Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Do attributes in the physical environment influence children's physical activity? A review of the literature

Do attributes in the physical environment influence children's physical activity? A review of the... Background: Many youth today are physically inactive. Recent attention linking the physical or built environment to physical activity in adults suggests an investigation into the relationship between the built environment and physical activity in children could guide appropriate intervention strategies. Method: Thirty three quantitative studies that assessed associations between the physical environment (perceived or objectively measured) and physical activity among children (ages 3 to 18-years) and fulfilled selection criteria were reviewed. Findings were categorized and discussed according to three dimensions of the physical environment including recreational infrastructure, transport infrastructure, and local conditions. Results: Results across the various studies showed that children's participation in physical activity is positively associated with publicly provided recreational infrastructure (access to recreational facilities and schools) and transport infrastructure (presence of sidewalks and controlled intersections, access to destinations and public transportation). At the same time, transport infrastructure (number of roads to cross and traffic density/speed) and local conditions (crime, area deprivation) are negatively associated with children's participation in physical activity. Conclusion: Results highlight links between the physical environment and children's physical activity. Additional research using a transdisciplinary approach and assessing moderating and mediating variables is necessary to appropriately inform policy efforts. low bone density [4], and low physical fitness [5]. Further- Background Many youth today are physically inactive. Considerable more, children who are not physically active are denied evidence documents that nearly 35% of youth in the US the positive social and emotional benefits of physical fail to meet the minimum physical activity guidelines, and activity including higher self esteem, lower anxiety, and another 14% are completely inactive [1,2]. Low levels of lower stress [6]. A comprehensive understanding of the physical activity and the failure to meet physical activity determinants of physical activity among youth is essential recommendations have notable health consequences for the identification of appropriate points of intervention among children including increased risk of obesity [3], to promote active lifestyles and their associated health Page 1 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 benefits. In this paper, we examine environmental influ- organizing schema that identifies the parties responsible ences on children's physical activity. Specifically, we for specific elements in the built environment. review research assessing the association between attributes of the physical environment and children and Methods adolescents' physical activity. Definition of the physical environment The physical environment is defined herein as objective The physical or built environment has come to the fore- and perceived characteristics of the physical context in front of public health research in the past 5 years, leading which children spend their time (e.g., home, neighbor- to a surge of research on environmental attributes and hood, school) including aspects of urban design (e.g., their associations with physical activity behaviors. A presence and structure of sidewalks), traffic density and number of reviews have examined links between the phys- speed, distance to and design of venues for physical activ- ical environment and adults' physical activity [7-12]. ity (e.g., playgrounds, parks and school yards), crime, Much less emphasis has been placed on research specific safety and weather conditions. While crime and safety are to children. One cannot assume that associations between not explicitly characteristics of the physical environment, the physical environment and physical activity among they are included in this review as both are intimately adults are applicable to children. As highlighted by Krizek, linked with multiple characteristics of the physical envi- Birnbaum & Levinson [13], children in contrast to adults, ronment including for example lighting, the condition of spend large parts of their day at school, have considerable buildings, and the presence of trash. They also have vicin- time for recreation, are more likely to accumulate physical ity effects such that a particular area can gain a reputation activity through play, are not able to drive, and are subject for safety or criminal activity. to restrictions placed on them by adults. Identification of studies Two reviews to date are specific to children. In 2000, Sallis Computer searches using PubMed, PsychInfo, EBSCO, et al. [14] published a comprehensive review of predictors CINAHL, and TRANSPORT were conducted in the Eng- of physical activity among youth. Studies published lish-language literature to identify published studies and between 1970 and 1998 were included in the review. reports examining relationships between the physical While this review does not focus on the physical environ- environment and children and adolescents' physical activ- ment, a small proportion of the 108 studies reviewed are ity. Transportation and urban planning reports were specific to the physical environment. More recently, in accessed using TRANSPORT and general internet searches 2005, McMillan [15] reviewed studies in both planning and by searching the bibliographies of papers. Search and public health literatures on urban form and children's terms included physical activity, exercise, recreation, trip to school. McMillan outlines policies and programs sport, walk/walking, cycle/cycling, transport, active com- that may promote walking and cycling to school (e.g., Safe muting, environment, environmental determinants, Routes to School) and highlights the lack of focus on chil- physical environment, built environment, perceived envi- dren in the transportation literature. In the absence of ronment, design, urban design, context, facilities, neigh- research on environmental factors that affect children's borhood, park, playground, situational factors, safety, trips to school, most of the studies reviewed by McMillan crime and weather. These search terms are a compilation focus on adult populations. of the terms used in previous reviews [7-9,11,12]. In addi- tion, the search terms children, child, adolescent, adoles- In this descriptive review, we build on the work of Sallis et cence, youth, family, and parent were added to the terms al. and McMillan by reviewing recent studies (published list to limit studies to children and adolescents. The bibli- between 1990 and 2006) that examine the association ographies of the identified studies were also reviewed for between children's physical activity and environmental additional references. attributes (perceived and objectively measured). In partic- ular, we provide specific information on the sample char- Studies were not further considered if they (a) did not acteristics and design of each study, evaluate consistencies measure or model the environment (perceived or objec- and inconsistencies in the literature, and identify gaps in tive) and physical activity behaviors, (b) were descriptive the current research and possible avenues for future in nature (e.g. qualitative studies were not reviewed), (c) research. In addition, we broaden the set of children's only used a composite score of the environment that com- behaviors from their trip to school as outlined by McMil- bined a number of environmental attributes and (d) did lan to physical activity in general. In order to serve both not report findings for children separately from those for the need for understanding the link the physical environ- adults. The database searches resulted in a total of 106 ment and physical activity among youth and the imple- "hits", of which 29 were relevant and were further consid- mentation of next steps based on these findings, we use an ered. Of the 29 studies, 23 met the selection criteria. Addi- tional studies were identified through searching the Page 2 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 bibliographies of articles. A total of 33 articles were iden- Definitions and examples of each domain are provided, tified for inclusion in the review. the responsible parties for each are identified, the key findings for each domain are summarized, and consisten- Recording and synthesizing research findings cies and inconsistencies are highlighted and possible For each study, the following were recorded: (a) first explanations are provided. author and year; (b) sample characteristics such as sample size, gender, and ethnic/racial group of the participants; Results (c) age of participants; (d) research design, including Recreational infrastructure whether it was cross sectional or longitudinal and whether Recreational infrastructure (play areas) for children can be perceived and/or objective environmental attributes were classified as private (provided by their parents in and assessed; (e) the environmental variables that were exam- around the home), public (community areas or schools) ined; (f) the type of physical activity behavior assessed or private-public (commercial play areas). Private recrea- and the method of assessment; and (g) a summary of the tional infrastructure may be subject to regulations associ- significant associations that were identified (see Table 1 ated with the property and issues of flexibility of use based for a summary). This information was recorded by both on tenure of ownership (e.g., renter or home-owner). authors and a graduate student and was cross checked to Public recreational infrastructure is primarily the respon- identify any inconsistencies. In instances in which multi- sibility of the municipality or agency charged with the ple aspects of the physical environment were assessed, provision of the original infrastructure, as is the mainte- results for each component were recorded. If multiple nance and continued monitoring of the safety and condi- measures of physical activity were included in the study, tion of such assets. School yards, playgrounds and open only results for the most comprehensive measure were space parks are most often considered public recreational recorded. For example, if a study included both objec- infrastructure. Private-public recreational infrastructure tively measured physical activity and a generalized self- includes youth camps, commercial clubs, and other busi- report measure, results for the objective measure are nesses providing places for children to participate in phys- reported. This simplification was necessary given the ical activity. All of these recreational infrastructures are breadth of measures of physical activity used across stud- subject to land use regulations, including zoning codes. ies. Simplifying the presentation of results for physical Twenty one studies that were reviewed examined the rela- activity also served to maintain the focus of this review on tionship between recreational infrastructure and chil- the environmental correlates of physical activity broadly dren's physical activity [16-36]. The overwhelming construed. majority (i.e., 19) of studies used a cross sectional design. One study used a 1-year longitudinal design and one used Findings from the studies are reviewed and synthesized an intervention design. Five of the 21 studies used an using three a priori categories of environmental attributes objective measure of physical activity, including either including: (1) recreational infrastructure (e.g., the availa- accelerometry or heart rate monitoring, four used direct bility of parks/playgrounds, equipment in the home); (2) observation, 13 used a self-report measure, and one study transport infrastructure (e.g., traffic speed/density, pres- used both objective and self-report measures. Seven stud- ence of sidewalks); and (3) local conditions (e.g., safety, ies used an objective measure of the environment (gener- crime, weather). These categories were chosen to facilitate ally based on Geographic Information Systems), 12 used the identification of the parties responsible for changing a self-report measure, and two studies used both method- an environmental attribute and consequently possible ologies. Finally, 12 of the studies were conducted in the avenues for intervention. Using a system similar to previ- US with the remaining studies being conducted in coun- ous reviews on physical activity [7,14] it was noted tries including Canada, England, Australia and Portugal. whether a finding was positive and significant (+), nega- tive and significant (-) or not statistically significant (0). Private recreational infrastructure This information is summarized in Table 2. In order to Home equipment facilitate the comparison of findings across studies, results Four out of six studies found no association between from bivariate (in contrast to multivariate) analyses are home equipment and children's physical activity. Specifi- recorded and, where possible, results from bivariate anal- cally, Sallis et al. [22] found no association between an yses controlling for basic demographic variables (i.e., SES) objective assessment of equipment available in the home are presented. Results are recorded separately for per- and observed levels of physical activity among preschool ceived and objectively measured attributes of the environ- children. Dunton et al.[19] and Trost et al. [25] found no ment. Results specific to the perceived environment were association between adolescents' reports of equipment in further separated according to children's and adults' (usu- the home and their self-reported physical activity. A sec- ally parents) reports of the environment. The narrative ond study by Trost et al.[26] found no association review below accompanies the data presented in Table 2. between adolescents' reports of home equipment and Page 3 of 17 (page number not for citation purposes) Table 1: Characteristics and main findings of the studies reviewed First Author Number/Gender/ Age group Design Environmental attributes Physical activity behavior Significant associations with ref# (year) Ethnicity/Country (independent variables) (outcome variable) outcome variable Adkins (2004) 52 F B USA 8- to 10-years CS, P There are playgrounds, parks and Objectively measured (accelerometer) No associations were identified gyms nearby, it is safe to play outside physical activity between environmental attributes and (parent and child report) physical activity. Baranowski (1993) 191 M/F B/W/H USA 3 and 4 years CS, O Month of the year (weather) Directly observed physical activity. Children were least active outdoors during the hottest months. 39 rd Boarnet (2005) 62 M/F W/B/H/A USA Parents of children in 3 CS, O Installation of sidewalks, crossing Parents' perceived change in child Greater increases in perceived rates th – 5 grade (8–10 years) signals, traffic control as part of a Safe walking/biking to school of children walking/riding to school for Routes to School (SR2S) program. children who passed a completed SR2S zone compared to those who did not pass a zone. 30 th Braza (2004) 105 students from 34 5 grade (ages 9 to 11 CS, O School size; population density; Rates of walking and biking to school Higher population density and a schools M/F W/B/H/A years) number of intersections per street among students surveyed in each greater number of intersections per USA Unit of analysis: schools mile in .5 mile buffer around school school street mile were associated with site. Data were obtained using higher rates of walking and biking to Geographic Information Systems. school in bivariate models. Brodersen (2005) 4320 M/F W/B/A 11 to 12 years CS, O Area deprivation; number of sport Self-reported days during past week Area deprivation (F) and total rainfall England pitches in borough; public spending on child performed hard exercise that (F) were associated with lower leisure facilities and open spaces; made him/her breathe heavily and physical activity. Colder temperatures weather conditions. sweat. (M) and number of sport pitches (F) were associated with higher physical activity. Burdette (2005) 3141 M/F W/B/H USA 3 years old CS, P Mothers' ratings of perceived Mothers' reports of the average time No associations between mothers' neighborhood safety per day their child played outdoors perception of neighborhood safety and their reports of the time their child spent playing outdoors Carver (2005) 347 M/F U Australia 12–13 years CS, P Parents' perceptions of good sports Child self-reported frequency and Adolescents walked or cycled more facilities for child, safe for child to duration of walking or cycling in the frequently when there were fewer walk/ride, good places for child to be neighborhood (for recreation, unattended dogs (M, F), there were active, traffic makes it difficult to walk. transport, exercise, get to school). good places to be active (F), traffic was Child perceptions of ease to get less problematic (M, F), there was around by bike, safety while walking/ lower perceived ease to cycle (M), riding, roads safe, unattended dogs, there were more sport facilities in the strangers, fast food and convenience area (M), the roads were perceived as stores near home. safe(F), and convenience stores were further from home (F). Above is a simplified summary of results given number of variables assessed and analyses performed (i.e., >400 associations assessed). 38 th Cohen (2006) 1554 F W/H/B/A USA 6 grade 12–13 years CS, O Distance to school along the shortest Objectively measured (accelerometer) A shorter distance to school was street network physical activity associated with greater MVPA during weekdays but not during the weekend International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Page 4 of 17 (page number not for citation purposes) Table 1: Characteristics and main findings of the studies reviewed (Continued) Dunton (2003) 87 G W/H/A USA 14–17 years CS, P Perceived activity-related equipment Self-reported vigorous physical No associations were identified in the home and activity-related activity, total energy expenditure, and between activity-related resources in resources in the community (e.g., leisure time activity the home or the community and girls' park, gym, biking trail) self-reported physical activity. 33 th Ewing (2004) 726 people and 709 Students K-12 grade CS, O Estimated walk/bike time between Likelihood of walking or biking to Students with shorter walk or bike school trips surveyed destinations; proportion of street school times to school, and students traveling U (gender) miles with street trees, bike lanes or through areas with sidewalks on main U(ethnicity) USA paved shoulders, or sidewalks; roads were more likely to walk or sidewalk width; accessibility of bike to school. School size was not attractions; neighborhood population related to the likelihood of walking/ density; school size biking to school. Fein (2004) 610 M/F W Canada Grades 9–12 Mean age CS, P Home environment; convenient Self-reported physical activity Perceived convenience of facilities and 15.5 years facilities (park, bike trails, gym, skating perceived home, neighborhood, and rink); School environment (gym space, school environment were significantly availability of exercise equipment, correlated with self-reported physical athletic facilities accessible). The activity. The perceived importance of perceived importance of each each of these constructs was also resource was also assessed. associated with higher physical activity. Felton (2002) 1668 F W/B USA 8th grade (approx age 13 CS, O Urban/rural residence Self-reported moderate and vigorous White girls living in urban areas and years) physical activity black girls in rural areas reported higher vigorous activity than their respective counterparts. Gomez (2004) 177 M, F H USA 7th grade (approx age 12 CS, P, Crime density (O); perceived Self-reported participation in outdoor Greater proximity to play areas (M), years) O neighborhood safety (P); distance to activities (not in school) lower crime density (F), and high nearest play areas (O) perceived safety (F) were associated with higher outdoor activity. th th Gordon-Larsen 17766 M, F W/B/H/A 7 to 12 grade (approx CS, O Urban/rural residence; crime; month Self-reported moderate to vigorous Lower reported crime was associated (2000) USA ages 12 – 17 years) of the year; region (South West, physical activity with higher moderate to vigorous Midwest, Northeast). activity. Hume (2005) 127 M, F U Australia 10 year olds CS, P Children drew maps of their home Objectively measured (accelerometer) Girls who drew a greater number of and neighborhood environments. The physical activity. opportunities for physical activity in frequency with which particular their neighborhood (e.g., the objects and locations were availability gyms, recreation and represented was coded including swimming centers, playgrounds) green space and outdoor areas and exhibited higher physical activity opportunities for physical activity in (specifically, low intensity physical the neighborhood (e.g., playgrounds activity). and facilities). Jago (2005) 210 M W/B/H USA 10–14 years CS, O Ease of walking/cycling; tidiness of Objectively measured (accelerometer) Sidewalk characteristics that foster neighborhood; sidewalk physical activity. walking (e.g., distance to curb, characteristics; street access and presence of trees as a buffer) were conditions positively associated with light- intensity physical activity. Molnar (2004) 1378 M/F W/B/H USA 11 to 16 years CS, P, Residents' perceived neighborhood Hours/week participated in More safe areas for children to play O safety and opportunities for children recreational physical activity (parent and lower social and physical disorder to play (P); social and physical disorder report). were associated with higher (O). recreational activity. International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Page 5 of 17 (page number not for citation purposes) Table 1: Characteristics and main findings of the studies reviewed (Continued) 32 th th Mota (2005) 1123 M/F U Portugal 7 – 12 grade Mean CS, P Adolescent reports of the activity- Self-reported physical activity In comparison to low active age: 14.6 ± 1.6 friendliness of their neighborhood (e.g. adolescents, high active adolescents access to destinations, connectivity of reported greater access to streets, infrastructure for walking and destinations such as stores and transit cycling, neighborhood safety, stops, higher neighborhood aesthetics, aesthetics, and recreational facilities). and more recreational facilities in their neighborhood. Norman (2006) 799 M/F W/H/B/A 11–15 years CS, O Number of private recreational Objectively measured (accelerometer) Significant bivariate associations were USA facilities, schools and parks within 1 physical activity. found between moderate-to-vigorous mile of home; walkability as assessed PA and the number of recreation by residential density, retail floor area, facilities (girls), the number of parks intersection density, and land use mix and measures of walkability including intersection density (girls), and retail floor area ratio (boys). Sallis (1993) 347 M/F W/H USA 4 years old CS, P, Number of specified play spaces (e.g., Directly observed physical activity. A greater number of specified play O friend's backyard, park) within walking spaces within walking distance of distance of home (P); equipment at home was associated with higher home (O). physical activity. 45 th th Sallis (1999) 732 M/F W/A/PI/H 4 – 5 grade (ages 9 to L (20 Neighborhood safety (parent report) Parent and child report of child No links were identified between USA 10 years) at baseline months physical activity. Objectively measured neighborhood safety and baseline ), P (accelerometer) physical activity physical activity or change in activity. Sallis (2001) 151 areas in 24 middle Middle-school-aged CS, O Type of play area (court space, open Directly observed physical activity of Higher levels of activity were noted schools USA students (approx ages 11 field space, indoor activity space); area students in each play area. when equipment was available in to 13 years) size; permanent activity structures outdoor play areas (F) when more (e.g., basketball hoops, tennis courts, permanent activity structures were soccer goals); equipment. available (M), and when such structures were available in combination with adult supervision (F). Sallis (2002) 781 M/F W (75%) Grades 1–12 (ages 6–18) CS, P Safe to play outdoors; access to parks/ Parents' reports of children's physical Among girls in grades 10–12, parents' USA playgrounds; distance to park; safety of activity and objectively measured perception of neighborhood safety nearest park. (accelerometer) physical activity (N = was associated with higher physical sub sample of 200) activity. Among girls in grades 7–9, parents' perception of park safety was negatively associated with children's physical activity. Unit of analysis = Elementary schools CS, O School urbanization and weather Rates of walking and cycling to school No associations between Sirard (2005) school (N = 8) USA conditions. for each school. environmental variables and active commuting were identified. Stratton (2005) 99 M/F U Wales and 4–11 years I, O Intervention in which school Heart rate telemeters were used to In comparison to control schools, England playgrounds were painted with murals, assess heart rate during physical time spent in MVPA and VPA hopscotch, fun trails, snakes and activity and converted to represent increased significantly in intervention ladders, and court markings (e.g., lines MVPA and VPA. schools as a result of playground for basketball). painting. th th Stucky-Ropp 240 M/F W USA 5 and 6 grade Mean CS, P Number of exercise-related items at Self reported physical activity A greater number of exercise-related (1993) age: 11.2 ± .7 home items in the home was associated with higher physical activity among girls but not boys. International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Page 6 of 17 (page number not for citation purposes) Table 1: Characteristics and main findings of the studies reviewed (Continued) Tappe (1989) 236 M/F W, B, A USA High school Mean age: CS, P Unsuitable weather as a barrier to Self-reported physical activity. No differences in weather as a 15 years 9 months exercise perceived barrier for physical activity among low and high active girls and boys. Timperio (2004) 1200 M/F U Australia 5–6 years and 10–12 CS, P Traffic density, road safety, strangers, Walking/riding to particular Among 5–6 year olds, parents' years. sporting facilities, and public destinations (e.g., friend's house, park, perception of heavy traffic (M), and transportation (parent report). school) 3 or more times/week (parent limited public transportation (F) were Children 10–12 years also reported report) associated with lower walking/cycling on perceived traffic, road safety, among children. Among 10–12 year strangers, and sport facilities. olds, youth who perceived no parks nearby (M, F) and whose parents believed that they had to cross many roads to get to play areas (M, F), that there were no lights or crossings (M), that there were few sporting arenas (F), and that there was limited public transportation (F) were less likely to bicycle/walk. Timperio (2006) † 912 M/F U Australia 5–6 years and 10–12 CS, O Distance to school, busy-road barrier, Walking or riding to school (parent In both age groups, children were less years route along busy road, pedestrian report) likely to actively commute to school if route directness (connectivity), steep their route as >800 m and a busy incline route barrier was present en route. Children with a steep incline (5–6 year olds) and a direct route to school (10– 12 year olds) were less likely to actively commute 25 th th Trost (1997) 202 (rural) M/F B/W 5 – 6 grade (ages 10 L (1 Availability of activity-related Self-reported physical activity No links between home equipment USA to 11) at baseline year), P equipment in the home. measured one year after determinants. and physical activity. 26 th Trost (1999) 108 M/F B USA 6 grade (approx age 11 CS, P Availability of activity-related Objectively measured (accelerometer) No links between home equipment years) equipment in the home. physical activity and physical activity 27 th th Zakarian (1994) 1634 M/F H/W/A/B 9 and 11 grade CS, P Number of facilities for sport and Self-reported vigorous exercise (20 Access to facilities was associated with USA (approx age 14 and 16 exercise; safe to exercise in minutes of activity that makes your higher vigorous exercise. years) neighborhood. heart rate and breathing increase) Zask (2001) 3912 M/F U Australia 5–12 years CS, O Direct observation of the availability of Direct observation of children's The presence of equipment (other activity-related equipment (e.g., balls, physical activity behavior in all school than balls) was not associated with fixed equipment) playground areas. children's physical activity. Note: Number/Gender/Ethnicity/Country: M, male; F, female; W, White; B = African American/Black; H, Hispanic/Mexican American; A, Asian; PI, Pacific Island; U = unknown (not mentioned); USA, United States of America. Design: CS, cross sectional; L, longitudinal; I, intervention; P, perceived environment; O, objectively measured environment. Physical activity behavior: MVPA, moderate to vigorous physical activity; VPA, vigorous physical activity. Significant associations: M, F, significant findings limited to males and females respectively. If an ethnic group made up ≤ 2% of the total sample, it was not included in the list of ethnic groups assessed. † Given that the same sample was used in Timperio (2006) and Timperio (2004) and there was an overlap in the measure of physical activity, only the novel findings are recorded for the more recent study. International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Page 7 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Table 2: Pattern of findings for links between environmental attributes (perceived and objective) and children's physical activity. Associations with physical activity Environmental Attribute Perceived Environment Objectively measured environment Adult report Child report Recreational infrastructure Private (19) (25) (26) (F)(31) (20) (22) Home equipment 0 0 0 + + 0 Public (B)(16) (28) (22) (B)(16) (24) (M)(21) Proximity of playgrounds and parks 0 0 + 0 + + (28) (18) (F)(19) (27) (32) (20) (F)(24) (F)(34) (F)(17) (F)(37) Availability recreation facilities 0 + 0 + + + + + + + (17) Spending on recreational infrastructure 0 (36) (F)(38) (33) Distance to school (school location) - - - (30) (33) School size 0 0 (20) (35) (23) (29) Equipment/play structures in school play areas + 0 + + Transport infrastructure Provision of amenities (32) (33) (39) + + Presence of sidewalks 0 (40) Street and sidewalk conditions + (M)(18) (40) (33) Presence of bike lanes/ease of cycling - 0 0 (M)(24) (39) Presence of controlled crossings + + (32) (30) (36) (F)(37) Connectivity of street network 0 + - + (32) (F)(18) (M)(37) Access to destinations + - + (F)(24) Availability of public transportation + Road hazards (24) Number of roads to cross - (M)(24) (18) (36) Traffic (density/speed) - - - (F)(18) Pedestrian and cyclist safety + (36) Steep terrain - Local conditions Safety and neighborhood disorder (45) (B)(16) (28) (48) (44) (27) (B)(16) (32) (F)(21) Perceived safety 0 0 0 0 + 0 0 0 + (F)(17) (F)(21) (43) Area deprivation and crime - - - (18) Roaming dogs - (24) (44) Social disorder/stranger danger 0 - (40) (44) Physical disorder/tidiness of area 0 - (32) Aesthetics of neighborhood + Region and weather (43) 11 (M)(17) Month of year (average temperature) 0 - - (46) (47) (17) Unsuitable weather 0 0 - (43) Region of the United States 0 (47) (B)(42) (W)(42) Rural/suburban versus urban 0 + - (30) (33) Population density + 0 Note: Associations identified with physical activity: +, significant positive association; -, significant negative association; 0, no association. Superscript numbers = reference number. Effects that are specific to particular demographic groups are noted as follows: M = males; F = females; B = Black/African American sample or subgroup their objectively measured physical activity using acceler- activity among adolescents girls and boys and young ado- ometers. In contrast, Fein and colleagues [20] and Stucky- lescent girls (but not boys). Ropp and DiLorenzo [31] found that the number pieces of exercise equipment in the home was positively and sig- Differences in the results outlined above cannot be nificantly associated with higher self-reported physical explained by differences in sample size, participant age, or Page 8 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 the operationalization of home equipment. Differences, home, and children's physical activity. Among Australian however, may be explained by differences in the ethnic samples, Timperio et al. [24] found that parents' reports of composition of the samples; both studies identifying a few sporting arenas in the area were linked with lower significant effect for home equipment used a predomi- rates of walking and cycling among girls and Carver et al. nantly white sample, whereas, studies that did not iden- [18] found that parents' reports of the presence of good tify an effect used either an exclusively African American sporting facilities nearby for their children were associated sample [26] or samples of mixed racial/ethnic back- with higher self-reported walking or cycling among ado- ground [3,19,22]. It should also be noted that both stud- lescent girls and boys (A simplified summary of the results ies identifying a significant positive effect used a self- from Carver et. al are presented throughout this review report measure of physical activity. The remaining studies given the extensive number of variables assessed. Only used self-report [19,25] or an objective assessment results for the frequency of walking/cycling in general are [22,26] of physical activity. Thus, any association identi- reported). In a study combining qualitative and quantita- fied between home equipment and children's physical tive methods, Hume et al. [34] found that, when children activity is limited to white adolescent samples and to self- were instructed to draw pictures of their home and their report measures of physical activity. neighborhood, girls who drew more opportunities for physical activity, including recreational facilities such as Public recreational infrastructure gyms and swimming centers, had higher objectively meas- Proximity of parks and playgrounds ured physical activity. Among US samples, Zakarian et al. A significant positive association between the proximity [27] found that a greater number of facilities for sport and of parks and playgrounds to the home and children's exercise in the area (based on self report) were associated physical activity was identified in three out of five studies. with higher adolescent self-reported vigorous activity and In an exclusively Hispanic sample, Gomez et al[21] found Brodersen et al. [17] found that the number of sport that objectively measured distance to the nearest play area pitches in the borough, as determined by objective assess- was inversely associated with adolescent boys', but not ment, was associated with higher self-reported vigorous girls', self-reported physical activity. Sallis et al. [22] found activity among girls but not boys. Similarly, Norman et. that parents' reports of the number of play areas within al. [37] found that objective measures of the number of walking distance of the home were positively associated recreational facilities and parks within a mile of the home with observed levels of physical activity among preschool were associated with higher objectively measured physical children. Furthermore, Timperio, et al. [24] found that activity among adolescent girls, but not boys. Finally, children who reported a lack of parks or sports grounds Mota et al. [32] and Fein et al.[20] using samples from near their home made fewer walking and cycling trips. In Portugal and Canada respectively, found that adolescents' contrast to these studies, Sallis et al. [28] and Adkins et al. reports of the availability of facilities such as swimming [16] (using an exclusively Black sample) found no associ- pools, playgrounds and parks were associated with higher ation between proximity of playgrounds and parks and self-reported physical activity. In contrast to the afore- children's objectively measured physical activity. mentioned studies, Dunton et al. [19] found no associa- tion between girls' reports of activity-related resources in Although a number of ethnic/racial groups were assessed the community and their self-reported physical activity across studies, no consistent ethnic/racial differences were and Sallis et al. [28] found no association between access identified. Differences in methods used to assess physical to facilities and children's objectively measured physical activity, however, were noted for studies that did and did activity. In addition, no association was identified not identify a significant association. Both studies that between spending on recreational infrastructure and chil- found no association [16,28] assessed physical activity dren's self-reported physical activity [17]. using accelerometers, which provide an aggregate meas- ure of physical activity across a number of days. In con- With one exception, there are no obvious differences in trast, studies that found a significant association relied on the designs of studies that did and did not identify a sig- self-reported or observed physical activity, both of which nificant association between the availability of recrea- are prone to reporter/observer bias, but which can be tai- tional areas and children's physical activity. Specifically, lored to provide a specific measure of physical activity there were no clear differences across studies in the defini- (e.g., walking or cycling trips). tion of recreational facilities (which usually included structures such as swimming pools, gyms, sporting arenas, Availability of recreation areas and spending on recreational and parks), the methods used to assess physical activity, or infrastructure the demographic characteristics of the samples. There In eight out of ten studies, a significant positive associa- were, however, clear differences is sample size across stud- tion was identified between the availability of recreation ies. The majority of studies that identified a significant areas, or the presence of such areas in the vicinity of the effect used samples of 1000 or more participants. In con- Page 9 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 trast, the two studies that found no effect used samples of activity. No associations, however, were found between approximately 100 participants, taking age and gender school size and children's physical activity. The lack of break-downs into consideration. This suggests that the effects of school size reported by Braza et al. [30] and association between the availability of facilities and phys- Ewing et al. [33] may be attributable to the use of aggre- ical activity among youth is relatively small and therefore gate data, or data collected at one level (e.g., a census only measurable with a large sample. While the availabil- track) that is then aggregated to a higher level (e.g., ity of facilities was assessed in all studies, no studies county). As a result of the process of aggregation, any directly asked children or parents whether they used such information pertaining to individual residences or spe- facilities. Consequently, the association between recrea- cific locations is lost. tional facilities and physical activity is indirect at best. Transport infrastructure School characteristics Two types of transport infrastructure were examined in Three out of three studies identified a negative association studies including the provision of amenities (e.g., side- between distance to school and children's physical activ- walks, crossings) and the presence of road hazards. Trans- ity. Timperio et al.[36] and Cohen et.al. [38] (girls only) portation infrastructure in urban areas is the found significant negative associations between an objec- responsibility of a number of agencies. For example, in tive measure of distance to school and children and ado- the United States, Metropolitan Planning Organization lescents' objectively measured moderate to vigorous (MPO) are generally charged with the preparation of plan- physical activity. Ewing et al. [33] found that lower walk/ ning documents and the allocation of funding for major cycle time to school, an indirect measure of distance, was programs and projects, whereas, the designation of cross- associated with higher rates of active commuting to walks, traffic signals, pedestrian signage, and other amen- school. In contrast to studies assessing distance to school, ities are in general the responsibility of various Braza et al. [30] and Ewing et al. [33] found no association transportation departments based on right-of-way and between school size, an indirect measure of whether or public ownership of property. Nine studies assessed asso- the school is located in a residential area and therefore ciations between transport infrastructure and children's close to homes, and the rates of walking and cycling to physical activity [18,24,30,32,33,36,37,39,40]. All nine school. studies used a cross sectional design. Two studies used an objective measure of physical activity and six studies used With respect to characteristics within schools, Sallis et al. an objective measure of the environment. The remaining [23] found that middle-school-aged children were more studies relied on self-report instruments. Five of the nine likely to be active during school recess periods when there studies were conducted in the US; the remaining studies was a larger number of activity-related equipment (e.g., were conducted in Australia and Portugal. balls) and the permanent activity structures (e.g., basket- ball hoops) available; these effects were most notable in Provision of amenities Presence and condition of sidewalks and bike lanes the presence of adult supervision. Similarly, Fein et al. [20] found that adolescents' reports of the availability of Results generally supported a positive association sports equipment, the functionability of equipment, and between the presence and condition of sidewalks and access to athletic facilities at school were associated with children's physical activity with three out of four studies higher self-reported physical activity. In contrast, Zask et identifying a significant positive effect. Ewing et al. [33] al. [35] found no association between the availability of found that the proportion of street miles with sidewalks playground equipment (with the exception of balls) and was positively associated with children's rates of walking children's physical activity. Finally, in an intervention or cycling to school. In an evaluation of the implementa- examining the effect of playground markings such as hop- tion of a Safe Routes to School program, Boarnet et al. scotch and court lines for basketball on children's physical [39] found that children who passed areas in which side- activity, Stratton and Mullan [29] found significant walks were installed were more likely to walk or cycle to increases in moderate to vigorous physical activity and school than children who did not pass such areas. In con- vigorous physical activity in intervention schools relative trast, Mota et al. [32] found no association between the to control schools. perceived presence of sidewalks on streets in the neigh- borhood and adolescents' self-reported activity. In the In sum, three out of three studies found that children who only study that assessed the impact of sidewalk condi- live close to schools are more likely to actively commute tions, Jago and colleagues [40] found that objectively to school and three out of four studies found that children assessed sidewalk characteristics such as the distance from were more active during play periods when characteristics the sidewalk to the curb, average height of trees, and side- of school play areas (e.g., access to equipment, permanent walk material and type were associated with higher objec- play structures, and marked courts) facilitated physical tively measured light intensity physical activity (e.g., slow Page 10 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 walking) among children. The studies that identified sig- self-reported activity. In contrast to what might be nificant effects used objective measures of the environ- expected, Timperio et al. [36] found that a more direct ment and measured children's walking (or low intensity route to school (i.e., higher connectivity, which was physical activity) as the outcome variable, which is the assessed using objective methods), was associated with most likely component of physical to be influenced by lower rates of walking and cycling to school among older sidewalk characteristics. In the only study that failed to children (10–12 years); no links were found between con- identify a significant effect, a self report measure of side- nectivity and active commuting to school among younger walk availability was used along with a generalized meas- children (5–6 years of age). ure of physical activity that may not reflect subtle differences in physical activity that result from the pres- The difference in findings reported by Mota et al. versus ence of sidewalks. Braza et al. and Norman et al. may reflect the possibility that effects of connectivity are only observed when objec- With respect to infrastructure for cycling, Jago et al. [40] tive measures of connectivity are used; it is possible that found no association between the ease of cycling (pres- individuals are not able to accurately recall and report the ence of bike lanes, attractiveness for cycling, number of level of street connectivity in their neighborhood. The read lanes) and objectively measured light intensity phys- findings outlined by Timperio et al., which were opposite ical activity in a sample of boys and Ewing et al. [33] to those expected (with higher connectivity or a more found no association between the presence of bike lanes direct route associated with lower rates of active commut- and children's walking/cycling to school. Furthermore, ing to school), are more difficult to explain. Timperio et Carver et al. [18] found that the perceived ease of cycling al. suggest that the counterintuitive effects of connectivity was associated with lower (rather than higher) rates of in their study may reflect the possibility that children's cycling among boys. Spurious findings for the presence of travel behavior is more influenced by traffic safety con- bike lanes or ease of cycling may be explained by a cerns than street networks. number of factors including the use of a measure of phys- ical activity that cannot detect cycling (i.e., accelerome- Three out of four studies identified a significant positive ters) [40], low rates of bicycling to school in general [33], association between access to destinations and children's and inflated type II error due to performing an extensive physical activity. This consistent pattern was noted number of analyses [18]. although a variety of measures of access were used across studies including the presence of destinations such as Presence of controlled crossings, street connectivity, and access to shops, access to public transportation, and retail floor area destinations ratio (i.e., ratio of retail building square footage to parcel Two studies examined the association between the pres- square footage). Timperio et al. [24] found that parents' ence of controlled crossings (e.g., presence of lights, cross- reports of a lack of public transportation were associated ings, or crosswalks) and children's physical activity, both with lower rates of walking and cycling among girls but of which identified significant positive effects. Timperio et not boys. Mota et al. [32] found that the ability to walk to al. [24] found that parents' reports of a lack of traffic lights destinations such as shops and transit stops was associ- and controlled crossings were associated with lower rates ated with higher physical activity among adolescents and of walking and cycling among boys, but not girls. In their Norman et. al.[37] found that a greater retail floor area evaluation of a Safe Routes to School program, Boarnet, et ratio (reflecting greater retail space and access to shops) al. [39] found that children who passed areas in which was associated with higher objectively measured moder- traffic control methods were installed were more likely to ate to vigorous physical activity among adolescent boys walk or cycle to school than children who did not pass but not girls. In contrast to expectations, Carver et al. [18] such areas. found that adolescent girls' reports of greater access to convenience stores reported lower, rather than higher, Conflicting results were found for studies assessing street rates of walking for transport. The general consistency of connectivity with only two out of four studies identifying results for access to destinations, despite differences in its a significant effect in the anticipated direction. Braza et al. operationalization, suggests that it should be considered [30] found that an objective measure of street connectivity further in future investigations. was associated with higher rates of walking or biking to school. Similarly, Norman et al.[37] found that higher Road hazards intersection density (also assessed using an objective A variety of road hazards have been examined across stud- measure) was associated with higher objectively measured ies including the number of roads to cross, the presence of moderate-to-vigorous physical activity among girls but a road barrier, traffic speed and density, pedestrian and not boys. Mota et al. [32], however, found no associations cyclist safety, and terrain. All three studies assessing road between perceived street connectivity and adolescents' hazards found a negative association between such haz- Page 11 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 ards and children's physical activity. Timperio et al. [24] their children's participation in recreational physical found that parents' reports that their children had to cross activity. Similarly, Gomez et al. [21] noted that adoles- many roads to get to a play area (girls and boys) and of cents' reports of perceived neighborhood safety were asso- high levels of traffic density in their local area (boys only) ciated with higher self-reported outdoor physical activity were associated with lower rates of walking and cycling for girls but not boys. The general lack of findings for per- among children. In a second study by Timperio et. al. ceived safety may reflect the fact that most of the studies [36], using the same sample but using an objective assess- measured general levels of physical activity, which may or ment of the environment, the presence of a busy road bar- may not be linked with neighborhood safety given that rier (e.g., a highway) en route to school (5–6 years olds children can be active outside their neighborhood. and 10–12 year olds) and the presence of a steep incline (5–6 year olds only) were associated with lower rates of In contrast to perceived safety, three out of three studies active commuting to school. Similarly, Carver et al. [18] identified a significant negative association between crime found that parents' reports of traffic impeding the ability or area deprivation and children's physical activity. Gor- to walk were associated with lower rates of walking or don-Larsen et al. [43] and Gomez et al. [21] (girls only) cycling among girls and boys, whereas, parents' percep- found significant inverse associations between objectively tion of the roads in the area being safe was associated with measured crime rates and adolescents' self-reported phys- a higher frequency of walking among girls (but not boys). ical activity. Similarly, Brodersen et al. [17] found that It is worth noting that all of these studies were conducted area deprivation (i.e., rates of car ownership, housing ten- with urban Australian samples. ure, unemployment and overcrowding in the district) was associated with lower self-reported physical activity Local conditions among 11–12 year old girls but not boys. Finally, Carver Both recreational and transport infrastructures exist et al. [18] found that the presence of roaming dogs were within the context of local community conditions. The associated with lower rates of walking or cycling among actions of other community members and agencies such adolescents. as police patrols, community clean-up programs, and/or transient populations, all exert influence at the local level. Social and physical disorder and neighborhood aesthetics Three studies assessed links between neighborhood disor- These conditions include both positive and negative envi- ronmental attributes such as general neighborhood safety, der and children's physical activity. Findings were mixed safety of play areas, crime rates, social disorder and across these studies, likely reflecting differences in the stranger danger, physical disorder and weather condi- operationalization of disorder. Molnar [44] objectively tions. Eighteen studies were identified that assessed links measured physical (e.g., graffiti, empty beer bottles) and between local conditions and children's physical activity social (e.g., alcohol in public, people selling drugs) disor- [16-18,21,24,27,28,30,32,40-48]. All but one study used der using coded video recordings and direct observation a cross sectional design. Four studies used an objective of neighborhoods. Both forms of disorder were associated measure of physical activity (accelerometry), one used with lower levels of parent-reported recreational activity direct observation, and fifteen studies relied on a self- among adolescents. Jago et al. [40], however, found no report measure of physical activity. With regard to meas- association between an objective measure of neighbor- ures of the environment, nine studies used a self-report hood tidiness and children's objectively measured physi- measure, seven studies used an objective measure and two cal activity. Likewise, Timperio et al. [24] found no studies used both methods. The vast majority of studies association between children's perceptions of stranger (13 out of 18) were conducted in the US. danger (a source of social disorder) and parents' reports of their walking and cycling to destinations. Thus, it appears Safety and neighborhood disorder that any association between neighborhood disorder and Safety, crime, and area deprivation physical activity may be limited to much higher levels of Nine studies examined the association between perceived disorder (or deviance) such as those measured by Molnar safety and children's physical activity. These studies over- et al. A general lack of tidiness or the perception that stran- whelming reported a null effect with seven gers can be dangerous but may not be enough to dissuade [16,27,28,32,45,48] of the nine studies showing no asso- youth from being active outdoors. In the only study that ciation between perceived safety and children's physical assessed perceived aesthetics, Mota et al. [32] found that activity. The lack of an association was not limited to a adolescents' reports of the aesthetics of their neighbor- particular research design or sample population. Two hoods (i.e., there are many interesting things to look at exceptions to the pattern of null findings are the studies while walking) were positively associated with their self- by Molnar, et al. [44] and Gomez et al. [21]. In Molnar et reported physical activity. al. residents' reports of the safety of children's local play areas were positively associated with parents' reports of Page 12 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Region and weather Felton et al. [42] found mixed results for location. White Weather girls in urban areas were more active (based on self- A significant association between weather and children's reports) than White girls in rural areas. The opposite was physical activity was identified in two out of five studies. found for Black girls; black girls living in rural areas were Baranowski et al. [41] and Brodersen et al. [17] found that more active than Black girls from urban areas. Although preschool children and 11–12-year-old boys respectively the difference was not discussed by the authors, it is pos- were less active during hotter months of the year. Brod- sible that White girls from urban areas lived in neighbor- ersen et al. [17] also found that higher rainfall was associ- hoods in which they could take advantage of the ated with lower self-reported physical activity among girls infrastructure for physical activity generally attributed to but not boys. Although "unsuitable" weather was urban areas such as the presence of sidewalks and accessi- reported by adolescents as a perceived barrier to physical ble parks. While Black girls may also have had access to activity in the study by Tappe et al. [46], such perceptions similar resources, their ability to use such resources may were not associated with lower levels of self-reported have been limited by neighborhood characteristics such as physical activity. Similarly, Gordon-Larsen et al. [43] crime. found no relationship between the month of the year and adolescents' self-reported physical activity, indicating a Two studies assessed links between population density lack of a seasonality effect. Finally, Sirard et al. [47] found and children's active commuting to school; no consistent no association between weather conditions and rates of effects were identified. Ewing et al. [33] and found no walking and biking to school. association between population density in the immediate area around children's homes and their rates of walking/ The effects of weather may have been underestimated in cycling to school. In contrast, Braza et al. [30] found that these studies due to the restricted time range in which the higher population density was associated with higher data were collected. For example, Gordon-Larsen et al. rates of active commuting to school. Neither study consid- [43] used data collected on physical activity between April ered whether children attended their local school, rather and December. It is possible that the effect of bad or than a magnet or private school outside of the local area, unsuitable weather was eliminated by the exclusion of the or the feasibility of children walking or riding to school. months of January through March (winter months in the northern hemisphere where the research was conducted). Discussion There was also limited variability in the geographic region In this paper we reviewed research on associations within each study. No studies collected data across multi- between the physical environment and children's physical ple regions that varied in the suitability of the climate for activity while highlighting the parties responsible for each outdoor activity. Consequently, inconsistent or non-sig- environmental attribute. This was achieved by classifying nificant effects could be explained by a general lack of var- and reviewing studies specific to recreational infrastruc- iability in the data by month of the year and/or location. ture, transport infrastructure and local conditions. The Furthermore, no studies considered the availability of most consistent pattern of findings was evident for trans- resources for indoor recreational activity in communities. port infrastructure, followed by recreational infrastruc- It is likely the unsuitable weather conditions will most ture, with the least consistent pattern of results noted for often be associated with low levels of physical activity in local conditions. Although there were no consistent differ- communities in which there are few opportunities for ences in results across age or ethnic groups, there was indoor physical activity. some indication that associations between environmental characteristics were more commonly noted for girls than Region, urban/rural location, and population density boys. Three studies examined associations between region and children's physical activity, with one of the three studies Summary of findings showing a significant effect. Gordon-Larsen et al. [43] Results from studies examining components of transport found that residence in the Northeast of the United States infrastructure showed that children were more active was associated with higher self-reported physical activity when there were sidewalks in their neighborhood, they among adolescents in comparison to residence in the had destinations to walk to, public transportation was South, West or Midwest. This effect of region could be available, there were fewer uncontrolled intersections to explained by a myriad of factors such as regional differ- cross, and traffic density was low. Results were more con- ences in weather, income, education, ethnic/racial make- sistent for the absence of roads hazards (i.e., roads to up, and access to community resources. When examining cross, traffic density/speed) than the provision of ameni- rural/suburban versus urban location, Sirard et al. [47] ties (i.e., sidewalks, presence of destinations, controlled found no differences in rates of walking and biking to intersections). In addition, findings were most consistent school for schools located in urban and suburban areas. for parents' reports of infrastructure followed by objective Page 13 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 measures; in general, null findings, or findings in the to pose. Greater rigor with regard to measurement of both opposite direction to those anticipated, were evident for physical activity and the environment, and the use of studies relying on children's reports of transport infra- more sophisticated designs will facilitate the establish- structure. No consistent differences by gender or ethnic ment of a transdisciplinary approach, which is imperative group emerged for transport infrastructure. to moving this body of research into the "next phase". Measurement issues Although findings were less consistent for recreational infrastructure, there were a number of instances in which The ability to measure characteristics of the physical envi- the majority of studies supported a particular relation- ronment is greatly facilitated by the use of geographic ship. The majority of studies showed that the availability information systems (GIS). Of the 33 studies reviewed, 6 of facilities in neighborhoods and the availability of used GIS-based methodologies. As the use of GIS becomes equipment and permanent activity structures in school more commonplace, it is imperative that the methods for play areas were associated with higher physical activity. In "creating" and displaying the data are recorded in detail addition, greater distances to school were associated with (the "meta data" – the data about the data). Currently, lower rates of walking and cycling to school. In contrast to there is little description of the various processing deci- expectations, most studies failed to identify an association sions that are made when using GIS in published research. between home equipment and children's physical activity The absence of such information slows research progress and results for the proximity to playgrounds were mixed. and inhibits the comparison of findings across studies Some gender differences in the reported associations were and research disciplines. While we advocate for the incor- apparent. Six out of seven effects specific to girls were sig- poration of GIS into research designs, the perceived envi- nificant and in the anticipated direction. Most of these ronment should also be taken into consideration because effects were noted for child reports of the environment. In people's perceptions may, in fact, motivate their behavior contrast, only one significant effect was specific to boys. In more than the true nature of the situation. two instances, associations specific to African Americans were reported. In both cases, no significant effects of rec- In contrast to the objective assessment of the environ- reational infrastructure were present for this demographic ment, objective measures of physical activity were more group. widely incorporated into the studies reviewed with 13 out of 33 studies reviewed using an objective measure of phys- Findings were least consistent for local conditions, reflect- ical activity (8 used accelerometers, 4 used direct observa- ing the broader range of characteristics assessed. In gen- tion, 1 used heat rate monitoring). Using accelerometers eral, no effects were found for perceived neighborhood to measure children's physical activity and/or directly safety or the perceived safety of play areas. However, both observed children's activity removes the possibility of studies that used objective measures of crime rates response bias, particularly among children [49]. Although reported a significant negative association between crime the use of objective measures of physical activity is prefer- and children's physical activity. Similarly, objectively able, because it allows greater confidence in the validity of measured area deprivation and the perceived presence of the assessment, objective measures may not be feasible in roaming dogs were associated with lower physical activity. large-scale survey research due to financial and logistical No consistent pattern of findings was evident for region or constraints. In addition, accelerometers provide only a weather conditions. With respect to differences noted by generalized measure of physical activity and do not pro- gender, three effects specific to girls were significant and in vide information on the type of activity or the location in the anticipated direction; only one effect was specific to which physical activity takes place. As is noted by Giles- boys and this was also in the anticipated direction. Of the Corti et al. [50], assessing context-specific behaviors is key three effects specific to African Americans, two were not to understanding associations between the physical envi- significant and one was in the opposite direction to that ronment and physical activity. In many of the studies expected. reviewed, generalized measures of physical activity were implemented that may not be sensitive to specific envi- Recommendations for future research ronmental attributes. Consequently, null effects may The research reviewed herein generally reflects the first reflect a lack of specificity in the measures used rather than "phase" of research on links between the physical envi- the absence of an association. New equipment that incor- ronment and children's physical activity. In this first porates Global Positioning Systems (GPS) into acceler- round of research, many of the methodologies were in a ometers may facilitate the ability to obtain context- developmental stage. As a result, there was little consist- specific measures of physical activity by making it possible ency in the methods used. In addition, in many cases, the to know exactly where (i.e., longitude/latitude data) and methods used and the designs adopted were opportunis- when (i.e., electronically time-stamped data) the physical tic as researchers grappled with which research questions activity occurred. Page 14 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Design issues that draws upon diverse research disciplines such as geog- With two exceptions, all studies relied on cross sectional raphy and planning, public health, exercise science, crim- analysis. Given the need to understand behavioral inal justice and human development. Such an approach changes associated with environmental attributes, longi- will require greater standardization of procedures and tudinal studies are more appropriate. Such studies will detailed reporting of these procedures than has generally help us determine whether the pattern of results reflects been exhibited in research to date. In addition, a transdis- the ability of the environment to constrain or facilitate ciplinary approach will require clear communication and certain behaviors or reflects the type of person/family who dialogue across research disciplines, including access to chooses to live in certain neighborhoods [9,51]. In addi- literatures across various research communities. In many tion to using a longitudinal design, more complex models cases, information on environmental attributes is con- need to be developed and tested. With the exception of tained in planning documents rather than in refereed gender, research has rarely examined factors that may journals. These documents are often available over the moderate the link between the environment and chil- Internet, but may not be included in databases or other dren's physical activity (i.e., interact with the environment researchable tools. to predict physical activity). The use of simplistic designs with little consideration of moderating factors as high- Summary and conclusion lighted by McMillan [15], may lead to simplistic and erro- In this review, we have found preliminary evidence that a neous conclusions. The most noteworthy example is the relationship exists between children's participation in general failure to consider ethnicity, family income, or physical activity and environmental attributes. Limita- neighborhood deprivation as possible moderating or con- tions of this review include the exclusion of studies not founding variables. Furthermore, many studies have published in English or searchable in English-based data- assessed children across a broad age range, which ignores bases, the overall bias against publishing studies with null the possibility that associations between the physical results, and the lack of research specific to children out- environment and physical activity may be age-specific due side the health sciences. Future work could enhance our to differences in parental control and children's inde- understanding of this important topic by assessing both pendent mobility. perceived and objective characteristics of the environ- ment, including objective measures of children's physical In addition to a lack of emphasis on children's age, the activity and the physical environment, adopting longitu- role that parents play in regulating children's use of the dinal designs, assessing the interaction between various physical environment has not been considered in research environmental attributes, and examining the important to date. Consequently, the assumption is generally that role that parents play as gate keepers to children's use of there is a direct link between the environment and chil- the physical environment. There is also a need for studies dren's physical activity. This is unlikely to be the case outside the US to determine whether results identified given children's lack of decision autonomy and the role using US samples can be generalized internationally. that parents play as gate keepers to children's use and Finally, we advocate the continued use of the classifica- exploration of the physical environment surrounding tion scheme outlined herein as this will allow us to deter- their home. Research shows that parents' decisions about mine the parties responsible for attributes found to their children's independent mobility are influenced by a influence children's physical activity and to make the nec- number of factors such as their perceptions of the safety of essary changes. We also strongly recommend the estab- the area, neighborhood relations, and proximity to a park lishment of a transdisciplinary research agenda [52]. Research designs and techniques will need to link sufficiently transparent to facilitate the sharing of infor- quantitative and qualitative data to successfully under- mation across a growing body of work being generated by stand the nature of parents' decision-making processes diverse research approaches. and their willingness to allow their children to participate in physical activity under a combination of environmen- Acknowledgements Drs. Kirsten Krahnstoever Davison and Catherine T. Lawson are grateful tal attributes. for the support of the Center for the Elimination of Minorities Health Dis- parities and the Center for Social and Demographic Analysis at the Univer- The need for a transdisciplinary approach sity at Albany. We want to extend special thanks to our research assistant Scientists from different research paradigms have largely Katie Haverly for her work on the literature search. approached research of children and physical activity from the perspective of their own discipline with little References integration of ideas and methods across disciplines. To 1. Centers for Disease Control and Prevention: Youth Risk Surveil- lance Survey - United States, 1997. Morbidity and Mortality most effectively assess the impact of the physical environ- Weekly Report 1998, 47(SS-3):1-89. ment on physical activity levels among youth, future 2. US Department of Health and Human Services: Healthy People research will need to adopt a transdisciplinary approach 2010. Washington, DC , US Government Printing Office; 2000. Page 15 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 3. Trost SG, Kerr LM, Ward DS, Pate RR: Physical activity and rural fifth-grade children. Preventive Medicine 1997, determinants of physical activity in obese and non-obese 26(2):257-263. children. International Journal of Obesity and Related Metabolic Disor- 26. Trost SG, Pate RR, Ward DS, Saunders R, Riner W: Determinants ders 2001, 25:822-829. of physical activity in active and low-active, sixth grade Afri- 4. Bailey DA, Martin AD: Physical activity and skeletal health in can-American youth. Journal of School Health 1999, 69(1):29-34. adolescents. Pediatric Exercise Science 1994, 6:330-347. 27. Zakarian JM, Hovell MF, Hofstetter CR, Sallis JF, Keating KJ: Corre- 5. Morrow JR, Freedson PS: Relationship between habitual physi- lates of vigorous exercise in a predominantly low SES and cal activity and aerobic fitness in adolescents. Pediatric Exercise minority high school population. Preventive Medicine 1994, Science 1994, 6:315-329. 23:314-321. 6. Calfras KJ, Taylor WC: Effects of physical activity on psycholog- 28. Sallis JF, Taylor WC, Dowda M, Freedson PS, Pate RR: Correlates ical variables in adolescents. Pediatric Exercise Science 1994, of vigorous physical activity for children in grades 1 through 6:406-423. 12: Comparing parent-reported and objectively measured 7. Humpel N, Owen N, Leslie E: Environmental factors associated physical activity. Pediatric Exercise Science 2002, 14:30-44. with adults' participation in physical activity: a review. Amer- 29. Stratton G, Mullan E: The effect of multicolor playground mark- ican Journal of Preventive Medicine 2002, 22(3):188-199. ings on children's physical activity level during recess. Preven- 8. Duncan MJ, Spence JC, Mummery WK: Perceived environment tive Medicine 2005, IN PRESS:. and physical activity: a meta-analysis of selected environ- 30. Braza M, Shoemaker W, Seeley A: Neighborhood design and mental characteristics. International Journal of Behavioral Nutrition rates of walking and biking to elementary school in 34 Cali- and Physical Activity 2005, 2(11):. fornia communities. American Journal of Health Promotion 2004, 9. Owen N, Humpel N, Leslie E, Bauman A, Sallis JF: Understanding 19(2):128-136. environmental influences on walking; Review and research 31. Stucky-Ropp R, DiLorenzo TM: Determinants of exercise in chil- agenda. American Journal of Preventive Medicine 2004, 27(1):67-76. dren. Preventive Medicine 1993, 22:880-889. 10. Sallis JF, Bauman A, Pratt M: Environmental and policy interven- 32. Mota J, Almeida M, Santos P, Ribeiro JC: Perceived neighborhood tions to promote physical activity. American Journal of Preventive environments and physical activity in adolescents. Preventive Medicine 1998, 15(4):379-397. Medicine 2005, IN PRESS:. 11. Saelens BE, Sallis JF, Frank LD: Environmental correlates of walk- 33. Ewing R, Schroeer W, Greene W: School location and student ing and cycling: findings from the transportation, urban travel. Transportation Research Record 2004, 1895:55-63. design, and planning literatures. Annals of Behavioral Medicine 34. Hume C, Salmon J, Ball K: Children's perceptions of their home 2003, 25(2):80-91. and neighborhood environments, and their association with 12. Lee C, Moudon AV: Physical activity and environment research objectively measured physical activity: a qualitative and in the health field: Implications for urban and transportation quantiative study. Health Education Research 2005, 20(1):1-13. planning, practice and research. Journal of Planning Literature 35. Zask A, van Beurden E, Barnett L, Brooks LO, Dietrich SEI: Active 2004, 19(2):147-181. school playgrounds - Myth or reality? Results of the "Move It 13. Krizek KJ, Birnbaum AS, Levinson DM: A schematic for focusing Groove It" project. Preventive Medicine 2001, 33(401-408):. on youth in investigations of community design and physical 36. Timperio A, Ball K, Salmon J, Roberts R, Giles-Corti B, Simmons D, activity. American Journal of Health Promotion 2004, 19(1):33-38. Baur LA, Crawford D: Personal, familial, social and environ- 14. Sallis J, Prochaska JJ, Taylor WC: A review of correlates of physi- mental correlates of active commuting to school. American cal activity of children and adolescents. Medicine and Science in Journal of Preventive Medicine 2006, 30(1):45-51. Sports and Exercise 2000, 32(2):963-975. 37. Norman GJ, Nutter SK, Ryan S, Sallis JF, Calfras KJ, Patrick K: Com- 15. McMillan TE: Urban form and a child's trip to school: The cur- munity design and access to recreational facilities as corre- rent literature and a framework for future research. Journal lates of adolescent physical activity and Body-Mass Index. of Planning Literature 2005, 19(4):440-456. Journal of Physical Activity and Health 2006, 3 (Suppl 1):S118-S128. 16. Adkins S, Sherwood NE, Story M, Davis M: Physical activity 38. Cohen DA, Ashwood S, Scott M, Overton A, Evenson KR, Voorhees among African-American Girls: The role of parents and the CC, Bedimo-Rung A, McKenzie TL: Proximity to school and home environment. Obesity Research 2004, 12 (supple- physical activity among middle school girls: The Trial of ment):38S - 45S. Activity for Adolescent Girls Study. Journal of Physical Activity and 17. Brodersen NH, Steptoe A, Williamson S, Wardle J: Sociodemo- Health 2006, 3(Suppl 1):S129-S138. graphic, developmental, environmental, and psychological 39. Boarnet MG, Anderson CL, Day K, McMillan T, Alfonzo M: Evalua- correlates of physical activity and sedentary behavior at age tion of the California Safe Routes to School legislation: urban 11 to 12. Annals of Behavioral Medicine 2005, 29(1):2-11. form changes and children's active transportation to school. 18. Carver A, Salmon J, Campbell K, Baur L, Garnett SCD: How do per- American Journal of Preventive Medicine 2005, 28(2 Suppl 2):134-140. ceptions of local neighborhood relate to adolescents' walk- 40. Jago R, Baranowski T, Zakeri I, Harris M: Observed environmen- ing and cycling? American Journal of Health Promotion 2005, tal features and the physical activity of adolescent males. 20(2):139-147. American Journal of Preventive Medicine 2005, 29(2):98-104. 19. Dunton GF, Jamner MS, Cooper DM: Assessing the perceived 41. Baranowski T, Thompson WO, DuRant RH, Baranowski J, Puhl J: environment among minimally active adolescent girls: valid- Observations on physical activity in physical locations: age, ity and relations to physical activity. American Journal of Health gender, ethnicity, and month effects. Research Quarterly for Exer- Promotion 2003, 18(1):70-73. cise and Sport 1993, 64(2):127-133. 20. Fein AJ, Plotnikoff RC, Wild C, Spence JC: Perceived environment 42. Felton GM, Dowda M, Ward DS, Dishman RK, Trost SG, Saunders R, and physical activity in youth. International Journal of Behavioral Pate RR: Differences in physical activity between black and Medicine 2004, 11(3):135-142. white girls living in rural and urban areas. Journal of School 21. Gomez JE, Johnson BA, Selva M, Sallis JF: Violent crime and out- Health 2002, 72(6):250-255. door physical activity among inner-city youth. Preventive Med- 43. Gordon-Larsen P, McMurray RG, Popkin BM: Determinants of icine 2004, 39(5):876-881. adolescent physical activity and inactivity patterns. Pediatrics 22. Sallis JF, Nader PR, Broyles SL, Berry CC, Elder JP, McKenzie TL, Nel- 2000, 105(6):E83. son JA: Correlates of physical activity at home in Mexican- 44. Molnar BE, Gortmaker SL, Bull FC, Buka SL: Unsafe to play? American and Anglo-American preschool children. Health Neighborhood disorder and lack of safety predict reduced Psychology 1993, 12(5):390-398. physical activity among urban children and adolescents. 23. Sallis JF, Conway TL, Prochaska JJ, McKenzie TL, Marshall MS, Brown American Journal of Health Promotion 2004, 18(5):378-386. M: The association of school environments with youth physi- 45. Sallis JF, Alcaraz JE, McKenzie TL, Hovell MF: Predictors of change cal activity. American Journal of Public Health 2001, 91(4):618-620. in children's physical activity over 20 months. Variations by 24. Timperio A, Crawford D, Telford A, Salmon J: Perceptions about gender and level of adiposity. American Journal of Preventive Med- the local neighborhood and walking and cycling among chil- icine 1999, 16(3):222-229. dren. Preventive Medicine 2004, 38(1):39-47. 46. Tappe MK, Duda JL, Ehrnwald PM: Perceived barriers to exercise 25. Trost SG, Pate RR, Saunders R, Ward DS, Dowda M, Felton G: A among adolescents. Journal of School Health 1989, 59(4):153-155. prospective study of the determinants of physical activity in Page 16 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 47. Sirard JR, Ainsworth BE, McIver KL, Pate RP: Prevalence of active commuting at urban and suburban elementary schools in Columbia, SC. American Journal of Public Health 2005, 95:236-237. 48. Burdette HL, Whitaker RC: A national study of neighborhood safety, outdoor play, television viewing and obesity in pre- school children. Pediatrics 2005, 116(3):657-662. 49. Sallis JF, Owen N: Physical activity and behavioral medicine. In Behavioral Medicine and Health Psychology 3 Edited by: Turner JR. Lon- don , Sage Publications; 1999. 50. Giles-Corti B, Timperio A, Bull F, Pikora T: Understanding Physi- cal Activity Environmental Correlates: Increased Specificity for Ecological Models. Exercise and Sport Science Reviews 2005, 33(4):175 -1181. 51. Transportation Research Board: Does the built environment influence physical activity? Examining the evidence. Wash- ington, DC , Institute of Medicine; 2005. 52. Prezza M, Pilloni S, Morabito C, Sersante C, Alparone FR, Giuliani MV: The influence of psychosocial and environmental factors on children's independent mobility and relationship to peer fre- quentation. Journal of Community and Applied Social Psychology 2001, 11:435-450. Publish with Bio Med Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 17 of 17 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Behavioral Nutrition and Physical Activity Springer Journals

Do attributes in the physical environment influence children's physical activity? A review of the literature

Loading next page...
 
/lp/springer-journals/do-attributes-in-the-physical-environment-influence-children-s-spqMkRDAk6

References (63)

Publisher
Springer Journals
Copyright
Copyright © 2006 by Davison and Lawson; licensee BioMed Central Ltd.
Subject
Chemistry; Nutrition; Behavioral Sciences; Health Promotion and Disease Prevention
eISSN
1479-5868
DOI
10.1186/1479-5868-3-19
pmid
16872543
Publisher site
See Article on Publisher Site

Abstract

Background: Many youth today are physically inactive. Recent attention linking the physical or built environment to physical activity in adults suggests an investigation into the relationship between the built environment and physical activity in children could guide appropriate intervention strategies. Method: Thirty three quantitative studies that assessed associations between the physical environment (perceived or objectively measured) and physical activity among children (ages 3 to 18-years) and fulfilled selection criteria were reviewed. Findings were categorized and discussed according to three dimensions of the physical environment including recreational infrastructure, transport infrastructure, and local conditions. Results: Results across the various studies showed that children's participation in physical activity is positively associated with publicly provided recreational infrastructure (access to recreational facilities and schools) and transport infrastructure (presence of sidewalks and controlled intersections, access to destinations and public transportation). At the same time, transport infrastructure (number of roads to cross and traffic density/speed) and local conditions (crime, area deprivation) are negatively associated with children's participation in physical activity. Conclusion: Results highlight links between the physical environment and children's physical activity. Additional research using a transdisciplinary approach and assessing moderating and mediating variables is necessary to appropriately inform policy efforts. low bone density [4], and low physical fitness [5]. Further- Background Many youth today are physically inactive. Considerable more, children who are not physically active are denied evidence documents that nearly 35% of youth in the US the positive social and emotional benefits of physical fail to meet the minimum physical activity guidelines, and activity including higher self esteem, lower anxiety, and another 14% are completely inactive [1,2]. Low levels of lower stress [6]. A comprehensive understanding of the physical activity and the failure to meet physical activity determinants of physical activity among youth is essential recommendations have notable health consequences for the identification of appropriate points of intervention among children including increased risk of obesity [3], to promote active lifestyles and their associated health Page 1 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 benefits. In this paper, we examine environmental influ- organizing schema that identifies the parties responsible ences on children's physical activity. Specifically, we for specific elements in the built environment. review research assessing the association between attributes of the physical environment and children and Methods adolescents' physical activity. Definition of the physical environment The physical environment is defined herein as objective The physical or built environment has come to the fore- and perceived characteristics of the physical context in front of public health research in the past 5 years, leading which children spend their time (e.g., home, neighbor- to a surge of research on environmental attributes and hood, school) including aspects of urban design (e.g., their associations with physical activity behaviors. A presence and structure of sidewalks), traffic density and number of reviews have examined links between the phys- speed, distance to and design of venues for physical activ- ical environment and adults' physical activity [7-12]. ity (e.g., playgrounds, parks and school yards), crime, Much less emphasis has been placed on research specific safety and weather conditions. While crime and safety are to children. One cannot assume that associations between not explicitly characteristics of the physical environment, the physical environment and physical activity among they are included in this review as both are intimately adults are applicable to children. As highlighted by Krizek, linked with multiple characteristics of the physical envi- Birnbaum & Levinson [13], children in contrast to adults, ronment including for example lighting, the condition of spend large parts of their day at school, have considerable buildings, and the presence of trash. They also have vicin- time for recreation, are more likely to accumulate physical ity effects such that a particular area can gain a reputation activity through play, are not able to drive, and are subject for safety or criminal activity. to restrictions placed on them by adults. Identification of studies Two reviews to date are specific to children. In 2000, Sallis Computer searches using PubMed, PsychInfo, EBSCO, et al. [14] published a comprehensive review of predictors CINAHL, and TRANSPORT were conducted in the Eng- of physical activity among youth. Studies published lish-language literature to identify published studies and between 1970 and 1998 were included in the review. reports examining relationships between the physical While this review does not focus on the physical environ- environment and children and adolescents' physical activ- ment, a small proportion of the 108 studies reviewed are ity. Transportation and urban planning reports were specific to the physical environment. More recently, in accessed using TRANSPORT and general internet searches 2005, McMillan [15] reviewed studies in both planning and by searching the bibliographies of papers. Search and public health literatures on urban form and children's terms included physical activity, exercise, recreation, trip to school. McMillan outlines policies and programs sport, walk/walking, cycle/cycling, transport, active com- that may promote walking and cycling to school (e.g., Safe muting, environment, environmental determinants, Routes to School) and highlights the lack of focus on chil- physical environment, built environment, perceived envi- dren in the transportation literature. In the absence of ronment, design, urban design, context, facilities, neigh- research on environmental factors that affect children's borhood, park, playground, situational factors, safety, trips to school, most of the studies reviewed by McMillan crime and weather. These search terms are a compilation focus on adult populations. of the terms used in previous reviews [7-9,11,12]. In addi- tion, the search terms children, child, adolescent, adoles- In this descriptive review, we build on the work of Sallis et cence, youth, family, and parent were added to the terms al. and McMillan by reviewing recent studies (published list to limit studies to children and adolescents. The bibli- between 1990 and 2006) that examine the association ographies of the identified studies were also reviewed for between children's physical activity and environmental additional references. attributes (perceived and objectively measured). In partic- ular, we provide specific information on the sample char- Studies were not further considered if they (a) did not acteristics and design of each study, evaluate consistencies measure or model the environment (perceived or objec- and inconsistencies in the literature, and identify gaps in tive) and physical activity behaviors, (b) were descriptive the current research and possible avenues for future in nature (e.g. qualitative studies were not reviewed), (c) research. In addition, we broaden the set of children's only used a composite score of the environment that com- behaviors from their trip to school as outlined by McMil- bined a number of environmental attributes and (d) did lan to physical activity in general. In order to serve both not report findings for children separately from those for the need for understanding the link the physical environ- adults. The database searches resulted in a total of 106 ment and physical activity among youth and the imple- "hits", of which 29 were relevant and were further consid- mentation of next steps based on these findings, we use an ered. Of the 29 studies, 23 met the selection criteria. Addi- tional studies were identified through searching the Page 2 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 bibliographies of articles. A total of 33 articles were iden- Definitions and examples of each domain are provided, tified for inclusion in the review. the responsible parties for each are identified, the key findings for each domain are summarized, and consisten- Recording and synthesizing research findings cies and inconsistencies are highlighted and possible For each study, the following were recorded: (a) first explanations are provided. author and year; (b) sample characteristics such as sample size, gender, and ethnic/racial group of the participants; Results (c) age of participants; (d) research design, including Recreational infrastructure whether it was cross sectional or longitudinal and whether Recreational infrastructure (play areas) for children can be perceived and/or objective environmental attributes were classified as private (provided by their parents in and assessed; (e) the environmental variables that were exam- around the home), public (community areas or schools) ined; (f) the type of physical activity behavior assessed or private-public (commercial play areas). Private recrea- and the method of assessment; and (g) a summary of the tional infrastructure may be subject to regulations associ- significant associations that were identified (see Table 1 ated with the property and issues of flexibility of use based for a summary). This information was recorded by both on tenure of ownership (e.g., renter or home-owner). authors and a graduate student and was cross checked to Public recreational infrastructure is primarily the respon- identify any inconsistencies. In instances in which multi- sibility of the municipality or agency charged with the ple aspects of the physical environment were assessed, provision of the original infrastructure, as is the mainte- results for each component were recorded. If multiple nance and continued monitoring of the safety and condi- measures of physical activity were included in the study, tion of such assets. School yards, playgrounds and open only results for the most comprehensive measure were space parks are most often considered public recreational recorded. For example, if a study included both objec- infrastructure. Private-public recreational infrastructure tively measured physical activity and a generalized self- includes youth camps, commercial clubs, and other busi- report measure, results for the objective measure are nesses providing places for children to participate in phys- reported. This simplification was necessary given the ical activity. All of these recreational infrastructures are breadth of measures of physical activity used across stud- subject to land use regulations, including zoning codes. ies. Simplifying the presentation of results for physical Twenty one studies that were reviewed examined the rela- activity also served to maintain the focus of this review on tionship between recreational infrastructure and chil- the environmental correlates of physical activity broadly dren's physical activity [16-36]. The overwhelming construed. majority (i.e., 19) of studies used a cross sectional design. One study used a 1-year longitudinal design and one used Findings from the studies are reviewed and synthesized an intervention design. Five of the 21 studies used an using three a priori categories of environmental attributes objective measure of physical activity, including either including: (1) recreational infrastructure (e.g., the availa- accelerometry or heart rate monitoring, four used direct bility of parks/playgrounds, equipment in the home); (2) observation, 13 used a self-report measure, and one study transport infrastructure (e.g., traffic speed/density, pres- used both objective and self-report measures. Seven stud- ence of sidewalks); and (3) local conditions (e.g., safety, ies used an objective measure of the environment (gener- crime, weather). These categories were chosen to facilitate ally based on Geographic Information Systems), 12 used the identification of the parties responsible for changing a self-report measure, and two studies used both method- an environmental attribute and consequently possible ologies. Finally, 12 of the studies were conducted in the avenues for intervention. Using a system similar to previ- US with the remaining studies being conducted in coun- ous reviews on physical activity [7,14] it was noted tries including Canada, England, Australia and Portugal. whether a finding was positive and significant (+), nega- tive and significant (-) or not statistically significant (0). Private recreational infrastructure This information is summarized in Table 2. In order to Home equipment facilitate the comparison of findings across studies, results Four out of six studies found no association between from bivariate (in contrast to multivariate) analyses are home equipment and children's physical activity. Specifi- recorded and, where possible, results from bivariate anal- cally, Sallis et al. [22] found no association between an yses controlling for basic demographic variables (i.e., SES) objective assessment of equipment available in the home are presented. Results are recorded separately for per- and observed levels of physical activity among preschool ceived and objectively measured attributes of the environ- children. Dunton et al.[19] and Trost et al. [25] found no ment. Results specific to the perceived environment were association between adolescents' reports of equipment in further separated according to children's and adults' (usu- the home and their self-reported physical activity. A sec- ally parents) reports of the environment. The narrative ond study by Trost et al.[26] found no association review below accompanies the data presented in Table 2. between adolescents' reports of home equipment and Page 3 of 17 (page number not for citation purposes) Table 1: Characteristics and main findings of the studies reviewed First Author Number/Gender/ Age group Design Environmental attributes Physical activity behavior Significant associations with ref# (year) Ethnicity/Country (independent variables) (outcome variable) outcome variable Adkins (2004) 52 F B USA 8- to 10-years CS, P There are playgrounds, parks and Objectively measured (accelerometer) No associations were identified gyms nearby, it is safe to play outside physical activity between environmental attributes and (parent and child report) physical activity. Baranowski (1993) 191 M/F B/W/H USA 3 and 4 years CS, O Month of the year (weather) Directly observed physical activity. Children were least active outdoors during the hottest months. 39 rd Boarnet (2005) 62 M/F W/B/H/A USA Parents of children in 3 CS, O Installation of sidewalks, crossing Parents' perceived change in child Greater increases in perceived rates th – 5 grade (8–10 years) signals, traffic control as part of a Safe walking/biking to school of children walking/riding to school for Routes to School (SR2S) program. children who passed a completed SR2S zone compared to those who did not pass a zone. 30 th Braza (2004) 105 students from 34 5 grade (ages 9 to 11 CS, O School size; population density; Rates of walking and biking to school Higher population density and a schools M/F W/B/H/A years) number of intersections per street among students surveyed in each greater number of intersections per USA Unit of analysis: schools mile in .5 mile buffer around school school street mile were associated with site. Data were obtained using higher rates of walking and biking to Geographic Information Systems. school in bivariate models. Brodersen (2005) 4320 M/F W/B/A 11 to 12 years CS, O Area deprivation; number of sport Self-reported days during past week Area deprivation (F) and total rainfall England pitches in borough; public spending on child performed hard exercise that (F) were associated with lower leisure facilities and open spaces; made him/her breathe heavily and physical activity. Colder temperatures weather conditions. sweat. (M) and number of sport pitches (F) were associated with higher physical activity. Burdette (2005) 3141 M/F W/B/H USA 3 years old CS, P Mothers' ratings of perceived Mothers' reports of the average time No associations between mothers' neighborhood safety per day their child played outdoors perception of neighborhood safety and their reports of the time their child spent playing outdoors Carver (2005) 347 M/F U Australia 12–13 years CS, P Parents' perceptions of good sports Child self-reported frequency and Adolescents walked or cycled more facilities for child, safe for child to duration of walking or cycling in the frequently when there were fewer walk/ride, good places for child to be neighborhood (for recreation, unattended dogs (M, F), there were active, traffic makes it difficult to walk. transport, exercise, get to school). good places to be active (F), traffic was Child perceptions of ease to get less problematic (M, F), there was around by bike, safety while walking/ lower perceived ease to cycle (M), riding, roads safe, unattended dogs, there were more sport facilities in the strangers, fast food and convenience area (M), the roads were perceived as stores near home. safe(F), and convenience stores were further from home (F). Above is a simplified summary of results given number of variables assessed and analyses performed (i.e., >400 associations assessed). 38 th Cohen (2006) 1554 F W/H/B/A USA 6 grade 12–13 years CS, O Distance to school along the shortest Objectively measured (accelerometer) A shorter distance to school was street network physical activity associated with greater MVPA during weekdays but not during the weekend International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Page 4 of 17 (page number not for citation purposes) Table 1: Characteristics and main findings of the studies reviewed (Continued) Dunton (2003) 87 G W/H/A USA 14–17 years CS, P Perceived activity-related equipment Self-reported vigorous physical No associations were identified in the home and activity-related activity, total energy expenditure, and between activity-related resources in resources in the community (e.g., leisure time activity the home or the community and girls' park, gym, biking trail) self-reported physical activity. 33 th Ewing (2004) 726 people and 709 Students K-12 grade CS, O Estimated walk/bike time between Likelihood of walking or biking to Students with shorter walk or bike school trips surveyed destinations; proportion of street school times to school, and students traveling U (gender) miles with street trees, bike lanes or through areas with sidewalks on main U(ethnicity) USA paved shoulders, or sidewalks; roads were more likely to walk or sidewalk width; accessibility of bike to school. School size was not attractions; neighborhood population related to the likelihood of walking/ density; school size biking to school. Fein (2004) 610 M/F W Canada Grades 9–12 Mean age CS, P Home environment; convenient Self-reported physical activity Perceived convenience of facilities and 15.5 years facilities (park, bike trails, gym, skating perceived home, neighborhood, and rink); School environment (gym space, school environment were significantly availability of exercise equipment, correlated with self-reported physical athletic facilities accessible). The activity. The perceived importance of perceived importance of each each of these constructs was also resource was also assessed. associated with higher physical activity. Felton (2002) 1668 F W/B USA 8th grade (approx age 13 CS, O Urban/rural residence Self-reported moderate and vigorous White girls living in urban areas and years) physical activity black girls in rural areas reported higher vigorous activity than their respective counterparts. Gomez (2004) 177 M, F H USA 7th grade (approx age 12 CS, P, Crime density (O); perceived Self-reported participation in outdoor Greater proximity to play areas (M), years) O neighborhood safety (P); distance to activities (not in school) lower crime density (F), and high nearest play areas (O) perceived safety (F) were associated with higher outdoor activity. th th Gordon-Larsen 17766 M, F W/B/H/A 7 to 12 grade (approx CS, O Urban/rural residence; crime; month Self-reported moderate to vigorous Lower reported crime was associated (2000) USA ages 12 – 17 years) of the year; region (South West, physical activity with higher moderate to vigorous Midwest, Northeast). activity. Hume (2005) 127 M, F U Australia 10 year olds CS, P Children drew maps of their home Objectively measured (accelerometer) Girls who drew a greater number of and neighborhood environments. The physical activity. opportunities for physical activity in frequency with which particular their neighborhood (e.g., the objects and locations were availability gyms, recreation and represented was coded including swimming centers, playgrounds) green space and outdoor areas and exhibited higher physical activity opportunities for physical activity in (specifically, low intensity physical the neighborhood (e.g., playgrounds activity). and facilities). Jago (2005) 210 M W/B/H USA 10–14 years CS, O Ease of walking/cycling; tidiness of Objectively measured (accelerometer) Sidewalk characteristics that foster neighborhood; sidewalk physical activity. walking (e.g., distance to curb, characteristics; street access and presence of trees as a buffer) were conditions positively associated with light- intensity physical activity. Molnar (2004) 1378 M/F W/B/H USA 11 to 16 years CS, P, Residents' perceived neighborhood Hours/week participated in More safe areas for children to play O safety and opportunities for children recreational physical activity (parent and lower social and physical disorder to play (P); social and physical disorder report). were associated with higher (O). recreational activity. International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Page 5 of 17 (page number not for citation purposes) Table 1: Characteristics and main findings of the studies reviewed (Continued) 32 th th Mota (2005) 1123 M/F U Portugal 7 – 12 grade Mean CS, P Adolescent reports of the activity- Self-reported physical activity In comparison to low active age: 14.6 ± 1.6 friendliness of their neighborhood (e.g. adolescents, high active adolescents access to destinations, connectivity of reported greater access to streets, infrastructure for walking and destinations such as stores and transit cycling, neighborhood safety, stops, higher neighborhood aesthetics, aesthetics, and recreational facilities). and more recreational facilities in their neighborhood. Norman (2006) 799 M/F W/H/B/A 11–15 years CS, O Number of private recreational Objectively measured (accelerometer) Significant bivariate associations were USA facilities, schools and parks within 1 physical activity. found between moderate-to-vigorous mile of home; walkability as assessed PA and the number of recreation by residential density, retail floor area, facilities (girls), the number of parks intersection density, and land use mix and measures of walkability including intersection density (girls), and retail floor area ratio (boys). Sallis (1993) 347 M/F W/H USA 4 years old CS, P, Number of specified play spaces (e.g., Directly observed physical activity. A greater number of specified play O friend's backyard, park) within walking spaces within walking distance of distance of home (P); equipment at home was associated with higher home (O). physical activity. 45 th th Sallis (1999) 732 M/F W/A/PI/H 4 – 5 grade (ages 9 to L (20 Neighborhood safety (parent report) Parent and child report of child No links were identified between USA 10 years) at baseline months physical activity. Objectively measured neighborhood safety and baseline ), P (accelerometer) physical activity physical activity or change in activity. Sallis (2001) 151 areas in 24 middle Middle-school-aged CS, O Type of play area (court space, open Directly observed physical activity of Higher levels of activity were noted schools USA students (approx ages 11 field space, indoor activity space); area students in each play area. when equipment was available in to 13 years) size; permanent activity structures outdoor play areas (F) when more (e.g., basketball hoops, tennis courts, permanent activity structures were soccer goals); equipment. available (M), and when such structures were available in combination with adult supervision (F). Sallis (2002) 781 M/F W (75%) Grades 1–12 (ages 6–18) CS, P Safe to play outdoors; access to parks/ Parents' reports of children's physical Among girls in grades 10–12, parents' USA playgrounds; distance to park; safety of activity and objectively measured perception of neighborhood safety nearest park. (accelerometer) physical activity (N = was associated with higher physical sub sample of 200) activity. Among girls in grades 7–9, parents' perception of park safety was negatively associated with children's physical activity. Unit of analysis = Elementary schools CS, O School urbanization and weather Rates of walking and cycling to school No associations between Sirard (2005) school (N = 8) USA conditions. for each school. environmental variables and active commuting were identified. Stratton (2005) 99 M/F U Wales and 4–11 years I, O Intervention in which school Heart rate telemeters were used to In comparison to control schools, England playgrounds were painted with murals, assess heart rate during physical time spent in MVPA and VPA hopscotch, fun trails, snakes and activity and converted to represent increased significantly in intervention ladders, and court markings (e.g., lines MVPA and VPA. schools as a result of playground for basketball). painting. th th Stucky-Ropp 240 M/F W USA 5 and 6 grade Mean CS, P Number of exercise-related items at Self reported physical activity A greater number of exercise-related (1993) age: 11.2 ± .7 home items in the home was associated with higher physical activity among girls but not boys. International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Page 6 of 17 (page number not for citation purposes) Table 1: Characteristics and main findings of the studies reviewed (Continued) Tappe (1989) 236 M/F W, B, A USA High school Mean age: CS, P Unsuitable weather as a barrier to Self-reported physical activity. No differences in weather as a 15 years 9 months exercise perceived barrier for physical activity among low and high active girls and boys. Timperio (2004) 1200 M/F U Australia 5–6 years and 10–12 CS, P Traffic density, road safety, strangers, Walking/riding to particular Among 5–6 year olds, parents' years. sporting facilities, and public destinations (e.g., friend's house, park, perception of heavy traffic (M), and transportation (parent report). school) 3 or more times/week (parent limited public transportation (F) were Children 10–12 years also reported report) associated with lower walking/cycling on perceived traffic, road safety, among children. Among 10–12 year strangers, and sport facilities. olds, youth who perceived no parks nearby (M, F) and whose parents believed that they had to cross many roads to get to play areas (M, F), that there were no lights or crossings (M), that there were few sporting arenas (F), and that there was limited public transportation (F) were less likely to bicycle/walk. Timperio (2006) † 912 M/F U Australia 5–6 years and 10–12 CS, O Distance to school, busy-road barrier, Walking or riding to school (parent In both age groups, children were less years route along busy road, pedestrian report) likely to actively commute to school if route directness (connectivity), steep their route as >800 m and a busy incline route barrier was present en route. Children with a steep incline (5–6 year olds) and a direct route to school (10– 12 year olds) were less likely to actively commute 25 th th Trost (1997) 202 (rural) M/F B/W 5 – 6 grade (ages 10 L (1 Availability of activity-related Self-reported physical activity No links between home equipment USA to 11) at baseline year), P equipment in the home. measured one year after determinants. and physical activity. 26 th Trost (1999) 108 M/F B USA 6 grade (approx age 11 CS, P Availability of activity-related Objectively measured (accelerometer) No links between home equipment years) equipment in the home. physical activity and physical activity 27 th th Zakarian (1994) 1634 M/F H/W/A/B 9 and 11 grade CS, P Number of facilities for sport and Self-reported vigorous exercise (20 Access to facilities was associated with USA (approx age 14 and 16 exercise; safe to exercise in minutes of activity that makes your higher vigorous exercise. years) neighborhood. heart rate and breathing increase) Zask (2001) 3912 M/F U Australia 5–12 years CS, O Direct observation of the availability of Direct observation of children's The presence of equipment (other activity-related equipment (e.g., balls, physical activity behavior in all school than balls) was not associated with fixed equipment) playground areas. children's physical activity. Note: Number/Gender/Ethnicity/Country: M, male; F, female; W, White; B = African American/Black; H, Hispanic/Mexican American; A, Asian; PI, Pacific Island; U = unknown (not mentioned); USA, United States of America. Design: CS, cross sectional; L, longitudinal; I, intervention; P, perceived environment; O, objectively measured environment. Physical activity behavior: MVPA, moderate to vigorous physical activity; VPA, vigorous physical activity. Significant associations: M, F, significant findings limited to males and females respectively. If an ethnic group made up ≤ 2% of the total sample, it was not included in the list of ethnic groups assessed. † Given that the same sample was used in Timperio (2006) and Timperio (2004) and there was an overlap in the measure of physical activity, only the novel findings are recorded for the more recent study. International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Page 7 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Table 2: Pattern of findings for links between environmental attributes (perceived and objective) and children's physical activity. Associations with physical activity Environmental Attribute Perceived Environment Objectively measured environment Adult report Child report Recreational infrastructure Private (19) (25) (26) (F)(31) (20) (22) Home equipment 0 0 0 + + 0 Public (B)(16) (28) (22) (B)(16) (24) (M)(21) Proximity of playgrounds and parks 0 0 + 0 + + (28) (18) (F)(19) (27) (32) (20) (F)(24) (F)(34) (F)(17) (F)(37) Availability recreation facilities 0 + 0 + + + + + + + (17) Spending on recreational infrastructure 0 (36) (F)(38) (33) Distance to school (school location) - - - (30) (33) School size 0 0 (20) (35) (23) (29) Equipment/play structures in school play areas + 0 + + Transport infrastructure Provision of amenities (32) (33) (39) + + Presence of sidewalks 0 (40) Street and sidewalk conditions + (M)(18) (40) (33) Presence of bike lanes/ease of cycling - 0 0 (M)(24) (39) Presence of controlled crossings + + (32) (30) (36) (F)(37) Connectivity of street network 0 + - + (32) (F)(18) (M)(37) Access to destinations + - + (F)(24) Availability of public transportation + Road hazards (24) Number of roads to cross - (M)(24) (18) (36) Traffic (density/speed) - - - (F)(18) Pedestrian and cyclist safety + (36) Steep terrain - Local conditions Safety and neighborhood disorder (45) (B)(16) (28) (48) (44) (27) (B)(16) (32) (F)(21) Perceived safety 0 0 0 0 + 0 0 0 + (F)(17) (F)(21) (43) Area deprivation and crime - - - (18) Roaming dogs - (24) (44) Social disorder/stranger danger 0 - (40) (44) Physical disorder/tidiness of area 0 - (32) Aesthetics of neighborhood + Region and weather (43) 11 (M)(17) Month of year (average temperature) 0 - - (46) (47) (17) Unsuitable weather 0 0 - (43) Region of the United States 0 (47) (B)(42) (W)(42) Rural/suburban versus urban 0 + - (30) (33) Population density + 0 Note: Associations identified with physical activity: +, significant positive association; -, significant negative association; 0, no association. Superscript numbers = reference number. Effects that are specific to particular demographic groups are noted as follows: M = males; F = females; B = Black/African American sample or subgroup their objectively measured physical activity using acceler- activity among adolescents girls and boys and young ado- ometers. In contrast, Fein and colleagues [20] and Stucky- lescent girls (but not boys). Ropp and DiLorenzo [31] found that the number pieces of exercise equipment in the home was positively and sig- Differences in the results outlined above cannot be nificantly associated with higher self-reported physical explained by differences in sample size, participant age, or Page 8 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 the operationalization of home equipment. Differences, home, and children's physical activity. Among Australian however, may be explained by differences in the ethnic samples, Timperio et al. [24] found that parents' reports of composition of the samples; both studies identifying a few sporting arenas in the area were linked with lower significant effect for home equipment used a predomi- rates of walking and cycling among girls and Carver et al. nantly white sample, whereas, studies that did not iden- [18] found that parents' reports of the presence of good tify an effect used either an exclusively African American sporting facilities nearby for their children were associated sample [26] or samples of mixed racial/ethnic back- with higher self-reported walking or cycling among ado- ground [3,19,22]. It should also be noted that both stud- lescent girls and boys (A simplified summary of the results ies identifying a significant positive effect used a self- from Carver et. al are presented throughout this review report measure of physical activity. The remaining studies given the extensive number of variables assessed. Only used self-report [19,25] or an objective assessment results for the frequency of walking/cycling in general are [22,26] of physical activity. Thus, any association identi- reported). In a study combining qualitative and quantita- fied between home equipment and children's physical tive methods, Hume et al. [34] found that, when children activity is limited to white adolescent samples and to self- were instructed to draw pictures of their home and their report measures of physical activity. neighborhood, girls who drew more opportunities for physical activity, including recreational facilities such as Public recreational infrastructure gyms and swimming centers, had higher objectively meas- Proximity of parks and playgrounds ured physical activity. Among US samples, Zakarian et al. A significant positive association between the proximity [27] found that a greater number of facilities for sport and of parks and playgrounds to the home and children's exercise in the area (based on self report) were associated physical activity was identified in three out of five studies. with higher adolescent self-reported vigorous activity and In an exclusively Hispanic sample, Gomez et al[21] found Brodersen et al. [17] found that the number of sport that objectively measured distance to the nearest play area pitches in the borough, as determined by objective assess- was inversely associated with adolescent boys', but not ment, was associated with higher self-reported vigorous girls', self-reported physical activity. Sallis et al. [22] found activity among girls but not boys. Similarly, Norman et. that parents' reports of the number of play areas within al. [37] found that objective measures of the number of walking distance of the home were positively associated recreational facilities and parks within a mile of the home with observed levels of physical activity among preschool were associated with higher objectively measured physical children. Furthermore, Timperio, et al. [24] found that activity among adolescent girls, but not boys. Finally, children who reported a lack of parks or sports grounds Mota et al. [32] and Fein et al.[20] using samples from near their home made fewer walking and cycling trips. In Portugal and Canada respectively, found that adolescents' contrast to these studies, Sallis et al. [28] and Adkins et al. reports of the availability of facilities such as swimming [16] (using an exclusively Black sample) found no associ- pools, playgrounds and parks were associated with higher ation between proximity of playgrounds and parks and self-reported physical activity. In contrast to the afore- children's objectively measured physical activity. mentioned studies, Dunton et al. [19] found no associa- tion between girls' reports of activity-related resources in Although a number of ethnic/racial groups were assessed the community and their self-reported physical activity across studies, no consistent ethnic/racial differences were and Sallis et al. [28] found no association between access identified. Differences in methods used to assess physical to facilities and children's objectively measured physical activity, however, were noted for studies that did and did activity. In addition, no association was identified not identify a significant association. Both studies that between spending on recreational infrastructure and chil- found no association [16,28] assessed physical activity dren's self-reported physical activity [17]. using accelerometers, which provide an aggregate meas- ure of physical activity across a number of days. In con- With one exception, there are no obvious differences in trast, studies that found a significant association relied on the designs of studies that did and did not identify a sig- self-reported or observed physical activity, both of which nificant association between the availability of recrea- are prone to reporter/observer bias, but which can be tai- tional areas and children's physical activity. Specifically, lored to provide a specific measure of physical activity there were no clear differences across studies in the defini- (e.g., walking or cycling trips). tion of recreational facilities (which usually included structures such as swimming pools, gyms, sporting arenas, Availability of recreation areas and spending on recreational and parks), the methods used to assess physical activity, or infrastructure the demographic characteristics of the samples. There In eight out of ten studies, a significant positive associa- were, however, clear differences is sample size across stud- tion was identified between the availability of recreation ies. The majority of studies that identified a significant areas, or the presence of such areas in the vicinity of the effect used samples of 1000 or more participants. In con- Page 9 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 trast, the two studies that found no effect used samples of activity. No associations, however, were found between approximately 100 participants, taking age and gender school size and children's physical activity. The lack of break-downs into consideration. This suggests that the effects of school size reported by Braza et al. [30] and association between the availability of facilities and phys- Ewing et al. [33] may be attributable to the use of aggre- ical activity among youth is relatively small and therefore gate data, or data collected at one level (e.g., a census only measurable with a large sample. While the availabil- track) that is then aggregated to a higher level (e.g., ity of facilities was assessed in all studies, no studies county). As a result of the process of aggregation, any directly asked children or parents whether they used such information pertaining to individual residences or spe- facilities. Consequently, the association between recrea- cific locations is lost. tional facilities and physical activity is indirect at best. Transport infrastructure School characteristics Two types of transport infrastructure were examined in Three out of three studies identified a negative association studies including the provision of amenities (e.g., side- between distance to school and children's physical activ- walks, crossings) and the presence of road hazards. Trans- ity. Timperio et al.[36] and Cohen et.al. [38] (girls only) portation infrastructure in urban areas is the found significant negative associations between an objec- responsibility of a number of agencies. For example, in tive measure of distance to school and children and ado- the United States, Metropolitan Planning Organization lescents' objectively measured moderate to vigorous (MPO) are generally charged with the preparation of plan- physical activity. Ewing et al. [33] found that lower walk/ ning documents and the allocation of funding for major cycle time to school, an indirect measure of distance, was programs and projects, whereas, the designation of cross- associated with higher rates of active commuting to walks, traffic signals, pedestrian signage, and other amen- school. In contrast to studies assessing distance to school, ities are in general the responsibility of various Braza et al. [30] and Ewing et al. [33] found no association transportation departments based on right-of-way and between school size, an indirect measure of whether or public ownership of property. Nine studies assessed asso- the school is located in a residential area and therefore ciations between transport infrastructure and children's close to homes, and the rates of walking and cycling to physical activity [18,24,30,32,33,36,37,39,40]. All nine school. studies used a cross sectional design. Two studies used an objective measure of physical activity and six studies used With respect to characteristics within schools, Sallis et al. an objective measure of the environment. The remaining [23] found that middle-school-aged children were more studies relied on self-report instruments. Five of the nine likely to be active during school recess periods when there studies were conducted in the US; the remaining studies was a larger number of activity-related equipment (e.g., were conducted in Australia and Portugal. balls) and the permanent activity structures (e.g., basket- ball hoops) available; these effects were most notable in Provision of amenities Presence and condition of sidewalks and bike lanes the presence of adult supervision. Similarly, Fein et al. [20] found that adolescents' reports of the availability of Results generally supported a positive association sports equipment, the functionability of equipment, and between the presence and condition of sidewalks and access to athletic facilities at school were associated with children's physical activity with three out of four studies higher self-reported physical activity. In contrast, Zask et identifying a significant positive effect. Ewing et al. [33] al. [35] found no association between the availability of found that the proportion of street miles with sidewalks playground equipment (with the exception of balls) and was positively associated with children's rates of walking children's physical activity. Finally, in an intervention or cycling to school. In an evaluation of the implementa- examining the effect of playground markings such as hop- tion of a Safe Routes to School program, Boarnet et al. scotch and court lines for basketball on children's physical [39] found that children who passed areas in which side- activity, Stratton and Mullan [29] found significant walks were installed were more likely to walk or cycle to increases in moderate to vigorous physical activity and school than children who did not pass such areas. In con- vigorous physical activity in intervention schools relative trast, Mota et al. [32] found no association between the to control schools. perceived presence of sidewalks on streets in the neigh- borhood and adolescents' self-reported activity. In the In sum, three out of three studies found that children who only study that assessed the impact of sidewalk condi- live close to schools are more likely to actively commute tions, Jago and colleagues [40] found that objectively to school and three out of four studies found that children assessed sidewalk characteristics such as the distance from were more active during play periods when characteristics the sidewalk to the curb, average height of trees, and side- of school play areas (e.g., access to equipment, permanent walk material and type were associated with higher objec- play structures, and marked courts) facilitated physical tively measured light intensity physical activity (e.g., slow Page 10 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 walking) among children. The studies that identified sig- self-reported activity. In contrast to what might be nificant effects used objective measures of the environ- expected, Timperio et al. [36] found that a more direct ment and measured children's walking (or low intensity route to school (i.e., higher connectivity, which was physical activity) as the outcome variable, which is the assessed using objective methods), was associated with most likely component of physical to be influenced by lower rates of walking and cycling to school among older sidewalk characteristics. In the only study that failed to children (10–12 years); no links were found between con- identify a significant effect, a self report measure of side- nectivity and active commuting to school among younger walk availability was used along with a generalized meas- children (5–6 years of age). ure of physical activity that may not reflect subtle differences in physical activity that result from the pres- The difference in findings reported by Mota et al. versus ence of sidewalks. Braza et al. and Norman et al. may reflect the possibility that effects of connectivity are only observed when objec- With respect to infrastructure for cycling, Jago et al. [40] tive measures of connectivity are used; it is possible that found no association between the ease of cycling (pres- individuals are not able to accurately recall and report the ence of bike lanes, attractiveness for cycling, number of level of street connectivity in their neighborhood. The read lanes) and objectively measured light intensity phys- findings outlined by Timperio et al., which were opposite ical activity in a sample of boys and Ewing et al. [33] to those expected (with higher connectivity or a more found no association between the presence of bike lanes direct route associated with lower rates of active commut- and children's walking/cycling to school. Furthermore, ing to school), are more difficult to explain. Timperio et Carver et al. [18] found that the perceived ease of cycling al. suggest that the counterintuitive effects of connectivity was associated with lower (rather than higher) rates of in their study may reflect the possibility that children's cycling among boys. Spurious findings for the presence of travel behavior is more influenced by traffic safety con- bike lanes or ease of cycling may be explained by a cerns than street networks. number of factors including the use of a measure of phys- ical activity that cannot detect cycling (i.e., accelerome- Three out of four studies identified a significant positive ters) [40], low rates of bicycling to school in general [33], association between access to destinations and children's and inflated type II error due to performing an extensive physical activity. This consistent pattern was noted number of analyses [18]. although a variety of measures of access were used across studies including the presence of destinations such as Presence of controlled crossings, street connectivity, and access to shops, access to public transportation, and retail floor area destinations ratio (i.e., ratio of retail building square footage to parcel Two studies examined the association between the pres- square footage). Timperio et al. [24] found that parents' ence of controlled crossings (e.g., presence of lights, cross- reports of a lack of public transportation were associated ings, or crosswalks) and children's physical activity, both with lower rates of walking and cycling among girls but of which identified significant positive effects. Timperio et not boys. Mota et al. [32] found that the ability to walk to al. [24] found that parents' reports of a lack of traffic lights destinations such as shops and transit stops was associ- and controlled crossings were associated with lower rates ated with higher physical activity among adolescents and of walking and cycling among boys, but not girls. In their Norman et. al.[37] found that a greater retail floor area evaluation of a Safe Routes to School program, Boarnet, et ratio (reflecting greater retail space and access to shops) al. [39] found that children who passed areas in which was associated with higher objectively measured moder- traffic control methods were installed were more likely to ate to vigorous physical activity among adolescent boys walk or cycle to school than children who did not pass but not girls. In contrast to expectations, Carver et al. [18] such areas. found that adolescent girls' reports of greater access to convenience stores reported lower, rather than higher, Conflicting results were found for studies assessing street rates of walking for transport. The general consistency of connectivity with only two out of four studies identifying results for access to destinations, despite differences in its a significant effect in the anticipated direction. Braza et al. operationalization, suggests that it should be considered [30] found that an objective measure of street connectivity further in future investigations. was associated with higher rates of walking or biking to school. Similarly, Norman et al.[37] found that higher Road hazards intersection density (also assessed using an objective A variety of road hazards have been examined across stud- measure) was associated with higher objectively measured ies including the number of roads to cross, the presence of moderate-to-vigorous physical activity among girls but a road barrier, traffic speed and density, pedestrian and not boys. Mota et al. [32], however, found no associations cyclist safety, and terrain. All three studies assessing road between perceived street connectivity and adolescents' hazards found a negative association between such haz- Page 11 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 ards and children's physical activity. Timperio et al. [24] their children's participation in recreational physical found that parents' reports that their children had to cross activity. Similarly, Gomez et al. [21] noted that adoles- many roads to get to a play area (girls and boys) and of cents' reports of perceived neighborhood safety were asso- high levels of traffic density in their local area (boys only) ciated with higher self-reported outdoor physical activity were associated with lower rates of walking and cycling for girls but not boys. The general lack of findings for per- among children. In a second study by Timperio et. al. ceived safety may reflect the fact that most of the studies [36], using the same sample but using an objective assess- measured general levels of physical activity, which may or ment of the environment, the presence of a busy road bar- may not be linked with neighborhood safety given that rier (e.g., a highway) en route to school (5–6 years olds children can be active outside their neighborhood. and 10–12 year olds) and the presence of a steep incline (5–6 year olds only) were associated with lower rates of In contrast to perceived safety, three out of three studies active commuting to school. Similarly, Carver et al. [18] identified a significant negative association between crime found that parents' reports of traffic impeding the ability or area deprivation and children's physical activity. Gor- to walk were associated with lower rates of walking or don-Larsen et al. [43] and Gomez et al. [21] (girls only) cycling among girls and boys, whereas, parents' percep- found significant inverse associations between objectively tion of the roads in the area being safe was associated with measured crime rates and adolescents' self-reported phys- a higher frequency of walking among girls (but not boys). ical activity. Similarly, Brodersen et al. [17] found that It is worth noting that all of these studies were conducted area deprivation (i.e., rates of car ownership, housing ten- with urban Australian samples. ure, unemployment and overcrowding in the district) was associated with lower self-reported physical activity Local conditions among 11–12 year old girls but not boys. Finally, Carver Both recreational and transport infrastructures exist et al. [18] found that the presence of roaming dogs were within the context of local community conditions. The associated with lower rates of walking or cycling among actions of other community members and agencies such adolescents. as police patrols, community clean-up programs, and/or transient populations, all exert influence at the local level. Social and physical disorder and neighborhood aesthetics Three studies assessed links between neighborhood disor- These conditions include both positive and negative envi- ronmental attributes such as general neighborhood safety, der and children's physical activity. Findings were mixed safety of play areas, crime rates, social disorder and across these studies, likely reflecting differences in the stranger danger, physical disorder and weather condi- operationalization of disorder. Molnar [44] objectively tions. Eighteen studies were identified that assessed links measured physical (e.g., graffiti, empty beer bottles) and between local conditions and children's physical activity social (e.g., alcohol in public, people selling drugs) disor- [16-18,21,24,27,28,30,32,40-48]. All but one study used der using coded video recordings and direct observation a cross sectional design. Four studies used an objective of neighborhoods. Both forms of disorder were associated measure of physical activity (accelerometry), one used with lower levels of parent-reported recreational activity direct observation, and fifteen studies relied on a self- among adolescents. Jago et al. [40], however, found no report measure of physical activity. With regard to meas- association between an objective measure of neighbor- ures of the environment, nine studies used a self-report hood tidiness and children's objectively measured physi- measure, seven studies used an objective measure and two cal activity. Likewise, Timperio et al. [24] found no studies used both methods. The vast majority of studies association between children's perceptions of stranger (13 out of 18) were conducted in the US. danger (a source of social disorder) and parents' reports of their walking and cycling to destinations. Thus, it appears Safety and neighborhood disorder that any association between neighborhood disorder and Safety, crime, and area deprivation physical activity may be limited to much higher levels of Nine studies examined the association between perceived disorder (or deviance) such as those measured by Molnar safety and children's physical activity. These studies over- et al. A general lack of tidiness or the perception that stran- whelming reported a null effect with seven gers can be dangerous but may not be enough to dissuade [16,27,28,32,45,48] of the nine studies showing no asso- youth from being active outdoors. In the only study that ciation between perceived safety and children's physical assessed perceived aesthetics, Mota et al. [32] found that activity. The lack of an association was not limited to a adolescents' reports of the aesthetics of their neighbor- particular research design or sample population. Two hoods (i.e., there are many interesting things to look at exceptions to the pattern of null findings are the studies while walking) were positively associated with their self- by Molnar, et al. [44] and Gomez et al. [21]. In Molnar et reported physical activity. al. residents' reports of the safety of children's local play areas were positively associated with parents' reports of Page 12 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Region and weather Felton et al. [42] found mixed results for location. White Weather girls in urban areas were more active (based on self- A significant association between weather and children's reports) than White girls in rural areas. The opposite was physical activity was identified in two out of five studies. found for Black girls; black girls living in rural areas were Baranowski et al. [41] and Brodersen et al. [17] found that more active than Black girls from urban areas. Although preschool children and 11–12-year-old boys respectively the difference was not discussed by the authors, it is pos- were less active during hotter months of the year. Brod- sible that White girls from urban areas lived in neighbor- ersen et al. [17] also found that higher rainfall was associ- hoods in which they could take advantage of the ated with lower self-reported physical activity among girls infrastructure for physical activity generally attributed to but not boys. Although "unsuitable" weather was urban areas such as the presence of sidewalks and accessi- reported by adolescents as a perceived barrier to physical ble parks. While Black girls may also have had access to activity in the study by Tappe et al. [46], such perceptions similar resources, their ability to use such resources may were not associated with lower levels of self-reported have been limited by neighborhood characteristics such as physical activity. Similarly, Gordon-Larsen et al. [43] crime. found no relationship between the month of the year and adolescents' self-reported physical activity, indicating a Two studies assessed links between population density lack of a seasonality effect. Finally, Sirard et al. [47] found and children's active commuting to school; no consistent no association between weather conditions and rates of effects were identified. Ewing et al. [33] and found no walking and biking to school. association between population density in the immediate area around children's homes and their rates of walking/ The effects of weather may have been underestimated in cycling to school. In contrast, Braza et al. [30] found that these studies due to the restricted time range in which the higher population density was associated with higher data were collected. For example, Gordon-Larsen et al. rates of active commuting to school. Neither study consid- [43] used data collected on physical activity between April ered whether children attended their local school, rather and December. It is possible that the effect of bad or than a magnet or private school outside of the local area, unsuitable weather was eliminated by the exclusion of the or the feasibility of children walking or riding to school. months of January through March (winter months in the northern hemisphere where the research was conducted). Discussion There was also limited variability in the geographic region In this paper we reviewed research on associations within each study. No studies collected data across multi- between the physical environment and children's physical ple regions that varied in the suitability of the climate for activity while highlighting the parties responsible for each outdoor activity. Consequently, inconsistent or non-sig- environmental attribute. This was achieved by classifying nificant effects could be explained by a general lack of var- and reviewing studies specific to recreational infrastruc- iability in the data by month of the year and/or location. ture, transport infrastructure and local conditions. The Furthermore, no studies considered the availability of most consistent pattern of findings was evident for trans- resources for indoor recreational activity in communities. port infrastructure, followed by recreational infrastruc- It is likely the unsuitable weather conditions will most ture, with the least consistent pattern of results noted for often be associated with low levels of physical activity in local conditions. Although there were no consistent differ- communities in which there are few opportunities for ences in results across age or ethnic groups, there was indoor physical activity. some indication that associations between environmental characteristics were more commonly noted for girls than Region, urban/rural location, and population density boys. Three studies examined associations between region and children's physical activity, with one of the three studies Summary of findings showing a significant effect. Gordon-Larsen et al. [43] Results from studies examining components of transport found that residence in the Northeast of the United States infrastructure showed that children were more active was associated with higher self-reported physical activity when there were sidewalks in their neighborhood, they among adolescents in comparison to residence in the had destinations to walk to, public transportation was South, West or Midwest. This effect of region could be available, there were fewer uncontrolled intersections to explained by a myriad of factors such as regional differ- cross, and traffic density was low. Results were more con- ences in weather, income, education, ethnic/racial make- sistent for the absence of roads hazards (i.e., roads to up, and access to community resources. When examining cross, traffic density/speed) than the provision of ameni- rural/suburban versus urban location, Sirard et al. [47] ties (i.e., sidewalks, presence of destinations, controlled found no differences in rates of walking and biking to intersections). In addition, findings were most consistent school for schools located in urban and suburban areas. for parents' reports of infrastructure followed by objective Page 13 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 measures; in general, null findings, or findings in the to pose. Greater rigor with regard to measurement of both opposite direction to those anticipated, were evident for physical activity and the environment, and the use of studies relying on children's reports of transport infra- more sophisticated designs will facilitate the establish- structure. No consistent differences by gender or ethnic ment of a transdisciplinary approach, which is imperative group emerged for transport infrastructure. to moving this body of research into the "next phase". Measurement issues Although findings were less consistent for recreational infrastructure, there were a number of instances in which The ability to measure characteristics of the physical envi- the majority of studies supported a particular relation- ronment is greatly facilitated by the use of geographic ship. The majority of studies showed that the availability information systems (GIS). Of the 33 studies reviewed, 6 of facilities in neighborhoods and the availability of used GIS-based methodologies. As the use of GIS becomes equipment and permanent activity structures in school more commonplace, it is imperative that the methods for play areas were associated with higher physical activity. In "creating" and displaying the data are recorded in detail addition, greater distances to school were associated with (the "meta data" – the data about the data). Currently, lower rates of walking and cycling to school. In contrast to there is little description of the various processing deci- expectations, most studies failed to identify an association sions that are made when using GIS in published research. between home equipment and children's physical activity The absence of such information slows research progress and results for the proximity to playgrounds were mixed. and inhibits the comparison of findings across studies Some gender differences in the reported associations were and research disciplines. While we advocate for the incor- apparent. Six out of seven effects specific to girls were sig- poration of GIS into research designs, the perceived envi- nificant and in the anticipated direction. Most of these ronment should also be taken into consideration because effects were noted for child reports of the environment. In people's perceptions may, in fact, motivate their behavior contrast, only one significant effect was specific to boys. In more than the true nature of the situation. two instances, associations specific to African Americans were reported. In both cases, no significant effects of rec- In contrast to the objective assessment of the environ- reational infrastructure were present for this demographic ment, objective measures of physical activity were more group. widely incorporated into the studies reviewed with 13 out of 33 studies reviewed using an objective measure of phys- Findings were least consistent for local conditions, reflect- ical activity (8 used accelerometers, 4 used direct observa- ing the broader range of characteristics assessed. In gen- tion, 1 used heat rate monitoring). Using accelerometers eral, no effects were found for perceived neighborhood to measure children's physical activity and/or directly safety or the perceived safety of play areas. However, both observed children's activity removes the possibility of studies that used objective measures of crime rates response bias, particularly among children [49]. Although reported a significant negative association between crime the use of objective measures of physical activity is prefer- and children's physical activity. Similarly, objectively able, because it allows greater confidence in the validity of measured area deprivation and the perceived presence of the assessment, objective measures may not be feasible in roaming dogs were associated with lower physical activity. large-scale survey research due to financial and logistical No consistent pattern of findings was evident for region or constraints. In addition, accelerometers provide only a weather conditions. With respect to differences noted by generalized measure of physical activity and do not pro- gender, three effects specific to girls were significant and in vide information on the type of activity or the location in the anticipated direction; only one effect was specific to which physical activity takes place. As is noted by Giles- boys and this was also in the anticipated direction. Of the Corti et al. [50], assessing context-specific behaviors is key three effects specific to African Americans, two were not to understanding associations between the physical envi- significant and one was in the opposite direction to that ronment and physical activity. In many of the studies expected. reviewed, generalized measures of physical activity were implemented that may not be sensitive to specific envi- Recommendations for future research ronmental attributes. Consequently, null effects may The research reviewed herein generally reflects the first reflect a lack of specificity in the measures used rather than "phase" of research on links between the physical envi- the absence of an association. New equipment that incor- ronment and children's physical activity. In this first porates Global Positioning Systems (GPS) into acceler- round of research, many of the methodologies were in a ometers may facilitate the ability to obtain context- developmental stage. As a result, there was little consist- specific measures of physical activity by making it possible ency in the methods used. In addition, in many cases, the to know exactly where (i.e., longitude/latitude data) and methods used and the designs adopted were opportunis- when (i.e., electronically time-stamped data) the physical tic as researchers grappled with which research questions activity occurred. Page 14 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 Design issues that draws upon diverse research disciplines such as geog- With two exceptions, all studies relied on cross sectional raphy and planning, public health, exercise science, crim- analysis. Given the need to understand behavioral inal justice and human development. Such an approach changes associated with environmental attributes, longi- will require greater standardization of procedures and tudinal studies are more appropriate. Such studies will detailed reporting of these procedures than has generally help us determine whether the pattern of results reflects been exhibited in research to date. In addition, a transdis- the ability of the environment to constrain or facilitate ciplinary approach will require clear communication and certain behaviors or reflects the type of person/family who dialogue across research disciplines, including access to chooses to live in certain neighborhoods [9,51]. In addi- literatures across various research communities. In many tion to using a longitudinal design, more complex models cases, information on environmental attributes is con- need to be developed and tested. With the exception of tained in planning documents rather than in refereed gender, research has rarely examined factors that may journals. These documents are often available over the moderate the link between the environment and chil- Internet, but may not be included in databases or other dren's physical activity (i.e., interact with the environment researchable tools. to predict physical activity). The use of simplistic designs with little consideration of moderating factors as high- Summary and conclusion lighted by McMillan [15], may lead to simplistic and erro- In this review, we have found preliminary evidence that a neous conclusions. The most noteworthy example is the relationship exists between children's participation in general failure to consider ethnicity, family income, or physical activity and environmental attributes. Limita- neighborhood deprivation as possible moderating or con- tions of this review include the exclusion of studies not founding variables. Furthermore, many studies have published in English or searchable in English-based data- assessed children across a broad age range, which ignores bases, the overall bias against publishing studies with null the possibility that associations between the physical results, and the lack of research specific to children out- environment and physical activity may be age-specific due side the health sciences. Future work could enhance our to differences in parental control and children's inde- understanding of this important topic by assessing both pendent mobility. perceived and objective characteristics of the environ- ment, including objective measures of children's physical In addition to a lack of emphasis on children's age, the activity and the physical environment, adopting longitu- role that parents play in regulating children's use of the dinal designs, assessing the interaction between various physical environment has not been considered in research environmental attributes, and examining the important to date. Consequently, the assumption is generally that role that parents play as gate keepers to children's use of there is a direct link between the environment and chil- the physical environment. There is also a need for studies dren's physical activity. This is unlikely to be the case outside the US to determine whether results identified given children's lack of decision autonomy and the role using US samples can be generalized internationally. that parents play as gate keepers to children's use and Finally, we advocate the continued use of the classifica- exploration of the physical environment surrounding tion scheme outlined herein as this will allow us to deter- their home. Research shows that parents' decisions about mine the parties responsible for attributes found to their children's independent mobility are influenced by a influence children's physical activity and to make the nec- number of factors such as their perceptions of the safety of essary changes. We also strongly recommend the estab- the area, neighborhood relations, and proximity to a park lishment of a transdisciplinary research agenda [52]. Research designs and techniques will need to link sufficiently transparent to facilitate the sharing of infor- quantitative and qualitative data to successfully under- mation across a growing body of work being generated by stand the nature of parents' decision-making processes diverse research approaches. and their willingness to allow their children to participate in physical activity under a combination of environmen- Acknowledgements Drs. Kirsten Krahnstoever Davison and Catherine T. Lawson are grateful tal attributes. for the support of the Center for the Elimination of Minorities Health Dis- parities and the Center for Social and Demographic Analysis at the Univer- The need for a transdisciplinary approach sity at Albany. We want to extend special thanks to our research assistant Scientists from different research paradigms have largely Katie Haverly for her work on the literature search. approached research of children and physical activity from the perspective of their own discipline with little References integration of ideas and methods across disciplines. To 1. Centers for Disease Control and Prevention: Youth Risk Surveil- lance Survey - United States, 1997. Morbidity and Mortality most effectively assess the impact of the physical environ- Weekly Report 1998, 47(SS-3):1-89. ment on physical activity levels among youth, future 2. US Department of Health and Human Services: Healthy People research will need to adopt a transdisciplinary approach 2010. Washington, DC , US Government Printing Office; 2000. Page 15 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 3. Trost SG, Kerr LM, Ward DS, Pate RR: Physical activity and rural fifth-grade children. Preventive Medicine 1997, determinants of physical activity in obese and non-obese 26(2):257-263. children. International Journal of Obesity and Related Metabolic Disor- 26. Trost SG, Pate RR, Ward DS, Saunders R, Riner W: Determinants ders 2001, 25:822-829. of physical activity in active and low-active, sixth grade Afri- 4. Bailey DA, Martin AD: Physical activity and skeletal health in can-American youth. Journal of School Health 1999, 69(1):29-34. adolescents. Pediatric Exercise Science 1994, 6:330-347. 27. Zakarian JM, Hovell MF, Hofstetter CR, Sallis JF, Keating KJ: Corre- 5. Morrow JR, Freedson PS: Relationship between habitual physi- lates of vigorous exercise in a predominantly low SES and cal activity and aerobic fitness in adolescents. Pediatric Exercise minority high school population. Preventive Medicine 1994, Science 1994, 6:315-329. 23:314-321. 6. Calfras KJ, Taylor WC: Effects of physical activity on psycholog- 28. Sallis JF, Taylor WC, Dowda M, Freedson PS, Pate RR: Correlates ical variables in adolescents. Pediatric Exercise Science 1994, of vigorous physical activity for children in grades 1 through 6:406-423. 12: Comparing parent-reported and objectively measured 7. Humpel N, Owen N, Leslie E: Environmental factors associated physical activity. Pediatric Exercise Science 2002, 14:30-44. with adults' participation in physical activity: a review. Amer- 29. Stratton G, Mullan E: The effect of multicolor playground mark- ican Journal of Preventive Medicine 2002, 22(3):188-199. ings on children's physical activity level during recess. Preven- 8. Duncan MJ, Spence JC, Mummery WK: Perceived environment tive Medicine 2005, IN PRESS:. and physical activity: a meta-analysis of selected environ- 30. Braza M, Shoemaker W, Seeley A: Neighborhood design and mental characteristics. International Journal of Behavioral Nutrition rates of walking and biking to elementary school in 34 Cali- and Physical Activity 2005, 2(11):. fornia communities. American Journal of Health Promotion 2004, 9. Owen N, Humpel N, Leslie E, Bauman A, Sallis JF: Understanding 19(2):128-136. environmental influences on walking; Review and research 31. Stucky-Ropp R, DiLorenzo TM: Determinants of exercise in chil- agenda. American Journal of Preventive Medicine 2004, 27(1):67-76. dren. Preventive Medicine 1993, 22:880-889. 10. Sallis JF, Bauman A, Pratt M: Environmental and policy interven- 32. Mota J, Almeida M, Santos P, Ribeiro JC: Perceived neighborhood tions to promote physical activity. American Journal of Preventive environments and physical activity in adolescents. Preventive Medicine 1998, 15(4):379-397. Medicine 2005, IN PRESS:. 11. Saelens BE, Sallis JF, Frank LD: Environmental correlates of walk- 33. Ewing R, Schroeer W, Greene W: School location and student ing and cycling: findings from the transportation, urban travel. Transportation Research Record 2004, 1895:55-63. design, and planning literatures. Annals of Behavioral Medicine 34. Hume C, Salmon J, Ball K: Children's perceptions of their home 2003, 25(2):80-91. and neighborhood environments, and their association with 12. Lee C, Moudon AV: Physical activity and environment research objectively measured physical activity: a qualitative and in the health field: Implications for urban and transportation quantiative study. Health Education Research 2005, 20(1):1-13. planning, practice and research. Journal of Planning Literature 35. Zask A, van Beurden E, Barnett L, Brooks LO, Dietrich SEI: Active 2004, 19(2):147-181. school playgrounds - Myth or reality? Results of the "Move It 13. Krizek KJ, Birnbaum AS, Levinson DM: A schematic for focusing Groove It" project. Preventive Medicine 2001, 33(401-408):. on youth in investigations of community design and physical 36. Timperio A, Ball K, Salmon J, Roberts R, Giles-Corti B, Simmons D, activity. American Journal of Health Promotion 2004, 19(1):33-38. Baur LA, Crawford D: Personal, familial, social and environ- 14. Sallis J, Prochaska JJ, Taylor WC: A review of correlates of physi- mental correlates of active commuting to school. American cal activity of children and adolescents. Medicine and Science in Journal of Preventive Medicine 2006, 30(1):45-51. Sports and Exercise 2000, 32(2):963-975. 37. Norman GJ, Nutter SK, Ryan S, Sallis JF, Calfras KJ, Patrick K: Com- 15. McMillan TE: Urban form and a child's trip to school: The cur- munity design and access to recreational facilities as corre- rent literature and a framework for future research. Journal lates of adolescent physical activity and Body-Mass Index. of Planning Literature 2005, 19(4):440-456. Journal of Physical Activity and Health 2006, 3 (Suppl 1):S118-S128. 16. Adkins S, Sherwood NE, Story M, Davis M: Physical activity 38. Cohen DA, Ashwood S, Scott M, Overton A, Evenson KR, Voorhees among African-American Girls: The role of parents and the CC, Bedimo-Rung A, McKenzie TL: Proximity to school and home environment. Obesity Research 2004, 12 (supple- physical activity among middle school girls: The Trial of ment):38S - 45S. Activity for Adolescent Girls Study. Journal of Physical Activity and 17. Brodersen NH, Steptoe A, Williamson S, Wardle J: Sociodemo- Health 2006, 3(Suppl 1):S129-S138. graphic, developmental, environmental, and psychological 39. Boarnet MG, Anderson CL, Day K, McMillan T, Alfonzo M: Evalua- correlates of physical activity and sedentary behavior at age tion of the California Safe Routes to School legislation: urban 11 to 12. Annals of Behavioral Medicine 2005, 29(1):2-11. form changes and children's active transportation to school. 18. Carver A, Salmon J, Campbell K, Baur L, Garnett SCD: How do per- American Journal of Preventive Medicine 2005, 28(2 Suppl 2):134-140. ceptions of local neighborhood relate to adolescents' walk- 40. Jago R, Baranowski T, Zakeri I, Harris M: Observed environmen- ing and cycling? American Journal of Health Promotion 2005, tal features and the physical activity of adolescent males. 20(2):139-147. American Journal of Preventive Medicine 2005, 29(2):98-104. 19. Dunton GF, Jamner MS, Cooper DM: Assessing the perceived 41. Baranowski T, Thompson WO, DuRant RH, Baranowski J, Puhl J: environment among minimally active adolescent girls: valid- Observations on physical activity in physical locations: age, ity and relations to physical activity. American Journal of Health gender, ethnicity, and month effects. Research Quarterly for Exer- Promotion 2003, 18(1):70-73. cise and Sport 1993, 64(2):127-133. 20. Fein AJ, Plotnikoff RC, Wild C, Spence JC: Perceived environment 42. Felton GM, Dowda M, Ward DS, Dishman RK, Trost SG, Saunders R, and physical activity in youth. International Journal of Behavioral Pate RR: Differences in physical activity between black and Medicine 2004, 11(3):135-142. white girls living in rural and urban areas. Journal of School 21. Gomez JE, Johnson BA, Selva M, Sallis JF: Violent crime and out- Health 2002, 72(6):250-255. door physical activity among inner-city youth. Preventive Med- 43. Gordon-Larsen P, McMurray RG, Popkin BM: Determinants of icine 2004, 39(5):876-881. adolescent physical activity and inactivity patterns. Pediatrics 22. Sallis JF, Nader PR, Broyles SL, Berry CC, Elder JP, McKenzie TL, Nel- 2000, 105(6):E83. son JA: Correlates of physical activity at home in Mexican- 44. Molnar BE, Gortmaker SL, Bull FC, Buka SL: Unsafe to play? American and Anglo-American preschool children. Health Neighborhood disorder and lack of safety predict reduced Psychology 1993, 12(5):390-398. physical activity among urban children and adolescents. 23. Sallis JF, Conway TL, Prochaska JJ, McKenzie TL, Marshall MS, Brown American Journal of Health Promotion 2004, 18(5):378-386. M: The association of school environments with youth physi- 45. Sallis JF, Alcaraz JE, McKenzie TL, Hovell MF: Predictors of change cal activity. American Journal of Public Health 2001, 91(4):618-620. in children's physical activity over 20 months. Variations by 24. Timperio A, Crawford D, Telford A, Salmon J: Perceptions about gender and level of adiposity. American Journal of Preventive Med- the local neighborhood and walking and cycling among chil- icine 1999, 16(3):222-229. dren. Preventive Medicine 2004, 38(1):39-47. 46. Tappe MK, Duda JL, Ehrnwald PM: Perceived barriers to exercise 25. Trost SG, Pate RR, Saunders R, Ward DS, Dowda M, Felton G: A among adolescents. Journal of School Health 1989, 59(4):153-155. prospective study of the determinants of physical activity in Page 16 of 17 (page number not for citation purposes) International Journal of Behavioral Nutrition and Physical Activity 2006, 3:19 http://www.ijbnpa.org/content/3/1/19 47. Sirard JR, Ainsworth BE, McIver KL, Pate RP: Prevalence of active commuting at urban and suburban elementary schools in Columbia, SC. American Journal of Public Health 2005, 95:236-237. 48. Burdette HL, Whitaker RC: A national study of neighborhood safety, outdoor play, television viewing and obesity in pre- school children. Pediatrics 2005, 116(3):657-662. 49. Sallis JF, Owen N: Physical activity and behavioral medicine. In Behavioral Medicine and Health Psychology 3 Edited by: Turner JR. Lon- don , Sage Publications; 1999. 50. Giles-Corti B, Timperio A, Bull F, Pikora T: Understanding Physi- cal Activity Environmental Correlates: Increased Specificity for Ecological Models. Exercise and Sport Science Reviews 2005, 33(4):175 -1181. 51. Transportation Research Board: Does the built environment influence physical activity? Examining the evidence. Wash- ington, DC , Institute of Medicine; 2005. 52. Prezza M, Pilloni S, Morabito C, Sersante C, Alparone FR, Giuliani MV: The influence of psychosocial and environmental factors on children's independent mobility and relationship to peer fre- quentation. Journal of Community and Applied Social Psychology 2001, 11:435-450. Publish with Bio Med Central and every scientist can read your work free of charge "BioMed Central will be the most significant development for disseminating the results of biomedical researc h in our lifetime." Sir Paul Nurse, Cancer Research UK Your research papers will be: available free of charge to the entire biomedical community peer reviewed and published immediately upon acceptance cited in PubMed and archived on PubMed Central yours — you keep the copyright BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 17 of 17 (page number not for citation purposes)

Journal

International Journal of Behavioral Nutrition and Physical ActivitySpringer Journals

Published: Jul 27, 2006

There are no references for this article.