Outdoor Time Is Associated with Physical Activity, Sedentary Time, and Cardiorespiratory Fitness in Youth Lee Schaefer, PhD1, Ronald C. Plotnikoff, PhD2, Sumit R. Majumdar, MD, MPH3,4, Rebecca Mollard, PhD5, Meaghan Woo, BSc5, Rashik Sadman5, Randi Lynn Rinaldi, MSc6, Normand Boule, PhD6, Brian Torrance, MSc7, Geoff D. C. Ball, PhD, RD8, Paul Veugelers, PhD3, Paul Wozny, EdD9, Linda McCargar, PhD10, Shauna Downs, MSc11, Richard Lewanczuk, MD, PhD5, Douglas Gleddie, PhD12, and Jonathan McGavock, PhD5 Objective To determine whether time spent outdoors was associated with increased moderate-to-vigorous physical activity (MVPA) and related health benefits in youth.
Study design We performed a cross-sectional study of 306 youth aged 13.6 1.4 years. The exposure of interest was self-reported time spent outdoors after school, stratified into three categories: none, some, and most/all of the time. The main outcome of interest was accelerometer-derived MVPA (Actical: 1500 to >6500 counts/min). Secondary outcomes included sedentary behavior, cardiorespiratory fitness, overweight status, and blood pressure. Results Among the 306 youth studied, those who reported spending most/all of their after-school time outdoors (n = 120) participated in more MVPA (61.0 24.3 vs 39.9 19.1 min/day; adjusted P < .001), were more likely to achieve the recommended minimum 60 min/day of MVPA (aOR 2.8; 95% CI, 1.3-6.4), spent less time in sedentary activities (539 97 min/day vs 610 146 min/day; adjusted P < .001), and had higher cardiorespiratory fitness (49 5 vs 45 6 mL/kg/min; adjusted P < .001) than youth who reported no time outdoors (n = 52). No differences in overweight/obesity or blood pressure were observed across the groups. Conclusions Time spent outdoors is positively associated with MVPA and cardiorespiratory fitness in youth and negatively associated with sedentary behavior. Experimental trials are needed to determine whether strategies designed to increase time spent outdoors exert a positive influence on physical activity and fitness levels in youth. (J Pediatr 2014;165:516-21).
ery few youth living in developed countries achieve the recommended minimum of 60 minutes of moderate-to-vigorous physical activity (MVPA) daily.1 The low levels of physical activity among children and adolescents are attributed to several factors, including changes in the built environment,2 increased availability of electronic gaming and computers,3 and possibly reductions in the time spent outdoors.4 Studies by our group5 and others6,7 reveal that youth accumulate the majority of their MVPA during school hours. Specifically, we found that children accumulated 25% more minutes of MVPA on school days relative to weekend days or holidays.5,6 In fact, children achieve on average only 10 minutes of MVPA during the after-school period (3:00-5:00 p.m.).7 Several expert groups have called for increasing outdoor time as one strategy to increase physical activity levels in children during the after-school time period.8,9 Two key observations have initiated calls for more outdoor activities: (1) children accumulate little MVPA during after-school hours; and (2) levels of outdoor play for children have decreased considerably over the past 20 years.8,10,11 Unfortunately, the evidence supporting these calls is lacking and few From the Faculty of Education, University of Regina, Regina, Saskatchewan, Canada; Priority Research well-controlled studies have documented a link between time spent outdoors Center in Physical Activity and Nutrition, The University of and health behaviors in children.12,13 The few studies published to date on this Newcastle, Newcastle, New South Wales, Australia; School of Public Health, Department of Medicine, topic are generally restricted to younger children, relied on parental report of University of Alberta, Edmonton, Alberta, Canada; Department of Pediatrics and Child Health, Manitoba outdoor activity, and have not adequately addressed possible confounding Institute of Child Health, University of Manitoba, 8,10,11 Winnipeg, Manitoba, Canada; Faculty of Physical factors (eg, sex, season, weight status). Education and Recreation, University of Alberta; Ever To overcome limitations of previous studies and to determine whether time Active Schools; Department of Pediatrics, University of Alberta, Edmonton, Alberta, Canada; Black Gold School spent outdoors during the after-school hours was positively associated with District, Alberta, Canada; Faculty of Agricultural, Life and Environmental Sciences, University of Alberta, MVPA and related health outcomes in youth, we performed a cross-sectional Edmonton, Alberta, Canada; Menzies Center for Health study of youth 9-17 years of age enrolled in the Healthy Hearts Prospective Policy, University of Sydney, Sydney, Australia; and Faculty of Education, University of Alberta, Edmonton, Cohort Study of Physical Activity and Cardiometabolic Health.14,15 We hypothAlberta, Canada esized that: (1) time spent outdoors would be positively associated with MVPA; Supported by the Canadian Diabetes Association, the 1
Alberta Center for Child, Family, and Community Research, the Canadian Institutes of Health Research, and the Manitoba Health Research Council. The authors declare no conflicts of interest.
Body mass index Moderate-to-vigorous physical activity
0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2014.05.029
Vol. 165, No. 3 September 2014 and (2) youth who reported spending most or all of their time outdoors after school would be less sedentary and have greater cardiorespiratory fitness than their peers who spent little or no time outdoors during this period.
Methods Data from youth who provided valid accelerometer data (minimum of 3 days of wear time with at least 8 hours of registered minutes per day14) and completed a questionnaire regarding time spent outdoors from the 2009 wave of the Healthy Hearts Prospective Cohort Study of Physical Activity and Cardiometabolic Health were used in this study.14,15 The prospective cohort study originally was designed to determine the association between cardiorespiratory fitness and the risk of overweight status in youth. The study involved annual measures of anthropometrics (height, weight, waist-to-hip ratio), cardiorespiratory fitness, and systolic blood pressure in a school-based cohort of children and adolescents (9-17 years of age) attending schools within the Black Gold School District, for 3 years (2008-2010). Data from this study as well as a more in depth description of the methods used have been published previously.14,15 A convenience sample of 591 youth aged 9-17 wore an accelerometer and completed an online survey (WebSpan)16 in the 2009 wave of data collection. Data for exposure and outcome variables were not available for a sufficient number of youth in all 3 waves; therefore, the data presented here are restricted to the wave with the largest sample size that provided complete data. Data-collection procedures for the larger cohort study have previously been described.14 All parents and youth provided written informed consent and assent, respectively, before initiation of the study, in accordance with the Declaration of Helsinki. This study was approved by the Human Research Ethics Board at the University of Alberta (Edmonton, Alberta, Canada). During class time, youth completed a validated web-based questionnaire of health behaviors that included a selfreported measure of time spent outdoors.16,17 This measure was meant to encompass all time spent out of doors, after the school period, whether taking part in organized activities or free play. Students were asked to recall the amount of time spent outdoors during the afterschool period over the past 7 days with 4 possible responses: (1) none of the time; (2) some of the time; (3) most of the time; and (4) all of the time. As there were few responses for the fourth option, we collapsed answers for option 3 and 4 into a single exposure of “most/all of the time.” The exact time period that makes up the after-school period was not specified in this question. The primary outcome measure (MVPA) was determined from 15-second epochs using waist-mounted accelerometers worn over a period of 7 days (Actical, serial nos. B101270B101375; Respironics, Bend, Oregon) in the winter and spring terms of the 2008/2009 academic calendar year. Raw physical activity counts were converted into minutes of physical activity with the Kinesoft software program (KineSoft,
Saskatoon, Saskatchewan, Canada).18,19 Participants who failed to obtain a minimum of 3 days of wear with at least 480 registered minutes (8 hours) per day and the absence of consecutive zero counts >60 minutes were excluded from analyses. Similar criteria have been used in previous studies by our group and others to estimate habitual physical activity from accelerometer-derived data.14,20-22 Standardized thresholds were used to classify raw counts per minute (cpm) into intensities of physical activity: (1) sedentary time (6500 cpm).1,23 Secondary outcomes included: (1) sedentary behavior derived from accelerometry; (2) cardiorespiratory fitness determined using the Leger shuttle run15,24; and (3) systolic blood pressure assessed in triplicate using an automated machine after 5 minutes of quiet sitting,15 and overweight status determined from age- and sex-specific thresholds for absolute body mass index (BMI) values to classify youth as “healthy weight,” “overweight,” or “obese” according to the International Obesity Task Force guidelines.25 Blood pressure also was treated as a categorical variable with youth stratified as high normal if systolic blood pressure was >90th percentile for age, sex, and height. Age,26 sex,5 weight status,7 and season27 (cold season: November-March; warm season: April-October) when physical activity was undertaken were considered potential confounders. Body weight was measured to the nearest 0.1 kg in duplicate using a digital scale (Seca 882 Digital Floor scale; Seca, Hamburg, Germany), and height was measured in duplicate to the nearest 0.1 cm using a medical standard stadiometer (Seca Portable Model 214). Absolute BMI (kg/m2) was converted to an age- and sex-specific Z-score using EpiInfo.28 Waist circumference was measured using a flexible tape at the top of the iliac crest as previously reported. Statistical Analyses Independent t tests were used to test for group-wise differences in characteristics between youth who were included in the analysis and those who were excluded for lack of sufficient accelerometer data to test for potential sampling bias. Normally distributed variables were compared across levels of outdoor time using ANOVA. Non-normally distributed variables were compared across levels of outdoor time using Kruskal-Wallis tests. Pearson c2 tests were used to determine differences in categorical variables across levels of outdoor time. Any significant differences in outcome measures in unadjusted tests were repeated with multivariable general linear or logistic regression models, adjusting for wear time, sex, age, BMI Z-score, and time of year. In an effort to improve precision of the association, we conducted a second series of analyses where the outcome variable (MVPA) was restricted to the window of time when they were queried about the time they spent outdoors (3:30-11:59 p.m.). A multivariable linear regression model was used to determine whether the time spent outdoors was associated with MVPA and secondary outcome measures independent of confounding variables. The odds of achieving 517
THE JOURNAL OF PEDIATRICS
$60 minutes of MVPA during the data collection period across categories of outdoor time were determined by a multivariable logistic regression adjusting for covariates. To adjust for the confounding effects of sex and season we employed a stratification approach,29 testing for associations between exposure and outcome variables for boys and girls and seasons (warm months vs cold months) separately. All analyses were completed using SPSS version 20 (SAS Institute Inc, Cary, North Carolina). A 2-tailed P < .05 was considered statistically significant.
Results Of the original 591 participants included in the 2009 wave of the cohort study, 306 provided valid accelerometer and outdoor activity data. Of those 306, 52 (17%) reported spending no time outdoors after school, 134 (44%) reported spending some time outdoors, and 120 (39%) reported to having most/all of their time spent outdoors. Compared with those who provided valid data, those who did not were younger (13.0 1.6 years vs 13.6 1.4 years, P < .001), but no differences in sex, cardiorespiratory fitness, BMI Z-score, or MVPA were observed between those included in the analysis and those who failed to provide adequate data. Among the youth retained in the analysis, 58% were girls, 24% were overweight or obese, 34% achieved an average of 60 min/day of MVPA, and they provided an average of 5.1 1.4 days and 790 138 min/day of valid accelerometer data. Participant characteristics stratified by self-reported time outdoors are presented in Table I. Compared with those
Vol. 165, No. 3 reporting never being active outside, youth who reported spending most/all of their time outdoors were younger (13.3 vs 14.3 years, P < .001), more likely to be male (51% vs 31%, P = .03), and contributed data during the warm months of data collection (60% vs 33%, P = .001). No group differences in BMI Z-score were noted. MVPA increased and sedentary time decreased in a doseresponse manner with increasing time spent outdoors (Table I). Specifically, youth reporting no time spent outdoors achieved 21 fewer minutes of MVPA daily (P < .001) and an additional 70 min/day of sedentary behavior (P < .001) compared with youth who reported spending most or all of their after-school time outdoors. The strength and dose-response nature of the association remained significant when analyses were repeated in subgroups of youth stratified by sex and season (Figure 1). In a multivariate regression analysis, outdoor time was a significant predictor of MVPA in youth, independent of age, sex, BMI Z-score, fitness level, sedentary time, and season (Table II). Time spent being sedentary (min/day) and cardiorespiratory fitness (run score classification) also were significant determinants of time spent in MVPA (Table II). The odds of achieving >60 minutes of MVPA was 2.8-fold greater for those who spent most or all of their after-school time outdoors vs those who spent no time outdoors; however, this relationship was dependent on whether the time spent outdoors occurred on weekdays or weekend days (Table III). When analyses were restricted to the after-school period, which we defined as the hours between 3:30 and 11:59 p.m., the strength of the associations between outdoor time
Table I. Participant characteristics compared across the tertiles of time spent outdoors Outdoor exposure Variables
None of the time (n = 52)
Some of the time (n = 134)
Most/all of the time (n = 120)
Age, y Female/male, % Weight, kg BMI Z-score Overweight/obese, % Waist circumference, cm Run score VO2max, mL/kg/min Fit/not fit, %† Warm/cold months, %z Valid days, no. Valid minutes, min/d Sedentary time, min/d Light activity, min/d Moderate activity, min/d Vigorous activity, min/d Total MVPA, min/d >60 min/d total Weekday MVPA, min/d > 60 min/d weekday Weekend MVPA, min/d >60 min/d weekend
14.3 1.5 69/31 54.8 11.6 0.28 .94 21.5 74 10 6.2 2.2 44.8 5.6 48/52 33/67 5.0 1.3 790.4 138.4 609.9 145.5 151.9 66.0 37.0 17.2 2.9 4.2 39.9 19.1 19% 41.6 18.5 19% 31.3 24.7 13%
13.5 1.3* 60/40 54.0 12.7 0.46 0.90 27.8 73 70 6.9 2.2 47.8 5.7* 54/46 41/59 5.0 1.4 784.3 123.7 573.1 108.0 164.5 64.7* 48.8 20.0* 3.9 5.8 52.7 23.3* 31% 57.0 23.0* 37%* 38.5 33.5 12%
13.3 1.3* 49/51* 52.0 14.3 0.36 0.94 20.8 72 10 7.1 2.1* 48.7 5.4* 55/45 60/40*,x 5.1 1.4 789.9 137.6 538.5 97.2* 180.7 94.3*,x 57.0 22.5*,x 4.0 4.0* 61.0 24.3*,x 44%* 66.5 29.7*,x 50%*,x 43.3 29.1 19%