T H E

S

C I E N C E

O F

L I F E S T Y L E

C H A N G E

Community Design

Evaluating the Implementation and Active Living Impacts of a State Government Planning Policy Designed to Create Walkable Neighborhoods in Perth, Western Australia Paula Hooper, PhD; Billie Giles-Corti, PhD; Matthew Knuiman, PhD

Abstract

INTRODUCTION

Purpose. Evaluate the implementation of a government planning policy (Liveable Neighbourhoods Guidelines) and its impacts on residents ' walking behaviors. Design. Cross-sectional study of participants from the RESIDential Environments project (RESIDE). Setting. Nineteen ''liveable' and 17 "conventionally designed' housing developments across Perth, Western Australia. Subjects. Five hundred ninety-four participants from. RESIDE who resided in 36 housing developments. Measures. Developed in geographic information systems to a.sse.ss the on-ground implementation of 43 policy requirements. Policy compliance was defined as the degree to which construction of the developments adhered to the standards outlined. Walking behaviors were measured using the Neighborhood Physical Activity Questionnaire. K-means cluster analyses identified groups C)f homogeneous develofimenls luith respect to policy implementation. Analysis. Logistic regression xoith generalized estimating equations estimated the odds ratios (ORs) and 95 % confidence intervals (95 % CIs)for the likelihood of undertaking any and > 60 minutes of transport and recreational walking associated with (1) policy compliance and (2) different clusters of developments. Results. There were few significant differences in on-ground outcomes between the tiuo development types. Despite incomplete implementation, the odds of walking for transport increased with overall levels of policy compliance (0R= 1.53, 95% CI 1.13-2.08) and compliance with the community design (0R = 1.3, 95% CI 1.13-1.42), movement netwmk (0R= 2.49, 95% CI 1.38-4.50), and lot layout elements (0R= 1.26, 95% CI 1.06-1.50). Conclusion. Consistent with the aims of the policy, residents in walkable (i.e., liveable) neighborhoods may be more physically active. (Am J Health Promot 2014;28[3s]:S5-S18.) Key Words: Built Environment, Liveable Neighborhoods, New Urbanism, Perth, Walking, Planning Policy, Prevention Research. Manuscript format: research; Research ptirpose: intervention testing/program evaluation; Study design: quasiexperimental; Outcome measure: behavioral; Setting: local community, state/ national; Health focus: Fitness/physical activity; Strategy: policy, built environment; Target population: adults; Target population circvimstances: geographic location

Government policy and planning initiatives determine the way in which cities and towns are built and therefore play a vital role in creating neighborhoods where residents can safely and conveniently be physically active. As planning policies are implemented, opportunities for natural experiments are emerging to monitor the impact of these built environmental interventions on active living behaviors''^ and provide more rigorous evidence than the cross-sectional sttidies that dominate this literature. To date, however, few physical activity studies have been designed to measure the implementation of planning policies aimed at creating environments more conducive to active living behaviors. In Febrtiary 1998, the Western Australian state government introduced the Liveable Neighbourhoods Community Design Guidelines (LNG). An interpretation of New Urbanism tailored to the Western Australian context, this was intended to replace conventional planning policies and

Paula Hooper, PhD, is with the Centre for the Built Environment and Health, School of Population Health, and Professor Matthetu Knuiman, PhD, is xuith the School of Population Health, University of Western Australia, Craiuley, Australia. Billie Giles-Corti, PhD, is with the McCaughey VicHealth Centre for Community Wellbeing, Melbourne School of Population Health, University of Melbourne, Australia. Send reprint requests to Paula Hooper, PhD, Centre for the Built Environment and Health, School of Population Health (M707), University of Western Atistralia, 35 Stirling Highway, Crawley, WA 6009, Australia; [email protected]. This ninnuscripí wa.'i submitted May 3, 2013; tlie manustript was accepted for pubticntion August 13, 2013. CopyrigtU © 2014 Ity Ameiican fcrumat of Health Promotion, Inc. 0890-1171/14/$S.00-hO noi: 10.4278/ajhp.l30503-QUAN-226

American Journal of Health Promotion

January/February 2014, Vol. 28, No. 3 Supplement

S5

create more sustainable suburban communities across metropolitan Perth. ' A key intended outcome of the LNG was to reduce suburban sprawl and car dependence and encourage more walking, cycling, and public transport tise.' The LNG consists of six general design topics ("elements"), four of which (community design, movement networks, lot layout, and ptiblic parkland) aimed to provide more compact, self-sufficient, pedestrian-friendly neighborhoods, with destinations and public transport links. The introduction of the LNG provided a unique opportunity for a natural experiment, and in 2003 the RESIDential Environments project (RESIDE) commenced with the aim of assessing the impact of the policy on the walking, cycling, and public transport behaviors of residents moving into 19 "liveable" developments (i.e., those deemed by the Department of Planning [DOP] to have complied with the LNG), compared with 45 "conventionally designed" neighborhoods and 11 "hybrid" developments.'' One aim of RESIDE was to quantify the levels of policy implementation. In order to accurately assess the impacts of the LNG on the walking behaviors of residents and evaluate whether the policy was achieving its intended walking outcomes, it was essential to first determine the degree to which the onground implementation actually reflected that intended by the policy. Without such quantification, it was impo.ssible to know whether any observed restilts, or lack thereof, reflected a failure of the policy itself (i.e., its principles) or a failure to implement the policy as intended.'^^^ A comprehensive process evaluation was devised to measure the levels of ongrotmd implementation of the LNG within the RESIDE developments and quantify the degree to which the onground implementation reflected that envisaged by the policy. The specific objectives of the process evaluation were to (1) develop objective lneasures for each LNG requirement in geographic information systems (GIS), (2) quantify the levels of on-ground implementation of all requirements within each of the four elements, and (3) quantify levels of compliance with each element and overall.

S6

American Journal of Health Promotion

The outcome evaluation investigated whether greater levels of implementation led to better walking outcomes. Associations between (adtilt) residents' walking behaviors and the levels of LNG implementation for each element and overall were examined. Further, we examined the cumulative effect of compliance with the LNG elements on walking behaviors. Using cluster analysis, homogeneous clusters of RESIDE developments were identified ba.sed on their implementation of the LNG requirements and walking behaviors compared across clusters. METHODS PROCESS EVALUATION METHODS Each element contained a list of requirements that presented a range of qualitative and quantitative responses or criteria to meeting the element objectives.' There were 128 different reqtiirements across the four elements. A review of the LNG document revealed substantial duplication and overlap of requirements within, and across, the elements. A total of 60 unique, quantifiable, and measurable reqtiirements that addressed actual feattires of the built environment and that were relevant for walking behavior were identified. Forty-three of these requirements (72%) that were meastirable using available GIS data within the evaluation's time frame were used in the LNG evaluation. These are illustrated in Table 1. Objective Measurement of LNG Requirements in GIS Measures specific to each of the 43 requirements were computed in GIS. Where possible, existing measures of the built environment reported in literature were applied. However, these often required refinement and modification to better meastire the specific policy requirement(s). Where no suitable meastire was available, new measures and methods were developed using a mixture of existing GIS data layers and automated processes and tools within GIS (ArcMap v9.3 and vlO.O, Environmental Systems Research Institute) and, where necessary, the development of new GIS data layers and protocols. Table 1 also outlines the

GIS variables developed under each of the four elements. Selection of Housing Developments The process for selection of housing developments in the RESIDE study is reported elsewhere.'' All 19 developments identified by the DOP as having been designed in accordance with the LNG were included in this evaluation. Seventeen of the 44 conventionally designed developments that matched the liveable developments on their size and location (i.e., distance from the ocean) were also included in this evaluation. The 36 developments varied in size from small subdivisions (300 ha). The evaluation was concerned with quantifying the levels of LNG implementation by developers, and as such the boundaries of the developments were used as the study areas of interest for generation of all measures. Timing of Evaluation and Stage of Construction The majority of the RESIDE developments were being constructed on greenfield sites (i.e., previously undeveloped land). The timing of the LNG evaluation was therefore chosen to coincide with the third time point of RESIDE data collection in 2009 (i.e., 5-6 years post—commencement of RESIDE and approval of the developments) as this allowed for the greatest amount of construction to have occurred. As the developments were still under construction at the time of evaluation and many of the policy requirements were area based (e.g., the proportion of the land area that should be set aside for ptiblic open space) or net residential density targets, all undeveloped areas of land were excluded. The constructed land area of each development was then calculated as a proportion of the gross development area and categorized: 0% to 25% to 50% to 75% to 100%. This ensured comparability across developments at different stages of construction. Quantifying Policy Compliance A simple scoring system was developed to quantify the extent to which the requirements had been imple-

January/February 2014, Vol. 28, No. 3 Supplement

mented as intended by the LNG (Figure 1). All measured requirements were assigned a score ranging from 0 to 1 according to their level of implementation evaluated against the policy specifications or standards. All requirements phrased as "matters that must be satisfied"' were deemed mandatory. Where these also stipulated a desired standard or threshold, the maximum score of 1 was assigned if the measured feature met or exceeded the LNG standard for that requirement and a score of 0 assigned if it failed to meet the standard. The majority of requirements, however, either failed to stipulate a standard against which compliance could be assessed and/or were indicated as "matters that should be considered."'' Here, varying levels of compliance were rewarded. Wliere existing measures identified from literature had recognized or accepted thresholds, these were adopted and scores from 0 to 1 applied accordingly. For example, the connected node ratio"'"^^ was used to determine compliance relating to the connectivity of the street networks within the developments. Scores were applied to its recommended categories: .5 to 60 min of Waiking Within the Neighborhood

Any Walking for Recreatior1

> 6 0 min of Transport

> 6 0 min of Recreation

Waiking

Waiking

OR

95%CI

P

OR

95%C1

P

OR

95%CI

P

OR

95%CI

P

1.53t 1.27t 2.49t 1.26t 0.78t

1.13-2.08 1.13-1.42 1.38-4.50 1.06-1.50 0.59-1.04

0.007 0.000 0.003 0.010 0.085

0.97t 1.03t 1.49t 1.06t 0.85t

0.73-1.30 0.92-1.14 0.83-2.65 0.88-1.28 0.69-1.03

0.848 0.618 0.180 0.511 0.099

1.14t 1.17t 2.01t 1.02t 0.76§

0.70-1.88 1.03-1.32 0.94-4.27 0.80-1.30 0.53-1.08

0.598 0.016 0.071 0.874 0.121

0.97t i.oit

0.74-1.27 0.92-1.11 0.83-2.20 0.81-1.19 0.75-1.06

0.799 0.822 0.230 0.863 0.202

1.35t 0.981: 0.89t

* OR indicates odds ratio; LNG, Liveable Neighbourhoods Guidelines; and 95% CI, 95% confidence interval. Elements examined separately and adjusted for demographic variables, stage of construction, size of development, and self-selection. Bolded values indicate statistically significant associations. t Five self-selection variables remained in the model: affordability/value; ease of walking; the streets have footpaths; the streets are designed to minimize traffic; and closeness to a variety of parks. t One self-selection variable/preference remained in the model: the streets are designed to minimize traffic. § One self-selection variable/preference remained in the model: closeness to a variety of parks.

American Journal of Health Promotion

January/February 2014, Vol. 28, No. 3 Supplement

S15

Table 5 Odds Ratios (ORs) for Cluster Development Types and Walking for Transport and Recreation* Walking for Transport Cluster Group Any walking 1 2 3 4 >60 min of walking 1 2 3 4

Description

OR

95% CI

Poor CD + MN + LL + PP compliance Higher MN + LL compliance Higher PP compliance Higher CD compliance

1.00 1.54 1.43 2.64

Poor CD -1- MN -1- LL + PP compiiance Higher MN + LL compliance Higher PP compiiance Higher CD compliance

1.00 1.287 0.83 1.98

Walking for Recreation P

OR

95% CI

P

0.79-2.99 0.53-3.91 1.38-5.06

0.209 0.482 0.003

1.00 1.74 3.53 0.96

1.22-2.48 2.02-6.17 0.68-1.36

0.002 0.000 0.817

0.56-2.93 0.31-2.22 1.09-3.61

0.556 0.710 0.026

1.00 2.05 3.37 0.96

1.46-2.88 1.98-5.74 0.75-1.22

0.000 0.000 0.716

* Adjusted for demographic variables, stage of construction. size of development, and self-selection factors. Cluster 1 is reference group. 95% CI indicates 95% confidence interval; CD, community design; MN, movement network; LL, lot layout; and PP, public parkiand. Bolded values indicate statistically significant associations.

scores of all LNG elements were entered into the same multivariate model, compliance with the community design remained the only element to be significantly and positively associated with walking for transport. The results of the cluster analysis reinforced these findings and highlighted the importance of implementing the combination of elements. Compliance with the movement network and lot layout elements was important. However, above and beyond these, the added combination of compliance with the community design element was critical. Consistent with previous evidence,^'~^® walkable community design ensured that residents had a diversity of utilitarian destinations to walk to, thereby supporting walking for transport within the neighborhood. In practice this suggests that simply providing sidewalks or cycle paths or even increasing density without providing local destinations will be insufficient to bring about behavior change and encourage active living. Further work will assess the relative influence of the individual community design requirements (i.e., centers, public transport stops, schools) on active living behaviors. Moreover, although compliance with the public parkland element alone was not associated with walking for recreation, compliance with the public park-

S16

American Journal of Health Promotion

land element in combination with the movement network element was essential for encouraging recreational walking in the neighborhood. Consistent with previous evidence, '" ' these findings suggest that to encourage walking for recreation the provision of parks (both as destinations for people to walk to and in and as enhancements of the aesthetics of the neighborhood) in combination with the provision of supportive movement networks and pedestrian infrastructure was critical. Because of the lack of existing infrastructure and the timelines required for the construction of developments on greenfield sites, it is important to recognize that construction was still unfolding in the developments. In fact, at the time of this evaluation, none of the developments were fully complete. This is important consideration for the observed lack of full compliance. As the developments are still undergoing construction, the "dose" of LNG may increase with future implementation of additional LNG requirements. The true influence of the LNG on the walking behaviors of residents may not, therefore, be fully apparent at this time. This also highlights the timelines required for the construction of new neighborhoods, with important iinplications for the design of future natural experiments (longitudinal and cross-sectional) and

the timelines for evaluation and assessment of planning policies or design codes with health behaviors. In particular, the installation of neighborhood centers is often withheld until the latter stages of construction and until sufficient population densities are present to support the shops and services. This offers some explanation for the low community design connpliance scores observed at the current time. Efforts are required to facilitate the timely introduction of Iocal destinations in neighborhoods to provide residents with local shops and services and somewhere to walk. Providing incentives that facilitate the early installation and establishment of mixeduse centers or mandating business development in specific areas is warranted. At the time of its introduction, use of the LNG was not mandated but relied on voluntary uptake by developers who saw merit in using its principles and design standards. Developers were able to choose whether they used the LNG or the conventional design policies and were permitted to incorporate selected (i.e., not all) LNG requirements into their designs.^'' Subsequent follow-up with policy makers revealed that many conventionally designed projects were in fact approved under the LNG if they adopted some requirements into their

January/February 2014, Vol. 28, No. 3 Supplement

designs. These flexible approval practices offer some explanation for the limited overall levels of compliance and lack of differences between the socalled liveable and conventionally designed developments selected for the RESIDE project. Additionally, a number of barriers to implementation were also identified: these included issues directly relating to the structure and content of the LNG document and conflicts with existing policies and regulation and engineering standards, as well as administrative and political processes affecting the developer's ability or willingness to implement different LNG requirements and a lack of incentives encouraging compliance The LNG sets out a comprehensive approach to the regional application of the policy through the strticturing of walkable neighborhoods clustered to form towns.^ However, its voluntary uptake also meant it was not always applied at the scale intended.^® Smaller subdivision developments are too small to warrant the inclusion of neighborhood centers and creation of complete neighborhoods as envisaged by LNG. More genuine attempts at applying LNG at structure plan and regional-scaled projects are thus needed. The development of smaller, piecemeal subdivisions shotild be regionally planned and their development coordinated to ensure shops and services are provided and distributed equitably and within walking distances in a timely manner. The process evaluation results provide important information helping to interpret previous RESIDE findings of no differences in the walking behaviors of residents relocating to these liveable or conventionally designed neighborhoods.^•'''''' This is despite the fact that, irrespective of whether a neighborhood was designated liveable or conventional, increased policy compliance was associated with increased walking. Moreover, our published longitudinal analyses provide support for the principles of the LNG policy.'"^''' Participants who gained access to transportrelated destinations increased their transport walking by around 6 min/wk for each type of utilitarian destination gained after relocation, and recreational walking increased by around 18 min/wk for each type of recreational

American Journal of Health Promotion

destination gained.'* These results were consistent with cross-sectional evidence of McCormack et al.^** that highlighted the importance of the mix of destinations to encotirage local walking. The scoring system provided a simple quantification of the fidelity of the policy in terms of the on-ground "dose delivered" in each of the developments as a proportion of the policy's intended level of on-ground implementation and outcomes. As all reqtiirements were treated as equal, it allowed for the identification of areas of LNG conformance and nonconformance to be detected and provided a basic barometer of how well the developments were complying with the policy—i.e., the number or range of different requirements and the degree to which these were implemented. However, the assumption that all of the LNG requirements (i.e., design features) may be of equal importance to walking or physical activity outcomes (or other health or environmental outcomes) is likely not to be the case. Future work will aim to unpack the policy further to assess the relative influence of the individual requirements, or combinations of requirements, that account for the greatest variance in walking outcomes with a view to establishing a hierarchy of the LNG requirements (i.e., essential, important, and supportive) across the elements that influence active living behaviors. This will assist in identifying which of the numerous LNG requirements are more, or less, important for different types of walking behavior. Nevertheless, the findings from this evaluation highlight the importance of including process evaluation measures of dose and fidelity in natural experiments to better evaluate the level of implementation, and hence explain the intervention's effect and potential associations with physical activity behaviors. This study is unique in its effort to comprehensively evaluate the implementation of a real-world government planning policy focused on creating walkable suburban neighborhoods and investigate its associations with walking behaviors. Results confirm previous research into the importance of the

built environment for active living behaviors, providing evidence for the impact of a planning policy on the walking behaviors of residents in suburban environments. The outcome evaluation was based on self-report measures of walking derived from NPAQ. Self-administered physical activity instruments have welldocumented limitations, including the tendency to overreport activity levels.^'^ However, NPAQ has also previously been shown to have acceptable reliability.^^ Moreover, through the use of NPAQ, RESIDE data differentiated between walking within and outside of the neighborhood and whether that walking was for transport or recreation, allowing examination of behavior-specific environmental characteristics. This evaluation moves beyond previous studies by developing objective measures specific to the planning policy and its respective requirements. Allender and colleagues found that evidence-based public health recommendations to planners and policy makers are often provided without links to existing policies or legislation, which may account for their lack of impact. Exddence on the effectiveness of actual planning policies was reportedly needed to influence planning professionals' practices. The specific measures used to evaluate the LNG have a high degree of ecological validity, better reflecting or approximating the real-world on-the-ground outcomes of the LNG, and also "speak" to planners in ways in which they conceptualize and seek to modify the built environment. The findings from this study are directly relevant to an important operational planning policy in Perth, Western Australia. It offers a timely and unique opportunity to provide directly relevant feedback from a public health perspective to the planners and policy makers responsible for implementing this policy. The RESIDE study has been conducted in partnership with the DOP. The findings from the LNG evaluation will provide feedback from an active living perspective to the DOP that will inform its ongoing development and dissemination of the LNG.

January/February 2014, Vol. 28, No. 3 Supplemenf

S17

s o WHAT? Implications for Health Promotion Practitioners and Researchers What is already known on this topic? A substantial body of literature identifying specific features of the built environment that are associated with physical activity and walking now exists. What does this article add? Few studies internationally have been designed to evaluate the effectiveness and impacts of environmental interventions and planning policies to increase levels of walking and physical activity. Evaluating the impacts of planning policies on active living behaviors is an identified research priority for thefield.'"'^This study is unique in its effort to rigorously evaluate the implementation of a real-world government planning policy focused on creating walkable suburban neighborhoods and investigate its associations with walking behaviors. What are the implications for health promotion practice or research? Results confirm previous research into the importance of the built environment for active living behaviors, providing evidence for the impact of a planning policy on the walking behaviors of residents in suburban enwronments. The findings also highlight the importance of including process evaluation measures of dose and fidelity in natural experiments to better evaluate the level of policy implementation, and hence explain the effects of planning policies and the potential associations with physical activity

Acknowledgments All funding bodies are gratefully acknowledged. RESIDE was funded by grants from the Western Australian Health Promotion Foundation (Mealthway) (Ul 1S28) and the Australian Research Council (ARC) (#W04S5453), and .supported by an Australian National Health ¿¡f Medical Research Council (NHMRC) Capacity Building Grant (#458688). Thefirstauthor was supported by a Scholarship for International Research Eees (SIRE) from the University of Western Australia and an NHMRC Population Health Capacity Building Crant (#458668); the srrond author is suj)ported Iry a NHMRC Principal Knriirch Eellowship (#1004900). CBEH's GIS team (Nick Miiidleton, Sharyn Hickey, Bridget Beasley, and Dr. Bryan Boruff) are gratefully acknowledged for their assistance and advice in the development of the GIS measures in this study. Jacinta Erancis, Claire Lauritsen, and Roseanne Barnes coordinated data collection of the RESIDE sunjey

S18

American Journal of Health Promotion

References 1. SchillingJ, Linlon LS. The public health roots of zoning: in search of active living's legal genealogy. AmJPrev Med. 2005;28(2, suppl 2):9&-104. 2. Story M, Giles-Corti B, Yaroch AL, et al. Work group FV: future directions for measures of the food and physical activity environments. Am] Prev Med. 2009; 36(suppl 4):S182-S188. 3. Ogilvie D, Charles EF, Helen R, et al. Inter\'entions to promote walking: systematic review. BMJ. 2007;334:1204. 4. Sallis J, Bauman A, Pratt M. Environmental and policy interventions to promote physical activity. AmJPreu Med. 1998;15:379-397. 5. Western Australian Planning Commission. Liveable, Neighbourhoods: A Western Australian Government Sustainable Cities Initiative.

6.

7.

8.

9. 10.

Edition 2. Perth, Western Australia: Western Australian Planning Commission; 2000. Giles-Corti B, Knuiman M, Timperio A, et al. Evaluation of the implementation of a state government community design policy aimed at increasing local walking: design issues and baseline results from RESIDE, Perth Western Australia. Prev Med. 2008:46:54. Basch CE, Sliepcevich EM, Cold RS, et al. Avoiding type III errors in health education program evaltiations: a case study. Health Edtic Behau 1985;12:315-331. Dehar M, Casswell S, Duignan P. Eomiative and process evaluation of health promotion and disease prevention programs. EvalRni. 1993:17:204-220. Talen E. Do plans get implemented? A review of evaluation in planning./P¿ann Lit. 1996:10:248-259. Steiner R, Bond A, Miller D, Shad P. Future Directions for Muttimodat Area Wide Level of Service Handbook Research and Development.

Tallahassee, Fla: Florida Dept of Transportation, Office of Planning Systems; 2004. 11. Criterion Planners/Engineers Inc. Smart Growth Index: A Sketch Tool for Community Planning. Indicator Dictionary. Version 2.0.

Portland, Ore: US Environmental Protection Agency; 2002. 12. Giles Corti B, Timperio A, Cutt H, et al. Development of a reliable measure of walking within and outside the local neighborhood: RESIDE's Neighborhood Physical Activity Questionnaire. Prev Med. 2006;42:455^59. 13. Christian H, Knuiman M, Bull F, et al. Evaluation of a natural experiment of the impact of a new urban planning code on walking: longitudinal results from RESIDE. Am f Public Health. 2013;103: 1219-1228. 14. Giles-Corti B, Bull F, Knuiman M, et al. The influence of tnban design on neighborhood walking following residential relocation: longitudinal results

from the RESIDE study. Soc Sei Med. 2013; 77:20-30. 15. Dill J. Evaluating a New Urbanist neighborhood. Berkeley Plann f 2006;19: 59-78. 16. Khattak A, Rodriguez D. Travel behaviour in neo-traditional neighborhood development: a case study in USA. Transportation Res A. 2005;39:481-500. 17. Rodriguez D, Kliattak A, Evenson K. Can New Urbanism encourage physical activity? Comparing a New Urbanist neighborhood with conventional suburbs. f Am Plann As.soc. 2006;72:43-54. 18. McCormack GR, Giles-Corti B, Bulsara M. The relationship between destination proximity, destination mix and physical activity behaviors. Prev Med. 2008;46:3S40. t9. Moudon A, Lee C, Cheadle A, et al. Operational definitions of walkable neighborhood: theoretical and empirical insights, f Phys Act Health. 2006;3(.suppl 1): S99-S117. 20. Sugiyama T, Francis J, Middleton NJ, et al. Associations between recreational walking and attractiveness, size, and proximity of neighborhood open spaces. Am] Public Health. 2010;100:1752-1757. 21. Giles-Cord B, Broomhall MH, Knuiman et al. Increasing walking: how important is distance to, attractiveness, and size of public open space? AmJ Prev Med. 2005; 28(2, suppl 2):169-176. 22. Pikora T, Giles-Corti B, Knuiman M, et al. Neighborhood environmental factors correlated with walking near home: using SPACES. Med Sei Sports Exerc. 2006;38:708714. 23. Owen N, Humpel N, Leslie F. et al. Understanding en\ironmental influences on walking: review and research agenda. AmJ Prev Med. 2004;27:67-76. 24. Saelens B, Handy S. Built environment correlates of walking: a review. Med Sei Sports Exerc. 2008;40(7S):S550-S566. 25. Saelens B, SallisJ, Frank L. Environmental correlates of walking and cycling: findings from the transportation, urban design and planning literatures. Ann Behav Med. 2003; 25:80-91. 26. Jones E. Liveable Neighborhoods Revieto: Discussion Paper for the Urban Development Institute of Australia. Perth, Western

Australia: Urban Development Institute of Australia; 2012. 27. Sallis JF, Saelens B. Assessment of physical activity by self-report: status, limitations and future directions. Res Q Exerc Sport. 2000:71:409. 28. Allender S, Cavill N, Parker M, Foster C. "Tell us something we don't already know or do!" The response of planning and transport professionals to public health guidance on the built environment and physical activity. J Public Health Policy 2009;

30:102-116.

January/February 2014, Vol. 28, No. 3 Supplement

Copyright of American Journal of Health Promotion is the property of American Journal of Health Promotion and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

Evaluating the implementation and active living impacts of a state government planning policy designed to create walkable neighborhoods in Perth, Western Australia.

Evaluate the implementation of a government planning policy (Liveable Neighbourhoods Guidelines) and its impacts on residents' walking behaviors...
15MB Sizes 0 Downloads 0 Views