Portal ENSP - Escola Nacional de Saúde Pública Sergio Arouca Portal FIOCRUZ - Fundação Oswaldo Cruz

Cadernos de Saúde Pública

ISSN 1678-4464

37 nº.5

Rio de Janeiro, Maio 2021


O papel de mediação da densidade residencial na associação entre fatores ambientais percebidos e deslocamento ativo à escola em adolescentes brasileiros

Arieli Fernandes Dias, Anelise Reis Gaya, Caroline Brand, Alex Antonio Florindo, Emilio Villa-González, Antonio García-Hermoso, Jorge Mota, Adroaldo Cezar Araujo Gaya


  • Artigo
  • Autores
  • Comentários (0)
  • Informações Suplementares

O estudo teve como objetivo verificar se a densidade residencial e a conectividade entre ruas são mediadores da associação entre fatores ambientais percebidos e deslocamento ativo casa-escola (DAE) em adolescentes brasileiros. Foi um estudo transversal com uma amostra randomizada de 1.130 adolescentes (52,7% do sexo feminino) entre 14 e 20 anos de idade, de Porto Alegre, Rio Grande do Sul, Brasil. Os adolescentes informaram seu modo de deslocamento casa-escola (ida e volta) através de um questionário, além dos atributos ambientais percebidos, de acordo com a Neighborhood Environment Walkability Scale for Youth (NEWS-Y). A densidade residencial e a conectividade entre ruas foram medidas por sistemas de informação geográfica (SIG), dentro de um raio de 1km da rede viária em torno do endereço residencial do participante. Foram ajustados modelos de regressão de acordo com os procedimentos para análises de mediação. Os resultados mostraram que a densidade residencial é um mediador da associação entre os fatores ambientais percebidos e o DAE, inclusive a diversidade do uso do solo (EI = 0,114; IC95%: 0,130, 0,311; 32% de mediação), infraestrutura para atividades de lazer na vizinhança (EI = 0,064; IC95%: 0,034, 0,105; 15% de mediação) e acesso a serviços (EI = 0,045; IC95%: 0,006, 0,104; 14% de mediação). A conectividade entre ruas não mostrou correlação com DAE, portanto não foi testado no modelo de mediação. Conclui-se que a densidade residencial é um mediador da relação entre fatores ambientais percebidos e DAE.

Ambiente Construído; Transportes; Exercício Físico; Estudantes



The prevalence of cardiometabolic risk factors, such as obesity, type 2 diabetes, and hypertension have been increasing in children and adolescents over the years 1. These non-communicable diseases are with low levels of physical activity in this population 2. Active commuting to and from school (ACS) (i.e., walking or cycling) is a reasonable form of physical activity among children and adolescents, recognized as an opportunity for health promotion 3,4. Increase in daily physical activity 4,5 and cardiorespiratory fitness 6, as well as the decrease in metabolic syndrome incidences 7 and waist circumference 8, are some of the positive health factors associated with ACS. Since most school children must travel to and from school twice a day, this behavior could be integrated into their individual routines. In this sense, previous review studies have identified numerous interventions of ACS promotion and its association with several health indicators in children and adolescents 9,10,11.

Adolescents' mode of commuting is influenced by personal, social, and environmental factors 12,13. For instance, Mandic et al. 13 showed that shorter distance to school, younger age, fewer vehicles, opportunity to chat with friends, nice scenery, and parental perceptions of safety when active commuting to school were positively associated with ACS in adolescents. Studies have also shown that walkability is a physical environmental factor that influences physical activity and active commuting in youths 14,15,16. Regarding the environmental perceived factors, results indicate that traffic and crime safety 17, land-use mix diversity 18, shorter distance to school, and access to destination 19 were associated with ACS in adolescents. Thus, we emphasize that safe neighborhoods with easy access and nice aesthetics could improve adolescent's perception, increasing ACS 20,21. In fact, good available environments is not enough, people need to perceive them as adequate for use.

Recently, studies have investigated how much the physical environment influenced ACS. A previous systematic review pointed out that higher objective accessibility, new infrastructure for walking and cycling, and public transportation were associated with increased ACS 21. However, all studies included in this review were conducted in high-income economies (North America, Europe, Australia, Asia, and New Zealand). Thus, it is important to emphasize that environmental characteristics associated with physical activity in high-income countries are different from low- and middle-income countries (LMIC) 22. Besides, few research is available regarding the characteristics of the environment and physical activity of young people in LMIC 22, emphasizing the need for more research in Latin American countries, specifically in Brazil, where the occurrence of children and adolescents who passively commute to school varies between 43.3% and 52% 23,24.

Considering the mentioned aspects, we emphasize that exploratory analyses about the relationship between environmental factors and ACS are important to create effective policies for physical activity promotion in regions where evidence about these aspects is still unavailable, especially in developing countries. Likewise, studies investigating the association between ACS with environmental variables in Brazilian youth are scarce. Thus, our study intends to add information to understand whether built environmental variables (objectively measured) influence the relationship between environment perception factors and ACS. Therefore, this study aimed to verify whether residential density and connectivity between streets are mediators on the association between perceived environmental factors and ACS in Brazilian adolescents.


Study design

A cross-sectional study with a quantitative approach was conducted in Porto Alegre, Rio Grande do Sul State, Brazil. This study is part of a wider project entitled Environmental Factors, Physical Activity and Health. Porto Alegre has approximately 1.4 million inhabitants (year 2010), a demographic density of 2,837.53 inhabitants/km2 and a territorial area of 496,681km2 (Brazilian Institute of Geography and Statistics. https://cidades.ibge.gov.br/brasil/rs/porto-alegre/panorama, accessed on 08/Sep/2019). This city is one of the most forested capitals in Brazil; it has approximately 630 squares, eight public parks, and three conservation units, according to data from the Porto Alegre Environment, Urbanism and Sustainability Department (http://www2.portoalegre.rs.gov.br/smam/default.php?p_secao=297, accessed on 09/May/2021). The population of the study was approximately 34,645 high school students from 71 public schools (Brazilian National Institute for Educational Studies and Research. http://portal.inep.gov.br/basica-censo-escolarmatricula, accessed on 08/Sep/2019). They were allocated in the following regions: 15,897 center, 8,057 north, 6,423 south, and 4,268 east.

The following criteria were considered to calculate the sample size: (a) estimated population of 34,645 students (N); (b) proportion of 50% (p); (c) complementary percentage of 100-P (q); (d) confidence level of 1.96 standard deviations (S); and (e) acceptable sampling error of 3% (e). After adopting these criteria and the formula presented below, it was estimated that 1,035 students should be evaluated. However, to avoid probable dropouts in the sample, an increase of 5% was assumed, resulting in 1,086 school-aged adolescents. The formula [n = S2.p.q.N/e2(N-1)+(S2.p.q)] was used to achieve a sample that represented the study population. This estimate was performed using the formula to have a representative sample of the population.

The software G*power version 3.1 (http://www.psycho.uni-duesseldorf.de/abteilungen/aap/gpower3) was used to estimate the sample size necessary for association analysis. Linear regression models were used with test power (1-β) = 0.85, significance level of α = 0.05, and effect size of 0.02. The considered number of predictors was 13 and the minimum sample size was established as 1,003.

Sample selection considered the proportion of adolescents enrolled in the schools per region. Thus, the sample was composed of 263 adolescents from four public schools in the north region (23.26%); 518 adolescents from seven public schools in the central region (45.88%); 140 adolescents from two public schools in the east region (12.32%); and 209 adolescents from three public schools in the south region (18.54%).

The sample selection was done by a procedure of multiple phases 25. Initially, the schools were selected according to each region. A number was assigned for each school and all numbers were placed in a box, mixed, and randomly reelected one by one. In the schools, the high school classes were selected randomly. Data was collected in one class from each year: first, second, and third.

The students from the selected classes were invited to participate in the study and the inclusion criteria were (a) belonging to the first, second, or third year of high school; (b) signing the assent document manifesting will to participate; and (c) handing in the informed consent document signed by a parent or guardian. Also, according to Sawyer et al. 26, a definition of 10-24 years corresponds more closely to adolescent growth and popular understandings of this life phase; thus, we used the term adolescents even when some students were over 18 years of age.

Data collection was performed during an 8-month period in 2017. First, the researchers provided the required information to the selected schools, explaining the aims of the study and, if the managers agreed to participate, they were asked to sign an acceptance term. Then, data collection was scheduled. Adolescents took, on average, 30 minutes to fill out the questionnaire, which was answered during a regular class. The study was approved by the Ethics Committee of Research with Human Beings of the Federal University of Rio Grande do Sul (n. 1,338.597).

Measurement procedures

Mode of commuting to and from school

Adolescents self-reported their usual mode of commuting using a questionnaire while supervised by a researcher (ratio 2 researchers per 30 students). Consequently, the question “How do you usually get to and from school?” was used. Response choices were (1) on foot, (2) by bicycle, (3) by bus, (4) by car and (5) other. Question about mode of commuting to and from school has been proposed as one of the most appropriate measurements for asking about mode of commuting to school in this population 27. Active commuters were students that commuted to and from school on foot or cycling, while passive commuters were those who commuted by bus or by car. The students who answered “other” were individually categorized and defined as active or passive commuters according to the transportation that they used.

Perceived environmental factors

To measure self-reported environment perception factors, the instrument Neighborhood Environmental Walkability Scale for Young (NEWS-Y) 28 was used, which was previously validated in Brazil for this population 29. This questionnaire evaluates environment perception factors that may influence adolescents' total physical activity 28. Questions were developed according to the dimensions (land-use mix diversity, neighborhood recreation facilities, access to services, street connectivity, places for walking, neighborhood aesthetics, neighborhood safety, and crime safety) proposed by the NEWS-Y.

Land-use mix diversity and neighborhood recreation facilities refer to distance perception from home to a variety of destinations, such as shops and school, as well as perceived distance from the student's house, on foot, to places for physical activity practice, such as walking/running tracks, or a large public park. The possible answers were 1-5 minutes, 6-10, 11-20, 21-30, more than 30 minutes, and don't know/there isn't. All items were reverse coded and employed mean values. Questions from the dimensions access to services, street connectivity, places for walking, neighborhood aesthetics, neighborhood and crime safety were measured using 4-point Likert scale (strongly disagree, partially disagree, partially agree, strongly agree). All dimensions were calculated following the NEWS-Y scoring guidelines 30 and the mean value was considered for each dimension. More information about the questions can be found in Lima et al. 29 and Rosenberg et al. 28, and on the James F. Sallis' website 30.

GIS-measured environmental factors

Adolescents' addresses, reported in the questionnaire, were geocoded in the geographic information system (GIS) with the ArcMap 10.3.1 software (https://www.esri.com). This software was used for the analyses with the shape file of the streets, parks, and squares provided by the Porto Alegre Environment, Urbanism and Sustainability Department. A 1km road network buffer was defined around the participant's residential address to estimate accessible neighborhood features 31.

The following variables were used: residential density and connectivity between streets. Residential density was calculated using the number of residential units in the area within each buffer. Connectivity between streets was assessed by the number of intersections between the streets in the buffer. These variables were transformed to standardized values (Z-scores).

Socioeconomic status

Socioeconomic status was assessed through a questionnaire, which included the number of owned items at the adolescents' residences and the level of schooling of the parent or guardian, according to the criteria established by the Brazilian Association of Research Companies (http://www.abep.org/criterio-brasil, accessed on 08/Sep/2019). Then, students answered questions about the existence and quantity of items in their home, such as car, washing machine, bathroom, among others. In addition, information was given on the origin of the house water (well or spring, general distribution register, or other means), and the street condition of his/her home (asphalted/paved or earth/gravel). For each answer, a score was attributed and the sum of the points identified each student's economic class. Then, subjects were classified into the following economic classes: A1, A2, B1, B2, C1, C2, D, E. For the analyses, the classes were grouped as high (A1+A2), middle (B1+B2), and low (C1+C2+D+E).

Statistical analysis

Descriptive data stratified by usual mode of commuting to and from school were calculated as absolute and relative values for sex, age, and socioeconomic status. Means and standard deviations (SD) were calculated for each GIS-measured objective (residential density and connectivity between street) and perceived environmental factors. These data were presented stratified by active and passive commuters. Differences between ways of commuting to school were determined through chi-square test for dichotomous variables and independent t-test for continuous variables.

The internal consistency from the variables of the NEWS-Y dimensions was verified with Cronbach's alpha (0.89), which indicated 0.89 as an acceptable reliability. Values for each dimension were land-use mix diversity (0.88), neighborhood recreation facilities (0.84), access to services (0.36), street connectivity (0.36), places for walking (0.30), neighborhood aesthetics (0.71), neighborhood safety (0.10), and crime safety (0.85).

Spearman correlation was used to determine association between GIS-measured objective and perceived environmental factors and ACS. Residential density and connectivity between streets (mediating variables) should be correlated with the outcome. Perceived environmental factors that presented association with both GIS-measured objective environmental factors and ACS were tested in mediation models.

To examine whether the association between perceived environmental factors and ACS was mediated by residential density and connectivity between streets, linear regression models were fitted using bootstrapped mediation procedures included in the PROCESS macro 32. For mediation hypotheses tests, a resampling procedure of 10,000 bootstrap samples was used. PROCESS estimates models with binary and continuous variables. Indirect effect was estimated through point estimates and 95% confidence intervals (95%CI). When the confidence interval did not contain zero, the point estimate was considered significant.

The following criteria were used to establish mediation: (1) the independent variable (perceived environmental factors) is significantly related to the mediators (equation a); (2) the mediator is significantly related to the dependent variable (ACS) (equation b); (3) the independent variable (perceived environmental factors) is significantly related to the dependent variable (ACS) (equation c); and (4) the association between the independent and dependent variable is attenuated when the mediator is included in the regression model (equation c'). The proportion mediated was calculated by the following formula: 1-(equation c'/equation c). Other important criteria are that the value of equation b must be greater than the equation c' and the value of equation c' must be smaller than the equation c.

The analyses were adjusted by sex and socioeconomic status. All analyses were carried out using the SPSS-IBM v.22.0 (https://www.ibm.com/), and a value of p < 0.05 was considered statistically significant.


Table 1 presents descriptive characteristics of the sample, stratified by usual mode of commuting to and from school, i.e., active vs. passive commuters. The adolescents (52.7% girls) attended 16 public high schools, their mean age was 16.49 (SD = 1.05) years old. The results indicated significant statistical difference between active and passive commuters by sex (χ2 = 13.97; p < 0.001). Most adolescents were middle class. Residential density and perceived environmental factors, such as land-use mix, neighborhood recreation facilities, access to services, and neighborhood aesthetics were statistically different between active and passive commuters.



Tab.: 1
Table 1 Sample's characteristics, descriptive and occurrence analysis according to usual mode of commuting to school in adolescents.


Table 2 shows correlations between GIS-measured objective and perceived environmental factors and ACS. It is observed that connectivity between streets was not associated with ACS. Thus, this variable cannot be tested as a mediator. Residential density presented positive correlation with ACS. Some perceived environmental factors (land-use mix diversity, neighborhood recreation facilities, access to services, and neighborhood aesthetics) presented correlation between both residential density and ACS.



Tab.: 2
Table 2 Correlation between active commuting to school, GIS-measured objective, and perceived environmental factors in adolescents.


The mediator role of residential density in the association between land-use mix diversity, neighborhood recreation facilities, access to services, and neighborhood aesthetics with ACS is in Figure 1. Overall, perceived environmental factors (land-use mix diversity, neighborhood recreation facilities, and access to services) were positively associated with ACS. Mediation analysis, including residential density, revealed that the association between these perceived environmental factors and ACS was mediated by residential density. The roles of this mediation for sex, socioeconomic status, region, and all perceived environmental factors were the following: residential density mediates 32% of land-use mix diversity for ACS (indirect effect [IE] = 0.114; 95%CI: 0.130, 0.311); 15% for neighborhood recreation facilities (IE = 0.064; 95%CI: 0.034, 0.105); and 14% for access to services (IE = 0.045; 95%CI: 0.006, 0.104) parameters. In Figure 1d, residential density did not mediate the relationship between neighborhood aesthetics with ACS.



Figure 1 Residential density mediation model of the relationship between land-use mix diversity (1a), neighborhood recreation facilities (1b), access to services (1c), and neighborhood aesthetics (1d) with active commuting to school in adolescents.



The main findings of this study indicate that residential density (objectively measured) and perceived environmental factors (land-use mix diversity, neighborhood recreation facilities, access to services, and neighborhood aesthetics) are associated with ACS. Results showed that residential density acts as a mediator on the ACS association with land-use mix diversity (32%), neighborhood recreation facilities (15%), and access to services (14%) in adolescents.

In the present study, adolescents that perceived a shorter distance from their home to several places - supermarkets, library, among others (land-use mix diversity) - as well as those who reported better access, nice neighborhood aesthetics, and a short distance from places to practice physical activity - sport clubs, parks, etc. (neighborhood recreation facilities) - actively commuted to school. In agreement with our results, previous studies with adolescents from Belgium, Nigeria, and New Zealand stated that land-use mix diversity 18, access to services 19, and nice scenery in the neighborhood 13 were associated with ACS. Duncan et al. 33 showed that a distance of approximately 2km was associated with the best results of physical activity related to active transport (9% to 15% increase on weekdays) in children and adolescents from New Zealand. However, distance to and from school is a very specific variable that differs in each context, even within the same country. For example, in a study conducted with Spanish adolescents 34, the limit distance for walking to school was 1,350m, so it is interesting to know what the real distance between home and school is, and what distance is the adolescent able to complete.

Studies developed with Colombian and Chilean children and adolescents indicated that ACS is associated with cultural and geographic factors, however, environmental factors were not considered 35,36. Studies in low-income countries are scarce, specially concerning the relationship between the specific context of commuting to school with environmental factors in adolescents. In the northeast of Brazil, the aspect “having places they like to go to” was reported by adolescents and positively associated with active commuting 37. In this sense, a possible explanation for our results is that adolescents are used to walking to different places in the neighborhood, which could be the reason why they actively commute to school.

GIS-measured environmental characteristics associated with ACS have increased in recent years 14,16,38. Studies in developed countries, such as United States, Spain, and Belgium, found results similar to our study, indicating that residential density was positively associated with active commuting in adolescents 14,15,18,39. On the other hand, we found that connectivity between streets was not associated with ACS. Queralt & Molina-García 39 showed that street intersection density was positively associated with active commuting in Spanish adolescents. This discrepancy may derive from the different assessment methods that were used for active commuting and the different characteristics of the countries.

Taking these relationships into account, the mediation analysis showed that the residential density variable (objectively measured) attenuated the association between three perceived environmental factors and ACS. That is, higher residential density in the neighborhood influences adolescents' perception of distance to different locations (land-use mix diversity), and places to practice physical activity, as well as better access in the neighborhood for ACS. In this sense, the magnitude of the mediation of residential density could support the hypothesis of the combined environmental measurements (objective and perceived), because higher residential density would not fully influence the relationship of perceived environmental factors (land-use mix diversity, neighborhood recreation facilities, and access to services) with ACS in adolescents. For example, it is not enough having high residential density and available infrastructure if people do not have a positive perception about the availability of such places, as well as good access in the neighborhood. It should be noted that this mediation role ranged from approximately 14% to 32% for the three investigated perceived environmental factors, which points out to the relevance of these factors for the development of public health strategies. It is important to emphasize that a systematic review indicated that studies providing combined information between perceived and objective environmental factors were unavailable in Brazil, involving adolescents 40.

Our findings could be relevant for professionals and stakeholders of the public health area. We emphasize that public and health managers, as well as transport agents and urban planning, can develop strategies to promote active commuting in the Brazilian population. Characteristics of urban design such as high-density residential locations, perception of proximity, and easy access to services can encourage students to actively commute to school and, consequently, increase their physical activity levels. Therefore, developing strategies to promote more active cities with built environment resources - as well as other structures, such as walking and running trails and bicycle paths - can have positive impacts on the health of the entire population.

Some limitations of our study should be mentioned. The transversal design of the study does not allow the determination of cause and effect. Also, we did not considered the period that the adolescents lived in that address or neighborhood, which could influence their perceptions. Moreover, the question about mode of commuting was not formally validated, although it has been proposed as one of the most appropriate measurements for determining mode of commuting to school in this population. Finally, larger buffers size was not used and adolescents' ACS time was not determined. Future proposals should include more objective measurements of this behavior 41. The strengths of the present study include a large sample of Brazilian adolescents and the use of environmental variables measured objectively and subjectively to understand how much this reflects in ACS. Finally, as far as we know, this is the first study using mediation analysis to understand the role of residential density in the association of perceived environment factors with ACS of Brazilian adolescents. Future studies should try to look for causal inferences and remember that understanding the impact of improvements in the infrastructure of built environment, and resources and incentives in the neighborhood or school can positively influence parents and adolescents' perception of the environment, thus increasing the prevalence of ACS. Another important issue is to know the reasons that lead adolescents to actively commute to school. Addressing environmental, social, individual, and political factors can also help elucidate these relationships.

In conclusion, residential density is a mediator on the relationship between perceived environmental factors and ACS in Brazilian adolescents. The connectivity between streets did not correlate with ACS, so it was not tested in the mediation model. Our findings also emphasize that self-reported environmental data, such as land-use mix diversity, neighborhood recreation facilities, and access to services seem to be important factors for ACS in Brazilian adolescents, considering residential density.


Scholarship from the Brazilian Graduate Studies Coordinating Board (CAPES). Research fellowship from the Brazilian National Research Council (CNPq; 305200/2013-5). J. Mota was supported by grants from the Portuguese Foundation for Science and Technology (FCT; SFRH/BSAB/142983/2018 and UID/DTP/00617/2019), as well as Scholarship Program of Santander Universities 2018.


1.   MacPherson M, de Groh M, Loukine L, Prud'Homme D, Dubois L. Prevalence of metabolic syndrome and its risk factors in Canadian children and adolescents: Canadian Health Measures Survey Cycle 1 (2007-2009) and Cycle 2 (2009-2011). Health Promot Chronic Dis Prev Can 2016; 36:32-40.
2.   Tarp J, Child A, White T, Westgate K, Bugge A, Grøntved A, et al. Physical activity intensity, bout-duration, and cardiometabolic risk markers in children and adolescents. Int J Obes 2018; 42:1639-50.
3.   Larouche R, Saunders TJ, Faulkner GEJ, Colley R, Tremblay M. Associations between active school transport and physical activity, body composition, and cardiovascular fitness: a systematic review of 68 studies. J Phys Act Health 2014; 11:206-27.
4.   Chillón P, Ortega FB, Ruiz JR, Veidebaum T, Oja L, Mäestu J, et al. Active commuting to school in children and adolescents: an opportunity to increase physical activity and fitness. Scand J Soc Med 2010; 38:873-9.
5.   Pizarro AN, Schipperijn J, Andersen HB, Ribeiro JC, Mota J, Santos MP. Active commuting to school in Portuguese adolescents: using PALMS to detect trips. J Transp Health 2016; 3:297-304.
6.   Lubans DR, Boreham CA, Kelly P, Foster CE. The relationship between active travel to school and health-related fitness in children and adolescents: a systematic review. Int J Behav Nutr Phys Act 2011; 8:5.
7.   Ramírez-Vélez R, García-Hermoso A, Agostinis-Sobrinho C, Mota J, Santos R, Correa-Bautista JE, et al. Cycling to school and body composition, physical fitness, and metabolic syndrome in children and adolescents. J Pediatr 2017; 188:57-63.
8.   Pizarro AN, Ribeiro JC, Marques EA, Mota J, Santos MP. Is walking to school associated with improved metabolic health? Int J Behav Nutr Phys Act 2013; 10:12.
9.   Chillón P, Evenson KR, Vaughn A, Ward DS. A systematic review of interventions for promoting active transportation to school. Int J Behav Nutr Phys Act 2011; 8:10.
10.   Villa-González E, Barranco-Ruiz Y, Evenson KR, Chillón P. Systematic review of interventions for promoting active school transport. Prev Med 2018; 111:115-34.
11.   Pang B, Kubacki K, Rundle-Thiele S. Promoting active travel to school: a systematic review (2010-2016). BMC Public Health 2017; 17:638.
12.   Sallis JF, Cervero RB, Ascher W, Henderson KA, Kraft MK, Kerr J. An ecological approach to creating active living communities. Annu Rev Public Health 2006; 27:297-322.
13.   Mandic S, Leon de la Barra S, García Bengoechea E, Stevens E, Flaherty C, Moore A, et al. Personal, social and environmental correlates of active transport to school among adolescents in Otago, New Zealand. J Sci Med Sport 2015; 18:432-7.
14.   Molina-García J, García-Massó X, Estevan I, Queralt A. Built environment, psychosocial factors and active commuting to school in adolescents: clustering a self-organizing map analysis. Int J Environ Res Public Health 2019; 16:83.
15.   Carlson JA, Saelens BE, Kerr J, Schipperijn J, Conway TL, Frank LD, et al. Association between neighborhood walkability and GPS-measured walking, bicycling and vehicle time in adolescents. Health Place 2015; 32:1-7.
16.   Kerr J, Rosenberg D, Sallis JF, Saelens BE, Frank LD, Conway TL. Active commuting to school: associations with environment and parental concerns. Med Sci Sports Exerc 2006; 38:787-94.
17.   Huertas-Delgado FJ, Mertens L, Chillon P, Van Dyck D. Parents' and adolescents' perception of traffic- and crime-related safety as correlates of independent mobility among belgian adolescents. PLoS One 2018; 13:e0204454.
18.   De Meester F, Van Dyck D, De Bourdeaudhuij I, Deforche B, Cardon G. Does the perception of neighborhood built environmental attributes influence active transport in adolescents? Int J Behav Nutr Phys Act 2013; 10:38.
19.   Oyeyemi AL, Ishaku CM, Deforche B, Oyeyemi AY, De Bourdeaudhuij I, Van Dyck D. Perception of built environmental factors and physical activity among adolescents in Nigeria. Int J Behav Nutr Phys Act 2014; 11:56.
20.   Dias AF, Gaya AR, Brand C, Pizarro AI, Fochesatto CF, Mendes TM, et al. Distance from home to the nearest park and the use of the parks for physical activity: the mediator role of road safety perception in adolescents. Public Health 2019; 168:9-16.
21.   Kärmeniemi M, Lankila T, Ikäheimo T, Koivumaa-Honkanen H, Korpelainen R. The built environment as a determinant of physical activity: a systematic review of longitudinal studies and natural experiments. Ann Behav Med 2018; 52:239-51.
22.   Bauman AE, Reis RS, Sallis JF, Wells JC, Loos RJF, Martin BW. Correlates of physical activity: why are some people physically active and others not? Lancet 2012; 380:258-71.
23.   Burgos MS, Tornquist D, Tornquist L, Reuter CP, Garcia EL, Renner JDP, et al. Cardiometabolic risk factors associated with active commuting to school. Rev Paul Pediatr 2019; 37:181-7.
24.   Silva KS, Nahas MV, Borgatto AF, Oliveira ES, Del Duca GF, Lopes AS. Factors associated with active commuting to school and to work among Brazilian adolescents. J Phys Act Health 2011; 8:926-33.
25.   Gaya ACA, Garlipp DC, Silva MF, Moreira RB. Ciências do movimento humano: introdução à metodologia da pesquisa. Porto Alegre: Artmed; 2008.
26.   Sawyer SM, Azzopardi PS, Wickremarathne D, Patton GC. The age of adolescence. Lancet Child Adolesc Health 2018; 2:223-8.
27.   Herrador-Colmenero M, Perez-Garcia M, Ruiz JR, Chillon P. Assessing modes and frequency of commuting to school in youngsters: a systematic review. Pediatr Exerc Sci 2014; 26:291-341.
28.   Rosenberg D, Ding D, Sallis JF, Kerr J, Norman GJ, Durant N, et al. Neighborhood Environment Walkability Scale for Youth (NEWS-Y): reliability and relationship with physical activity. Prev Med 2009; 49:213-8.
29.   Lima AV, Rech CR, Reis RS. Equivalência semântica, de itens e conceitual da versão brasileira do Neighborhood Environment Walkability Scale for Youth (NEWS-Y). Cad Saúde Pública 2013; 29:2547-53.
30.   Sallis JF. Scoring for the Neighborhood Environment Walkability Scale - Youth (NEWS-Y). http://sallis.ucsd.edu/Documents/Measures_documents/NEWS_Y_scoring.pdf (accessed on 08/Sep/2019).
31.   Sallis JF, Cerin E, Conway TL, Adams MA, Frank LD, Pratt M, et al. Physical activity in relation to urban environments in 14 cities worldwide: a cross-sectional study. Lancet 2016; 387:2207-17.
32.   Preacher KJ, Hayes AF. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods 2008; 40:879-91.
33.   Duncan S, White K, Mavoa S, Stewart T, Hinckson E, Schofield G. Active transport, physical activity, and distance between home and school in children and adolescents. J Phys Act Health 2016; 13:447-53.
34.   Rodríguez-López C, Salas-Fariña ZM, Villa-González E, Borges-Cosic M, Herrador-Colmenero M, Medina-Casaubón J, et al. The threshold distance associated with walking from home to school. Health Educ Behav 2017; 44:857-66.
35.   Rodríguez-Rodríguez F, Jara OP, Kuthe NM, Herrador-Colmenero M, Ramírez-Vélez R, Chillón P. Influence of distance, area, and cultural context in active commuting: continental and insular children. PLoS One 2019; 14:e0213159.
36.   Ramírez-Vélez R, Beltrán CA, Correa-Bautista JE, Vivas A, Prieto-Benavidez DH, Martínez-Torres J, et al. Factors associated with active commuting to school by bicycle from Bogotá, Colombia: the FUPRECOL study. Ital J Pediatr 2016; 42:97.
37.   Mendonça G, Florindo AA, Rech CR, Freitas DKS, Farias Júnior JC. Perceived neighborhood environmental characteristics and different types of physical activity among Brazilian adolescents. J Sports Sci 2017; 36:1068-75.
38.   Wong BY-M, Faulkner G, Buliung R. GIS measured environmental correlates of active school transport: a systematic review of 14 studies. Int J Behav Nutr Phys Act 2011; 8:39.
39.   Queralt A, Molina-García J. Physical activity and active commuting in relation to objectively measured built-environment attributes among adolescents. J Phys Act Health 2019; 16:371-4.
40.   Manta SW, Silva KS, Minatto G, Lopes MVV, Mello GT, Barbosa Filho VC. Community and environment for physical activity among young people: a systematic review of the Report Card Brazil 2018. Rev Bras Cineantropom Desempenho Hum 2018; 20:543-62.
41.   Villa-González E, Rosado-López S, Barranco-Ruiz Y, Herrador-Colmenero M, Cadenas-Sánchez C, Santos MP, et al. Objective measurement of the mode of commuting to school using GPS: a pilot study. Sustainability 2019; 11:5395.

This is an open-access article distributed under the terms of the Creative Commons Attribution License


Cadernos de Saúde Pública | Reports in Public Health

Rua Leopoldo Bulhões 1480 - Rio de Janeiro RJ 21041-210 Brasil

Secretaria Editorial +55 21 2598-2511.

  • APOIO:

©2015 | Cadernos de Saúde Pública - Escola Nacional de Saúde Pública Sergio Arouca | Fundação Oswaldo Cruz. - Ministério da Saúde Governo Federal | Desenvolvido por Riocom Design