Developmental disparities in sedentary time by period of the day among US youth: a cross-sectional study

Sedentary time (ST) in children is associated with adverse health risk factors including obesity, poor cardiometabolic markers [1], and poor physical fitness [1, 2]. Addressing ST in youth may be a more effective way of addressing adult chronic disease, rather than waiting until adulthood to treat the issues. Thus, it is important to understand sedentary behavior in youth across stages of development. ST levels begin to rise at age seven and worsen each year with a dramatic increase between 13 and 14 years [3]. This period corresponds with the transition from childhood to adolescence, when new roles, greater independence, and physical maturation coalesce. It has been estimated that US youth, children and adolescents from 6–18 years old [4], spend an estimated 7 h [5] to 9 h per day [6] in ST. ST is substantially higher in adolescents than in children. A study using National Health and Nutrition Examination Survey [7], estimated children between 6 and 11 years old accumulate approximately 6 h per day of ST; this increases to nearly 8 h per day in 12–15 year-olds, and 8.5 h per day in 16 to18 years olds. Although there have increasingly been calls to limit ST [8], the data suggest major development transitions correspond with increases in ST and these increases are unwavering.

Youth showed lower levels of ST during weekdays compared to weekend days over a two-year period in 970 youth (10 to 13 years of age) in Finland [9]. One potentially vital insight is an understanding of the periods of the day during which ST is accumulating in children and adolescents as well as where disparities are highest between adolescents and children. This information may be helpful for determining when (i.e., before school, during school, afterschool, and evening) and on which type of day (weekday vs weekend day) adolescents are more sedentary. Such information would allow interventionists to target periods of the day with the most potential for improvement. To our knowledge, three studies have evaluated ST patterns in children and adolescents by period of the day; one in Belgium, one in Spain, and one in Norway [10‐12].

In Belgium, investigators examined the frequency of device-based ST bouts in children (N = 740, mean age = 10.9 years) of various durations (i.e., 5–10 min, 10–20 min, and 30 + minutes) during each period of the weekday (i.e., before school, during school, afterschool, and evening) [10]. In other words, when researchers assessed the frequency of ST in 10-min bouts during school hours, they captured the number of times a child remained sedentary for 10 consecutive minutes during school hours. The authors found that—regardless of bout duration—children engaged in more sedentary bouts during evening hours (6:00–10:00 p.m.) than school hours (8:30 a.m. – 4:00 p.m.), and more sedentary bouts during school hours than after-school hours (4:00 – 6:00 p.m.). In this case, researchers operationalized ST in a way that is difficult to translate into minutes of ST. Minutes of ST is the most intuitive framing for parents, teachers, and children seeking to reduce ST. Lastly, although the researchers defined the periods of the day, they did not take into account before school hours, did not compare weekdays vs weekend days, nor did they include adolescents.

In Spain, researchers studied disparities in ST between school hours (e.g. recess, physical education class, and lunchtime) and out of school hours [11]. They recruited 826 children attending 1^st to 4^th grades and 678 adolescents attending 7^th to 10^th grades to measure ST with accelerometers over a three-year period. Their findings revealed adolescents were more sedentary than children during school hours, out of school hours, during recess, during physical education class, and during the weekend. In this case, authors reported out of school time and weekend time as single values, rather than breaking them down into their constituent parts. This makes it hard to determine the extent to which out of school ST was driven by before school time, afterschool time, or evening hours, which limits our ability to inform interventions to maximize ST reductions.

Finally, in Norway, a large nationally representative cross-sectional study [12] tested the changes in ST from 2005 to 2018 in children (9-year-old) and adolescents (15-year-old) by periods of the day (before school, during school, and afterschool). They reported there was a significant increase in ST afterschool in 9-year-old boys from 2005 to 2018, but not in the morning or during school. Also, there was a significant increase in ST during school from 2005 to 2018 among 15-year-old boys, but not in the morning or afterschool time. Findings suggest ST is higher in adolescents than children but some limitations remain. First, authors determined changes in ST across the period of the day within each developmental stage but did not test the difference in ST across developmental stages. Second, ST at age nine may not be representative of ST throughout childhood and ST at age 15 may not represent ST behavior across adolescence. Lastly, authors did not breakdown the weekend day into periods, precluding determination of when ST disparities by developmental stage take place on weekends.

Despite the body of knowledge about ST differences by developmental stage, some gaps in knowledge remain. First, previous studies were conducted in Belgium, Spain, and Norway but none have been conducted in the US. Second, none of the studies used a nationally representative sample including a broad age range; limiting the generalizability of ST comparisons by developmental stage. Third, no study broke down weekend time into component parts. We contend that addressing these gaps will provide an understanding of children’s ST patterns that will inform interventions by elucidating when youth developmental stage disparities emerge and how much potential each period of the day has for addressing these disparities. These are important insights given periods of the day correspond with variations in youth location, activities, and adult supervision.

Hence, the purpose of this study was to examine device-assessed ST engagement by periods of day in a nationally representative sample of US youth. Our first aim was to determine the periods of the weekday (before school, during school, afterschool, and evening) during which adolescents were more sedentary than children. We hypothesized children (ages 6–12 years) would be less sedentary during school and afterschool hours compared to adolescents (ages 13–18 years)—but not before school or during the evening—on an average weekday. Our second aim was to determine the times of day (morning, afternoon, and evening) during which adolescents were more sedentary than children during an average weekend day. It was hypothesized that children would be less sedentary during the afternoon and evening compared to adolescents—but not in the morning—on an average weekend day. Our hypotheses are based on data showing adolescents are more sedentary than children when evaluating the entire day [3] regardless of a weekday or weekend day [12]; and that adolescents are more sedentary than children during particular periods within the school day such as recess and lunchtime [12]. Similarly, previous studies have reported children are less sedentary during afterschool hours compared to during school hours [11]. Although no previous study has provided data about the before school period, we do not anticipate any difference since children and adolescents are expected to have similar routines before school (e.g., having breakfast, getting dressed, brushing their teeth).

This study is a secondary analysis of cross-sectional National Health and Nutrition Examination Survey (NHANES) data aiming to test developmental stage disparities by periods of the weekday and weekend day.

The study sample consisted of 2,972 children and adolescents with a mean age of 12.5 ± 3.6 years and a mean body mass index of 21.5 ± 5.7 kg/m². Overall, youth spent an average of 474.6 ± 149.0 min/day (≈ 7.9 h/day) in ST. Children’s overall ST was 408.8 ± 127.1 min/day (≈ 6.8 h/day) and adolescents’ overall ST was 533.6 ± 142.4 min/day (≈ 8.9 h/day). Table 1 provides descriptive characteristics of the sample partitioned by developmental stage. Body mass index was transformed using the natural logarithm function. All other assumptions were met.

Sensitivity analyses showed that regression models remained significant when shifting the time frames one hour earlier and later for all periods of the day besides before school. The unadjusted model including ST before school with a shift of one hour earlier (5:00 a.m. to 6:59 a.m.) showed developmental stage was not a significant predictor. This was not a concern since no difference in this earlier hour shift may reflect time in which youth were sleeping, which is not considered ST. Thus, it was decided to define the before school period as the time between 6:00 a.m. and 7:59 a.m., as originally planned.

Findings from this study provide insight about the periods of day during which adolescents are more sedentary than children during the week and weekend using a representative sample of US youth between six and 18 years old. Overall, adolescents were more sedentary than children in all periods besides before school. For children (mean ST = 408.8 min/day ≈ 6.8 h/day) and adolescents (mean ST = 533.6 min/day ≈ 8.9 h/day), similar levels of overall ST were observed when compared to previous ST data in US youth which reported that children spend up to 6.0 h/day and adolescents spend up to 8.5 h/day in ST [3].

Within a weekday, the afterschool period showed the largest difference in ST (8.4 min/hour) by developmental stage. However, when the differences at each weekday period were extrapolated, the during school period emerged as the most potentially promising time to reduce ST developmental stage disparities since the daily difference between children and adolescents almost reached one hour (56.7 min/day or 35% difference). On the other hand, the weekend day data by developmental stage revealed that the smallest difference in ST occurred during the weekend morning period (7.1 min/hour) and the largest difference occurred during the weekend afternoon period (9.2 min/hour). Similarly, when the differences at each weekend day period were extrapolated, the afternoon period emerged as the most potentially promising time to reduce these disparities since the daily difference between childhood and adolescence was almost an hour (57.6 min/day or 14% difference). Previous research has shown that if youth substitute 60 min of ST with 60 min of light physical activity [24], moderate to vigorous physical activity [25], or 60 min of sleep [24], it can result in improved body composition [25] and mood [24]. Hence, addressing during school ST disparities (or weekend afternoon ST disparities) alone would substantially improve adolescent health.

The accelerometer recorded ST disparities by developmental stage were much larger on weekdays afterschool and during weekend afternoon periods. This may reflect a shift in structured programming between childhood and adolescence. Adolescents are much less likely to be enrolled in structured extracurricular physical activity programs. In adolescence, academic demands also increase and homework responsibilities may displace the physical activity that adolescents tend to participate in during the afternoon hours when they are younger.

Evidence supports the potential role of increasing schoolwork demands on ST as children age [26]. A Canadian qualitative study conducted a survey with open-ended questions among 102 fulltime undergraduate students and found that attending classes as well as studying for classes at home was their main barrier to reduce ST [26]. Although the current study included few adolescents that could attend college (mainly those between 17 and 18 years of age), the transition from primary school to secondary school also results in greater schoolwork demands.

Our weekday findings provided evidence to support our first hypothesis that children would be less sedentary than adolescents during school and afterschool periods. However, it was surprising that developmental stage was a significant predictor of ST during the evening. In terms of the weekday evening difference in ST, the first hours (6:00 p.m. – 7:30 p.m.) of the weekday evening period can be an extension of the afterschool period, in which some youth are still participating in afterschool activities, though perhaps, accessibility to screens and homework demands also play a role here. Similarly, weekend data supported our second hypothesis that children would be less sedentary during the afternoon and evening compared to adolescents. However, the differences during the weekend morning period were unexpected and can be explained, in part, by adolescents’ accessibility to electronic devices that promote ST compared to children’s accessibility to such devices. For instance, some parents ask children to make their bed, take a shower, and have breakfast before starting the day, and if screen time is part of the routine, it is probable that they will engage in screen time under the adult’s supervision.

Our findings confirm previous literature stating that adolescents are more sedentary than children during the weekday periods and weekend. In Spain, a cross-sectional study revealed male adolescents spend 255.6 ± 32.3 min/day sedentary compared to male children who spent 192.7 ± 27.4 min/day [11]. A similar trend was reported by the authors for female adolescents (275.3 ± 35.6 min/day) and children (201.4 ± 27.6 min/day). The same authors reported children were less sedentary than adolescents during the weekend; here again, our US sample findings are similar to the Spanish sample. However, our findings provide further information by breaking down the prolonged periods into shorter periods that coincide with shifts in setting, activity, and supervision. In the long run (weekly and monthly differences), targeting ST during school and weekend afternoons, present the most profound opportunity to close the ST gap by developmental stage. Although previous interventions have included ST components, achieved health benefits, and been proven to reduce ST (mainly screen time) [27, 28], none have targeted specific periods of the day. The current study suggests future interventions on ST in adolescence should focus on the periods of the day with greater opportunity to help reduce ST disparities by developmental stage such as during school, afterschool, and weekend afternoon periods.

It is believed that recreational screen time is a likely contributor to disparities in ST between adolescents and children. With age, parents give children more autonomy and discretion to decide how they use their leisure time. This can promote the use of smartphones, desktop, and laptop computers, which are very accessible to adolescents. Since 2014 – 2015, there has been a 22% increase (from 77 to 95%) in the number of US adolescents who have access to a smartphone and a 3% increase (from 87 to 90%) in those who have access to desktop and laptop computers [29]. Furthermore, accessibility to these devices can be problematic because adolescents are likely to spend more time in passive activities (e.g., shows, movies, gaming, social media), than cognitively demanding enriching activities (e.g., reading books, playing chess, doing math problems). Most passive activities are designed to grab our attention and not let it go, using features such as never-ending scroll, streaks, and notifications which can result in developing addictive behaviors toward screen media [30]. On average, 95% of US adolescents reported using YouTube, followed by TikTok (67% of US adolescents) and Instagram (62% of adolescents). More than half (54%) of US adolescents report it would be hard for them to give up social media [29].

Although this study is the first to use a US-based nationally representative sample to test differences in ST by developmental stage and periods of the day, it is not exempt of limitations. Thigh-worn accelerometers are more accurate than waist-worn accelerometers for measuring ST (due to their ability to assess participant posture). However, it has been reported that thigh-worn accelerometers reduce wear-time compliance [31, 32] due to skin irritation [31, 32] or having a sweaty thigh [31]. Thus, the hip-worn accelerometer provides a more comfortable placement, which increases adherence, while still providing a valid measure of ST when using validated cut-points. In our study, we used Evenson et al. cut-points [33], which have been validated for our age range. Second, our data were collected between 2003 and 2006, which raises concerns related to the relevance of findings in 2022, since societal ST patterns can change over time. Indeed, observational studies [12, 34] have shown increases in the quantity of ST with the easy accessibility to internet access. Dalene et al. [12] measured device-assessed ST in 5,168 Norwegian children and adolescents in years 2005, 2011, and 2018. The authors reported increases of weekly ST between 2005 and 2018 of 6% in 9-year-old boys, 1% in 9-year-old girls, and 4% in 15-year-old boys and girls [12]. Similarly, Schroeder et al. [34] collected ST data in US youth in 2017. They reported that children (8 – 12 years old) spent an average of 8.3 ± 2.1 h/day in ST. This suggests that in the decade between when NHANES distributed hip-worn accelerometers (2003–2006) and 2017, there may have been a ~ 5% increase in ST among US youth. Lastly, the present study does not fill the gap about why adolescents are more sedentary during each period of the day. Therefore, future studies should be conducted to gather information about the barriers and facilitators of ST in adolescence specific to each period of the weekday and weekend day. The design of the study also does not allow us to determine causality; therefore, we cannot say that too much ST is caused by being an adolescent, though the inverse (ST causes adolescence) is impossible.

The current study’s analyses accounted for the NHANES complex sampling design to account for stratification, clustering, and weighting as well as sex, race/ethnicity, annual family income, and body mass index. Although previous cross-sectional and longitudinal studies have shown adolescents are more sedentary than children, to our knowledge, this is the first study to examine differences by periods of the day in a US sample. Findings confirm adolescents are more sedentary than children during most of the weekday periods (during school, afterschool, and the evening) and all weekend day periods (weekend mornings, afternoons, and evenings), though there is substantial variability in the magnitude of the disparities by period of the day (weekly disparities ranging from 9 to 35%). Evidence suggests that during school and afterschool periods contributed most to sedentary disparities overall, though substantial disparities also exist in periods of the day that correspond with time at home. This provides insight for future interventions, on when and where interventions should seek to reduce ST levels in adolescents.