O the places rural children will go…to get physical activity: a cross sectional analysis

Abstract

Purpose

Youth living in rural areas have higher risk for overweight/obesity. It is important to understand where these children engage in moderate-to-vigorous physical activity (MVPA) and sedentary time to encourage or intervene on activity in specific locations. This study compared MVPA and sedentary time across locations among children with overweight/obesity in the rural Midwest of the U.S.

Methods

Participants wore an accelerometer and Global Positioning System tracker over 7-days to collect data on MVPA, sedentary time, and location. Locations were categorized as Home, Home Neighborhood, School, School Neighborhood, and ‘Other’. Differences based on school and non-school days were examined.

Results

Participants (n = 44; 8.8 ± 0.8 years; 61.4% females) engaged in an average of 41.4 min of MVPA/day and 6.7 h of sedentary time/day. In total, most MVPA was obtained at School (18.2 min/day, 44.2% of total MVPA), followed by Other (22.7%) and Home (20.5%). Participants accrued most of their sedentary time at School (141.7 min/day, 35.3%) followed by Home (31.2%) and Other locations (20.3%). Relative to time spent in location, participants were least active in their School Neighborhood (3.7% of time was in MVPA) and most active in Other locations (7.0%). When comparing non-school and school days, participants obtained 95.7 more minutes sedentary time at Home and were in Other locations for almost 2.5 more hours more on non-school days than school days. During school days, participants obtained 25.0 min/day of MVPA at school.

Conclusion

The School location is supportive of MVPA and high amounts of sedentary time. In addition to supporting children to travel to locations where they are likely to be active, efforts are needed to increase activity in locations where children spend substantial time. Providing more opportunities for activity in/around the home and reducing sedentary time during school may be promising targets for improving health among rural children.

Introduction

Engaging in moderate to vigorous physical activity (MVPA) and reducing sedentary time have a wide variety of health benefits for children, especially for children with overweight/obesity to promote a healthy weight [1, 2, 3]. Despite the known benefits, less than 25% of children ages 6–11 years in the United States meet the physical activity guidelines of engaging in 60 min of MVPA each day [4]. Research observes that elementary school age children and adolescents in rural areas are 46% less likely than their urban peers to meet physical activity guidelines and tend to engage in more sedentary time [5, 6, 7]. Lower activity levels may be a contributing factor to the higher rates of obesity observed in rural areas [8]. Understanding factors related to physical activity behaviors in children is essential, especially in rural areas with greater disparities.

Research demonstrates an individual’s environment can influence their physical activity and sedentary time [9]. Schools (e.g., physical education classes), playgrounds, neighborhoods, and green spaces may be settings that promote MVPA in children [6, 10, 11, 12, 13], whereas schools (e.g., classrooms) or homes may be settings that constrain or limit support of activity and promote sedentary time [11]. Most research in this area is performed in urban settings, and the few studies that examine rural populations note differences in the locations where children accumulate MVPA between rural and urban settings [6, 12, 14]. These studies suggest elementary school aged children and adolescents in rural areas accumulate more of their MVPA in the school environment and outside of the home compared to their urban counterparts [6, 12]. Differences in child activity levels in urban and rural settings may be due to environmental determinants as urban areas have a higher concentration of supports for activity (e.g., sidewalks/walking paths, access to parks, short distances between commercial areas, neighbors close by to play with). In contrast, environmental supports for activity in rural settings are more geographically dispersed and thus may be less readily accessed and may contribute to an obesogenic environment [7, 12, 14, 15]. Currently, there are no studies that examine locational patterns of physical activity in rural children with overweight and obesity or the location of sedentary time of children living in rural areas using objective measures (e.g., GPS).

The objective of this cross-sectional study was to compare MVPA and sedentary time across Home, Home Neighborhood, School, School Neighborhood, and “Other” locations among elementary school aged children (6–10 years old) with overweight or obesity in rural areas. We also examine whether children’s activity in these locations differed between sexes and on school days compared to non-school days. This study fills important gaps in the literature by focusing on rural children, incorporating sedentary time in conjunction with MVPA, specifically examining children with overweight/obesity who may be at risk for being less active, and examining the types of trips (i.e., cycling, pedestrian, or vehicle) these children take.

Methods

Participants and procedures

Participants for the present analysis were part of the iAmHealthy Schools study evaluating the impact of a pediatric obesity intervention in rural families that include a child with overweight/obesity [16]. Researchers recruited children with overweight/obesity in 2nd − 4th grade (6–10 years old) from rural areas of Kansas, USA between 2017 and 2019. Rural areas were defined using the U.S. Census Bureau’s American Community Survey for ‘1-year Supplemental estimates’ as geographic areas with a population < 20,000 individuals [17]. Measures for the present analysis were collected at baseline of the intervention, which occurred only during the school year, between the months of September and March. Full description of the parent study and recruitment procedures were previously published [16]. The recruited sample was representative of the schools they were recruited from based on sex, race, and eligibility for free and reduced-price lunch [18]. The local Institutional Review Board approved all study protocols. All enrolled children provided informed assent and their parents provided informed consent to participate in the study.

Once enrolled in the study, child participants were asked to wear an ActiGraph GT3x-BT accelerometer (ActiGraph LLC, Pensacola, FL) and Qstarz BT-Q1000XT Global Positioning System (GPS; Qstarz International Co., Ltd., Taipei, Taiwan) tracker for 7 days during waking hours. The research team also collected demographic information on the child and family via surveys.

A total of 148 participants across 18 rural elementary schools enrolled in the iAmHealthy study. The present analysis includes participants from 8 schools who were asked to wear the GPS tracker as part of this ancillary study. In total, GPS trackers were distributed to 74 participants. Participants were excluded if the participant declined to wear the GPS monitor (n = 6) or the accelerometer (n = 7), the liaison or participant accidentally set the monitor to the incorrect recording mode or the monitor was not turned on (i.e., no GPS file available; n = 2), the participant had a home address that was not able to be geocoded (n = 4), or the participant did not wear both devices together for ≥ 10 h on ≥ 4 days, including ≥ 1 valid school day and ≥ 1 valid weekend day (n = 11). Therefore, the current sample included 44 children from 8 schools in 6 counties.

Measures

Demographics

Parent reported demographic information included child age, sex, and race/ethnicity. The parent participating in the larger study also reported their highest level of education, marital status, and approximate annual household income.

School information

School personnel provided information on start and end time during school days and a list of vacation days during the school year to help the research team identify and define school days vs. non-school days and in school vs. out of school time.

Physical activity

Physical activity was measured via ActiGraph accelerometers worn on the hip. Data were collected in 40 Hz and processed using Evenson cut points applied to 30-second epochs. The Evenson cut points provide a valid assessment of physical activity across a range of intensity levels in youth [19, 20]. Additionally, high-light intensity activity was examined because it may also provide cardiometabolic health benefits [21]. High-light intensity was defined as the highest half of the Evenson range for light intensity (1147–2295 counts per minute [cpm]).

GPS tracking and geographic information systems (GIS) location assessment

GPS location, obtained every 30 s, was used to calculate how much time participants spent in various locations. ArcGIS Pro software was used to geocode the locations of the participant’s homes and schools. Network buffers were drawn around their home and school to a distance of 1 km along roads to create ‘Home Neighborhood’ and ‘School Neighborhood’ locations for each participant. This buffer size was selected to broadly reflect walkable distances and align with prior neighborhood environment research [22, 23]. Time spent in an overlapping area of both School and Home Neighborhood was divided in half and split between the two locations (e.g., 10 min in ‘both’ was categorized as 5 min in Home Neighborhood and 5 min in School Neighborhood). County parcel data was used to draw 30-m buffers around each participant’s home parcel and school parcel to create ‘Home’ and ‘School’ locations. GPS points not within one of the aforementioned 4 locations were classified as occurring in ‘Other’ locations. Parcel land use information was used to distill the Other locations into the following categories: Agriculture (agricultural use), Residential (residential, farm homestead), Commercial (commercial, industrial), Community (nonprofit, public exempt, utility), and Unknown (participant was outside of the county). The Other: Community category included locations such as churches, recreation centers, sports complexes, community swimming pools, lakes, parks and public open spaces. MVPA points accumulated in Other: Community were then plotted on maps to visualize and describe where these points occurred. The parcel land use files were available for all but one county, which resulted in omission of 4 participants from analyses on the categories with the Other locations.

GPS and accelerometer data were processed and merged using the Human Activity Behavior Identification Tool and data Unification System (HABITUS). HABITUS is an updated version of the Personal Activity and Location Measurement System (PALMS) [24, 25]. This processing also allowed for identification of trips; trip classification parameters were vehicle if trip speed was > 35 km per hour (kph), bicycle if trip speed was between 10 and 34 kph, or pedestrian if trip speed was between 1 and 9 kph. To be counted as a trip, at least 100 m needed to be covered in 120 s. The maximum pause allowed during a trip was 180 s. Time spent in trips was removed from overall and location-specific activity estimations.

Activity variable derivation

Daily minutes of total time, MVPA, high light intensity activity, sedentary time, and time in each trip mode were calculated for each location and overall (across all locations). Proportion of time in the location that was spent in MVPA and sedentary time were also computed. Each variable was aggregated to the participant level by computing the weighted average daily value, defined as ([mean value across valid weekdays × 5] + [mean value across valid weekend days × 2]) ÷ 7. The exception was that school and non-school days were examined separately in some analyses.

Data analysis

Participant characteristics were examined through means and standard deviations or frequencies. Chi-square tests were used to compare excluded and included participants in the analyses based on sex, Hispanic/Latino ethnicity, eligibility for free or reduced-price lunch and meeting MVPA recommendations at baseline. Mixed effect models, accounting for nesting of participants within counties, were run to compare MVPA and sedentary time, as well as the proportion of time in the location that was spent in MVPA and sedentary time, across locations. Similar models were used to compare these 4 dependent variables between sexes and between school days/non-school days. Covariates in the models included child age (in months), sex (female/male), school, race/ethnicity (white/non-Hispanic), income (categorized into <$30,000, $30,000- $80,000, and >$80,000), total accelerometer wear days, proportion of school days accelerometer was worn, and average minutes of wear time per day. A power analysis was not conducted a priori for this current sub-study.

Results

Participant and family characteristics are presented in Table 1. Participants included in the analyses were similar to participants who were excluded from analyses based on sex (X² (1) = 0.005 p = 0.945), Hispanic/Latino ethnicity (X² (1) = 0.881 p = 0.348) free or reduced-price lunch eligibility (X² (1) = 2.504 p = 0.114) and meeting MVPA recommendations at baseline (X² (1) = 0.601 p = 0.438). On average, participants were 8.8 ± 0.8 years and wore the accelerometer and GPS devices for an average of 12.6 ± 0.9 h/day over 6.9 ± 0.7 days. Half of the participants (n = 22, 50%) had overlapping time in both the home and school neighborhood buffers (i.e., the child lived within ~ 2km of the school); and participants lived an average of 4.2 ± 7.3 km from their school (using a straight-line distance). Of note, only an average of 0.05 ± 0.34 min/day was spent in Parks, with a maximum of 3 min/day in one individual.

Table 1 Participant demographics (N = 44)

Based on device wear time, participants spent the most time at Home, followed by School and Other locations (Table 2). On average, participants engaged in 41.4 min of MVPA per day and 9 (20%) met physical activity recommendations. Participants obtained most of their MVPA at School, totaling 18.2 min/day 44.2% of their overall MVPA when considering both during and outside of school hours, followed by Other (9.4 min/day; 22.7% of overall MVPA) and Home (8.5 min/day; 20.5% of overall MVPA). Limited MVPA was accrued in the Home and School Neighborhoods (1.6 and 1.3 min/day, respectively) when considering non-trip time. When adding MVPA and active trips, children spent 4.2 and 5.9 min/day active in the Home and School Neighborhoods, respectively. Considering when a participant was in a location, the proportion of time that was spent in MVPA within that location was highest for the Other location (7.0%), followed by School, both during and outside of the school day (6.2–6.3%). The proportion of location time that was spent in MVPA was lowest for the School Neighborhood location (3.7%). Participants spent 6.7 h of their day in sedentary time, with most of this time being accrued at Home followed by School during school hours. The proportion of location time spent sedentary was highest in the Home Neighborhood, followed by School during school hours. Results for differences in high light intensity activity across locations were similar to the results for MVPA and are presented in Table S1.

Table 2 Differences in children’s daily time and activity across locations (N = 44 children)

When MVPA time in Other: Community locations was visually examined, the specific settings included parks, open green spaces, recreation centers (e.g., YMCA), sports complexes, churches, school campuses (different from the school the child attended), and sidewalks and parking lots (activity that was not considered to be part of a trip). Examples of MVPA settings within the Other: Community category are shown in Fig. 1.

Fig. 1

Examples of specific settings where physical activity occurred within the Other: Community category. footnote: Each blue point on the map indicates 30 s of moderate to vigorous physical for a study participant

Regarding sex differences, males obtained more MVPA than females in all locations for both minutes/day and proportion of time; these differences were statistically significant for School during school hours (minutes/day and proportion of time) and for Other locations (minutes/day; Table 3). The largest discrepancy in MVPA between sexes was observed in Other locations, with males getting almost 10 more minutes/day of MVPA compared to females. Males were more active at School than females, but School sedentary time was similar between the sexes. The largest discrepancy in sedentary time between sexes was observed at Home, with females accumulating almost one hour more than males. Within the Other locations, males obtained more MVPA than females in Other: Unknown locations (minutes/day) but amounts and proportions of time were comparable between sexes in Other locations (Table 4).

Table 3 Differences in children’s daily time and activity between males and females (N = 44 children, 17 males, 27 females)

Table 4 Differences in males’ and females’ time and activity in ‘other’ locations categorized by land use (n = 40 children, 14 males, 26 females)

Children obtained comparable amounts of overall MVPA and sedentary time on school days compared to non-school days (Table 5). On school days, children obtained an average of 25.0 min/day of MVPA at School when considering time during and before/after school. On non-school days, the extra time at Home was typically spent in sedentary activities (~ 1.5 h more compared to school days). Children spent almost 2.5 additional hours in Other locations on non-school days compared to school days and those locations contributed a large amount of MVPA. Within the Other locations, more sedentary time was accrued in Other: Unknown locations during non-school days compared to school days (Table 6).

Table 5 Differences in children’s daily time and activity between school days and non-school days (n = 44 children)

Table 6 Time spent in MVPA and SED in ‘other’ location based on land use (n = 40 children)

The distribution of trips is displayed in Table 7. Overall, children spent little time in pedestrian (walking) and cycling trips (5.6 and 4.3 min/day, respectively) relative to vehicle time (33 min/day). A majority of the pedestrian and cycling trips occurred in the school neighborhood and on school days. Of note, 20 (45.4%) participants had on average > 5 min daily of pedestrian trips and 18 (40.9%) participants had on average > 5 min daily of cycling trips.

Table 7 Children’s time in trips by location, gender, and school day or non-school day (n = 44 children)

Discussion

Supporting healthy levels of physical activity for youth, including children with overweight/obesity, is a critical public health initiative [26]. In this study, we examined locational differences in physical activity in rural children with overweight and obesity to understand how best to support these youth accumulate physical activity across their environments. Consistent with children in more urban environments, rural children’s MVPA and sedentary time differed across locations [6, 10, 11, 13]. In this sample, children obtained the most MVPA at School during school hours followed by Other locations, then Home, School outside of school hours, the Home Neighborhood, and finally the School Neighborhood. These findings are similar to those found in previous studies where children were most active at School and areas that make up Other locations [6, 10, 11, 13].

Recommendations for elementary school children are that they obtain 30 min/day of MVPA during the school day (i.e., physical education classes) with additional opportunities for activity (e.g., recess, before/after school activities) [27, 28]. State laws in Kansas at the time of the study did not specify a minimum amount of physical education or physical activity that schools were required to provide [29]. Children in the present study obtained almost 20 min/day of MVPA during the school day and an additional 7 min/day of MVPA at school outside of school hours. These findings highlight the importance of MVPA and demonstrate how MVPA can be accumulated at School both during and outside of school hours. When accounting for the proportion of time spent in different locations, there was little-to-no significant difference in MVPA accumulation across locations. Amount of time spent in sedentary activities was highest at Home with School during school hours closely behind. When considering the proportion of time spent in different locations, children were most sedentary at School during school hours. This suggests that while the School location is supportive of MVPA, the School location also confers risk for high sedentary time, which is consistent with other studies examining MVPA and sedentary time at school [11]. Since the children in this sample spent most of their time at Home, School (during school hours), and Other locations, intervening at these three locations may lead to the most impact in MVPA and sedentary time [13]. These also tended to be locations in which MVPA or sedentary time were the highest.

This study was the first to quantify the amount of time rural youth with overweight/obesity from the United States are spending in bicycle, pedestrian, and vehicle trips along with School neighborhoods. The average amount of time spent in active commuting (bicycle and pedestrian trips) was generally low (approximately 10 min/day) but consistent with previous literature in an urban population [13]. Present findings suggest that active travel to school may account for a large amount of youth’s active commuting activity given that active trips were greater on school days and generally occurred in the School and Home neighborhood. However, only half of the participants lived within ~ 2 km of their school and only 45% walked > 5 min/day. Thus, promoting active trips may be an excellent way to increase physical activity among rural youth, especially youth with overweight/obesity. This could be accomplished by recommending new types of active trips (e.g., walking trips around their Home Neighborhood, active transport to school) or by capitalizing on the current trips occurring in the School Neighborhood by encouraging longer trips (e.g., encouraging them to take an extra lap or take a longer way home). Additionally, given the low amount of activity accrued in the Home and School Neighborhood (4.2 and 5.9 min/day when combining MVPA and active trips), more efforts are needed to provide physical activity opportunities near children’s home and school. The home neighborhood is often an activity supporting environment for children in urban and suburban areas, but this may not be the case for many rural children with overweight/obesity. A previous study reported that children living in rural areas accrued most of their MVPA (57.2%, 23 min/day) within their Home Neighborhood compared to outside of their home neighborhood [12]. This is in contrast to the present study observed that children with overweight/obesity obtained 10.1 min/day of MVPA in the Home and Home Neighborhood (23.8% of total MVPA). It is also possible that children who live outside the town, for example, can have poorer access to streetscape and recreational features often present in urban neighborhoods and thus obtain lower amounts of MVPA within their Home Neighborhood [30, 31].

Regarding vehicle trips, youth spent an average of 30 min on school days and 41 min on non-school days in vehicle trips. Previous research has indicated that youth from large metropolitan areas typically spend approximately 24 min per day in vehicles (note that this value from Carlson et al. (2015) is a median while our study present means) [32]. Thus, our study suggests that youth from rural areas may spend greater time commuting in motor vehicles. This may be due to the greater distances that residents have to travel for tasks such as grocery shopping or healthcare [33, 34]. Interestingly, time spent in vehicle trips was higher among males, who also experienced overall greater MVPA. One hypothesis is that this greater vehicle time was occurring for males as they were being driven to Other and Unknown (i.e., outside of the county) locations to potentially engage in in MVPA. This is further supported by previous findings suggesting that greater time spent in a vehicle was related to more MVPA and less sedentary time [32].

Within Other locations, several types of settings appeared to be relevant for youth’s physical activity. Residential locations outside of the child’s home, where a large proportion of Other activity occurred, are likely to reflect the homes of friends and family members. These settings highlight the importance of peers and family members in supporting physical activity among rural youth with overweight/obesity and show opportunities for leveraging social support. The Other: Community category appeared to include several settings that are known to be important for physical activity. Although participants spent limited time in Other: Community locations, an average of 18 min/day on non-school days, visits to these locations were often associated with physical activity. Around 10% of youth’s time in these community settings involved physical activity, which was the highest among all locations, even higher than the home and school neighborhoods, which were ~ 6%. Thus, efforts to increase availability and use of community settings such as parks, open greenspace, churches, and recreation centers appear promising for supporting increased physical activity among rural youth with overweight/obesity [35].

This study highlights that rural youth with overweight/obesity spend very little of their time at Home engaging in MVPA. Future research and intervention projects should focus on promoting children’s MVPA and reducing sedentary time within the Home. This could include encouraging families to procure resources to increase MVPA at home (e.g., workout or playground equipment), to better utilize technology to improve MVPA at home (e.g., active video games, phone applications that encourage MVPA), or ensuring sedentary time is reduced by creating new family habits (e.g., having MVPA competitions between TV shows) [36, 37, 38]. These Home interventions should especially focus on including tailored components for females who engaged in less MVPA and more sedentary time than their boy counterparts in the Home. Within these environments, involving parents as intervention targets (e.g., encouraging parents to plan and initiate MVPA at home) may be effective in fostering more active behaviors in children [27]. In fact, inclusion of parents in treatment is a central tenet of pediatric obesity treatments [39]. Further, more efforts should be made to intervene on sedentary time in all locations (Home, School, and Other). For example, school-based interventions could create an active classroom environment which could reduce children’s sedentary time [28]. Providing teachers with more training opportunities and resources may support this effort. Finally, interventions should strive for implementation across multiple locations. For those that are delivered at School, including a substantial Home component to involve parents may be helpful for children to attain and sustain adequate MVPA and reduce sedentary time even after the intervention ends.

This study includes both strengths and limitations. A strength is the use of objective measures for physical activity and assessment of location rather than using subjective, self-report measures. Additionally, multiple intensities of activity behavior (i.e., MVPA and sedentary time) were reported. A limitation of this study is that data were only collected during the school year and are therefore not representative of the whole calendar year nor potential impact of seasonality on findings. This may be important as physical activity behaviors in school-age children vary between seasons, likely due to differences in the structure of the day [40]. Additionally, the sample comes from a rural Midwest state in the United States and may not represent other rural settings. The small sample size may also limit generalizability and provided limited power for detecting differences (e.g., by sex). Finally, the Other locations (agricultural, commercial, residential, community, and unknown) were higher-level categories based on the available data provided by each individual county and lacked detail on specific types of facilities known to be important for physical activity (e.g., recreation areas, sports complexes).

Conclusion

The current study presents data on locations where children with overweight/obesity living in rural areas accumulate their MVPA and sedentary time, as well as the amount of time they spend in during active trips. The School location is a location both supportive of MVPA and one where a high amount of sedentary time occurs. In the Home location, children obtained the lowest amount of MVPA and highest proportion of time spent sedentary. Thus, the Home should be targeted for intervening on MVPA and sedentary time, especially for females who may accumulate more sedentary time at home than their male counterparts. Interventions should continue to target both increasing MVPA and reducing sedentary time in Home and School locations for children with overweight/obesity in rural areas.