Abstract
Study Objectives:
To determine if associations between presleep arousal and sleep disturbance reported in adults are also characteristic of children.
Design:
Linear regression analyses examined whether somatic and cognitive presleep arousal predicted sleep disturbances.
Setting:
Two inner city schools, London, UK.
Participants:
One hundred twenty-three children aged 8 to 10 years, 49% boys, from ethnically diverse backgrounds.
Interventions:
N/A
Measurements and Results:
Children completed the Sleep Self-Report and the Pre-sleep Arousal Scale (comprising somatic and cognitive subscales). Parents completed the Child Sleep Habits Questionnaire. In separate models, both somatic (~ = 0.44, P < 0.001, R2 = 0.19) and cognitive (~ = 0.48, P < 0.001, R2 = 0.23) presleep arousal predicted the Sleep Self-Report total score. Somatic (~ = 0.28, P <.01, R2 = 0.08) and cognitive (~ = 0.37, P < 0.001, R2 = 0.14) arousal also predicted Sleep Self-Report insomnia items in separate models. These results were partially replicated when using the parent report of the Child Sleep Habits Questionnaire. When somatic and cognitive items were included in the same models, cognitive but not somatic arousal significantly predicted (most definitions of) sleep disturbance.
Conclusions:
Cognitive, and to a lesser extent somatic, presleep arousal appears to be associated with sleep disturbances in children. This suggests that further research into cognitive aspects of sleep disturbance in children is warranted—as incorporating this information into treatments may eventually prove fruitful.
Citation:
Gregory AM; Willis TA; Wiggs L; Harvey AG. Presleep arousal and sleep disturbances in children. SLEEP 2008;31(12):1745–1747.
Keywords: Children, cognitive processes, insomnia, presleep arousal
PRESLEEP AROUSAL REFERS TO BOTH PHYSIOLOGIC PROCESSES (SUCH AS A RAPIDLY BEATING HEART) AND MENTAL PROCESSES (SUCH AS BEING UNABLE to stop thinking). Such arousal is known to play an important role in insomnia in adults. Experiments that activate and deactivate somatic and cognitive presleep arousal have been associated with longer and shorter sleep-onset latency, respectively.1,2 Targets of cognitive behavioral therapies for insomnia in adults include cognitive and somatic arousal. These approaches aim to help sufferers to be better equipped to manage presleep arousal (e.g., engaging in relaxation prior to bedtime, using techniques from cognitive therapy to manage presleep worry). This approach to treatment, in adults, is effective and durable.3
In contrast to a growing knowledge base in adults, research has not systematically examined the associations between sleep disturbance and presleep arousal in children. It is not possible to assume that adult data, models, or treatments are applicable to children because of developmental differences in the maturity of several cognitive skills, as well as in age-related changes in the manifestations of sleep disturbances. Accordingly, the aim of this study was to investigate whether there is an association between presleep arousal and sleep disturbances in children. We focused on children aged 8 to 10 years because research suggests that this is around the youngest age at which children can accurately report on their own symptoms.4 Based on the adult literature, it was hypothesized that significant positive associations would be observed between presleep arousal and sleep disturbances in children. We also predicted that, as has been found in adult populations, sleep disturbances would show particularly strong associations with presleep cognitiveas opposed to somatic arousal.5
METHODS
Participants
Junior schools in central London, UK, were contacted until 2 agreed to participate in the current investigation. We requested that 8 classes take part in the current study (in the hope of obtaining a sample size of 100). The schools selected the specific classes to be involved. Children from 6 year—4 classes (8 to 9 years old) and 2 year—5 classes (9 to 10 years old) were presented with information about the study and provided with further information, questionnaire booklets, and consent forms to take home to their parents or guardians. After repeat reminders, 123 (65%) of the 189 targeted children returned consent forms and their parent questionnaire booklet and took part in the investigation. Each child was questioned individually by a trained research assistant in a quiet room, and questions were read to the children. Sixty of the participants (49%) were boys, and the children were all between age 8 and 10 years of age (mean = 8 years 8 months).
Of the parents who provided information about their child's ethnicity (n = 95, 77%), the largest proportion described their children as African (n = 19, 20%), followed by white British (n = 15, 16%), Caribbean (n = 11, 12%), white and black Caribbean (n = 4, 4%) and white and black African (n = 4, 4%). A large group of parents (n = 36, 38%) provided their own description of their child's ethnicity, as they did not consider any of the categories available to be accurate representations. Numerous descriptions were provided, with the most common being Portuguese (n = 10, 11%), followed by black British (n = 5, 5%) and Polish (n = 4, 4%). Information was provided about 87 mothers—who were most likely to be reported to be housewives (n = 27, 31%) or unemployed (n = 12, 14%). Information was provided about 60 fathers—and the largest proportion worked in the service industry (n = 15, 25%) or in craft or manual jobs (n = 14, 23%).
Measures
Sleep Disturbances
Children provided information concerning their own sleep disturbances using the Sleep Self-Report (SSR),6 which focuses on the past week. Items are rated on a 3-point scale (1 = rarely to 3 = usually). This measure was designed to assess a wide range of sleep behaviors and difficulties (e.g., Do you wake up at night when your parents think you're asleep?). The internal consistency of the total 13-item score was reasonable (α = 0.71).
Sleep disturbances were also examined using the abbreviated version of the parent-report Child Sleep Habits Questionnaire (CSHQ).7 This measure consists of 33 items, such as “child struggles at bedtime (cries, refuses to stay in bed, etc.),” each rated on a 3-point scale (1 = rarely to 3 = usually). The CSHQ consists of 8 subscales, focusing upon bedtime resistance, sleep-onset delay, sleep duration, sleep anxiety, night wakings, parasomnias, sleep disordered breathing, and daytime sleepiness (total scale α = 0.83). Parents were asked to report on the most recent typical week. The reliability and validity of this measure have been demonstrated elsewhere.6,7
Presleep Arousal
Presleep arousal was examined using the Presleep Arousal Scale,5 which comprises 16 items rated on a 5-point scale (1 = not at all to 5 = extremely). Eight of the items measure somatic arousal (e.g., “cold feeling in your hands, feet or your body in general”), and the other half measure cognitive arousal (e.g., “worry about falling asleep”). Because this measure was originally designed for use with adults, we undertook a careful adaptation for children. First, psychologists with research experience with 8-year-olds considered each item and recommended (1) several word substitutions (e.g., “perspiration” was replaced with “wetness”) and (2) that definitions be provided to explain concepts (e.g., following “being mentally alert, active,” further text was added “i.e. thinking a lot”). In this study, the child version of the scale showed reasonable internal consistency (total scale α = 0.85; each subscale α = 0.75).
Data Preparation and Statistics
With a lack of consensus in the literature as to the best way to conceptualize sleep disturbances in children, we used 4 separate definitions. First, we focused on the standard total 13-item score for the SSR.6 Second, given that the adult literature focuses on the links between presleep arousal and insomnia, we focused exclusively on the insomnia items in the SSR (we summed the 4 items that focus upon difficulties falling asleep, night waking, and trouble falling back to sleep). Third, we focused on the standard total 33-item CSHQ score.6,7 Finally, we focused on insomnia items in the CSHQ (we summed 4 items in the sleep-onset delay and night-waking scales). For each of the 4 definitions of sleep disturbance, we ran 2 types of regression model. First, we predicted sleep disturbance from somatic arousal and from cognitive arousal in separate models. Second, we predicted sleep disturbance from a model including both somatic and cognitive arousal.
RESULTS
The mean for the child-report SSR total score was 22.44 (SD = 4.68), and 7.69 (SD = 1.78) for the child-report SSR insomnia items. Scores on theses scales can potentially range from 13 to 39 and 4 to 12, respectively. The mean parent-report CSHQ total score was 43.31 (SD = 7.56) and 5.37 (SD = 1.38) for the parent-report CSHQ insomnia items. Scores on these scales can potentially range from 31 to 97 and 4 to 12, respectively. Boys and girls did not differ significantly on these scales, nor were there any differences between the main ethnic groups (African and white British). The correlation between the total SSR and CSHQ scores was small and nonsignificant (r85= 0.14, P = 0.20).
The results of the linear regression analyses are presented in Table 1. In separate models, both somatic and cognitive arousal predicted child-report SSR total score and insomnia items. Furthermore, both somatic and cognitive arousal predicted the parent-report CSHQ total score. Parent-report CSHQ insomnia items were not predicted by somatic arousal but were predicted by cognitive arousal. Of note, cognitive arousal appears to be a stronger predictor than somatic arousal of sleep disturbance. Indeed, for each conceptualization of sleep disturbances, cognitive arousal appears to have a stronger β value and is able to explain a greater proportion of the variance, as compared with somatic arousal. Furthermore, when somatic and cognitive items were both included in the regression models (Table 1, Models 2), cognitive but not somatic arousal predicted sleep disturbance (according to most definitions of the latter).
Table 1.
Prediction of Sleep Disturbances from Presleep Arousal
| Model | B (SE) | β | T |
|---|---|---|---|
| Predicting child-reported sleep disturbances (total SSR score) | |||
| Model 1 | |||
| Somatic arousal (R2 = 0.19) | 0.35 (0.06) | 0.44 | 5.35d |
| Cognitive arousal (R2 = 0.23) | 0.34 (0.06) | 0.48 | 5.96d |
| Model 2 (R2 = 0.26) | |||
| Somatic arousal | 0.17 (0.08) | 0.21 | 1.99b |
| Cognitive arousal | 0.24 (0.08) | 0.33 | 3.10c |
| Predicting child-report insomnia items (insomnia items from the SSR) | |||
| Model 1 | |||
| Somatic arousal (R2 = 0.08) | 0.08 (0.03) | 0.28 | 3.19c |
| Cognitive arousal (R2 = 0.14) | 0.10 (0.02) | 0.37 | 4.38d |
| Model 2 (R2 = 0.14) | |||
| Somatic arousal | 0.02 (0.03) | 0.05 | 0.45 |
| Cognitive arousal | 0.09 (0.03) | 0.34 | 2.91c |
| Predicting parent-reported sleep disturbances (total CSHQ score) | |||
| Model 1 | |||
| Somatic arousal (R2 = 0.09) | 0.38 (0.14) | 0.29 | 2.80c |
| Cognitive arousal (R2 = 0.11) | 0.37 (0.12) | 0.33 | 3.22c |
| Model 2 (R2 = 0.12) | |||
| Somatic arousal | 0.15 (0.20) | 0.11 | 0.74 |
| Cognitive arousal | 0.28 (0.17) | 0.25 | 1.68a |
| Predicting parent-reported insomnia items (insomnia items from the CSHQ) | |||
| Model 1 | |||
| Somatic arousal (R2 = 0.01) | 0.001 (0.002) | 0.07 | 0.64 |
| Cognitive arousal (R2 = 0.05) | 0.003 (0.001) | 0.21 | 2.00b |
| Model 2 (R2 = 0.06) | |||
| Somatic arousal | −0.002 (0.002) | −0.158 | −1.07 |
| Cognitive arousal | 0.004 (0.002) | 0.32 | 2.18b |
The parent-report insomnia items scale was transformed because of skew (elsewhere in the manuscript the untransformed variable is reported for ease of interpretation). Model 1 provides the unadjusted values for presleep arousal predicting sleep disturbances. Model 2 provides the adjusted values for the presleep arousal predicting sleep disturbances as both somatic and cognitive arousal are included in the models. A similar pattern of results is obtained when adjusting for age and sex (unreported). SSR refers to sleep self-report; CSHQ, Child Sleep Habits Questionnaire. B, unstandardized coeffi cient; SE, standard error; β, standardized coefficient.
p < 0.10;
p < 0.05;
p < 0.01;
p < 0.001
DISCUSSION
As has been reported in adult populations, there appear to be associations between presleep arousal and sleep disturbances in children. Furthermore, (as with adults5) cognitive, as compared with somatic, presleep arousal appears to be more robustly associated with sleep disturbance. Although children should not be thought of as “little adults,” we believe that the results of this investigation suggest that there should be further investigation into whether cognitive factors known to be associated with insomnia in adults are also associated with children's sleep disturbances.
Strengths of this study include the reasonable participation rate (65%) and the ethnically diverse sample—perhaps making the results of our study more generalizable, as compared with studies focusing on less representative, more homogeneous samples. Although we did not find any ethnic differences with regard to sleep, others have reported differences,8 and we concur that further research employing ethnically diverse samples is advantageous. The limitations of this study also need consideration. The first concerns our sleep measures. Although the CSHQ and the SSR are considered among the best questionnaires to investigate sleep disturbances in children, there was disagreement between parents' and children's reports of sleep disturbances. This discrepancy reflects a standard finding in other areas of developmental psychopathology.9 It is noteworthy, however, that the current results were partially replicated when using different conceptualizations of sleep disturbances and different raters. Second, we recognize that the children with sleep disturbances in our sample were not seeking treatment nor were their sleep disturbances formally diagnosed. Of note, a clinical cutoff of 41 has been proposed for the CSHQ total scale score,7 and we found (unreported) that 56% of our sample scored above this point. This high percentage is somewhat surprising and could partially reflect the established widespread nature of sleep disturbances in children and the demographic of our sample, which included children from economically disadvantaged backgrounds. (There are known links between sleep disturbances and socioeconomic status.10) It is important to test for replications of these findings in treatment-seeking samples. Additionally, future research needs to move beyond employing a correlational design to understand more about the directions of effects with regard to the association between presleep arousal and sleep disturbances. Should observed associations be replicated and extended, it may be worth considering whether aspects of cognitive behavioral therapies— which have been so successful in treating adults with insomnia— can be adapted for use with children. Indeed, helping children to physically unwind before the sleep-onset period and teaching children strategies to manage cognitive arousal could have long-reaching positive implications.
ACKNOWLEDGMENTS
The STEPS study was funded by a small grant from the British Academy to the first author who is currently supported by a Leverhulme Research Fellowship
We thank the participants, parents, and staff involved in the STEPS study.
Footnotes
Disclosure Statement
This was not an industry supported study. Dr. Harvey has consulted for Actelion and participated in a speaking engagement for Sleep Medicine Initiative. The other authors have indicated no financial conflicts of interest.
REFERENCES
- 1.Bonnet MH, Arand DL. Caffeine use as a model of acute and chronic insomnia. Sleep. 1992;15:526–36. [PubMed] [Google Scholar]
- 2.Tang NKY, Harvey AG. Effects of cognitive arousal and physiological arousal on sleep perception. Sleep. 2004;27:69–78. doi: 10.1093/sleep/27.1.69. [DOI] [PubMed] [Google Scholar]
- 3.Morin CM, Bootzin R, Buysse DJ, Edinger JD, Espie C, Lichstein KL. Psychological and behavioral treatment of insomnia: update of recent evidence (1998–2004) Sleep. 2006;29:1398–1414. doi: 10.1093/sleep/29.11.1398. [DOI] [PubMed] [Google Scholar]
- 4.Merrell KW, McClun LA, Kempf KKG, Lund J. Using self-report assessment to identify children with internalizing problems: validity of the internalizing symptoms scale for children. J Psychoeducational Assess. 2002;20:223–39. [Google Scholar]
- 5.Nicassio PM, Mendlowitz DR, Fussell JJ, Petras L. The phenomenology of the pre-sleep state—the development of the pre-sleep arousal scale. Behav Res Ther. 1985;23:263–71. doi: 10.1016/0005-7967(85)90004-x. [DOI] [PubMed] [Google Scholar]
- 6.Owens JA, Maxim R, Nobile C, McGuinn M, Msall M. Parental and self-report of sleep in children with attention-deficit/hyperactivity disorder. Arch Pediatr Adolesc Med. 2000;154:549–55. doi: 10.1001/archpedi.154.6.549. [DOI] [PubMed] [Google Scholar]
- 7.Owens JA, Spirito A, McGuinn M. The Children's Sleep Habits Questionnaire (CSHQ): psychometric properties of a survey instrument for school-aged children. Sleep. 2000;23:1043–51. [PubMed] [Google Scholar]
- 8.Durrence HH, Lichstein KL. The sleep of African Americans: a comparative review. Behav Sleep Med. 2006;4:29–44. doi: 10.1207/s15402010bsm0401_3. [DOI] [PubMed] [Google Scholar]
- 9.Achenbach TM, McConaughy SH, Howell CT. Child adolescent behavioral and emotional problems: implications of crossinformant correlations for situational specificity. Psychol Bull. 1987;101:213–32. [PubMed] [Google Scholar]
- 10.Buckhalt JA, El-Sheikh M, Keller P. Children's sleep and cognitive functioning: race and socioeconomic status as moderators of effects. Child Dev. 2007;78:213–31. doi: 10.1111/j.1467-8624.2007.00993.x. [DOI] [PubMed] [Google Scholar]