Sleep is a crucial component of childhood and adolescent development and health. Similar to healthy diet and exercise, sleep is critical for children to stay healthy, grow, learn, perform well in school, and function at their best (1). Late sleep timing, chronic sleep issues, and wakefulness disorders are associated with poorer health outcomes in children and adolescents, including physical and mental health, cognitive function or academic performance, eating behaviors, and risk of other mental health issues (2–4). Teenagers aged 14–17 years require 8–10 h of sleep per day (5). However, children and adolescents get fewer than the recommended sleep hours (6).

Many internal and external factors have been identified as detrimental to children's sleep. A clinical review reported that tobacco, digital media, evening lights, a negative family environment, and caffeine are detrimental to sleep (7). Sleep disorders are usually assessed using objective measures such as polysomnography (PSG), actigraphy, and sleep logs. Although PSG and actigraphy are useful tools for physiologically assessing sleep quality and quantity, they are highly specialized and expensive, whereas sleep logs and questionnaires are inexpensive and simple subjective measures for detecting sleep duration and sleep problems (8). Spruyt et al. conducted a comprehensive review and suggested that many of the tools used for children had not undergone careful and methodical psychometric evaluations (9). The Child and Adolescent Sleep Checklist for Parents (CASC-P) was designed to identify sleep habits and screen for sleep problems in preschoolers, elementary school children, and junior high school students in Japan. However, few Japanese sleep questionnaires have been internationally standardized, and the CASC-P has not yet been validated. This study aimed to validate the Japanese version of the CASC-P for junior high school students and determine its internal consistency.

This cross-sectional survey was conducted in Kumakogen Town (population: 6,700), a rural region of Ehime Prefecture (population: 1,290,000) in southwest Japan. The participants were parents with children enrolled in three different public junior high schools in Kumakogen Town. All parents were recruited based on whether their children were attending school. In September 2012, questionnaires were distributed to the parents by their children. The parents completed the questionnaires at home.

A total of 218 parents (108 male, 110 female) of students in grades 7–9, aged 12–15 years, were included after excluding two incomplete questionnaires. In the present sample (N = 218), Cronbach's α for the overall scale was 0.771, and the subscale coefficients were 0.592 (bedtime problems), 0.561 (sleep breathing and unstable sleep), 0.550 (parasomnia and sleep movement), and 0.626 (daytime problems). The results of the ratio of students by gender and the total CASC-P score by grade are presented in Table 1. There were no significant differences according to grade. The average CASC-P scores are presented in Table 2. According to the CASC-P cut-off score, the prevalence of sleep problems was 15.6%. Factorial construct validity was assessed using the four-factor model used in the original CASC-P. The sample size of 218 satisfied the estimated size requirement (15). The value of KMO is 0.71, and the Bartlett's test of sphericity (Chi-square = 1,181.8; P < 0.001). Multiple fit indices were used for the CFA. The selected indices were the chi-square statistic (CMIN/DF), standardized root mean square residual (SRMR), adjusted goodness of fit index (AGFI), comparative fit index (CFI), and root mean square error of the approximation (RMSEA). As shown in Table 3, AGFI and CFI did not indicate a good fit for the data (16). For the CASC-P, nearly all items loaded meaningfully onto their designated factors, with standardized factor loadings ranging from 0.278 to 0.878. Exceptions included items 1 (0.159), 2 (0.186), 9 (0.072), 16 (0.199), and 21 (0.170). Detailed results are shown in Figure 1 and the supplementary materials in Supplementary Table 1A.

This study is the first to verify the factor structure of the CASC-P using CFA in a general population of parents of junior high school students in Japan. The CASC-P showed satisfactory internal consistency (Cronbach's alpha: 0.771), similar to a previous study on the self-reported CASC in junior high school students (Cronbach's alpha = 0.786) (17). However, this study found that some items did not have a good fit.

In general, a modified model needs to remove the non-good-fit items and attempt the next model using CFA. However, these non-good-fit items were difficult for parents to grasp. Items 1 and 2 are activities for children, such as drinking caffeine and playing games or using the Internet before bedtime, and parents find it difficult to answer these for their children. According to a Chinese survey, over 80% of early adolescents (average age = 10.83) sleep in separate bedrooms from their parents (18). Furthermore, most children use digital devices such as smartphones before bedtime. Although children generally have better age-appropriate sleep in the presence of household rules and regular sleep-wake routines, sleep deficiency is more likely to occur when parents and children have electronic devices in their bedrooms (1). According to a questionnaire survey involving primary schools in Japan, the use of digital devices was linked to children's emotional and psychological development (19). Item 2, “I play video games, surf the Internet, or send texts ≤1 h before going to bed,” had the highest score in the current study, indicating that many parents took this issue seriously. However, the relationship between sleep problems and digital device use is complicated; it may be difficult for parents to report their children's use of devices with a single question.

Item 9, “I stop breathing while I sleep,” received a very low score in this study. However, it is an important issue that some children stop breathing while sleeping. The exact prevalence of sleep-disordered breathing (SDB) in children is unknown but may be as high as 11% (20). SDB encompasses a spectrum of sleep disorders, including primary snoring at the mild end and obstructive sleep apnea at the severe end. Screening with a brief questionnaire is necessary to determine the risk of SDB and carefully inspect children at risk for SDB (21). Although the factor loading for this item was low, this may be due to underrecognition by parents rather than a lack of clinical relevance. Parental awareness of SDB symptoms is often limited, especially when children sleep in separate rooms. Furthermore, SDB is known to be associated with serious consequences such as impaired neurocognitive development, behavioral problems, and enuresis (22). Therefore, despite the statistical weakness of item 9, we consider it clinically important to retain this item for screening purposes.

Item 16, “I urinate in my sleep,” was also marked “very rarely” in the current study; this item did not obtain a good fit by CFA. The CASC-P was designed to assess the weekly frequency of enuresis comorbidity with SDB. Hence, the nocturnal enuresis score was low. For a diagnosis of nocturnal enuresis, children older than 6 years of age should have one or more wetting episodes per month (23). Nocturnal enuresis is commonly observed in children with SDB, and adenotonsillar hypertrophy is a significant contributor to SDB; consequently, adenotonsillectomy is the mainstay treatment for SDB (24, 25). To data, several pediatric sleep questionnaires have been developed internationally, including the Children's Sleep Habits Questionnaire (CSHQ), and the Pediatric Sleep Questionnaire (PSQ), each with different focuses and lengths. The CASC-P offers a concise structure while covering a full range of sleep-related behaviors throughout the day and night, making it well-suited for large-scale screening.

The present study has several limitations. First, the original CASC has not itself undergone comprehensive validation. In this study, the validation approach employed was limited, relying exclusively on validity using CFA, and reliability was assessed using Cronbach's alpha. In addition to structural validity, key aspects such as content validity, test-retest reliability, and construct validity should be addressed. Second, the assessment relied on parental reports. Only parental data are limited for such reasons as self-report bias and over-estimation of the associations between variables. Third, we did not control for confounding factors such as family background, economic status, and academic performance. Family background and dynamics, including bed-sharing, are important for understanding children's sleep duration. Fourth, there is a potential for selection bias. The study participants were recruited from public junior high schools in suburban areas, which may have skewed the results. Fifth, the data for this study were collected in 2012. Since then, major societal changes, particularly the COVID-19 pandemic, have significantly affected adolescent routines, screen time, emotional well-being, and sleep habits (26). These changes may influence the current applicability of our findings, and future research should evaluate the CASC-P using updated post-pandemic data. Finally, while the overall model fit may be considered acceptable in Table 3, the results suggest there is potential for refinement in CFA.

Our data indicate that the CASC-P is a suitable questionnaire for assessing parents' perspectives on adolescent sleep behavior. As sleep problems among early adolescents are common, to examine the reliability and validity of such measures are necessary. Future studies should include clinical populations, such as adolescents diagnosed with sleep disorders or receiving therapeutic interventions. Comparing CASC-P results across general and clinical samples would provide valuable information on the checklist's sensitivity and its ability to distinguish clinical cases from normative patterns.