Sleep Patterns and Sleep Problems Among Schoolchildren in the United States and China
http://www.100md.com
《小儿科》
Department of Family and Human Development and Prevention Research Center, Arizona State University, Tempe, Arizona
Shandong Mental Health Center, Jinan, People's Republic of China
Division of Pediatric Ambulatory Medicine, Rhode Island Hospital, Brown University School of Medicine, Providence, Rhode Island
Prevention Research Center, Arizona State University, Tempe, Arizona
ABSTRACT
Objectives. Sleep patterns and sleep problems in children are not only influenced by a large number of biological and psychologic factors but also by cultural and social factors. Little is known about similarities and differences in sleep patterns and sleep problems among children across countries. We attempted to compare sleep patterns and sleep problems among schoolchildren from 2 countries with distinctive cultural contexts: the United States and China.
Methods. The data come from 2 cross-sectional surveys in 3 elementary schools of Jinan City, People's Republic of China, and 3 elementary schools from a suburban school district in southeastern New England, United States. The Chinese sample consisted of 517 elementary school children (grades 1 to 5), and the US sample consisted of 494 elementary school children (grades kindergarten through 4). We used the Children's Sleep Habits Questionnaire (CSHQ) to assess children's sleep patterns and sleep problems as reported by parents. Parents of the Chinese sample completed a Chinese version of the CSHQ.
Results. For children in both the US and Chinese samples, reported bedtime was delayed and sleep duration decreased with increasing age. Compared with the US children (grades 1–4), Chinese children went to bed approximately half an hour later (9:02 vs 8:27 PM) and woke up half an hour earlier (6:28 vs 6:55 AM), resulting in an average sleep duration that was 1 hour less (9.25 vs 10.15 hours). Chinese children were rated significantly higher than the US children on almost all CSHQ scales, indicating more sleep problems in Chinese children. Common sleep problems observed for all children were difficulty falling asleep, having a fear of sleeping in the dark, sleep talking, restless sleep, teeth grinding during sleep, and daytime sleepiness. Shorter daily sleep duration was associated with difficulty falling asleep, struggling at bedtime, and trouble sleeping away for the US children, and with going to bed at different times and having a fear of sleeping alone for Chinese children. Short sleep duration was a main predictor of daytime sleepiness for Chinese children, whereas restless sleep and snoring predicted daytime sleepiness for the US children.
Conclusions. As reported by parents, children in China went to bed later and woke up earlier and their sleep duration was 1 hour shorter than the US children. Chinese children were reported to have more sleep problems than their US counterparts. Daytime sleepiness was determined by sleep duration only for those who slept insufficiently. Unique school schedules and sleep practices may contribute to the differences in the sleep patterns and sleep problems of children from the United States and China.
Key Words: sleep patterns sleep problems sleep duration daytime sleepiness
Abbreviations: SES, socioeconomic status CSHQ, Children's Sleep Habits Questionnaire MANCOVA, multivariate analysis of covariance CI, confidence interval OR, odds ratio
Sleep plays very important roles in children's development. Sleep not only impacts on physical growth, behavior, and emotional development but also is closely related to cognitive functioning, learning, and attention.1–3 Sleep disorders in children are not uncommon. In a number of survey studies, parents report sleep disturbances throughout childhood, with the prevalence seeming to be 20% to 30%.1,4–8 Sleep patterns and sleep problems in children are not only influenced by a large number of biological and psychologic factors but also by cultural, social, and family factors.9–11 However, most epidemiologic studies of sleep patterns and sleep problems have been of Western children.6,11–13 Although several studies have examined ethnocultural differences in sleep practices and/or sleep disturbances in US children and adolescents,11,14,15 few studies have compared sleep patterns and sleep problems among children across countries using the same standardized sleep instruments.16 This study represents the first comparison of sleep patterns and sleep problems among children from the United States and China, representing 2 distinctive cultural contexts: Western and Asian.
Western child-rearing practices emphasize the development of independence and individualism, whereas Asian culture stresses the development of interpersonal relations, collectivism, family closeness, and social harmony.17 These child-rearing beliefs and cultural characteristics may determine child-rearing and family sleep practices. For example, cosleeping with parents in young children is accepted and very common in many Asian countries. In China, all-night cosleeping during the infant and early childhood years is the norm, decreases in prevalence with increasing age, and is reportedly unusual after puberty. A recent study of school-aged children in urban China reported that the prevalence of regular bed sharing was 18.2% and as high as 55.8% in 7-year-olds.18 In a study of bed sharing in Korean children, Yang and Hahn19 found that 73.5% of mothers approved of bed sharing between 3 to 6 years of age. The main reasons for bed sharing were "to look after the child while sleeping" and "the child is too young to sleep alone."19(p155) In Western countries, however, cosleeping has been considered somewhat of a cultural taboo, and in many of these countries it is a norm for a child to sleep separately from the parents after birth. In the United States, with the exception of certain ethnic and racial groups (particularly Hispanics and African Americans11), regular cosleeping with parents in non-Hispanic white families is uncommon in these societies. A recent US national survey between 1993 and 2000 indicated that only 9.2% of infants usually shared a bed with parents.20 Many Western child health professionals emphasize the hazards of cosleeping, such as impediment of the development of autonomy and independence, increased dependence on parents, inappropriate sexual arousal, risk of sleep problems, and interference with parental sexuality and intimacy.19,21 There is evidence that cosleeping is related to children's sleep patterns, sleep disturbances, and daytime sleepiness.18,22
Research has indicated that there seems to be marked cultural variations in the time of going to bed and amount of time slept.23 In a study of cultural differences in sleep habits and complaints of 40303 11- to 16-year-old European children from 11 countries, Tynjala et al16 reported that sleep habits varied significantly between countries, with Israeli children sleeping the shortest and Swiss children sleeping the longest. School start time may contribute to the differences in sleep duration between countries.24 Compared with the United States, elementary schools in China start at an earlier time, at 7:30 AM, and children spend more time in school and studying at home because of increased homework demands and the emphasis on academic achievement in Chinese culture.25,26 Obviously, an early start time for school requires an even earlier waking time, and nighttime homework may result in a later bedtime, which in turn may result in shorter sleep duration and consequent daytime sleepiness in Chinese children.
Another cultural characteristic of sleep habits in China is napping after lunch. In China, almost all schools have a long lunch break, 2 hours, between 12:00 and 2:00 PM. Although there are no statistical data available, anecdotally, many students have a 30-minute to 1-hour nap after lunch either at school or at home. Some students go home for lunch and may have a nap at home; others have lunch and may nap at school. Generally speaking, most people in China perceive the purpose of taking a nap after lunch not specifically as compensation for a shortened night sleep duration but for restoration and recovery of energy and spirit. Recent experimental studies have demonstrated favorable effects of brief naps on following alertness and cognitive performance in young adults.27–29 On the other hand, it is known that long naps may also have negative effects on night sleep, such as sleep onset and sleep quality and quantity.
In addition, China (as a developing country) differs greatly from the United States in socioeconomic status (SES), which may have a unique contribution to sleep practices and sleep problems among Chinese families. For example, compared with the United States, urban residents of China have few choices of housing, and living space is very crowded because of financial restriction and rapid population growth. Most urban families with young children are residing in an apartment, possibly with only 1 bedroom, which may prevent the child from having his/her own room or bed and increase the likelihood of bed sharing with parents and sleep problems caused by parental night activities. Moreover, almost all the families in urban China have only 1 child because of the 1-child family-planning policy. Only children may be more likely to be overindulged ("the little emperor") and/or overprotected, and parents (and grandparents) may have more difficulties setting limits.30 These characteristics may increase the likelihood of the child cosleeping with his or her parents and may decrease sleep quality and duration.
Despite these cultural and socioeconomic differences, little is known about the similarities and differences in sleep patterns and sleep problems between the US and Chinese children. We report here 2 school-based studies of children in the United States and China with the same sleep measure that allows us to (1) compare sleep patterns (bedtime, morning wakeup time, sleep duration) and sleep problems as reported by parents between children from China and the United States, (2) examine sleep problems that are associated with sleep duration among children from different cultural contexts, and (3) examine the relationship between sleep duration and sleep problems and daytime sleepiness for the US and Chinese children. We hypothesized that Chinese children would go to bed later, wake up earlier, and sleep less than the US children because of earlier school start time and heavier homework load. We also hypothesized that the parents of Chinese children would report more sleep problems than parents of the US children because of potentially increased parental awareness of sleep issues as a result of bed sharing or room sharing with the child. Finally, we expected that daytime sleepiness for both groups of children would be predicted by shorter sleep duration.
METHODS
Subjects
Chinese Sample
The Chinese participants consisted of 517 children 7 to 13 years old sampled from 3 public elementary schools in Jinan City. Jinan is the capital city of Shandong Province located in the eastern part of mainland China. In Jinan City, children are generally admitted to school at 6 or 7 years old. In elementary schools in Jinan, school starts at 7:30 AM, and classes are scheduled for 45 minutes, with a 15-minute break between classes. Morning classes conclude at 12:00 PM, and there is a lunch break for 90 minutes in winter and spring and 120 minutes in summer and fall. Most students go home for lunch, and some of them have a nap afterward. Some students, especially first- or second-graders, often stay at school during the lunch break and may nap after lunch. This survey was conducted in the spring of 2001, at which time the lunch break was 90 minutes. Students have to do 1 to 3 hours of homework almost every day including weekends and holidays. Homework increases with grade level. Most students also watch television after school, before, during, or after dinner. Unfortunately, no statistical data are available.
Detailed sampling procedures and methodology have been reported elsewhere.18 Briefly, the aims of the survey were explained to school principals and teachers in the target schools, and permission was obtained to conduct the survey, as is standard practice in China. One class from each grade for each school was selected randomly, yielding a pool of 565 children as potential participants. The children were asked by teachers to take a questionnaire to their parents with a letter explaining the project and requesting the parents to complete the questionnaire. The parents also were told that participation in the survey was voluntary. Of the 565 sampled children, 517 (92%) returned completed questionnaires for statistical analysis. This sample consisted of 47% boys, and mean age was 11.0 years (SD: 1.9; range: 7–13 years). Because in China there are no established systems to assess the SES of individual families, parental education levels were used as an indicator of family SES. In this sample, parental education levels were distributed as middle school or less (10.8%), high school (31.7%), some college (22.4%), and university (35.0%) for fathers and middle school or less (18.8%), high school (38.6%), some college (24.6%), and college graduate (17.9%) for mothers. The higher-education levels of the parents indicate that this sample represents a group of children from families of at least middle or high SES in urban China.
US Sample
The US sample consisted of 494 children who were enrolled in kindergarten through fourth grade from 3 public elementary schools in a predominantly white, middle-income, English-speaking suburban school district in southeastern New England.6,31,32 In the school system studied, kindergarten consisted of half-day sessions, with students equally distributed between morning (8:50–11:30 AM) and afternoon (12:30–2:45 PM) sessions for a half year each. For children in grades 1 to 4, school hours were between 8:50 AM and 2:45 PM. There was a half-hour lunch period between the morning and afternoon classes. Although children were encouraged to nap for a 1-hour period in kindergarten, data were not collected specifically regarding the duration of any scheduled naps.
Of 1099 questionnaire packets mailed to the parents of children in the 3 elementary schools, 494 children returned usable questionnaires for data analysis (response rate: 46.9%). Mean age of the sample was 7.56 years (SD: 1.51; range: 4.83–11.00 years). There were 240 boys (51.2%) and 229 girls (48.8%). SES was determined by using the Hollingshead Form Index of Social Status (A.B. Hollingshead, PhD, Four-Factor Index of Social Status, unpublished paper, 1975), which is based on occupation and education. The mean Hollingshead SES was 45.5 (SD: 11.3; class 4).
Daily television-viewing data for this sample were also obtained at the time of the sleep study; mean reported television-viewing time on weekdays was 2.07 hours (SD: 1.32). The amount of television viewed in this sample did not differ significantly by age but was higher in boys (2 = 5.67, P < .05) and in families of lower SES (t = 3.62, P < .001).32
Measure of Sleep Patterns and Sleep Problems
The Children's Sleep Habits Questionnaire (CSHQ)31 was used to assess sleep patterns and sleep problems of children in the United States and China. The CSHQ consists of 33 sleep-disturbance items and 3 items asking for information about sleep patterns (ie, bedtime, morning wakeup time, and daily total sleep duration). Parents are asked to recall the child's sleep behaviors over a "typical" recent week. Items are rated on a 3-point scale: "usually" if the sleep behavior occurred 5 to 7 times per week; "sometimes" for 2 to 4 times per week; and "rarely" for 0 to 1 time per week. The 33 sleep-disturbance items were conceptually grouped into 8 subscales: bedtime resistance, sleep-onset delay, sleep duration, sleep anxiety, night wakings, parasomnias, sleep-disordered breathing, and daytime sleepiness. The CSHQ has satisfactory psychometric properties in US children.31
We developed a Chinese version of the CSHQ by translation and back translation for this study. Cronbach's using data from these Chinese children was .80 for the entire scale and .49 to .72 for the subscales. The psychometric properties of the Chinese CSHQ with this sample are comparable to those found in US children for most of the subscales.18,31
Statistical Analysis
We performed multivariate analyses of covariance (MANCOVAs) with gender as a between-subjects factor and grade as a covariate to assess the effects of gender and grade on sleep patterns for the US and Chinese children, respectively. We conducted MANCOVAs with nationality and gender as between-subjects factors and age as covariate for pooled children in grades 1 to 4. Multivariate logistic regression analyses were conducted to examine the differences in frequencies of sleep problems between the 2 groups of children after adjustment for the effects of age and gender. Odds ratios and 95% confidence intervals (CIs) were reported to show the differences. Finally, stepwise regression analyses were performed to examine the sleep problems that could predict sleep duration and the association between sleep duration and sleep problems that could predict daytime sleepiness for the US and Chinese children, respectively. Standardized regression coefficients were reported to present the effect size of each significant predictor.
All analyses were performed by using the Statistical Program for Social Sciences (SPSS) for Windows, version 11.5. All statistical significance was set at P < .05.
RESULTS
Sleep/Wake Patterns
For the US children, mean night bedtime was 8:27 PM (SD: 30.6 minutes), mean morning wakeup time was 6:55 AM (SD: 25.8 minutes), and mean sleep duration was 10.64 hours (SD: 0.67). Although boys went to bed later and woke up earlier than girls, multivariate analysis of variance did not show significant differences in bedtime and wake-up time between genders (P > .05). Girls slept significantly longer than boys (10.31 ± 0.70 vs 10.17 ± 0.64; F = 6.45, P = .011). With advancing grade, children's bedtime was significantly postponed ( = .102, t = 6.12, P < .001) and total sleep duration declined ( = –.137, t = 5.78, P < .001).
For Chinese children, mean night bedtime was 9:06 PM (SD: 44.4 minutes), mean morning wakeup time was 6:24 AM (SD: 19.8 minutes), and mean sleep duration was 9.13 hours (SD: 0.74). No significant gender differences in bedtime, wakeup time, and sleep duration were observed. With advancing grade, bedtime was significantly postponed ( = .05, t = 2.66, P = .01), wakeup time advanced ( = –.045, t = 5.48, P < .001), and total sleep duration significantly declined ( = –.094, t = 5.14, P < .001).
Comparisons of CSHQ Scale Scores
Comparisons of Individual Sleep Problems
Frequencies for individual sleep problems are presented in Table 2 for the US and Chinese children. For the US children, the most prevalent specific sleep problems (defined as occurring sometimes or usually in at least 20% of the sample) were: awakened by others in the morning (68%); hard time getting out of bed (38.7%); restless sleep (35.1%); waking up in a bad mood (34.1%); falling asleep while riding in a car or bus (27.1%); being afraid of sleeping in the dark (25.9%); waking once during the night (25.1%); being tired during the day (24.0%); taking a long time to be alert in the morning (23.5%); sleep talking (23.2%); difficulty falling asleep in 20 minutes (21%); teeth grinding during sleep (20.7%); and getting too little sleep (20.2%). Rare sleep problems (occurring in <5% of the sample) were: sleepwalking (3.8%); waking more than once during the night (2.5%); holding breath (1.3%); and awakening screaming or sweating (1.3%).
For Chinese children, prevalent sleep problems were: awakening by others in the morning (58.2%); hard time getting out of bed (43.5%); getting too little sleep (41.4%); falling asleep while riding in a car or bus (36.6%); restless sleep (36.3%); waking once during the night (35%); being afraid of sleeping alone (29.5%); being afraid of sleeping in the dark (29.1%); waking up in a bad mood (26.1%); needing a parent in the room to sleep (25.3%); falling asleep in someone else's bed (23%); taking a long time to be alert (22.6%); teeth grinding during sleep (21.6%); trouble sleeping away from home (21.5%); being tired during the day (21%); sleep talking (20.5%); and difficulty falling asleep in 20 minutes (20.2%). There were no sleep problems that happened in <5% of the Chinese children.
A series of multivariate logistic regression analyses were conducted to examine the differences in frequencies of sleep problems between the 2 groups of children, adjusting for the effects of age and gender. In the logistic regression model, the dependent variable was defined as 1 if the item rated as "sometimes" or "usually" and 0 if rated as "no" or "rarely." As shown in Table 2, most sleep problems were more prevalent in Chinese children than in the US children. The following sleep problems that Chinese children were 2 times more likely to have than the US children and remained significant after adjustment of age and gender were: going to bed at different times (odds ratio [OR]: 2.5); needing a parent in the room to sleep (OR: 3.5); being afraid of sleeping alone (OR: 3.3); waking more than once during the night (OR: 4.6); getting too little sleep (OR: 3.1); not sleeping the same amount each day (OR: 4.3); holding breath during sleep (OR: 9.4); snorting and gasping (OR: 3.0); and awakening screaming or sweating (OR: 6.3). A total of 13 sleep problems were commonly reported for both groups of children but did not significantly differ: difficulty falling asleep in 20 minutes; being afraid of sleeping in the dark; waking once during the night; sleep talking; restless sleep; teeth grinding; and all items regarding daytime sleepiness.
Associations of Sleep Problems With Sleep Duration
Stepwise linear regression analyses were conducted to examine the associations of individual sleep problems with sleep duration for the US and Chinese children, separately. In the regression analysis, items regarding daytime sleepiness were not selected as predicting variables, because we hypothesized that sleep duration could predict daytime sleepiness as described below. All variables in Table 3 were entered for stepwise regression analysis. As indicated in Table 3, 4 sleep problems in the final model predicted sleep duration for the US children: difficulty falling asleep in 20 minutes ( = –0.209); struggling at bedtime ( = –0.254); trouble sleeping away ( = –0.117); and bedwetting ( = 0.149). The 4 sleep problems accounted for 15.6% of variance of sleep duration. Bedwetting was associated with increased sleep duration, but 3 other sleep problems were associated with decreased sleep duration. For Chinese children, only 2 sleep problems remained significant in the final model: going to bed at different times ( = –0.213) and being afraid of sleeping alone ( = –0.146), accounting for 7.2% variance of sleep duration.
Association of Daytime Sleepiness With Sleep Duration and Sleep Problems
Daytime sleepiness was very prevalent in both US and Chinese children. Stepwise regression analyses were performed to examine the association between sleep duration and sleep problems and daytime sleepiness for the US and Chinese children, with the CSHQ daytime-sleepiness scale score as the dependent variable. Table 4 presents all variables that were entered initially for stepwise regression and standardized regression coefficients for those variables that remained significant in the final model.
For the US children, 4 sleep problems significantly predicted daytime sleepiness in the final model: difficulty falling asleep in 20 minutes ( = 0.138), being afraid of sleeping in the dark ( = 0.130), snoring ( = 0.178), and restless sleep ( = 0.193), accounting for 12.1% variance. All 4 sleep problems were associated with increased likelihood for daytime sleepiness. Sleep duration was not a predictor. For Chinese children, sleep duration ( = –0.205) and 5 sleep problems were significant predictors of daytime sleepiness, accounting for 21.5% variance. The 5 sleep problems were: needing a parent in the room to sleep ( = 0.145); snorting and gasping ( = 0.133); restless sleep ( = 0.125); teeth grinding ( = 0.148); and awakening screaming ( = 0.166). As expected, sleep duration was negatively associated with daytime sleepiness, and the 5 sleep problems were positively related to daytime sleepiness.
DISCUSSION
This is the first study to directly compare sleep patterns and sleep problems among elementary school children between the United States and China. Children's sleep patterns and sleep problems were reported by parents using the same well-established instrument, the CSHQ.31 The main findings of this study can be summarized as follows: (1) for all children in the United States and China, bedtime was increasingly delayed and sleep duration decreased with increasing age; (2) compared with the same-aged US children, Chinese children went to bed half an hour later, woke up half an hour earlier, and slept 1 hour less; (3) Chinese children were rated significantly higher than the US children on almost all CSHQ scales; (4) difficulty falling asleep, being afraid of sleeping in the dark, sleep talking, restless sleep, and teeth grinding during sleep and daytime sleepiness were common sleep problems for both groups of children, although most sleep problems were more frequent in Chinese children than in the US children; (5) there was a relationship between shorter sleep duration and difficulty falling asleep, struggling at bedtime, and trouble sleeping away from home for the US children, whereas going to bed at different times and having a fear of sleeping alone were related to shorter sleep duration in the Chinese sample; and (6) short sleep duration was a main predictor of daytime sleepiness for Chinese children, whereas restless sleep and snoring during sleep were main predictors for the US children.
In the current study, we found that sleep patterns in Chinese children were significantly different from the US children, including later bedtime, earlier morning wake time, and shorter sleep duration by 1 hour. The difference in sleep patterns in children between the 2 countries may be attributed to different school schedules and homework load and sleep practices of the 2 countries. First, school started at 7:30 AM in the sampled school district of China, 1 hour and 20 minutes earlier than the sampled school district of the United States (8:50 AM). The earlier school start time necessitates the Chinese children to get up earlier for the morning class. Second, it is well known that school children in China have more homework than most Western countries because of the high academic expectations of their parents and schools.25,26 Most children in China have to do homework after dinner for 1 or 2 hours depending on the amount of homework, whereas the usual homework load for the US children was not >30 minutes. Third, in the Chinese sample, 48% of children routinely shared a bed (26.4%) or room (21.6%) with their parents. Room sharing or bed sharing with parents may also contribute to later bedtimes and earlier morning wakeup times in Chinese children, because children's sleep patterns are more likely to be influenced by the sleep patterns of cosleeping parents.
It is important to note that Chinese children slept 1 hour less than the US children according to parent report. There are several potential explanations for the difference. First, Chinese children went to bed later and woke up earlier as described above. This is the most likely possibility to explain why Chinese children slept less. Second, Chinese parents may use night-sleep duration instead of total daily sleep duration, ignoring naptime. Napping is a well-accepted sleep habit in China. Most schools schedule lunch breaks of 2 hours, which allows students to have a brief nap of 30 minutes in winter and spring, and 1 hour in summer and fall. Thus, to some extent, the nap may explain the difference in sleep duration between the US and Chinese children. Unfortunately, in this study we did not collect specific data regarding naps. Alternatively, shorter sleep duration in Chinese children may be explained by sleep problems, because most sleep problems were more prevalent in Chinese children. However, stepwise regression analysis revealed only 2 sleep problems that were significantly associated with sleep duration and accounted for only 7% variance. Another possibility is that Chinese children require less sleep than the US children. This can be interpreted by the fact that daytime sleepiness did not differ between the US and Chinese children. It is also possible that the difference in sleep duration between the 2 groups of children may be caused by parental overestimation of sleep duration for the US children and/or underestimation of sleep duration for Chinese children (ie, reporting bias). The US parents may know less about their child's real sleep than Chinese parents because of the high frequency of room sharing or bed sharing with parents in Chinese children. The US parents may overestimate their child's sleep duration using bedtime instead of the time of actual sleep onset, which may be significantly later than bedtime because of difficulty falling asleep and other bedtime-resistance problems.6 Finally, although we did not collect specific information regarding differences in weekend (non–school days) bedtimes and wake times, it is possible that the Chinese children attempt to "make up" for relatively less sleep during the week by increasing sleep duration on the weekends.
Most sleep problems were more frequent in Chinese children than in the US children, although difficulty falling asleep, being afraid of sleeping in the dark, sleep talking, restless sleep, teeth grinding during sleep, and daytime sleepiness were very common for both groups of children. Cultural characteristics of sleep practices and environment may contribute to more sleep problems in Chinese children, such as cosleeping, being an only child in the family, crowded housing, and chronic stress caused by homework overload. For example, crowded housing in urban residents of China is very common, and many families with young children may have only 1 bedroom. In this sample, >60% of families owned a living space <20 m2 per person and 35% owned a living space <15 m2. Crowded housing limits children having their own room and even a separate bed at home, and parents have to bed share with their children. There is evidence that crowded housing and bed sharing affect children's sleep quality and quantity,18,22 which may explain why Chinese children are more likely to go to bed at different times, wake in the night, and sleep different amounts each day. Another potential factor that may contribute to higher rates of sleep problems among Chinese children should not be ignored: being the only child in the family because of the family-planning policy. An only child is frequently regarded as "a little emperor" or "a little sun," namely a spoiled/overprotected child. Only children in China are described as overdependent, egocentric, less cooperative, timid, and unadjusted.30 These characteristics may increase the likelihood of children bed sharing with parents and sleep-resistance problems such as needing a parent in the room to sleep and being afraid of sleeping alone. Another explanation for the difference is that Chinese children are more likely to develop sleep problems because of genetic effects. However, to date there has been no evidence to indicate that specific genes are linked to ethnic differences in sleep patterns and sleep problems. It is also possible that the difference is caused by information bias. The US parents may underestimate their child's sleep problems because the US parents and their children slept in separate rooms and subsequently know less about their children's sleep problems, which may be why some parasomnia and sleep-disordered breathing behaviors were less likely to be reported by the US parents.
Daytime sleepiness was very common for both groups of children. Approximately 60% of children needed to be awakened by parents in the morning at least 2 to 4 times a week, 40% had a hard time getting out of bed in the morning, 30% woke up in a bad mood, 26% to 36% fell asleep while riding in a car, 23% took a long time to be alert, 22% seemed tired during the day, and 10% fell asleep while watching television. To compare sleep predictors of daytime sleepiness between the US and Chinese children, stepwise regression analyses were conducted separately for both groups of children. It is interesting to note that short sleep duration was a main determinant of daytime sleepiness for Chinese children but not for the US children. Other important determinants for daytime sleepiness in Chinese children were episodic nocturnal behaviors such as teeth grinding and awakening screaming (possible night terrors). Daytime sleepiness in the US children was related to restless sleep, snoring, sleep-onset delay, and being afraid of sleeping in the dark, which all together accounted for a small variance (12%). This finding has important implications for effective assessment of and intervention for daytime sleepiness for the US and Chinese children, respectively. For the US children, screening and intervention should be focused on treatment of nighttime sleep problems, but for Chinese children, assessment and intervention should be focused on lengthening night-sleep duration either through advancing bedtime or delaying wake time (eg, school start times).
The main limitation of this study is that children's sleep was assessed by reports from parents rather than objective instruments, and parents were the only informants of children's sleep patterns and sleep problems. The differences in sleep patterns and sleep problems may be caused by reporting bias as described above. Another limitation of the comparative study is that we are not sure if the differences in sleep patterns and sleep problems can be generalized to all children in the United States and China, because only 3 schools were sampled from 1 school district in Jinan of China and southeastern New England of the United States. Furthermore, it should be noted that response rates differed greatly between the US and Chinese samples (47% vs 92%). It is unclear whether and to what extent the comparison of children's sleep could be affected by the difference in response rates of the 2 samples.
Four future directions for cross-cultural study of children's sleep patterns and sleep problems are suggested by the results of this study. Additional research is needed to investigate what cultural factors (child-rearing beliefs, attitudes, and practices) contribute to sleep practices and sleep problems among children of different ethnic and racial groups. Multiple informants (including teachers and self-report) as well as more objective methodologies such as actigraphy are needed to assess children's sleep in future comparative studies to reduce information bias. Third, prospective studies are needed to understand how sleep patterns and sleep problems in different cultural contexts exert effects on cognitive, behavioral, emotional, and physical development of children. Finally, our study showed that children's sleep duration differed between the United States and China. Large-scale epidemiologic studies with objective measures are needed to confirm this finding. If this conclusion is confirmed, additional studies are needed to investigate the effects of different sleep duration on children's development under different cultural contexts, and the optimal sleep duration for children from different cultural backgrounds should be recommended accordingly.
ACKNOWLEDGMENTS
This study was supported in part by the Shandong Medical Research Foundation, People's Republic of China, the Multi-Investigator Proposal Development Grant Program, Arizona State University, and National Institutes of Health National Heart, Lung, and Blood Institute grant HL 03896.
FOOTNOTES
Accepted Aug 5, 2004.
No conflict of interest declared.
REFERENCES
Anders TF, Eiben LA. Pediatric sleep disorders: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry. 1997;36 :9 –20
Sadeh A, Gruber R, Raviv A. Sleep, neurobehavioral functioning, and behavior problems in school-age children. Child Dev. 2002;73 :405 –417
Sadeh A, Gruber R, Raviv A. The effects of sleep restriction and extension on school-age children: what a difference an hour makes. Child Dev. 2003;74 :444 –455
Mindell JA. Sleep disorders in children. Health Psychol. 1993;12 :151 –162
Liu XC, Sun ZX, Uchiyama M, Shibui K, Kim K, Okawa M. Prevalence and correlates of sleep problems in Chinese schoolchildren. Sleep. 2000;23 :1053 –1062
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 –555
Sadeh A, Raviv A, Gruber R. Sleep patterns and sleep disruptions in school-age children. Dev Psychol. 2000;3 :291 –301
Stores G. Practitioner review: assessment and treatment of sleep disorders in children and adolescents. J Child Psychol Psychiatry. 1996;37 :907 –925
Stores G. Children's sleep disorders: modern approaches, developmental effects, and children at special risk. Dev Med Child Neurol. 1999;41 :568 –573
Rona RJ, Li L, Gulliford MC, Chinn S. Disturbed sleep: effects of sociocultural factors and illness. Arch Dis Child. 1998;78 :20 –25
Lozoff B, Askew GL, Wolf AW. Cosleeping and early childhood sleep problems: effects of ethnicity and socioeconomic status. J Dev Behav Pediatr. 1996;17 :9 –15
Archbold KH, Pituch KJ, Panabi P, Chervin RD. Symptoms of sleep disturbances among children at two general pediatric clinics. J Pediatr. 2002;140 :97 –102
Hayes MJ, Parker KG, Sallinen B, Davare AA. Bedsharing, temperament, and sleep disturbance in early childhood. Sleep. 2001;24 :657 –662
Roberts RE, Roberts CR, Chen IG. Ethnocultural differences in sleep complaints among adolescents. J Nerv Ment Dis. 2000;188 :222 –229
Quan SF, Goodwin JL, Babar SI, et al. Sleep architecture in normal Caucasian and Hispanic children aged 6–11 years recorded during unattended home polysomnography: experience from the Tucson Children's Assessment of Sleep Apnea Study (TuCASA). Sleep Med. 2003;4 :13 –19
Tynjala J, Kannas L, Valimaa R. How young Europeans sleep. Health Educ Res. 1993;8 :69 –80
Akimoto SA, Sanbonmatsu DM. Differences in self-effacing behavior between European and Japanese Americans. J Cross Cult Psychol. 1999;30 :159 –177
Liu XC, Liu LQ, Wang RZ. Bedsharing, sleep habits, and sleep problems among Chinese school-aged children. Sleep. 2003;26 :89 –94
Yang CK, Hahn HM. Cosleeping in young Korean children. J Dev Behav Pediatr. 2002;23 :151 –157
Willinger M, Ko CW, Hoffman HJ, Kessler RC, Corwin MJ. Trends in infant bed sharing in the United States, 1993–2000. Arch Pediatr Adolesc Med. 2003;157 :43 –49
Stein MT, Colarusso CA, McKenna JJ, Powers NG. Cosleeping (bedsharing) among infants and toddlers. J Dev Behav Pediatr. 1997;18 :408 –412
Blader JC, Koplewicz HS, Abikoff H, Foley C. Sleep problems of elementary school children. A community survey. Arch Pediatr Adolesc Med. 1997;151 :473 –480
Ottaviao S, Giannotti F, Cortesi F, Bruni O, Ottaviano C. Sleep characteristics in healthy children from birth to 6 years of age in the urban area of Rome. Sleep. 1996;19 :1 –3
Epstein R, Chillag N, Lavie P. Starting times of school: effects on daytime functioning of fifth-grade children in Israel. Sleep. 1998;21 :250 –256
Liu XC, Kurita H, Uchiyama M, Okawa M, Liu LQ, Ma DD. Life events, locus of control, and behavioral problems among Chinese adolescents. J Clin Psychol. 2000;56 :1565 –1577
Gau SF, Soong WT. Sleep problems of junior high school students in Taipei. Sleep. 1995;18 :667 –673
Mednick S, Nakayama K, Stickgold R. Sleep-dependent learning: a nap is as good as a night. Nat Neurosci. 2003;6 :697 –698
Takahashi M. The role of prescribed napping in sleep medicine. Sleep Med Rev. 2003;7 :227 –235
Takahashi M, Arito H. Maintenance of alertness and performance by a brief nap after lunch under prior sleep deficit. Sleep. 2000;23 :813 –819
Tao K. The only-child-per-family policy: a psychological perspective. In: Tseng WS, Wu DYH, eds. Chinese Culture and Mental Health. Orlando, FL: Academic Press; 1985:153–166
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;15 :1043 –1051
Owens J, Maxim R, Nobile C, McGuinn M, Alario A, Msall M. Television viewing habits and sleep disturbances in school-aged children. Pediatrics. 1999;104(3) . Available at: www.pediatrics.org/cgi/content/full/104/3/e27(Xianchen Liu, MD, PhD, Li)
Shandong Mental Health Center, Jinan, People's Republic of China
Division of Pediatric Ambulatory Medicine, Rhode Island Hospital, Brown University School of Medicine, Providence, Rhode Island
Prevention Research Center, Arizona State University, Tempe, Arizona
ABSTRACT
Objectives. Sleep patterns and sleep problems in children are not only influenced by a large number of biological and psychologic factors but also by cultural and social factors. Little is known about similarities and differences in sleep patterns and sleep problems among children across countries. We attempted to compare sleep patterns and sleep problems among schoolchildren from 2 countries with distinctive cultural contexts: the United States and China.
Methods. The data come from 2 cross-sectional surveys in 3 elementary schools of Jinan City, People's Republic of China, and 3 elementary schools from a suburban school district in southeastern New England, United States. The Chinese sample consisted of 517 elementary school children (grades 1 to 5), and the US sample consisted of 494 elementary school children (grades kindergarten through 4). We used the Children's Sleep Habits Questionnaire (CSHQ) to assess children's sleep patterns and sleep problems as reported by parents. Parents of the Chinese sample completed a Chinese version of the CSHQ.
Results. For children in both the US and Chinese samples, reported bedtime was delayed and sleep duration decreased with increasing age. Compared with the US children (grades 1–4), Chinese children went to bed approximately half an hour later (9:02 vs 8:27 PM) and woke up half an hour earlier (6:28 vs 6:55 AM), resulting in an average sleep duration that was 1 hour less (9.25 vs 10.15 hours). Chinese children were rated significantly higher than the US children on almost all CSHQ scales, indicating more sleep problems in Chinese children. Common sleep problems observed for all children were difficulty falling asleep, having a fear of sleeping in the dark, sleep talking, restless sleep, teeth grinding during sleep, and daytime sleepiness. Shorter daily sleep duration was associated with difficulty falling asleep, struggling at bedtime, and trouble sleeping away for the US children, and with going to bed at different times and having a fear of sleeping alone for Chinese children. Short sleep duration was a main predictor of daytime sleepiness for Chinese children, whereas restless sleep and snoring predicted daytime sleepiness for the US children.
Conclusions. As reported by parents, children in China went to bed later and woke up earlier and their sleep duration was 1 hour shorter than the US children. Chinese children were reported to have more sleep problems than their US counterparts. Daytime sleepiness was determined by sleep duration only for those who slept insufficiently. Unique school schedules and sleep practices may contribute to the differences in the sleep patterns and sleep problems of children from the United States and China.
Key Words: sleep patterns sleep problems sleep duration daytime sleepiness
Abbreviations: SES, socioeconomic status CSHQ, Children's Sleep Habits Questionnaire MANCOVA, multivariate analysis of covariance CI, confidence interval OR, odds ratio
Sleep plays very important roles in children's development. Sleep not only impacts on physical growth, behavior, and emotional development but also is closely related to cognitive functioning, learning, and attention.1–3 Sleep disorders in children are not uncommon. In a number of survey studies, parents report sleep disturbances throughout childhood, with the prevalence seeming to be 20% to 30%.1,4–8 Sleep patterns and sleep problems in children are not only influenced by a large number of biological and psychologic factors but also by cultural, social, and family factors.9–11 However, most epidemiologic studies of sleep patterns and sleep problems have been of Western children.6,11–13 Although several studies have examined ethnocultural differences in sleep practices and/or sleep disturbances in US children and adolescents,11,14,15 few studies have compared sleep patterns and sleep problems among children across countries using the same standardized sleep instruments.16 This study represents the first comparison of sleep patterns and sleep problems among children from the United States and China, representing 2 distinctive cultural contexts: Western and Asian.
Western child-rearing practices emphasize the development of independence and individualism, whereas Asian culture stresses the development of interpersonal relations, collectivism, family closeness, and social harmony.17 These child-rearing beliefs and cultural characteristics may determine child-rearing and family sleep practices. For example, cosleeping with parents in young children is accepted and very common in many Asian countries. In China, all-night cosleeping during the infant and early childhood years is the norm, decreases in prevalence with increasing age, and is reportedly unusual after puberty. A recent study of school-aged children in urban China reported that the prevalence of regular bed sharing was 18.2% and as high as 55.8% in 7-year-olds.18 In a study of bed sharing in Korean children, Yang and Hahn19 found that 73.5% of mothers approved of bed sharing between 3 to 6 years of age. The main reasons for bed sharing were "to look after the child while sleeping" and "the child is too young to sleep alone."19(p155) In Western countries, however, cosleeping has been considered somewhat of a cultural taboo, and in many of these countries it is a norm for a child to sleep separately from the parents after birth. In the United States, with the exception of certain ethnic and racial groups (particularly Hispanics and African Americans11), regular cosleeping with parents in non-Hispanic white families is uncommon in these societies. A recent US national survey between 1993 and 2000 indicated that only 9.2% of infants usually shared a bed with parents.20 Many Western child health professionals emphasize the hazards of cosleeping, such as impediment of the development of autonomy and independence, increased dependence on parents, inappropriate sexual arousal, risk of sleep problems, and interference with parental sexuality and intimacy.19,21 There is evidence that cosleeping is related to children's sleep patterns, sleep disturbances, and daytime sleepiness.18,22
Research has indicated that there seems to be marked cultural variations in the time of going to bed and amount of time slept.23 In a study of cultural differences in sleep habits and complaints of 40303 11- to 16-year-old European children from 11 countries, Tynjala et al16 reported that sleep habits varied significantly between countries, with Israeli children sleeping the shortest and Swiss children sleeping the longest. School start time may contribute to the differences in sleep duration between countries.24 Compared with the United States, elementary schools in China start at an earlier time, at 7:30 AM, and children spend more time in school and studying at home because of increased homework demands and the emphasis on academic achievement in Chinese culture.25,26 Obviously, an early start time for school requires an even earlier waking time, and nighttime homework may result in a later bedtime, which in turn may result in shorter sleep duration and consequent daytime sleepiness in Chinese children.
Another cultural characteristic of sleep habits in China is napping after lunch. In China, almost all schools have a long lunch break, 2 hours, between 12:00 and 2:00 PM. Although there are no statistical data available, anecdotally, many students have a 30-minute to 1-hour nap after lunch either at school or at home. Some students go home for lunch and may have a nap at home; others have lunch and may nap at school. Generally speaking, most people in China perceive the purpose of taking a nap after lunch not specifically as compensation for a shortened night sleep duration but for restoration and recovery of energy and spirit. Recent experimental studies have demonstrated favorable effects of brief naps on following alertness and cognitive performance in young adults.27–29 On the other hand, it is known that long naps may also have negative effects on night sleep, such as sleep onset and sleep quality and quantity.
In addition, China (as a developing country) differs greatly from the United States in socioeconomic status (SES), which may have a unique contribution to sleep practices and sleep problems among Chinese families. For example, compared with the United States, urban residents of China have few choices of housing, and living space is very crowded because of financial restriction and rapid population growth. Most urban families with young children are residing in an apartment, possibly with only 1 bedroom, which may prevent the child from having his/her own room or bed and increase the likelihood of bed sharing with parents and sleep problems caused by parental night activities. Moreover, almost all the families in urban China have only 1 child because of the 1-child family-planning policy. Only children may be more likely to be overindulged ("the little emperor") and/or overprotected, and parents (and grandparents) may have more difficulties setting limits.30 These characteristics may increase the likelihood of the child cosleeping with his or her parents and may decrease sleep quality and duration.
Despite these cultural and socioeconomic differences, little is known about the similarities and differences in sleep patterns and sleep problems between the US and Chinese children. We report here 2 school-based studies of children in the United States and China with the same sleep measure that allows us to (1) compare sleep patterns (bedtime, morning wakeup time, sleep duration) and sleep problems as reported by parents between children from China and the United States, (2) examine sleep problems that are associated with sleep duration among children from different cultural contexts, and (3) examine the relationship between sleep duration and sleep problems and daytime sleepiness for the US and Chinese children. We hypothesized that Chinese children would go to bed later, wake up earlier, and sleep less than the US children because of earlier school start time and heavier homework load. We also hypothesized that the parents of Chinese children would report more sleep problems than parents of the US children because of potentially increased parental awareness of sleep issues as a result of bed sharing or room sharing with the child. Finally, we expected that daytime sleepiness for both groups of children would be predicted by shorter sleep duration.
METHODS
Subjects
Chinese Sample
The Chinese participants consisted of 517 children 7 to 13 years old sampled from 3 public elementary schools in Jinan City. Jinan is the capital city of Shandong Province located in the eastern part of mainland China. In Jinan City, children are generally admitted to school at 6 or 7 years old. In elementary schools in Jinan, school starts at 7:30 AM, and classes are scheduled for 45 minutes, with a 15-minute break between classes. Morning classes conclude at 12:00 PM, and there is a lunch break for 90 minutes in winter and spring and 120 minutes in summer and fall. Most students go home for lunch, and some of them have a nap afterward. Some students, especially first- or second-graders, often stay at school during the lunch break and may nap after lunch. This survey was conducted in the spring of 2001, at which time the lunch break was 90 minutes. Students have to do 1 to 3 hours of homework almost every day including weekends and holidays. Homework increases with grade level. Most students also watch television after school, before, during, or after dinner. Unfortunately, no statistical data are available.
Detailed sampling procedures and methodology have been reported elsewhere.18 Briefly, the aims of the survey were explained to school principals and teachers in the target schools, and permission was obtained to conduct the survey, as is standard practice in China. One class from each grade for each school was selected randomly, yielding a pool of 565 children as potential participants. The children were asked by teachers to take a questionnaire to their parents with a letter explaining the project and requesting the parents to complete the questionnaire. The parents also were told that participation in the survey was voluntary. Of the 565 sampled children, 517 (92%) returned completed questionnaires for statistical analysis. This sample consisted of 47% boys, and mean age was 11.0 years (SD: 1.9; range: 7–13 years). Because in China there are no established systems to assess the SES of individual families, parental education levels were used as an indicator of family SES. In this sample, parental education levels were distributed as middle school or less (10.8%), high school (31.7%), some college (22.4%), and university (35.0%) for fathers and middle school or less (18.8%), high school (38.6%), some college (24.6%), and college graduate (17.9%) for mothers. The higher-education levels of the parents indicate that this sample represents a group of children from families of at least middle or high SES in urban China.
US Sample
The US sample consisted of 494 children who were enrolled in kindergarten through fourth grade from 3 public elementary schools in a predominantly white, middle-income, English-speaking suburban school district in southeastern New England.6,31,32 In the school system studied, kindergarten consisted of half-day sessions, with students equally distributed between morning (8:50–11:30 AM) and afternoon (12:30–2:45 PM) sessions for a half year each. For children in grades 1 to 4, school hours were between 8:50 AM and 2:45 PM. There was a half-hour lunch period between the morning and afternoon classes. Although children were encouraged to nap for a 1-hour period in kindergarten, data were not collected specifically regarding the duration of any scheduled naps.
Of 1099 questionnaire packets mailed to the parents of children in the 3 elementary schools, 494 children returned usable questionnaires for data analysis (response rate: 46.9%). Mean age of the sample was 7.56 years (SD: 1.51; range: 4.83–11.00 years). There were 240 boys (51.2%) and 229 girls (48.8%). SES was determined by using the Hollingshead Form Index of Social Status (A.B. Hollingshead, PhD, Four-Factor Index of Social Status, unpublished paper, 1975), which is based on occupation and education. The mean Hollingshead SES was 45.5 (SD: 11.3; class 4).
Daily television-viewing data for this sample were also obtained at the time of the sleep study; mean reported television-viewing time on weekdays was 2.07 hours (SD: 1.32). The amount of television viewed in this sample did not differ significantly by age but was higher in boys (2 = 5.67, P < .05) and in families of lower SES (t = 3.62, P < .001).32
Measure of Sleep Patterns and Sleep Problems
The Children's Sleep Habits Questionnaire (CSHQ)31 was used to assess sleep patterns and sleep problems of children in the United States and China. The CSHQ consists of 33 sleep-disturbance items and 3 items asking for information about sleep patterns (ie, bedtime, morning wakeup time, and daily total sleep duration). Parents are asked to recall the child's sleep behaviors over a "typical" recent week. Items are rated on a 3-point scale: "usually" if the sleep behavior occurred 5 to 7 times per week; "sometimes" for 2 to 4 times per week; and "rarely" for 0 to 1 time per week. The 33 sleep-disturbance items were conceptually grouped into 8 subscales: bedtime resistance, sleep-onset delay, sleep duration, sleep anxiety, night wakings, parasomnias, sleep-disordered breathing, and daytime sleepiness. The CSHQ has satisfactory psychometric properties in US children.31
We developed a Chinese version of the CSHQ by translation and back translation for this study. Cronbach's using data from these Chinese children was .80 for the entire scale and .49 to .72 for the subscales. The psychometric properties of the Chinese CSHQ with this sample are comparable to those found in US children for most of the subscales.18,31
Statistical Analysis
We performed multivariate analyses of covariance (MANCOVAs) with gender as a between-subjects factor and grade as a covariate to assess the effects of gender and grade on sleep patterns for the US and Chinese children, respectively. We conducted MANCOVAs with nationality and gender as between-subjects factors and age as covariate for pooled children in grades 1 to 4. Multivariate logistic regression analyses were conducted to examine the differences in frequencies of sleep problems between the 2 groups of children after adjustment for the effects of age and gender. Odds ratios and 95% confidence intervals (CIs) were reported to show the differences. Finally, stepwise regression analyses were performed to examine the sleep problems that could predict sleep duration and the association between sleep duration and sleep problems that could predict daytime sleepiness for the US and Chinese children, respectively. Standardized regression coefficients were reported to present the effect size of each significant predictor.
All analyses were performed by using the Statistical Program for Social Sciences (SPSS) for Windows, version 11.5. All statistical significance was set at P < .05.
RESULTS
Sleep/Wake Patterns
For the US children, mean night bedtime was 8:27 PM (SD: 30.6 minutes), mean morning wakeup time was 6:55 AM (SD: 25.8 minutes), and mean sleep duration was 10.64 hours (SD: 0.67). Although boys went to bed later and woke up earlier than girls, multivariate analysis of variance did not show significant differences in bedtime and wake-up time between genders (P > .05). Girls slept significantly longer than boys (10.31 ± 0.70 vs 10.17 ± 0.64; F = 6.45, P = .011). With advancing grade, children's bedtime was significantly postponed ( = .102, t = 6.12, P < .001) and total sleep duration declined ( = –.137, t = 5.78, P < .001).
For Chinese children, mean night bedtime was 9:06 PM (SD: 44.4 minutes), mean morning wakeup time was 6:24 AM (SD: 19.8 minutes), and mean sleep duration was 9.13 hours (SD: 0.74). No significant gender differences in bedtime, wakeup time, and sleep duration were observed. With advancing grade, bedtime was significantly postponed ( = .05, t = 2.66, P = .01), wakeup time advanced ( = –.045, t = 5.48, P < .001), and total sleep duration significantly declined ( = –.094, t = 5.14, P < .001).
Comparisons of CSHQ Scale Scores
Comparisons of Individual Sleep Problems
Frequencies for individual sleep problems are presented in Table 2 for the US and Chinese children. For the US children, the most prevalent specific sleep problems (defined as occurring sometimes or usually in at least 20% of the sample) were: awakened by others in the morning (68%); hard time getting out of bed (38.7%); restless sleep (35.1%); waking up in a bad mood (34.1%); falling asleep while riding in a car or bus (27.1%); being afraid of sleeping in the dark (25.9%); waking once during the night (25.1%); being tired during the day (24.0%); taking a long time to be alert in the morning (23.5%); sleep talking (23.2%); difficulty falling asleep in 20 minutes (21%); teeth grinding during sleep (20.7%); and getting too little sleep (20.2%). Rare sleep problems (occurring in <5% of the sample) were: sleepwalking (3.8%); waking more than once during the night (2.5%); holding breath (1.3%); and awakening screaming or sweating (1.3%).
For Chinese children, prevalent sleep problems were: awakening by others in the morning (58.2%); hard time getting out of bed (43.5%); getting too little sleep (41.4%); falling asleep while riding in a car or bus (36.6%); restless sleep (36.3%); waking once during the night (35%); being afraid of sleeping alone (29.5%); being afraid of sleeping in the dark (29.1%); waking up in a bad mood (26.1%); needing a parent in the room to sleep (25.3%); falling asleep in someone else's bed (23%); taking a long time to be alert (22.6%); teeth grinding during sleep (21.6%); trouble sleeping away from home (21.5%); being tired during the day (21%); sleep talking (20.5%); and difficulty falling asleep in 20 minutes (20.2%). There were no sleep problems that happened in <5% of the Chinese children.
A series of multivariate logistic regression analyses were conducted to examine the differences in frequencies of sleep problems between the 2 groups of children, adjusting for the effects of age and gender. In the logistic regression model, the dependent variable was defined as 1 if the item rated as "sometimes" or "usually" and 0 if rated as "no" or "rarely." As shown in Table 2, most sleep problems were more prevalent in Chinese children than in the US children. The following sleep problems that Chinese children were 2 times more likely to have than the US children and remained significant after adjustment of age and gender were: going to bed at different times (odds ratio [OR]: 2.5); needing a parent in the room to sleep (OR: 3.5); being afraid of sleeping alone (OR: 3.3); waking more than once during the night (OR: 4.6); getting too little sleep (OR: 3.1); not sleeping the same amount each day (OR: 4.3); holding breath during sleep (OR: 9.4); snorting and gasping (OR: 3.0); and awakening screaming or sweating (OR: 6.3). A total of 13 sleep problems were commonly reported for both groups of children but did not significantly differ: difficulty falling asleep in 20 minutes; being afraid of sleeping in the dark; waking once during the night; sleep talking; restless sleep; teeth grinding; and all items regarding daytime sleepiness.
Associations of Sleep Problems With Sleep Duration
Stepwise linear regression analyses were conducted to examine the associations of individual sleep problems with sleep duration for the US and Chinese children, separately. In the regression analysis, items regarding daytime sleepiness were not selected as predicting variables, because we hypothesized that sleep duration could predict daytime sleepiness as described below. All variables in Table 3 were entered for stepwise regression analysis. As indicated in Table 3, 4 sleep problems in the final model predicted sleep duration for the US children: difficulty falling asleep in 20 minutes ( = –0.209); struggling at bedtime ( = –0.254); trouble sleeping away ( = –0.117); and bedwetting ( = 0.149). The 4 sleep problems accounted for 15.6% of variance of sleep duration. Bedwetting was associated with increased sleep duration, but 3 other sleep problems were associated with decreased sleep duration. For Chinese children, only 2 sleep problems remained significant in the final model: going to bed at different times ( = –0.213) and being afraid of sleeping alone ( = –0.146), accounting for 7.2% variance of sleep duration.
Association of Daytime Sleepiness With Sleep Duration and Sleep Problems
Daytime sleepiness was very prevalent in both US and Chinese children. Stepwise regression analyses were performed to examine the association between sleep duration and sleep problems and daytime sleepiness for the US and Chinese children, with the CSHQ daytime-sleepiness scale score as the dependent variable. Table 4 presents all variables that were entered initially for stepwise regression and standardized regression coefficients for those variables that remained significant in the final model.
For the US children, 4 sleep problems significantly predicted daytime sleepiness in the final model: difficulty falling asleep in 20 minutes ( = 0.138), being afraid of sleeping in the dark ( = 0.130), snoring ( = 0.178), and restless sleep ( = 0.193), accounting for 12.1% variance. All 4 sleep problems were associated with increased likelihood for daytime sleepiness. Sleep duration was not a predictor. For Chinese children, sleep duration ( = –0.205) and 5 sleep problems were significant predictors of daytime sleepiness, accounting for 21.5% variance. The 5 sleep problems were: needing a parent in the room to sleep ( = 0.145); snorting and gasping ( = 0.133); restless sleep ( = 0.125); teeth grinding ( = 0.148); and awakening screaming ( = 0.166). As expected, sleep duration was negatively associated with daytime sleepiness, and the 5 sleep problems were positively related to daytime sleepiness.
DISCUSSION
This is the first study to directly compare sleep patterns and sleep problems among elementary school children between the United States and China. Children's sleep patterns and sleep problems were reported by parents using the same well-established instrument, the CSHQ.31 The main findings of this study can be summarized as follows: (1) for all children in the United States and China, bedtime was increasingly delayed and sleep duration decreased with increasing age; (2) compared with the same-aged US children, Chinese children went to bed half an hour later, woke up half an hour earlier, and slept 1 hour less; (3) Chinese children were rated significantly higher than the US children on almost all CSHQ scales; (4) difficulty falling asleep, being afraid of sleeping in the dark, sleep talking, restless sleep, and teeth grinding during sleep and daytime sleepiness were common sleep problems for both groups of children, although most sleep problems were more frequent in Chinese children than in the US children; (5) there was a relationship between shorter sleep duration and difficulty falling asleep, struggling at bedtime, and trouble sleeping away from home for the US children, whereas going to bed at different times and having a fear of sleeping alone were related to shorter sleep duration in the Chinese sample; and (6) short sleep duration was a main predictor of daytime sleepiness for Chinese children, whereas restless sleep and snoring during sleep were main predictors for the US children.
In the current study, we found that sleep patterns in Chinese children were significantly different from the US children, including later bedtime, earlier morning wake time, and shorter sleep duration by 1 hour. The difference in sleep patterns in children between the 2 countries may be attributed to different school schedules and homework load and sleep practices of the 2 countries. First, school started at 7:30 AM in the sampled school district of China, 1 hour and 20 minutes earlier than the sampled school district of the United States (8:50 AM). The earlier school start time necessitates the Chinese children to get up earlier for the morning class. Second, it is well known that school children in China have more homework than most Western countries because of the high academic expectations of their parents and schools.25,26 Most children in China have to do homework after dinner for 1 or 2 hours depending on the amount of homework, whereas the usual homework load for the US children was not >30 minutes. Third, in the Chinese sample, 48% of children routinely shared a bed (26.4%) or room (21.6%) with their parents. Room sharing or bed sharing with parents may also contribute to later bedtimes and earlier morning wakeup times in Chinese children, because children's sleep patterns are more likely to be influenced by the sleep patterns of cosleeping parents.
It is important to note that Chinese children slept 1 hour less than the US children according to parent report. There are several potential explanations for the difference. First, Chinese children went to bed later and woke up earlier as described above. This is the most likely possibility to explain why Chinese children slept less. Second, Chinese parents may use night-sleep duration instead of total daily sleep duration, ignoring naptime. Napping is a well-accepted sleep habit in China. Most schools schedule lunch breaks of 2 hours, which allows students to have a brief nap of 30 minutes in winter and spring, and 1 hour in summer and fall. Thus, to some extent, the nap may explain the difference in sleep duration between the US and Chinese children. Unfortunately, in this study we did not collect specific data regarding naps. Alternatively, shorter sleep duration in Chinese children may be explained by sleep problems, because most sleep problems were more prevalent in Chinese children. However, stepwise regression analysis revealed only 2 sleep problems that were significantly associated with sleep duration and accounted for only 7% variance. Another possibility is that Chinese children require less sleep than the US children. This can be interpreted by the fact that daytime sleepiness did not differ between the US and Chinese children. It is also possible that the difference in sleep duration between the 2 groups of children may be caused by parental overestimation of sleep duration for the US children and/or underestimation of sleep duration for Chinese children (ie, reporting bias). The US parents may know less about their child's real sleep than Chinese parents because of the high frequency of room sharing or bed sharing with parents in Chinese children. The US parents may overestimate their child's sleep duration using bedtime instead of the time of actual sleep onset, which may be significantly later than bedtime because of difficulty falling asleep and other bedtime-resistance problems.6 Finally, although we did not collect specific information regarding differences in weekend (non–school days) bedtimes and wake times, it is possible that the Chinese children attempt to "make up" for relatively less sleep during the week by increasing sleep duration on the weekends.
Most sleep problems were more frequent in Chinese children than in the US children, although difficulty falling asleep, being afraid of sleeping in the dark, sleep talking, restless sleep, teeth grinding during sleep, and daytime sleepiness were very common for both groups of children. Cultural characteristics of sleep practices and environment may contribute to more sleep problems in Chinese children, such as cosleeping, being an only child in the family, crowded housing, and chronic stress caused by homework overload. For example, crowded housing in urban residents of China is very common, and many families with young children may have only 1 bedroom. In this sample, >60% of families owned a living space <20 m2 per person and 35% owned a living space <15 m2. Crowded housing limits children having their own room and even a separate bed at home, and parents have to bed share with their children. There is evidence that crowded housing and bed sharing affect children's sleep quality and quantity,18,22 which may explain why Chinese children are more likely to go to bed at different times, wake in the night, and sleep different amounts each day. Another potential factor that may contribute to higher rates of sleep problems among Chinese children should not be ignored: being the only child in the family because of the family-planning policy. An only child is frequently regarded as "a little emperor" or "a little sun," namely a spoiled/overprotected child. Only children in China are described as overdependent, egocentric, less cooperative, timid, and unadjusted.30 These characteristics may increase the likelihood of children bed sharing with parents and sleep-resistance problems such as needing a parent in the room to sleep and being afraid of sleeping alone. Another explanation for the difference is that Chinese children are more likely to develop sleep problems because of genetic effects. However, to date there has been no evidence to indicate that specific genes are linked to ethnic differences in sleep patterns and sleep problems. It is also possible that the difference is caused by information bias. The US parents may underestimate their child's sleep problems because the US parents and their children slept in separate rooms and subsequently know less about their children's sleep problems, which may be why some parasomnia and sleep-disordered breathing behaviors were less likely to be reported by the US parents.
Daytime sleepiness was very common for both groups of children. Approximately 60% of children needed to be awakened by parents in the morning at least 2 to 4 times a week, 40% had a hard time getting out of bed in the morning, 30% woke up in a bad mood, 26% to 36% fell asleep while riding in a car, 23% took a long time to be alert, 22% seemed tired during the day, and 10% fell asleep while watching television. To compare sleep predictors of daytime sleepiness between the US and Chinese children, stepwise regression analyses were conducted separately for both groups of children. It is interesting to note that short sleep duration was a main determinant of daytime sleepiness for Chinese children but not for the US children. Other important determinants for daytime sleepiness in Chinese children were episodic nocturnal behaviors such as teeth grinding and awakening screaming (possible night terrors). Daytime sleepiness in the US children was related to restless sleep, snoring, sleep-onset delay, and being afraid of sleeping in the dark, which all together accounted for a small variance (12%). This finding has important implications for effective assessment of and intervention for daytime sleepiness for the US and Chinese children, respectively. For the US children, screening and intervention should be focused on treatment of nighttime sleep problems, but for Chinese children, assessment and intervention should be focused on lengthening night-sleep duration either through advancing bedtime or delaying wake time (eg, school start times).
The main limitation of this study is that children's sleep was assessed by reports from parents rather than objective instruments, and parents were the only informants of children's sleep patterns and sleep problems. The differences in sleep patterns and sleep problems may be caused by reporting bias as described above. Another limitation of the comparative study is that we are not sure if the differences in sleep patterns and sleep problems can be generalized to all children in the United States and China, because only 3 schools were sampled from 1 school district in Jinan of China and southeastern New England of the United States. Furthermore, it should be noted that response rates differed greatly between the US and Chinese samples (47% vs 92%). It is unclear whether and to what extent the comparison of children's sleep could be affected by the difference in response rates of the 2 samples.
Four future directions for cross-cultural study of children's sleep patterns and sleep problems are suggested by the results of this study. Additional research is needed to investigate what cultural factors (child-rearing beliefs, attitudes, and practices) contribute to sleep practices and sleep problems among children of different ethnic and racial groups. Multiple informants (including teachers and self-report) as well as more objective methodologies such as actigraphy are needed to assess children's sleep in future comparative studies to reduce information bias. Third, prospective studies are needed to understand how sleep patterns and sleep problems in different cultural contexts exert effects on cognitive, behavioral, emotional, and physical development of children. Finally, our study showed that children's sleep duration differed between the United States and China. Large-scale epidemiologic studies with objective measures are needed to confirm this finding. If this conclusion is confirmed, additional studies are needed to investigate the effects of different sleep duration on children's development under different cultural contexts, and the optimal sleep duration for children from different cultural backgrounds should be recommended accordingly.
ACKNOWLEDGMENTS
This study was supported in part by the Shandong Medical Research Foundation, People's Republic of China, the Multi-Investigator Proposal Development Grant Program, Arizona State University, and National Institutes of Health National Heart, Lung, and Blood Institute grant HL 03896.
FOOTNOTES
Accepted Aug 5, 2004.
No conflict of interest declared.
REFERENCES
Anders TF, Eiben LA. Pediatric sleep disorders: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry. 1997;36 :9 –20
Sadeh A, Gruber R, Raviv A. Sleep, neurobehavioral functioning, and behavior problems in school-age children. Child Dev. 2002;73 :405 –417
Sadeh A, Gruber R, Raviv A. The effects of sleep restriction and extension on school-age children: what a difference an hour makes. Child Dev. 2003;74 :444 –455
Mindell JA. Sleep disorders in children. Health Psychol. 1993;12 :151 –162
Liu XC, Sun ZX, Uchiyama M, Shibui K, Kim K, Okawa M. Prevalence and correlates of sleep problems in Chinese schoolchildren. Sleep. 2000;23 :1053 –1062
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 –555
Sadeh A, Raviv A, Gruber R. Sleep patterns and sleep disruptions in school-age children. Dev Psychol. 2000;3 :291 –301
Stores G. Practitioner review: assessment and treatment of sleep disorders in children and adolescents. J Child Psychol Psychiatry. 1996;37 :907 –925
Stores G. Children's sleep disorders: modern approaches, developmental effects, and children at special risk. Dev Med Child Neurol. 1999;41 :568 –573
Rona RJ, Li L, Gulliford MC, Chinn S. Disturbed sleep: effects of sociocultural factors and illness. Arch Dis Child. 1998;78 :20 –25
Lozoff B, Askew GL, Wolf AW. Cosleeping and early childhood sleep problems: effects of ethnicity and socioeconomic status. J Dev Behav Pediatr. 1996;17 :9 –15
Archbold KH, Pituch KJ, Panabi P, Chervin RD. Symptoms of sleep disturbances among children at two general pediatric clinics. J Pediatr. 2002;140 :97 –102
Hayes MJ, Parker KG, Sallinen B, Davare AA. Bedsharing, temperament, and sleep disturbance in early childhood. Sleep. 2001;24 :657 –662
Roberts RE, Roberts CR, Chen IG. Ethnocultural differences in sleep complaints among adolescents. J Nerv Ment Dis. 2000;188 :222 –229
Quan SF, Goodwin JL, Babar SI, et al. Sleep architecture in normal Caucasian and Hispanic children aged 6–11 years recorded during unattended home polysomnography: experience from the Tucson Children's Assessment of Sleep Apnea Study (TuCASA). Sleep Med. 2003;4 :13 –19
Tynjala J, Kannas L, Valimaa R. How young Europeans sleep. Health Educ Res. 1993;8 :69 –80
Akimoto SA, Sanbonmatsu DM. Differences in self-effacing behavior between European and Japanese Americans. J Cross Cult Psychol. 1999;30 :159 –177
Liu XC, Liu LQ, Wang RZ. Bedsharing, sleep habits, and sleep problems among Chinese school-aged children. Sleep. 2003;26 :89 –94
Yang CK, Hahn HM. Cosleeping in young Korean children. J Dev Behav Pediatr. 2002;23 :151 –157
Willinger M, Ko CW, Hoffman HJ, Kessler RC, Corwin MJ. Trends in infant bed sharing in the United States, 1993–2000. Arch Pediatr Adolesc Med. 2003;157 :43 –49
Stein MT, Colarusso CA, McKenna JJ, Powers NG. Cosleeping (bedsharing) among infants and toddlers. J Dev Behav Pediatr. 1997;18 :408 –412
Blader JC, Koplewicz HS, Abikoff H, Foley C. Sleep problems of elementary school children. A community survey. Arch Pediatr Adolesc Med. 1997;151 :473 –480
Ottaviao S, Giannotti F, Cortesi F, Bruni O, Ottaviano C. Sleep characteristics in healthy children from birth to 6 years of age in the urban area of Rome. Sleep. 1996;19 :1 –3
Epstein R, Chillag N, Lavie P. Starting times of school: effects on daytime functioning of fifth-grade children in Israel. Sleep. 1998;21 :250 –256
Liu XC, Kurita H, Uchiyama M, Okawa M, Liu LQ, Ma DD. Life events, locus of control, and behavioral problems among Chinese adolescents. J Clin Psychol. 2000;56 :1565 –1577
Gau SF, Soong WT. Sleep problems of junior high school students in Taipei. Sleep. 1995;18 :667 –673
Mednick S, Nakayama K, Stickgold R. Sleep-dependent learning: a nap is as good as a night. Nat Neurosci. 2003;6 :697 –698
Takahashi M. The role of prescribed napping in sleep medicine. Sleep Med Rev. 2003;7 :227 –235
Takahashi M, Arito H. Maintenance of alertness and performance by a brief nap after lunch under prior sleep deficit. Sleep. 2000;23 :813 –819
Tao K. The only-child-per-family policy: a psychological perspective. In: Tseng WS, Wu DYH, eds. Chinese Culture and Mental Health. Orlando, FL: Academic Press; 1985:153–166
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;15 :1043 –1051
Owens J, Maxim R, Nobile C, McGuinn M, Alario A, Msall M. Television viewing habits and sleep disturbances in school-aged children. Pediatrics. 1999;104(3) . Available at: www.pediatrics.org/cgi/content/full/104/3/e27(Xianchen Liu, MD, PhD, Li)