Excessive sedentary behavior in childhood, such as habitual television (TV) viewing or video game play, is associated with weight gain/obesity [1], cardiometabolic disease risk [2], and poor mental health [3], independent of physical activity levels [4]. Despite these risks, the proportion of children who spend excessive time in sedentary behavior is increasing worldwide [4]. Reducing sedentary behavior among children is critical in order to stem the current increase in lifestyle-associated diseases (e.g., type-2 diabetes and hypertension). Current activity guidelines [5, 6] recommend no more than 2 h per day of recreational screen time (i.e., watching TV, DVDs, or videos, playing TV games, and computer use), and limiting sedentary transportation and reducing prolonged periods of sitting or time indoors [4]. However, the development of more effective intervention strategies requires a greater understanding of behavioral context, such as the places and social situations associated with inactivity, as well as the individual times spent on specific sedentary behaviors in target populations.

Numerous studies have examined sedentary behavior in school-age children, and many have focused on relationships to diet, obesity, diabetes, and cardiovascular disease [7–10] or sex differences [11, 12]. Others have examined domain-specific sedentary times, particularly time spent in TV viewing [13–17] or total screen time [18–23]. Such studies have revealed regional differences in domain-specific sedentary times. For instance, the proportion of children meeting the <2 h/day recreational screen time guideline differed between study populations in Norway [24] and Canada [25]. Other types of sedentary behaviors also vary by country [26], as these behaviors may be heavily dependent on culture, the socioeconomic status, and lifestyle.

While documenting behavior characteristics in target populations is critical for effective targeted interventions [27], few studies have comprehensively assessed differences in domain-specific sedentary behaviors and compliance with activity or screen-time guidelines between sexes and among different age groups in childhood. Descriptive epidemiological studies have suggested that differences in sedentary behavior between boys and girls are domain-specific [28–30], and that sedentary behavior increases with age [7, 31]. Therefore, the aim of this present study was to examine domain-specific sedentary behaviors and compliance with screen-time guidelines in Japanese school children from 6 to 11 years of age in order to develop targeted interventions according to age (or grade level) and sex.

The present study describes the weekly time spent in specific sedentary behaviors by Japanese elementary school children. Boys spent more time playing TV games, while girls spent more time reading or listening to music, doing homework or assignments, and car travel. Higher-grade students spent more time reading or listening to music, on computer or Internet use, and doing homework or assignments even after adjusted for the geographic areas or BMI. Moreover, approximately 60% of children spent time ≥ 2 h per day in sedentary behavior, the majority being devoted to screen time (TV watching and gaming). These findings indicate that programs for decreasing sedentary behavior in Japanese school children should be designed specifically for different age groups of boys and girls.

Sedentary behavior was high on both weekdays and weekends in the presented domains, particularly watching TV or videos and TV game use. Moreover, even compared with a previous study that reported a TV viewing time of 1.42 h (426 min/wk) on weekdays and 1.57 h (188.4 min/wk) on weekends among aged 5 years old [36], sedentary behavior was high. According to a national self-reported survey of 3210 children in 10 regions [37], Japanese children spend a substantial amount of time watching TV, playing video games, and using computers when home alone. Therefore, instructing children how to spend time when alone at home might be one effective strategy for decreasing sedentary behavior. However, the total average daily time spent on sedentary activity was even higher on weekends than weekdays. For instance, average TV viewing time was approximately more than 99 min daily on weekdays and 146 min daily on weekends. Alternatively, relatively little time was spent in computer use in this age group. Previous studies [38, 39] also found that leisure sedentary behavior time was higher on weekend days than weekdays. Therefore, family-focused weekend activities are a promising way to reduce sedentary behavior in children.

On weekdays, boys were more likely to spend time playing TV games and less likely to spend time reading, listening to music, and doing homework or assignments than girls. On weekends, boys were also more likely to spend time playing TV games and less likely to spend time in car travel than girls. These sex differences by weekday or weekend are consistent with findings from a previous cross-sectional study on a random sample of children aged 12.6-16.7 years from Scottish schools [39]. Several studies have reported that boys spend more time engaged in TV games than girls [12, 21, 23] and that girls spend more time engaged in reading and homework than boys [6, 10, 12]. Other studies have also reported greater TV video game play by boys than girls [18, 19] and more time spent reading by girls [9], although the differences were not statistically significant. As children’s leisure time activities clearly differ by sex [11], domain-focused approaches for reducing sedentary behavior must also be tailored specifically for boys and girls. Car travel on weekends frequently occurs with the family. According to a national self-reported survey of 3000 randomly sampled children [40], girls tend to spend more time on weekends with their families than do boys, which may explain the difference in car travel.

Higher grade students were more likely to spend time reading or listening to music, using a computer, and doing homework or assignments than were lower grade students. Other studies conducted on children aged 4-11.99 years have shown similar results in that higher-grade students spent more time on TV viewing (2.4 h/ day and 2.3 h/day in 9- to 11-year-old and 4- to 5-year-old boys, respectively, and 2.4 h/day and 2.2 h/day in 9- to 11-year-old and 4- to 5-year-old girls, respectively) and computer/video games (1.2 h/ day and 0.7 h/day in 4-5 years in 9- to 11-year-old and 4- to 5-year-old boys, respectively, and 0.8 h/day, 0.6 h/day in 9- to 11-year-old and 4- to 5-year-old girls, respectively) [18]. Increased time spent on computer use with advancing school grade or age is consistent across studies [7, 9, 17]. In contrast to numerous studies reporting that time spent on TV viewing increased with school grade or age [8, 13, 15, 16, 18], the present study found no significant differences. However, the average time spent watching TV in the present population (1.6 to 1.9 h per day on weekdays and 2.4 to 3.1 h per day on weekends) was low compared to previous estimates that ranged from 2.0 to 3.7 h per day in 35 countries and regions [14]. Nonetheless, TV viewing time still accounted by far the largest proportion of total sedentary time in Japanese children. No significant sex and grade differences were observed in the present study, which suggests that changing TV viewing habits should be a focus for reducing sedentary behavior in all Japanese children. Furthermore, sedentary behaviors in childhood are maintained in adolescence and into adulthood [22]; therefore, targeted strategies to reduce excessive screen time in childhood are crucial for encouraging lasting behavioral changes.

A national self-reported survey of 53,458 randomly selected Japanese children found that the proportion of time spent in learning activities in sedentary behavior outside of school hours without family increased with school grade [41]. Another self-reported survey of 3210 children from 10 areas of Japan found that lower-grade students spent more time with families on weekends than did higher-grade students [37], suggesting that lower-grade students received more social support from family than higher-grade students. Therefore, the present study showed that lower-grade students engaged in less sedentary behavior than higher-grade students and that family support is vital for reducing these sedentary behaviors, because they frequently occur at the family’s discretion, particularly in younger children.

In the present study, only 35.5% of boys and 40.7% of girls met the recommendation that children spend less than 2 h on screen time daily. This trend is comparable to that found in a study from China that asked about time spent watching videos, CDs, and DVDs; playing video games; and using a computer (46.7% and 42.3% of boys and girls, respectively) [42], study from the United States that asked about time spent watching TV and playing video games (26.7% and 35.0% of boys and girls, respectively) [19], and study from Canada that asked about the time spent watching TV and using a computer or playing video games (38.1% and 56.0% of boys and girls, respectively) [43]. Even after adjusted for z-scores of BMI, number of days with moderate to vigorous physical activity (MVPA) for at least 60 min, and geographic area, compared to lower-grade students, higher-grade students in our survey were 2.09 times during the whole week, 2.08 times on the weekday, and 1.88 times on the weekend more likely to spend ≥2 h/day on screen time. A study of Canadian children were asked to report the number of hours per day during the week that they spend watching TV or movies or playing video or computer games; in grades 5 − 8, the authors found that boys were 2.2 times more likely than girls to spend ≥3 h/day on screen time [25]. While the measuring methods of screen time differed among studies, making direct comparison difficult, the proportion of children who attained the 2-h limit was low in the present study group, especially for the higher-grade students. Effective strategies to decrease screen time are needed, especially in higher-grade students, including family-based interventions and removing TVs from children’s bedrooms [44, 45]. Further studies are required to identify factors that predict meeting or exceeding these guidelines.

The present study has some limitations. First, the study relied on self-reported measures, which have the potential for error due to different interpretations of the questions. However, objective measures such as accelerometers are free of behavioral context [46]. In addition, the present study did not gather data regarding all the domains of the children’s sedentary behavior, such as school time sitting, to give a more comprehensive overview of total sedentary time. Furthermore, the reliability and validity of the scales used in this study are unknown as it is not reported in previous studies [34]. However, it was used in a national survey of Japanese elementary school children [34]. In a previous study [34], parents/guardians of children in the first through third grades (6 − 8 years) were asked to complete the questionnaire after consulting with their children. Children in the fourth grade and above (9 − 11 years) were asked to complete the questionnaire without parental/guardian assistance to ensure reliability and validity. Although the present study used a previous representative survey of Japanese school children [34], the self-reported questionnaire was likely to be underappreciated. Second, data from lower-grade school children were collected with the assistance of a parent or guardian, whereas higher-grade children completed the questionnaire independently, which present two distinct sources of bias. Third, the questionnaire asked about each domain-specific sedentary behavior separately. However, when responding about domain-specific sedentary behaviors, the participants may not have reported on only one domain, e.g., they could have reported the time spent using a reading or listening to music while watching TV or video. Thus, it is necessary to consider that the self-reported sedentary time is overestimated. Fourth, there were differences in the grade level, BMI, and sedentary time between participants from Okayama in 2010 and Tokyo in 2014. Moreover, the difference in the period of measurements in 2010 and 2014 may have biased the data. Moreover, it is likely that Japanese children who are largely sedentary also have parents/guardians who are largely sedentary. However, the present study did not include items on parent’s/guardian’s sedentary behavior to adjust for these effects. Despite these limitations, the present study consider that the present study is meaningful because it is the first study to highlight sex and grade differences in domain-specific sedentary behaviors and attainment of screen-time guidelines in Japanese children. The present results provide an important foundation for the development of interventions to decrease sedentary behavior in Japanese children.

Domain-specific sedentary behaviors differed according to sex and school grade. Higher-grade students were more likely to spend ≥2 h per day on screen time than lower-grade students. These findings highlight the need for strategies targeting specific domains in order to more effectively decrease sedentary behaviors in Japanese school-age children.