) | Reference |
| 15–16 yrs | 2,528 | 16.3 (1.3) | 3.20 (2.13 – 4.81) | |
| 17–18 yrs | 1,344 | 12.9 (1.8) | 2.43 (1.49 – 3.96) | |
| Race/Ethnicity | Hispanic | 758 | 11.0 (1.6) | 0.93 (0.60 – 1.45) |
| Non-Hispanic Black | 1,097 | 12.5 (2.0) | 1.07 (0.69 – 1.65) | |
| Other | 371 | 11.7 (2.6) | 1.00 (0.57 – 1.76) | |
| Non-Hispanic White | 4,257 | 11.7 (1.0) | Reference | |
| Family Income | Low | 925 | 10.2 (1.4) | 0.76 (0.55 – 1.04) |
| Low average | 1,218 | 10.8 (1.2) | 0.81 (0.59 – 1.10) | |
| High average | 2,139 | 11.7 (1.3) | 0.89 (0.66 – 1.19) | |
| High | 2,201 | 13.0 (1.0) | Reference | |
| Urbanicity | Metropolitan | 2,645 | 13.1 (0.8) | 1.20 (0.77 – 1.85) |
| Other urban | 2,242 | 10.4 (1.2) | 0.93 (0.59 – 1.46) | |
| Rural | 1,596 | 11.2 (2.0) | Reference | |
| Parental Education | Less than high school | 746 | 8.7 (2.1) | 0.75 (0.45 – 1.25) |
| High school grad | 1,852 | 12.7 (1.4) | 1.14 (0.86 – 1.51) | |
| Some college | 1,364 | 12.8 (1.7) | 1.14 (0.80 – 1.64) | |
| College grad + | 2,521 | 11.4 (0.9) | Reference | |
| Parental marital status | Married/Cohabiting | 4,602 | 11.5 (0.7) | Reference |
| Previously married | 1,009 | 14.0 (2.0) | 1.25 (0.89 – 1.76) | |
| Never married | 308 | 7.9 (2.6) | 0.66 (0.33 – 1.34) | |
| Unknown | 564 | 11.6 (1.7) | 1.01 (0.71 – 1.44) | |
| Body mass index a | Underweight | 137 | 6.2 (2.4) | 0.47 (0.19 – 1.15) |
| Normal | 4,309 | 12.3 (1.0) | Reference | |
| Overweight | 972 | 10.1 (1.8) | 0.80 (0.48 – 1.34) | |
| Obese | 821 | 12.7 (1.7) | 1.03 (0.71 – 1.49) | |
Note: % standard error indicates prevalence. Odds ratio (OR) with 95% CI indicates unadjusted OR based on logistic regression models using demographic characteristics as exposure of interest and dichotomous hypersomnolence as outcome. Family income was assessed relative to the federal poverty line, which was categorized as low income (less than 1.5 times the poverty line), low-average income (1.5−3.0 times the poverty line), high-average income (3.0-6.0 times the poverty line), and high income (greater than 6.0 times the poverty line).
unknown body mass index (n=244) were excluded
Hypersomnolence was more prevalent in adolescent girls compared to boys (OR=1.40; 95% CI=1.09 – 1.78). The prevalence of hypersomnolence varied by age group in both boys and girls, with teens between the ages of 15–16 years endorsing the highest rates of hypersomnolence (0R=3.20; 95% CI=2.13 – 4.81), followed by 17–18 year-olds (OR=2.43; 95% CI=1.49 – 3.96) (Figure 2). Race, family income, urbanicity, parental education, BMI and marital status were not associated with hypersomnolence.
Figure 2.
Prevalence of Hypersomnolence by Age and Sex Note: Prevalence of hypersomnolence was significantly different by sex (p=0.011). There were differences across age groups in both males (p<0.001) and females (p=0.001)
Hypersomnolence, insomnia and mental disorders
The prevalence of various mental disorders by hypersomnolence status, with hypersomnolence defined in accordance with DSM-5 criterion A (daytime sleepiness plus at least one other subsymptom, despite adequate sleep duration), is detailed in table 3. After accounting for demographic characteristics and insomnia, hypersomnolence was significantly associated with having a mental disorder (0R=1.99; 95% CI=1.42 – 2.77). Among individual mental disorders, anxiety, behavior, eating and mood disorders were significantly associated with hypersomnolence after adjusting for demographic characteristics and other DSM-IV disorders. However after further adjustment for insomnia, only behavior (0R=1.69; 95% CI=1.21 – 2.36) and eating disorders (0R=2.20; 95% CI=1.32 – 3.66) were associated with hypersomnolence. These associations were unchanged after additional analyses accounting for BMI (results not shown).
Table 3.
Past-Year DSM-IV Mental Disorders and Psychotropic Medication Use by Hypersomnolence Status, National Comorbidity Survey Replication Adolescent Supplement (NCS-A) (n=6,483)
| OUTCOME | n | % (SE) | Hypersomnolence | Reference (Hypersomnolence = No) | |||
|---|---|---|---|---|---|---|---|
| No | Yes | ||||||
| % (SE) | % (SE) | OR (95% CI) a | OR (95% CI) b | OR (95% CI) c | |||
| Past-year DSM-IV disorder | |||||||
| Anxiety | 1,188 | 19.7 (0.8) | 18.2 (0.8) | 30.8 (2.4) | 2.04 (1.54 – 2.71) | 1.67 (1.22 – 2.28) | 1.39 (0.98 – 1.97) |
| Behavior | 959 | 15.5 (0.9) | 14.1(0.9) | 25.7 (3.0) | 2.28 (1.71 – 3.04) | 1.87 (1.35 – 2.58) | 1.69 (1.21 – 2.36) |
| Eating | 191 | 3.2 (0.5) | 2.6 (0.5) | 7.4(1.1) | 3.01 (1.78 – 5.07) | 2.32 (1.36 – 3.95) | 2.20 (1.32 – 3.66) |
| Mood | 673 | 10.1 (0.8) | 9.0 (0.7) | 18.5 (2.3) | 2.08 (1.57 – 2.75) | 1.42 (1.02 – 1.98) | 1.18 (0.84 – 1.66) |
| Substance use | 684 | 10.9 (0.7) | 10.0(0.7) | 17.6 (2.6) | 1.67 (1.13 – 2.47) | 1.13 (0.69 – 1.85) | 1.04 (0.64 – 1.70) |
| Any Mental Disorder | 2,452 | 38.8 (1.1) | 36.0 (1.0) | 59.4 (3.3) | 2.55 (1.90 – 3.42) | 2.55 (1.90 – 3.42) | 1.99 (1.42 – 2.77) |
| Use of medication in past-year | |||||||
| Antidepressant | 232 | 3.7 (0.3) | 3.4 (0.3) | 5.9 (1.4) | 1.63 (0.92 – 2.89) | 1.05 (0.58 – 1.91) | 1.02 (0.58 – 1.78) |
| Antipsychotic | 36 | 0.5 (0.1) | 0.4 (0.1) | 1.5 (0.6) | 5.51 (2.04 – 14.88) | 3.83 (1.44 – 10.21) | 3.71 (1.17 – 11.77) |
| Stimulant | 155 | 2.8 (0.4) | 2.2 (0.3) | 7.2 (2.1) | 4.28 (2.23 – 8.21) | 3.02 (1.64 – 5.56) | 2.78 (1.52 – 5.09) |
| Any medication | 403 | 6.6 (0.5) | 5.7 (0.5) | 13.0 (2.3) | 2.51 (1.57 – 4.01) | 1.69 (1.01 – 2.83) | 1.61 (0.97 – 2.66) |
NOTE: % standard error indicates prevalence. Odds ratio (OR) with 95% CI was based on logistic regression models using dichotomous hypersomnolence as exposure of interest, DSM-IV disorders and medications as outcomes entering models one at a time; Anxiety = any of agoraphobia, generalized anxiety disorder, social phobia, specific phobia, panic disorder, posttraumatic stress disorder, separation anxiety disorder; Behavior = any of attention deficit hyperactivity disorder, conduct disorder, oppositional defiant disorder; Eating = any of anorexia, bulimia, binge eating; Mood =any of major depressive disorder, dysthymia, bipolar I or II; Substance use = any of alcohol abuse/dependence, drug abuse/dependence, nicotine dependence; Any medication included antidepressant, antipsychotic, stimulant, mood stabilizer/anticonvulsant, sedative,Z drug, Benzo, Non-Benzo, and Barbiturate
adjusted for demographic characteristics (age and sex)
additionally adjusted for past year DSM-IV disorders
additionally adjusted for insomnia
Hypersomnolence was also significantly associated with insomnia after accounting for demographic characteristics and other DSM-IV disorders (OR=2.45; 95% CI=1.87 – 3.21).
Hypersomnolence and prescription medication use
Overall, 6.2% of adolescents reported the use of psychotropic medication(s) in our sample. Adolescents with hypersomnolence were more likely to report using psychotropic medication after adjustment for demographic characteristics and other DSM-IV disorders (OR= 1.69; CI=1.01 – 2.83)(table 3). After further analyses accounting for the presence of insomnia, use of psychotropic medication was no different among those with and without hypersomnolence (OR= 1.61; CI=0.97 – 2.66).
Discussion
In this large nationally representative sample of 6,483 adolescents aged 13 to18 years, 41.5% of teenagers reported sleepiness during the daytime and 11.7 % met criteria for hypersomnolence per the DSM-5 criterion A. Hypersomnolence was more common in adolescent girls and prevalence rates varied depending on age. Hypersomnolence was significantly associated with insomnia and mental disorders. These results indicate that daytime sleepiness and hypersomnolence are common in U.S. teenagers and that inadequate sleep duration and delayed bedtime increase the risk of daytime sleepiness and other sub-symptoms of hypersomnolence, i.e. having a tendency to fall asleep during the daytime, not feeling rested upon awakening and having difficulty waking in the morning.
Our finding that daytime sleepiness is reported by the nearly half of U.S. adolescents confirm the high prevalence of sleepiness in adolescents from previous telephone6 and local surveys5, in the first large nationally representative sample of U.S. adolescents. Up to 51% of youth ages 11–17 years reported feeling tired or sleepy during the day in a telephone survey by the National Sleep Foundation6 and 45.7% of 13–15 year-olds in a small survey of a single school district in Colorado reported daytime sleepinesss5. Additionally, our study further examined the specific DSM-5 symptoms for hypersomnolence, as well as the association between hypersomnolence and insomnia. This study is among the first to report rates of hypersomnolence in an adolescent population and a significant number of teenagers met criteria for hypersomnolence as defined by the DSM-5 criterion A after excluding those with inadequate sleep duration.
The American Academy of Sleep Medicine consensus statement recommends that teenagers obtain between 8–10 hours of sleep each night. However, concerns have been raised that 8 hours of sleep is likely inadequate for 13 and 14 year olds12, who require more sleep than older teenagers, especially on weekdays13. In this study we chose a cut off of 9 hours of sleep for 13 and 14 year-olds and 8 hours of sleep for 15 to 17 year-olds on weekdays in order to define adequate sleep duration. A critical finding from our study was that obtaining less than 8 hours of sleep on weekdays and bedtimes later than 11 p.m. on weekdays, and obtaining less than 7 hours of sleep and bedtimes later than 12 a.m. on weekends, were associated with excessive sleepiness and hypersomnolence sub-symptoms in teens.
Earlier school start times have been shown to result in inadequate sleep duration and daytime sleepiness in adolescents; even moderate delays in start times have been found to result in an increase in sleep duration, reduction in sleepiness/fatigue and improvement in mood without a significant change in bedtime14–16. These findings add impetus to the argument that inadequate sleep duration on school nights has a negative impact on alertness in the adolescent age group. Previous data from the NCS-A has shown that a delay in school start times, until 8 a.m. or later is associated with increased nighttime sleep duration8. Thus, concerted efforts toward delaying school start times and allowing for greater sleep opportunity could potentially help mitigate the risk of sleepiness in teenagers.
Other potential causes for excessive sleepiness in teenagers who report adequate sleep include central disorders of hypersomnolence such as narcolepsy and idiopathic hypersomnia and sleep- related breathing disorders. While the incidence of narcolepsy peaks in the teens, it is a relatively rare disorder17. On the other hand, sleep-related breathing disorders are more common and prevalence rates of up to 5.7% have been reported in this age group18. Although the NCS-A did not have adequate data to rule out these causes of hypersomnolence we were able to control for BMI, a significant risk factor for sleep-related breathing disorders. Adolescents with excessive sleepiness should be screened for these disorders. Another potential explanation for excessive sleepiness, particularly in older adolescents, could be circadian rhythm sleep-wake disorders19. These disorders, especially delayed sleep-wake phase disorder, are common in this age group, and could present both as excessive sleepiness and insomnia. The effects of delayed sleep-wake phase disorder could be compounded by early school start times. The NCS-A did not assess circadian rhythm sleep-wake disorders, so this explanation cannot be ruled out.
Older adolescents exhibited higher rates of hypersomnolence compared to younger adolescents in our sample. This confirms findings of greater sleepiness in older adolescents based on objective assessment with the mean sleep latency test (MSLT), which demonstrated that pre-pubertal and early pubertal adolescents had a higher mean sleep latency compared to mid and late pubertal adolescents, indicating that older adolescents tend to be sleepier20,21. These studies did not report sex differences in objective sleepiness. Another potential cause for excessive sleepiness, particularly in older adolescents, could be circadian rhythm sleep-wake disorders. These disorders, especially delayed sleep-wake phase disorder, are common in this age group, and could present both as excessive sleepiness and insomnia. The effects of delayed sleep-wake phase disorder could be compounded by early school start times. The NCS-A did not collect information about circadian preferences and the role of circadian rhythm sleep-wake disorders as a potential confounder cannot be ruled out.
We also found that adolescent females reported greater rates of hypersomnolence than males. In this particular population, menarche and associated hormonal changes could potentially influence sleep architecture, circadian preference and daytime sleepiness 22. The female predominance of excessive daytime sleepiness has been previously reported in studies of Korean high school students23, and a Finnish study of 11, 13 and 15 year olds24. Thus, overall, there appear to be sex differences with an increased propensity for hypersomnolence/sleepiness among adolescent girls. While few studies have examined the association between hypersomnolence and other demographic characteristics in adolescents, a previous study in adults reported an increased prevalence of hypersomnolence in Black individuals and those with a lower household income25. In our study there was no association between hypersomnolence and ethnicity, parental education, family income and urbanicity.
Our finding of a strong association between insomnia and hypersomnolence symptoms has not been previously reported in adolescents; however, a previous study showed that this association was evident in adults from the general population25. Although our finding may be in part attributable to the structure of the diagnostic probes for sleep conditions, co-occurrence of excessive sleepiness and insomnia should be pursued in future studies While hypersomnolence has been described in adolescents with depression and shown to be associated with heavy episodic drinking as well as alcohol-related adverse outcomes, the association between hypersomnolence and other mental disorders in adolescents has not been previously well delineated26. In our study, after adjustment for demographic characteristics, hypersomnolence was associated with increased risk of all major classes of mental disorders except substance use disorders. Previous studies in adults have revealed a strong association between hypersomnolence and depressive disorders. However, these studies did not account for the presence of insomnia25,27. Studies in children and adolescents have reported an increased association between excessive sleepiness and depressive symptoms but these studies utilized rudimentary and inconsistent definitions of hypersomnolence28. In this study, after accounting for insomnia, and other mental disorders, hypersomnolence was associated with having any mental disorder and specifically with having only behavior and eating disorders. These associations persisted after further analyses accounting for BMI.
Studies in children with narcolepsy indicate that they have higher rates of ADHD compared to healthy controls. However, ADHD symptoms were not associated with self-reported daytime sleepiness in this cohort29. Children with ADHD and daytime sleepiness may have higher rates of emotional and behavior problems30. Further analyses in our sample showed that hypersomnolence was not significantly associated with ADHD but with oppositional defiant and conduct disorders. Thus, hypersomnolence could potentially be an independent treatment target, similar to insomnia, in youth with oppositional defiant and conduct disorders.
As reported previously in this study, about 6.2% of adolescents reported receiving prescription medication(s) in our sample. This is similar to reports from other general population surveys31,32. Although antidepressants have been shown to be associated with hypersomnolence in adults, they were not associated with an increased rate of hypersomnolence in our cohort of adolescents25. In our sample, youth using any psychotropic medications were not at increased risk for hypersomnolence after analyses accounting for insomnia. Thus, data from our study indicate that prescription medications may be unlikely to influence the symptoms of hypersomnolence in adolescents.
Strengths of our study include the use of a large, nationally representative sample of US adolescents, relatively high response rates, comprehensive assessment of sleep patterns on both weekdays and weekends and systematic in-person evaluation of DSM-IV disorders. However, our study needs to be viewed in light of some limitations. First, our interview did not include the full criteria for the diagnosis of DSM-5 Hypersomnia that requires direct clinical assessment. Second, there was overlap in the interview items covering insomnia and hypersomnia so we could not distinguish the actual source of daytime sleepiness. Third, the data are cross-sectional so we could not discriminate between mental disorders preceding or postdating sleep problems. Fourth, we relied on subjective assessments of sleep duration which could potentially overestimate sleep times33. These data were collected prior to the significant growth in the use of technology devices by adolescents which could further reduce sleep duration/quality and conceivably promote daytime sleepiness and hypersomnolence34.
Close to half of U.S. teenagers report feeling sleepy during the day and one in nine adolescents meets criteria for hypersomnolence. After adjusting for insomnia and other confounders, hypersomnolence was associated with aggregate mental disorders but not with psychotropic medication use. These findings, in conjunction with other evidence on the adverse consequences of sleep disturbances on school performance and other health-related outcomes in teens, highlight the importance of promoting adequate sleep at both the societal and individual level in youth.
Acknowledgments
The National Comorbidity Survey - Adolescent Supplement (NCS-A) was supported by grant U01-MH60220 from the National Institute of Mental Health (NIMH) with supplemental support from the National Institute of Drug Abuse (NIDA; R01 DA016558). Manuscript preparation and analysis were supported by the Intramural Research Program of the National Institutes of Health, National Institute of Mental Health (He and Merikangas ZIAMH002808). The views and opinions expressed in this article are those of the authors and should not be construed to represent the views of any of the sponsoring organizations, agencies, or U.S. Government.
Footnotes
Ms. He served as the statistical expert for this research.
Disclosure:
Dr. Mansukhani is the principal investigator on a research grant funded by ResMed™ Foundation evaluating the effects of adaptive servoventilation treatment of central apnea syndromes on healthcare utilization that is not relevant to the current work. She is the recipient of a benefactor-sponsored career development award at Mayo Clinic. She has received honorarium as a reviewer for Lancet Respiratory Medicine.
Dr. Kotagal has received royalties from UpToDateTM.
Dr. Frye has received grant support from Assurex Health, Mayo Foundation, Myriad, the National Institute of Alcohol Abuse and Alcoholism, the NIMH, and Pfizer. He has served as a consultant to Actify Neurotherapies, Allergan, Intra-Cellular Therapies, Inc., Janssen, Mitsubishi Tanabe Pharma Corporation, Myriad, Neuralstem Inc., Otsuka American Pharmaceutical, Sunovion, Takeda, and Teva Pharmaceuticals.
Drs. Kolla and Merikangas, and Ms. He report no biomedical financial interests or potential conflicts of interest.
Financial disclosure statements should encompass the past 24 months (2 years) and foreseeable future for all authors. JAACAP requires all authors on all types of articles to specify the nature of potential conflicts of interest, financial or otherwise on the Manuscript Submission Form at the time of submission. This disclosure includes direct or indirect financial or personal relationships, interests, and affiliations whether or not directly related to the subject of the paper that have occurred over the last two years, or that are expected in the foreseeable future. This disclosure includes, but is not limited to, grants or funding, employment, affiliations, patents (in preparation, filed, or granted), inventions, honoraria, consultancies, royalties, stock options/ ownership, or expert testimony.
NOTE: If an author (or authors) has/have no conflicts of interest to declare, this must be explicitly stated. For example, Dr. Stearns reports no biomedical financial interests or potential conflicts of interest. Authors should contact the Editorial Office with questions or concerns but should err on the side of inclusion when in doubt.
Financial Disclosures:
Dr. Kolla reports no financial interests or potential conflicts of interest. Ms. Jian-Ping He reports no financial interests or potential conflicts of interest. Dr. Mansukhani is the principal investigator on a research grant funded by ResMed™ Foundation evaluating the effects of adaptive servoventilation treatment of central apnea syndromes on healthcare utilization that is not relevant to the current work. Dr. Mansukhani is the recipient of a benefactor-sponsored career development award at Mayo Clinic. She has received honorarium as a reviewer for the journal Lancet Respiratory Medicine. Dr. Kotagal reports royalties from Uptodate™. Dr Frye reports these potential conflicts of interest:Grant Support: Assurex Health, Mayo Foundation, Myriad, National Institute of Alcohol Abuse and Alcoholism (NIAAA),National Institute of Mental Health (NIMH), Pfizer.Consultant: Actify Neurotherapies, Allergan, Intra-Cellular Therapies, Inc., Janssen, Mitsubishi Tanabe Pharma Corporation, Myriad, Neuralstem Inc., Otsuka American Pharmaceutical, Sunovion, Takeda, Teva Pharmaceuticals. Dr. Merikangas reports no financial interests or potential conflicts of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
REFERENCES:
- 1.Adolescent Sleep Working Group. School start times for adolescents. Pediatrics. 2014;134(3):642–649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Millman RP. Excessive Sleepiness in Adolescents and Young Adults: Causes, Consequences, and Treatment Strategies. Pediatrics. 2005;115(6):1774–1786. [DOI] [PubMed] [Google Scholar]
- 3.Pack AI, Pack AM, Rodgman E, Cucchiara A, Dinges DF, Schwab CW. Characteristics of crashes attributed to the driver having fallen asleep. Accident Analysis & Prevention. 1995;27(6):769–775. [DOI] [PubMed] [Google Scholar]
- 4.Diagnostic and statistical manual of mental disorders : DSM-5. Arlington, VA: American Psychiatric Association; 2013. [Google Scholar]
- 5.Pagel JF, Forister N, Kwiatkowki C. Adolescent sleep disturbance and school performance: the confounding variable of socioeconomics. J Clin Sleep Med. 2007;3(1):19–23. [PubMed] [Google Scholar]
- 6.National Sleep Foundation.Sleep Polls. Available at: https://sleepfoundation.org/sites/default/files/2006_summary_of_findings.pdf. Accessed May 1, 2017.
- 7.Zhang J, Paksarian D, Lamers F, Hickie IB, He J, Merikangas KR. Sleep Patterns and Mental Health Correlates in US Adolescents. JPediatr. 2017;182:137–143. [DOI] [PubMed] [Google Scholar]
- 8.Paksarian D, Rudolph KE, He JP, Merikangas KR. School Start Time and Adolescent Sleep Patterns: Results From the U.S. National Comorbidity Survey--Adolescent Supplement. American journal of public health. 2015;105(7): 1351–1357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Merikangas KR, Avenevoli S, Costello EJ, Koretz D, Kessler RC. National Comorbidity Survey Replication Adolescent Supplement (NCS-A): I. Background and Measures. Journal of the American Academy of Child & Adolescent Psychiatry. 2009;48(4):367–379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Paruthi S, Brooks LJ, D’Ambrosio C, et al. Consensus Statement of the American Academy of Sleep Medicine on the Recommended Amount of Sleep for Healthy Children: Methodology and Discussion. J Clin Sleep Med. 2016; 12(11):1549–1561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hirshkowitz M, Whiton K, Albert SM, et al. National Sleep Foundation’s sleep time duration recommendations: methodology and results summary. Sleep Health: Journal of the National Sleep Foundation ..1): 40–43. [DOI] [PubMed] [Google Scholar]
- 12.Lewin DS, Wolfson AR, Bixler EO, Carskadon MA. Duration Isn’t Everything. Healthy Sleep in Children and Teens: Duration, Individual Need and Timing. Journal of Clinical Sleep Medicine : JCSM: Official Publication of the American Academy of Sleep Medicine. 2016;12(11):1439–1441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Olds T, Blunden S, Petkov J, Forchino F. The relationships between sex, age, geography and time in bed in adolescents: A meta-analysis of data from 23 countries. Sleep Medicine Reviews. 14(6):371–378. [DOI] [PubMed] [Google Scholar]
- 14.Carskadon MA, Wolfson AR, Acebo C, Tzischinsky O, Seifer R. Adolescent sleep patterns, circadian timing, and sleepiness at a transition to early school days. Sleep. 1998;21(8):871–881. [DOI] [PubMed] [Google Scholar]
- 15.Owens JA, Belon K, Moss P. Impact of delaying school start time on adolescent sleep, mood, and behavior. Archives of Pediatrics & Adolescent Medicine. 2010;164(7):608–614. [DOI] [PubMed] [Google Scholar]
- 16.Minges KE, Redeker NS. Delayed school start times and adolescent sleep: A systematic review of the experimental evidence. Sleep Medicine Reviews.28:86–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Silber MH, Krahn LE, Olson EJ, Pankratz VS. The Epidemiology of Narcolepsy in Olmsted County, Minnesota: A Population-Based Study. Sleep. 2002;25(2):197–202. [DOI] [PubMed] [Google Scholar]
- 18.Marcus CL, Brooks LJ, Draper KA, et al. Diagnosis and management of childhood obstructive sleep apnea syndrome. Pediatrics. 2012;130(3):e714–755. [DOI] [PubMed] [Google Scholar]
- 19.Morgenthaler TI, Lee-Chiong T, Alessi C, et al. Practice parameters for the clinical evaluation and treatment of circadian rhythm sleep disorders. An American Academy of Sleep Medicine report. Sleep. 2007;30(11):1445–1459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Campbell IG, Burright CS, Kraus AM, Grimm KJ, Feinberg I. Daytime Sleepiness Increases With Age in Early Adolescence: A Sleep Restriction Dose-Response Study. Sleep. 2017;40(5): zsx046–zsx046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Carskadon MA, Harvey K, Duke P, Anders TF, Litt IF, Dement WC. Pubertal changes in daytime sleepiness. Sleep. 1980;2(4):453–460. [DOI] [PubMed] [Google Scholar]
- 22.Pengo MF, Won CH, Bourjeily G. Sleep in Women Across the Life Span. Chest:. 2018;154(1):196–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Shin C, Kim J, Lee S, Ahn Y, Joo S. Sleep habits, excessive daytime sleepiness and school performance in high school students. Psychiatry and clinical neurosciences. 2003;57(4):451–453. [DOI] [PubMed] [Google Scholar]
- 24.Tynjala J, Kannas L, Levalahti E. Perceived tiredness among adolescents and its association with sleep habits and use of psychoactive substances. Journal of sleep research. 1997;6(3):189–198. [DOI] [PubMed] [Google Scholar]
- 25.Ohayon MM, Dauvilliers Y, Reynolds CF 3rd, Operational definitions and algorithms for excessive sleepiness in the general population: implications for DSM-5 nosology. Archives of general psychiatry. 2012;69(1):71–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Miller MB, Janssen T, Jackson KM. The Prospective Association Between Sleep and Initiation of Substance Use in Young Adolescents. The Journal of adolescent health : official publication of the Society for Adolescent Medicine. 2017;60(2):154–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Bixler EO, Vgontzas AN, Lin HM, Calhoun SL, Vela-Bueno A, Kales A. Excessive daytime sleepiness in a general population sample: the role of sleep apnea, age, obesity, diabetes, and depression. The Journal of clinical endocrinology and metabolism. 2005;90(8):4510–4515. [DOI] [PubMed] [Google Scholar]
- 28.Kaplan KA, Harvey AG. Hypersomnia across mood disorders: A review and synthesis. Sleep Medicine Reviews. 2009;13(4):275–285. [DOI] [PubMed] [Google Scholar]
- 29.Lecendreux M, Lavault S, Lopez R, et al. Attention-Deficit/Hyperactivity Disorder (ADHD) Symptoms in Pediatric Narcolepsy: A Cross-Sectional Study. Sleep.2015;38(8): 1285–1295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Lucas I, Mulraney M, Sciberras E. Sleep problems and daytime sleepiness in children with ADHD: Associations with social, emotional, and behavioral functioning at school, a cross-sectional study. Behavioral sleep medicine. 2017:1–12. [DOI] [PubMed] [Google Scholar]
- 31.National Center for Health Statistics. Psychotropic Medication Use Among Adolescents: United States, 2005–2010. Available at:https://www.cdc.gov/nchs/products/databriefs/db135.htm. Accessed May 1, 2017.
- 32.Merikangas KR, He J, Rapoport J, Vitiello B, Olfson M. Medication use in us youth with mental disorders. JAMA Pediatrics. 2013;167(2):141–148. [DOI] [PubMed] [Google Scholar]
- 33.Arora T, Broglia E, Pushpakumar D, Lodhi T, Taheri S. An Investigation into the Strength of the Association and Agreement Levels between Subjective and Objective Sleep Duration in Adolescents. PLOS ONE. 2013;8(8):e72406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kenney EL, Gortmaker SL. United States Adolescents’ Television, Computer, Videogame, Smartphone, and Tablet Use: Associations with Sugary Drinks, Sleep, Physical Activity, and Obesity. The Journal of Pediatrics. 2017;182:144–149. [DOI] [PubMed] [Google Scholar]