Reliability of the Children's Sleep Habits Questionnaire Hebrew Translation
and Cross Cultural Comparison of the Psychometric Properties

Orna Tzchishinsky, D.Sc.,1 Dubi Lufi, Ph.D.1 and Tamar Shochat, D.Sc.2 Your Idea: Protect It Before It's Too Late!

Abstract

Objective: Sleep disturbances and poor sleep habits in children have been associated with reduced daytime functioning. To investigate cross-cultural differences in sleep habits and disorders in school aged children, we assessed internal consistency of the Children's Sleep Habits Questionnaire (CSHQ) translated to Hebrew for use in the Israeli population, and compared standardized sleep assessments in an Israeli sample with both Chinese and U.S. samples using the CSHQ.

Methods: The questionnaire was translated based on standard criteria. Parents of 98 healthy school aged children in the 1st to 6th grades (43 males, 55 females, mean age 9.2 +/- 1.8) in northern Israel completed the questionnaire. Eight sub scores and a total score were computed. Comparisons were made with reported Chinese and U.S. samples.

Results: Interitem reliability (Cronbach's alpha) was 0.81. Strong correlations were found between most subscores. Sleep disturbances and poor sleep habits were most prevalent in Israeli children and least prevalent in the U.S. children; Chinese children were generally similar to the Israeli children.

Conclusions: Reliability of the translated CSHQ was found to be high in an Israeli sample of school aged children. This study is the first to present a comparison of sleep habits and disturbances between Israeli, Chinese and U.S. children, using standardized measurement.

Sleep disturbances in school-aged children is a topic which has received growing attention in recent years. Sleep disorders and insufficient sleep have been detrimentally associated with neuropsychological and neurobehavioral development, cognitive and academic functioning, as well as health related quality of life.^{3–5, 8, 11–13, 19–21, 24, 27}

Prevalence studies based on parental questionnaires and interviews have demonstrated that sleep disorders are common in school aged children.^{1, 2, 15, 17, 22, 23} Thus, bedtime resistance was reported in 15–27% ^{2, 17} and difficulty rising in the morning during weekdays in 17–21%.^{1, 2, 17} Sleep onset delay was reported in 6–12% ^{1, 2, 22, 23} and night waking in 4–16%.^2,22,23 Snoring was reported in 8–12%,^15,22,23 witnessed sleep disordered breathing in 1.5–3.7%, ^{15, 17, 23} nocturnal enuresis in 5% ,^{15, 22, 23} tooth grinding in 8–20%^{15, 23}and nightmares in 2–3%.^{22, 23} Finally, daytime fatigue was reported with a prevalence of 17–38% ^{1, 2} and daytime napping in 29–45%.¹ Reported overall prevalence of sleep problems in school children aged 4–12 ranged between 11% 23 to 37%.¹⁷

------------------------------------------------
¹Emek Yezreel Academic College, Emek Yezreel, Israel.
²Faculty of Social Welfare and Health Sciences, University of Haifa, Mount Carmel, Haifa, Israel.
Dr. Orna Tzischinsky, Department of Behavioral Sciences, Emek Yezreel Academic College, Emek Yezreel, Israel.
Email: orna@yvc.ac.il
------------------------------------------------

These studies clearly demonstrate that although prevalence rates are high, it is difficult to draw conclusions regarding rates of specific and overall sleep disturbances, or to perform meaningful cultural comparisons due to the lack of standardized measurement.

In a multicultural comparison of sleep and health habits in over 40,000 adolescent school children aged 11–16 in 11 European countries,²⁶ Israeli children had the shortest sleep times while Swiss children slept the longest; Finnish children reported the highest rates of sleep onset difficulties and daytime fatigue. Factors related to late bedtimes and sleep onset difficulties included the use of alcohol and tobacco, lack of physical activity, watching TV and video and spending evenings outside the home.

Few normative studies on sleep patterns in healthy Israeli elementary school children have been reported, and standardized measurement is lacking. One such study investigated the effects of school start times in Israel on sleep duration and fatigue in 572 fifth grade students based on a self report questionnaire.⁷ Sleep duration was significantly shorter for the early risers, who complained significantly more than the late risers of daytime fatigue, sleepiness and poor concentration at school. Yet another study sought to establish normative developmental data of sleep patterns in Israeli children in the 2nd, 4th and 6th grades using actigraphy, sleep diaries and a questionnaire based on parent and child reports.¹⁹ Questionnaire data revealed increased morning drowsiness and a tendency for daytime napping with age. We found no other normative studies performed in the Israeli elementary school population.

Epidemiologic studies of sleep habits and disorders in school children call for a more comprehensive and standardized tool, that would allow the estimation of prevalence, incidence, correlates and cross-cultural comparisons. The Children's Sleep Habits Questionnaire (CSHQ) is a pediatric screening tool for sleep habits and disorders in elementary school children with good psychometric properties.¹⁸ It is based on parental reports and encompasses sleep habits and prevalent physiological and behavioral symptoms of pediatric sleep disorders.

The CSHQ has been translated to Chinese, yielding high overall internal consistency in a public school sample.¹⁴ Mean sub and total scores were compared with an equivalent U.S. sample, showing increased sleep disturbance in Chinese children including later bedtimes, earlier rise times and shorter sleep times than their U.S. counterparts. Differences between samples were related to earlier school schedules, increased homework load, parent-child room sharing and report bias due to differences in sleeping arrangements.

Due to the need for a standardized assessment tool in Hebrew with good psychometric properties for both sleep habits and sleep disorders in children, our aims in the present study were to assess internal consistency of the Hebrew translation of the CSHQ in healthy elementary school aged children, and to compare standardized sleep assessments in the Israeli sample with both Chinese and U.S. samples using the CSHQ.

Methods

Subjects

The parents of 98 healthy, normal elementary school aged children in the 1st to 6th grades (43 males, 55 females, mean age 9.2 +/- 1.8) from urban and rural middle class communities in Northern Israel participated in the study.

Measures

The CSHQ: The original 45 item questionnaire was reduced to 33 items, which are divided into eight subscales representing different domains of sleep disorders, including bedtime resistance, sleep onset delay, sleep duration, sleep anxiety, night wakings, parasomnias, sleep disordered breathing (SDB) and daytime sleepiness. Item scores in each subscale are summed to obtain 8 subscores, and a total score is the sum of all 33 items. Parents are asked to recall the child's sleep behaviors over a typical recent week. Items are rated on a 3 –points scale: "usually" if the sleep behavior occurred 5–7 times per week; "sometimes" for 2–4 times per week; and "rarely" for 0–1 time per week. Psychometric properties of the questionnaire in clinical and non clinical populations has yielded overall internal consistency values of 0.68 and 0.78 respectively.¹⁸ In addition to the 8 component subscores, sleep patterns were assessed based on parent reported bedtime and wake time, and computed total sleep time based on bed/wake time reports. The CSHQ was translated to Hebrew by two sleep researchers fluent in both Hebrew and English. The translation was then evaluated independently by co investigators. Following modifications, the questionnaire was back translated from Hebrew to English, and additional adjustments were made. All co-authors reviewed and approved the final translation.

Procedure

The final Hebrew version (CSHQ-H) was administered to and subsequently collected from the parents by B.A. psychology and nursing students as part of the requirements in their research seminar on sleep disorders. The study was approved by the Haifa University ethics committee, and by the Israeli Ministry of Education.

Statistical Analyses

To assess reliability of the CSHQ-H, Cronbach's alpha coefficients were computed for all 33 items and for each subscale separately. Comparison between the Israeli, USA, and the Chinese sample were performed by analysis of variance (ANOVA), and post hoc Bonferroni test.

Results

Interitem consistency: Table 1 presents Cronbach's alpha coefficients for each of the 8 subscale items and for all items (total 22.00 CHSQ) for the Israeli, the U.S. clinical and non clinical 18 and the Chinese 14 samples. Cronbach's alpha values for all items were 0.81, 0.78, 0.68 and 0.80 for the four samples respectively. Interrelationships among subscores: Table 2 presents Pearson correlation coefficients matrix between the CHSQ-H sub-scores. Significant correlations were found between most subscores, excluding SDB which was weakly correlated only with parasomnias.

Comparisons of the sleep patterns and CSHQ subscales: Table 3 presents comparisons for sleep/wake patterns (bedtime, wake time, total sleep time), CSHQ sub and total scores (means +/- SD) between the three groups (Israeli, U.S. non clinical and Chinese) with post hoc Bonferroni tests. Most subscores were similar for the Chinese and Israeli samples and higher than the U.S. sample, indicating more disturbed sleep in the former (excluding sleep onset delay and parasomnia scores which were similar for all three samples). Bedtime was latest for the Israeli sample and earliest for the U.S. sample, wake time was latest for the U.S. sample and earliest for the Chinese sample. Total sleep time was longest for the U.S. sample and similar in the Chinese and Israeli samples, and total score showed highest values for the Israeli sample followed by the Chinese and lowest scores for the U.S. sample.

Figure 1(a-c) shows reported sleep/wake patterns (bedtime, wake time and total sleep time) comparing the U.S. non clinical and Chinese samples with the Israeli sample divided to two age groups: ages 6–9 and 10–12. Israeli children had later bedtimes than U.S. and Chinese children; bedtimes were on average 19 and 36 minutes later for the younger Israeli group and 53 and 70 minutes later for the older Israeli group than the Chinese and U.S. groups respectively. Wake times for the two Israeli groups were identical, and were 13 minutes later than the Chinese sample and 13 minutes earlier than the U.S. sample. Total sleep time was lowest for the older Israeli group (8.76), followed by the Chinese sample (9.25) and younger Israeli group (9.31), and the longest for the U.S. sample (10.15).

Comparisons of the individual sleep problems: Table 4 shows Israeli children ages 6–9 and ages 10–12, and Chinese and U.S. the frequencies of the individual CSHQ items comparing the Israeli (ages 6–9 and 10–12), Chinese and U.S. non clinical samples. The percentages of sleep problems occurring sometimes or usually are presented. Prevalent sleep problems were defined as those reported as occurring sometimes or usually by at least 20% of the sample.¹⁴

Discussion

In the present study, reliability of a pediatric sleep questionnaire based on parental report translated to Hebrew, the CSHQ-H, was found to be high in a non clinical community sample of elementary school children. In addition, this study is the first to present a comparison of sleep habits and disturbances between Israeli children with Chinese and U.S. children, using standardized measurement.

Reliability measures of the subscales in the present study were highly similar to those reported by the U.S. non clinical sample, with highest internal consistency coefficients for bedtime resistance, sleep duration, sleep anxiety and daytime sleepiness.¹⁸ However, reliability of the entire questionnaire was considerably higher for the Hebrew sample compared to the U.S. non clinical sample, but similar to that of the Chinese 14 and the U.S. clinical samples.¹⁸ As sleep disorders were generally higher in the Israeli and Chinese samples, these findings may indicate that internal consistency is higher in populations with more disturbed sleep.

Additionally, school grade differences exist between the samples. The Israeli and Chinese samples included elementary school children from first (ages 6–7) to fifth (Chinese) and sixth (Israeli) grades (ages 11–12), while the American non clinical sample included younger children in kindergarten (ages 4–5) up to fourth grade only (ages 9–10). The kindergartners had fewer hours at school than the first to fourth graders, and their schedule was split to morning or afternoon sessions. Clearly, the American sample was more heterogeneous than the Israeli and Chinese samples. This may explain the lower reliability in the U.S. sample, possibly reflecting differences in sleep/wake patterns and in parental assessment of sleep disturbances between kindergarteners and children in grades 1–4.

Correlations among the subscales in the Israeli sample were generally high, with the highest correlation between bedtime resistance and sleep anxiety, similarly to the U.S. study.18 The strong relationships between bedtime resistance, sleep anxiety, sleep duration, sleep onset delay and night wakings may reflect childhood insomnia; whereas strong relationships between parasomnias, night wakings and sleep anxiety are likely to reflect arousal parasomnias, e.g., night terrors and confusional arousals.

Israeli children had significantly more sleep disturbances overall as well as later bedtimes than Chinese and U.S. children. Levels of most subscales were similar between the Israeli and Chinese children, and higher than the U.S. children. Thus, Israeli and Chinese children exhibited more bedtime resistance, sleep anxiety, nighttime wakings, SDB, as well as less sleep duration compared to U.S. children. Wake times for the Israeli children were later than the Chinese children, and earlier than the U.S. children, likely reflecting the younger age range and the different school schedule for the kindergartners. It is important to note that we could not control for age differences, as we did not have access to their raw data. Finally, daytime sleepiness was higher for the Israeli children than both the Chinese and U.S. children.

When comparing the prevalence of specific sleep problems, the young Israeli group demonstrated the most sleep problems with a prevalence of at least 20%, especially problems of daytime sleepiness. Interestingly, the older Israeli group showed a higher prevalence of snoring loudly and snoring and gasping compared to other groups, and reports of falling asleep while watching television were highly prevalent for the young Israeli group.

Similarly to the U.S. and Chinese groups,¹⁴ as Israeli children get older, they go to bed later and sleep less. Nevertheless, Israeli children go to bed latest and have the highest amount of daytime sleepiness compared to their Chinese and U.S. counterparts. In the Chinese sample, later bedtimes and earlier rise times (compared to U.S. sample) were attributed to different school schedules, homework load and sleeping arrangements.¹⁴ In Israel, school start times vary widely,⁷ ranging from 07:10–08:30, and a limitation of the present study is the lack of control for school start times in our sample. Furthermore, although homework load and academic expectations have been indicated as important factors for late sleep times in Chinese school children,^9,14 in European countries including Israel, recreational habits such as TV/video games, and use of psychoactive substances have been associated with poor sleep habits.²⁶

Regarding sleeping arrangements, unlike urban China, room sharing of parents and school aged children is uncommon in Israeli families; Thus it is difficult to explain the later sleep onsets in Israeli compared to Chinese children in terms of sleeping arrangements. However, Israeli households are typically smaller than those in the U.S., possibly contributing to increased evening noise and activity in Israeli households on the one hand, as well as overestimation of children's sleep times in the U.S. sample, due to more private and separate bedroom quarters on the other hand.

An alternative explanation may be found in recreational habits such as television viewing and computer activities. The prevalence of such habits and their relationships with sleep patterns and daytime sleepiness have been reported elsewhere ^{6,16,25,26,28} and have yet to be investigated in Israeli children.

An obvious limitation of this study is the reliance on parental reports rather than objective data such as actigraphy and/or polysomnography (PSG). In a comparison of sleep duration and sleep onset latency based on parental reports and PSG recordings, parents significantly overestimated both total sleep time and sleep onset latency compared to the recorded measures.10 In a normative study comparing actigraphy with subjective reports of sleep patterns and disturbances, almost 20% of the sample had some sleep disturbance based on objective criteria, a finding that was not corroborated by neither the parent's nor the child's subjective reports.¹⁹ As poor and insufficient sleep have been related to poor health and performance in school aged children, these studies are disturbing and emphasize the need to increase parents' awareness of sleep need and sleep disorders in their children. For validation purposes, future studies may compare the CSHQ with objective measurements.

Conclusions

In summary, the CSHQ-S is a valid questionnaire with good psychometric properties for the assessment of sleep habits and disturbances in Israeli elementary school aged children. When comparing sleep disorders in Israeli, Chinese and U.S. samples, it is evident that Israeli children, have the most sleep disturbances, go to sleep the latest, and suffer from high levels of daytime sleepiness. Future studies may employ the CSHQ-H for larger normative studies in the Israeli population as well as for the assessment of sleep in clinical populations. Furthermore, it may be used for studies investigating the functional, emotional and health related consequences of poor sleep in children, and as a follow up tool for treatment.

Journal List