Health Promotion International Advance Access published online on June 15, 2007
Health Promotion International, doi:10.1093/heapro/dam015
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Article |
Use of local area facilities for involvement in physical activity in Canada: insights for developing environmental and policy interventions
1 GRIS – Groupe de recherche interdisciplinaire en santé (Interdisciplinary Research Group on Health) 2 Department of Social and Preventive Medicine 3 Faculty of Nursing, University of Montreal, PO Box 6128, Downtown station, Montreal, Quebec, Canada H3C 3J7 4 The Léa-Roback Research Centre on Social Inequalities of Health of Montreal, Canada
* Corresponding author. E-mail: mylene.riva{at}umontreal.ca
 |
SUMMARY |
|---|
 TOP SUMMARY INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Despite growing literature showing associations of availability and accessibility of facilities to greater levels of physical activity, considerably less is known about the actual extent of use of these facilities. The purpose of this study was to examine the individual (sex, age, education and extent of involvement in vigorous physical activity) and local area characteristics (socioeconomic status, locations and number of physical activity organizations per 1000 residents) associated with the use of local facilities for involvement in physical activity. A telephone survey was conducted with 3191 randomly selected adults in 22 non-contiguous areas across Canada. Use of local facilities for involvement in physical activity was examined among a subset of 1006 physically active adults. Data were analyzed using multilevel modeling. Findings revealed significant variation across areas in likelihood of use of local facilities among women but not men. Women in the 25–34 and 45–55 age categories were significantly more likely to use local facilities than women of 35–44 years of age. Women reporting greater levels of involvement in vigorous physical activity were more likely to use local area facilities. Higher area affluence and living in areas located in small urban towns were associated with greater use of local facilities among women only. None of the individual and local area characteristics was associated with the outcome among men. Understanding the processes associated with differential use of local area facilities for physical activity is essential for the implementation of effective environmental and policy interventions to increase physical activity in the population.
Key words: use of physical activity facilities; multilevel analyses; small area analysis
|
INTRODUCTION |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
The role of environmental and policy interventions to promote physical activity is emerging as a critical component of an overall strategy to increase physical activity levels in the population (Task Force on Community Preventive Services, 2002
; Brownson et al., 2006; Sallis et al., 2006). Several studies demonstrate that greater availability and accessibility to exercise facilities and to environments conducive to physical activity are associated with more frequent and lengthier involvement in physical activity (Brownson et al., 2001; Giles-Corti and Donovan, 2002a,b; Huston et al., 2003; Fisher et al., 2004; Wendel-Vos et al., 2004; Li et al., 2005; van Lenthe et al., 2005). Greater availability to parks and to open green spaces have been associated with greater levels of walking (Giles-Corti and Donovan, 2002a; Fisher et al., 2004; Li et al., 2005), cycling (Wendel-Vos et al., 2004) and overall physical activity (Giles-Corti and Donovan, 2002b; Huston et al., 2003). One study reported the association of proximity to sport and recreational facilities to sport participation but not to walking or cycling (van Lenthe et al., 2005). These studies draw attention towards accessibility to facilities and to environmental activity-friendliness as viable targets for population-based interventions promoting regular involvement in physical activity (Sallis et al., 1998). In this regard, results of a systematic review of physical activity interventions shows that people are more likely to become physically active if exposed to improved access to places to engage in physical activity (Task Force on Community Preventive Services, 2002), thus suggesting that environmental and policy interventions represent a viable means of promoting increased levels of physical activity.
However, the presence and availability of facilities for physical activity does not necessarily entail that facilities will be used. Despite the growing literature showing associations between availability and accessibility of facilities and greater levels of physical activity, considerably less is known about the actual extent of use of these facilities. One group of researchers examined the use of local area resources for recreational physical activity among Australian adults (Giles-Corti and Donovan, 2002a,b). Results indicated that among those reporting exercising vigorously, 100% reported using at least one location or facility near their home; this proportion was only 40.6% among light to moderate exercisers. Furthermore, the likelihood of using local area resources for recreational physical activity was substantially lower for respondents living in areas characterized by populations with low socioeconomic status (SES). Obtaining information on the use of physical activity facilities and proximity of use to the home is critical since public policies and environmental interventions to increase physical activity are often place-based entities and likely to be more far-reaching if implemented in the locales wherein people are most likely to use them.
Towards this end, the purpose of this study was to examine the individual and local area characteristics associated with the use of local facilities for engaging in physical activity among a sample of physically active adults. Facilities were defined as locations where physical activity programs and services are offered such as gyms, pools and fitness classes. Facilities for physical activity differ from environments that are conducive to activity. Facilities for physical activity are organizational structures with a method of operation that allows for delivering physical activity programs and services. In contrast environments that are conducive to physical activity refer to locations such as streets and parks where people may engage in leisure physical activity as well as alternative forms of physical activity including travel and occupational activities. These locales are typically devoid of an organizational structure for dispensing programs and services.
|
METHODS |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Design
Data reported in this study are part of a larger project designed to study public health infrastructures, policies and practices for the promotion of physical activity in Canada (Gauvin et al., 2001
). Telephone interviews were conducted in the fall of 2000 with 3191 adults living in 22 local areas in three Canadian provinces (Alberta, Ontario and Quebec). Interviewers were trained by members of the research team to appropriately address questions to respondents. Respondents were asked to report on their involvement in physical activity (frequency and duration) and their use of facilities in their local area to engage in physical activity. Participants also provided socio-demographic information. To insure quality control, members of the research team monitored about 25% of interviews. Response rates for the overall sample were about 75% (based on acceptance rates of persons answering the phone) but were lower (61%) when including total number of telephone calls dialed (including calls with completed interviews, refusals, no answers, absent upon follow-up call).
Sample
Sampling of local areas
To obtain a broad representation of Canadian residential local areas, 22 areas located in large urban, small urban, suburban and rural centers, were sampled. Local areas in large urban residential areas were selected in cities with a population of more than 500 000 inhabitants; local areas were delimited by municipality-identified boroughs. Local areas were selected in suburban cities that were located on the periphery of the large urban areas, and in small urban cities with populations <100 000 inhabitants located more than 100 km from the large urban center. The suburban and small urban local areas were delimited by city limits. Finally, rural local areas were selected in cities with a population <10 000 inhabitants located in agricultural areas and were delimited by the contours of the city and township. Overall, 13 local areas were sampled in large urban centers, and a further three local areas were sampled in each of the suburban, small urban and rural environments. Population size of the local areas ranged between 1 604 in one rural local area to 93 442 in one small urban local areas, for an average population of 25 588. Data on the socioeconomic profile of the local areas were obtained from the 1996 Canadian census.
Sampling of persons within local areas
In order to insure accurate representation of persons, within each of the 22 local areas, participants were randomly sampled across high, average and low affluence census-defined areas (census tracts), through the application of specialized computer software, which allows linking telephone numbers to census-defined areas. Criteria for inclusion in the sample was age (between 25 and 55 years old), having lived at their current address for at least 12 months and being able to respond to a series of questions in either French or English.
Study population
For the purpose of this study, use of local facilities for involvement in physical activity was examined among a subsample of active adults because physically inactive persons evidently do not use any facilities for involvement in physical activity. Our interest was in identifying differential patterns of local facility use among persons who would necessarily use at least one facility to be physically active. Active individuals were defined as those meeting minimum requirement for involvement in vigorous physical activity and thus those persons who would require access to some type of physical activity facility. Extent of involvement in vigorous physical activity was measured by the question: ‘Over the past three months, have you been involved in one or many of the following activities: cycling, swimming, jogging/running, cross-country skiing, tennis, ice hockey, fitness classes, racquetball, squash?’ Response options were ‘yes’ and ‘no’. Individuals who responded ‘yes’ were asked questions about the duration and the frequency of involvement in these activities. Those reporting involvement in vigorous physical activity between 15 and 30 min three times or more per week and those reporting 30–60 min or over an hour of activity two times or more per week were selected for this study. This measure was adapted from the Canadian Community Health Survey in which questions on participation, duration, and frequency of involvement were asked for each enumerated activity.
Overall, 31.6% of the sampled individuals were considered physically active. Among Canadians aged between 20 and 64 years old, 41% of women and 42% of men were considered physically active in their leisure time (Statistics Canada, 2002; www.statcan.ca). Discrepancies between our sample and the Canadian population may be explained by different thresholds for levels of involvement in physical activity.
Measures
Dependent variable
Use of local area facilities for involvement in physical activity was assessed by asking respondents whether or not they used programs, services or facilities located in their local area to participate in vigorous physical activities (i.e. cycling, swimming, jogging/running, cross-country skiing, tennis, ice hockey, fitness classes, racquetball, squash).
Independent individual characteristics
Given the dearth in knowledge on individual correlates associated with the use of facilities for engaging in physical activity, selected individual variables were measured based on their known association with involvement in physical activity (Sallis and Owen, 1999). Age was recoded in three categories: 25–34, 35–44 and 45–55 years, from which dummy variables were created. Individuals' SES was measured by educational attainment as individual income contained a large proportion of missing data (25.5%). Educational attainment was recorded into four categories, from which dummy variables were created: those who did not complete high school, those who completed high school and had a few years of college; those who completed college and had a few years of university; and those who held a university degree. Finally, dummy variables were created to distinguish between lower (70–90 min per week), average (135–180 min per week) and higher volume vigorous exercisers (270–360 min per week).
Forty-nine percent of the sample were women (average age = 38.4 years), and 51% were men (average age = 37.6 years). More than 40% held a university degree. Overall, 53.8% of the respondents reported using local facilities for engaging in physical activity.
Independent local area characteristics
Three characteristics of residential environments were measured: SES, location and number of physical activity facilities per 1000 inhabitants. Local area SES was measured by the average household income. Local areas were categorized into three groups: those having an average household income below $40 000; those between $40 000 and $60 000; and those having an average household income above $60 000. Local areas were situated in four types of communities: in large urban centers, or in either small urban, suburban and rural cities, from which dummy variables were created. As part of the larger research project (Gauvin et al., 2000), a list of organizations offering physical activity programs and services to the adult population in each of the 22 local areas was compiled by searching the phone book and the internet, by word of mouth and by conducting walking-tours of each area. A ratio of number of facilities per 1000 inhabitants was computed and modeled as a continuous variable.
Average household income in the residential areas varied between $25 241 and $98 940 (Canadian dollars), with an average of $52 919. On average, there was one physical activity facility per 1000 residents; however, a majority of local areas (n = 16) had less than one physical activity facility per 1000 residents.
Statistical analyses
Given the hierarchical structure of the dataset, i.e. individuals nested within local areas, multilevel analyses were conducted using HLM software (Raudenbush et al., 2001). The model building followed a step-up approach as suggested by Raudenbush and Bryk (2002). The first model, in which no predictor variables was specified, allowed to explore whether or not there were variations between local areas in likelihood of local facility use (model 1). In the second model, individual correlates were examined. In the third model, the association of local area characteristics on the likelihood of use of local facilities was examined without adjusting for individual-level correlates. Finally, the fourth model adjusted for both individual and local area characteristics. Non-linear Bernoulli analyses for a dichotomous outcome variable were used.
As previous results indicate that individual correlates of physical activity differ across women and men (Bauman et al., 2002; Humpel et al., 2004a,b; Suminski et al., 2005), models without intercepts were specified to allow for the simultaneous estimation of separate equations for women and men (Barnett et al., 1995; Raudenbush and Bryk, 2002). Results are reported in Table 1 for women and in Table 2 for men.
|
|
Data are reported for a sample of 1006 active adults nested within 22 local areas. Within area samples varied between 20 respondents in one rural local areas to 74 respondents in one suburban area. Average within area sample comprised 46 individuals.
|
RESULTS |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Local area variation in the likelihood of use of local facilities for involvement in physical activity
Results for local area variation are reported in Figure 1 and in model 1 (Tables 1 and 2). Findings showed significant between-area variation in the likelihood of use of local facilities among women [
2(21) = 47.3; P < 0.001] but not among men [2(21) = 26.7; P = 0.18]. Among women the average probability of using local facilities for physical activity was 55.8%. Computation of the 95% plausible value range (Raudenbush and Bryk, 2002) indicated that this probability varied between 32.5 and 76.9% across areas. Among men, the average probability of using local facilities was 51.1%; this probability did not differ significantly across areas. For this reason, the random effect of men was set as fixed (i.e. zero) across residential areas.
|
Individual characteristics associated with the likelihood of use of local facilities for physical activity
Among men, none of the selected individual characteristics was significantly associated with the likelihood of using local facilities for physical activity.
Compared with middle aged women, younger (25–34 years) and older women (45–55 years) were more likely to use local facilities for physical activity (younger women: OR = 1.58; 95% CI: 1.01, 2.47; older women: OR = 1.69; 95% CI: 1.05, 2.78) (Model 2, Table 1). Yet, the difference in facility use between younger and older women was not statistically significant [2(1) = 0.06; P > 0.50; results not shown]. Women reporting average and higher involvement in vigorous physical activity were more likely to use facilities to engage in physical activity than lower exercisers [the difference in facility use between average and high exercisers was not statistically significant; 2(1) = 0.05; P > 0.50; results not shown]. Educational attainment was not significantly associated with likelihood of facility use; nonetheless it was kept in the final model for control purposes. After introducing individual characteristics, there remained significant random between-areas variation in women's likelihood of use of local facilities for physical activity.
Local area characteristics associated with the likelihood of use of local facilities for involvement in physical activity
Effects of local area SES, location and the number of physical activity organizations per 1000 residents were modeled simultaneously, in models unadjusted (model 3) and adjusted for individuals' variables (model 4).
In unadjusted models, significant associations between local area characteristics and likelihood of using local facilities for physical activity were observed among women only. Women living in local areas located in small urban areas were significantly more likely to use facilities in their local area for involvement in physical activity than women residing elsewhere (OR = 2.62; 95% CI: 1.12, 6.16). In more affluent local areas, women were significantly more likely to use local facilities (OR = 1.93; 95% CI: 1.01, 3.66). The number of physical activity organizations per 1000 residents was not significantly associated with the outcome. These effects remained statistically significant in models adjusting for individuals' characteristics (model 4, Table 1).
After introducing local area characteristics, both in models unadjusted and adjusted for individuals' characteristics, the between area variation in women's likelihood of use of local facilities for physical activity was no longer statistically significant [unadjusted model: 2(15) = 23.16; P = 0.08; adjusted model: 2(15) = 20.76; P = 0.15].
|
DISCUSSION |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Although availability of resources for physical activity has been associated with greater levels of physical activity, availability of such resources is not an indicator of their use. For this reason, the aim of this study was to examine the association between individual and area characteristics with the use of local facilities for involvement in physical activity among active adults.
Results showed that although women and men overall appeared to use facilities for physical activity in the same proportion, the predicted probability of using local facilities for physical activity varied significantly across areas, but only among women. These data suggest that the determinants of access to facilities may be different for men and women.
Furthermore, among women, individual-level characteristics did not account for the between-area variation which was explained by selected local area characteristics. Living in more affluent areas and in areas located in small urban centers was associated with a higher likelihood of using local facility for engaging in physical activity. These associations remained statistically significant after accounting for the effect of individual characteristics.
Results of the study also indicate gender differences in the correlates of use of facilities: among men, none of the individual and local area characteristics were significantly associated with the outcome. Previous research supports the existence of gender differences in the correlates of physical activity (Bauman et al., 2002; Humpel et al., 2004a,b; Suminski et al., 2005). Other studies showed gender and socioeconomic differences in the perception of accessibility to places to exercise (Browning et al., 2001; Giles-Corti and Donovan, 2002a). With respect to using local facilities for physical activity, objectively assessing area characteristics and characteristics of facilities for physical activity may contribute to better explaining the observed gender discrepancies.
Individual SES might be an important characteristic associated with facility use since some of them may have a pay-for-use access. Although no associations were observed between the individual measure of SES (educational attainment) and local facility use, there appears to be a gradient in facility use, with those with lower educational attainment being less likely to use local facilities for physical activity than those with higher education. Associations between the likelihood of use of local facilities for physical activity and other measures of individual SES might have yielded different results.
Interestingly, no association was observed between the number of physical activity facilities and the likelihood of use of local facilities for physical activity. Although recent findings support the association between availability and accessibility of resources and involvement in physical activity, these findings are nonetheless inconsistent as associations appear to vary depending on the type of facilities and physical activity levels measured (Wendel-Vos et al., 2004; van Lenthe et al., 2005).
Results support those of Giles-Corti and Donovan (Giles-Corti and Donovan 2002a,b) who observed significant associations between-area SES and using facilities for physical activity. However, whereas these authors observed a lower likelihood of using facilities for physical activity in less-affluent areas, in the current study, the opposite was observed, i.e. a greater use of local facilities was more likely in more affluent local areas, though only among women.
Limitations
The current findings should be interpreted in light of certain limitations. First, the small number of local areas limits the accuracy of parameter estimates. Since there was random between-area variation in the likelihood of facility use among women, a larger sample of local areas would have yielded more accurate estimates (Raudenbush and Bryk, 2002). In future investigations, the intricacies of clustered designs should be examined more explicitly to determine the optimal number of areas and persons within areas to sample (Raudenbush, 1997).
In this study, we asked respondents whether or not they had used facilities for physical activity located in their neighborhood. As such, we relied on residents perceptions of their neighborhood boundaries which may or may not correspond to the boundaries of the local areas defined for the purpose of this study. Future studies should strive to define neighborhood boundaries and explore the extent to which different contours do or do not correspond to residents' perceptions (Coulton et al., 2001; Galster, 2001; Gauvin et al., in press). In addition, systematically and objectively assessing local area characteristics and characteristics of the local physical activity delivery system may contribute to better understanding the correlates of greater facility use for physical activity among both women and men.
The cross-sectional nature of the data limits the ability to establish whether or not local area characteristics are the catalyst for greater involvement in physical activity and whether or not persons who are involved in greater amounts of physical activity choose to live in an environment that has more physical activity resources. In other words, little can be said about the directionality of the associations observed. Despite this limitation, the current findings represent a first step in ascertaining the individual and ecological correlates of use of local facilities for physical activity.
Further research is needed to disentangle the correlates of the use of facilities for physical activity among both women and men, among individuals of differential SES and across different residential environments.
|
CONCLUSION |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Findings reported herein are novel in that the design of the study allowed for the joint examination of individual and local area characteristics associated with the likelihood of use of local facilities for physical activity. What seems to emerge from the results is that individual and ecological correlates of local facility use differ across women and men. Furthermore, selected local area characteristics explained the between area variations in outcome which suggests that characteristics of the built and social environment might be crucial in investigating local facility use for physical activity over and above individual characteristics.
Results reported in this paper indicate that both individual and local area characteristics are associated with the likelihood of using local facilities to engage in physical activity, but that the correlates differ across women and men. This suggests that policies and environmental interventions to promote physical activity might be more effective if a gender-specific approach is adopted. Also, using facilities for physical activity appears to be influenced by local area characteristics, independent of the characteristics of local population. This implies that environmental and policy interventions aimed at creating opportunities for engaging in physical activity, in conjunction with individual-oriented interventions, holds promise in increasing activity levels of population.
|
ACKNOWLEDGEMENT |
|---|
This study was funded by National Health Research and Development Program (grant no. R6605-5241-004). M.R. is a recipient of a Canada Graduate Scholarships Doctoral Award from the Canadian Institutes of Health Research (CGD-16386).
|
REFERENCES |
|---|
|
TOPSUMMARYINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONREFERENCES |
|---|
Barnett R. C., Raudenbush S. W., Brennan R. T., Pleck J. H., Marshall N. L. Change in marital experiences and change in psychological distress: A longitudinal study of dual-earner couples. Journal of Personality and Social Psychology (1995) 69:839–850.[CrossRef][Web of Science][Medline]
Bauman A. E., Sallis J. F., Dzewaltowski D. A., Owen N. Towards a better understanding of the influences on physical activity: the role of determinants, correlates, causal variables, mediators, moderators, confounders. American Journal of Preventive Medicine (2002) 23:5–14.[CrossRef][Web of Science][Medline]
Brownson R. C., Baker E. A., Houseman R. A., Brennan L. K., Bacak S. J. Environmental and policy determinants of physical activity in the United States. American Journal of Public Health (2001) 91:1995–2003.
Brownson R. C., Haire-Joshu D., Luke D. A. Shaping the context of health: a review of environmental and policy approaches in the prevention of chronic diseases. Annual Review of Public Health (2006) 27:341–370.[CrossRef][Web of Science][Medline]
Coulton C. J., Korbin J., Chan T., Su M. Mapping resident's perceptions of neighborhood boundaries: a methodological note. American Journal of Community Psychology (2001) 29:371–383.[CrossRef][Web of Science][Medline]
Fisher K. J., Li F., Michael Y., Cleveland M. Neighborhood-level influences on physical activity among older adults: a multi-level analysis. Journal of Aging and Physical Activity (2004) 12:45–63.[Web of Science][Medline]
Galster G. On the nature of neighbourhood. Urban Studies (2001) 38:2111–2124.
Gauvin L., Richard L., Potvin L., Craig C. L., Quinney H. A., Spence J. C., et al. Final Report Project R6605–5241-004. In: Public health infrastructure, policies, and practices for the promotion of physical activity in Canada (2001) Ottawa, Ont: National Health Research and Development Project (NHRDP).
Gauvin L., Robitaille E., Riva M., McLaren L., Dassa C., Potvin L. Conceptualizing and operationalizing neighbourhoods: the conundrum of identifying territorial units. Canadian Journal of Public Health. in press.
Giles-Corti B., Donovan R. J. Socio-economic status differences in recreational physical activity levels real perceived access to a supportive physical environment. Preventive Medicine (2002a) 35:601–611.[CrossRef][Web of Science][Medline]
Giles-Corti B., Donovan R. J. The relative influence of individual, social and physical environment determinants of physical activity. Social Science and Medicine (2002b) 54:1793–1812.[CrossRef]
Humpel N., Owen N., Iverson D., Leslie E., Bauman A. Perceived environment attributes, residential location, and walking for particular purposes. American Journal of Preventive Medicine (2004a) 26:119–125.[CrossRef][Web of Science][Medline]
Humpel N., Owen N., Leslie E., Marshall A. L., Bauman A. Associations of location and perceived environmental attributes with walking in neighbourhoods. American Journal of Health Promotion (2004b) 18:239–242.[Web of Science][Medline]
Huston S. L., Evenson K. R., Bors P., Gizlice Z. Neighborhood environment, access to places for activity, and leisure-time physical activity in a diverse North Carolina population. American Journal of Health Promotion (2003) 18:58–69.[Web of Science][Medline]
Li F., Fisher K. J., Brownson R. C., Bosworth M. Multilevel modelling of built environment characteristics related to neighbourhood walking activity in older adults. Journal of Epidemiology and Community Health (2005) 59:558–564.
Raudenbush S. W. Statistical analysis and optimal design for cluster randomized trials. Psychological Methods (1997) 2:173–185.[CrossRef][Web of Science]
Raudenbush S. W., Bryk A. Hierarchical Linear Models: Applications and Data Analysis Methods (2002) 2nd edition. Newbury Park, CA: Sage.
Raudenbush S. W., Bryk A., Congdon R. HLM: Hierarchical Linear and Nonlinear Modeling (2001) Chicago, IL: Scientific Software International.
Sallis J. F., Bauman A., Pratt M. Environmental and policy interventions to promote physical activity. American Journal of Preventive Medicine (1998) 15:379–397.[CrossRef][Web of Science][Medline]
Sallis J. F., Cervero R. B., Ascher W., Henderson K. A., Kraft M. K., Kerr J. An ecological approach to creating active living communities. Annual Review of Public Health (2006) 27:297–322.[CrossRef][Web of Science][Medline]
Sallis J. F., Owen N. Physical Activity and Behavioral Medicine (1999) Thousand Oaks, CA: Sage.
Suminski R. R., Carlos Poston W. S., Petosa R. L., Stevens E., Katzenmoyer L. M. Features of the neighborhood environment and walking by U.S. adults. American Journal of Preventive Medicine (2005) 28:149–155.[CrossRef][Web of Science][Medline]
Task Force on Local area preventive Services. Recommendations to increase physical activity in local areas. American Journal of Preventive Medicine (2002) 22:67–72.[CrossRef][Web of Science][Medline]
van Lenthe F. J., Brug J., Mackenbach J. Neighbourhood inequalities in physical inactivity: the role of neighbourhood attractiveness, proximity to local facilities and safety in the Netherlands. Social Science and Medicine (2005) 60:763–775.[CrossRef]
Wendel-Vos G. C. W., Sschuit A. J., De Niet R., Boshuizen H. C., Saris W. H. M., Kromhout D. Factors of the physical environment associated with walking and bicycling. Medicine and Science in Sports and Exercise (2004) 36:725–730.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||