The influence of urban design on neighbourhood walking following residential relocation: Longitudinal results from the RESIDE study
Highlights
► The built environment determined recreational (RW) and transport walking (TW). ► Following relocation, TW declined, as did access to transport-related destinations. ► Yet for each type of TW destination gained, TW increased 5.8 min/week. ► Following relocation, RW increased, as did access to recreational destinations. ► For each type of recreational destination gained, RW increased 17.6 min/week.
Introduction
Recognition that city design impacts public health was first established in the 19th century in efforts to control the outbreak of communicable disease (Corburn, 2007). With communicable disease control, the links between urban planning and public health attenuated, and only recently have there been calls for the two disciplines to reconnect (Corburn, 2007): this time in an effort to create healthy and sustainable cities that facilitate health behaviours that reduce the risk of non-communicable diseases.
For over two decades the World Health Organisation's Healthy City Movement (Duhl, 1996) has promoted urban design features required to create health-enhancing cities (Rydin et al., 2012). Healthy and sustainable communities create the conditions that optimize physical and mental health and well-being (Marmott, 2011) by impacting social determinants of health including access to public transport, public open space, local amenities and other social infrastructure. Nevertheless, relatively little systematic research has examined the influence of contextual factors on health outcomes (Macintyre & Ellaway, 2003). As a consequence, there is little understanding of the potential of urban environments to deliver and equitably distribute health benefits (Rydin et al., 2012).
Planning, transport and urban design policies and regulations directly influence the location and proximity of activities required for daily living (e.g. shops, workplaces and school, facilities places to socialize and recreate) and the ease with which places can be reached using active forms of transport (i.e., walking, cycling and public transport). Physical inactivity is a common risk factor for major non-communicable diseases, yet less than one half of adults in many developed countries are sufficiently active to protect their health (WHO, 2010). This poses a substantial public health risk and an economic burden to national health care budgets in both the developed (Department of Health, 2009) and the developing world (Rydin et al., 2012). Increasing levels of physical activity through popular activities such as walking and cycling is a practical means of improving health (Department of Health, 2009), as well as producing co-benefits across other sectors by reducing car use, traffic congestion and air pollution (Giles-Corti, Foster, Shilton, & Falconer, 2010; Haines et al., 2009).
To support policy-reform, recently there has been a plethora of reviews on the impact of the built environment on various health outcomes, including physical activity and obesity (CDC, 2007; Kopelman, Jebb, & Butland, 2007; National Institute for Health and Clinical Excellence, 2008; National Preventative Health Taskforce, 2009). The evidence suggests that community design affects travel mode choices, (National Institute for Health and Clinical Excellence, 2008; Transportation Research Board, 2005) and levels of walking and/or cycling (Durand, Andalib, Dunton, Wolch, & Pentz, 2011; McCormack et al., 2004; Ogilvie et al., 2007; Owen, Humpel, Leslie, Bauman, & Sallis, 2004; Panter, Jones, & van Sluijs, 2008; Saelens & Handy, 2008; Transportation Research Board, 2005) with mixed evidence on its impact on obesity (Dunton, Kaplan, Wolch, Jerrett, & Reynolds, 2009; Feng, Glass, Curriero, Stewart, & Schwartz, 2010; Papas et al., 2007; Robertson-Wilson & Giles-Corti, 2010; Van Cauwenberg et al., 2011). Specifically, walking for transport appears to be associated with increased residential density, high street connectivity, mixed land-use and proximity to destinations. Recreational walking on the other hand is associated with access to public open space, neighbourhood attractiveness and the accessibility and functionality of local facilities. Thus, creating pedestrian-friendly neighbourhoods with access to local amenities and well-designed public open space has the potential to benefit health (Marmott, 2011).
Despite the complexities associated with implementation (Rydin et al., 2012), urban planning that promotes walking, cycling and transit use is now being recommended by multiple sectors including public health (CDC, 2007; Kopelman et al., 2007; National Institute for Health and Clinical Excellence, 2008; National Preventative Health Taskforce, 2009), transport (Transportation Research Board, 2005) and planning authorities (Planning Institute of Australia & Heart Foundation, 2008). Nevertheless, most of the evidence to date is cross-sectional, and most evidence reviews conclude that stronger longitudinal evidence is required to better inform urban planning and practice. A limitation of cross-sectional evidence is self-selection: i.e., people may choose to live in neighbourhoods that reflect their active-living preferences, rather than neighbourhood design changing their behaviour. To address this limitation, some cross-sectional studies control for participants' reasons for moving into their current neighbourhood (Cao, Mokhtarian, & Handy, 2009; Frank, Saelens, Powell, & Chapman, 2007; Handy, Cao, & Mokhtarian, 2005; McCormack & Shiell, 2011; Owen et al., 2007). However recall bias cannot be ruled out when asking retrospectively about factors influencing decisions to relocate. Longitudinal evidence is therefore required to examine the influence of changes in urban form on health outcomes and individual lifestyle behaviours (Rydin et al., 2012).
Studies examining the impacts of changes to neighbourhood design and transport infrastructure are difficult to design and implement, and randomised controlled trials (RCTs) are rarely possible. Consequently, there are now a number of large established cohort studies examining the impact of neighbourhood design on health outcomes by identifying participants who relocate during follow-up (Berry et al., 2010; Boone-Heinonen, Evenson, Taber, & Gordon-Larsen, 2009; Boone-Heinonen, Guilkey, Evenson, & Gordon-Larsen, 2010; Hou et al., 2010; Krizek, 2000; Lee, Ewing, & Sesso, 2009; Ludwig et al., 2011). To date, mixed results have been reported, and frequently authors highlight methodological flaws that may contribute to their findings including: a small sample of ‘movers’; an inability to account for length of exposure to residential environments; and the use of behavioural and built environment measures that lack specificity or were chosen for another purpose. In addition, few of these studies have been grounded within a broader ecological framework. Ecological frameworks consider multilevel factors that influence behaviour including intrapersonal, interpersonal, and physical environment factors (Sallis et al., 2006; Stokols, 1996), allowing both contextual and compositional variables to be captured (Macintyre & Ellaway, 2003). Use of this framework enables consideration and adjustment for changes in demographic, intrapersonal (e.g., attitudes), and interpersonal, (e.g., social support) factors, as well as changes in built environmental factors following relocation.
Importantly, very few studies (Wells & Yang, 2008) have been specifically designed to study residential relocation within the context of a natural experiment, although this has been identified as a gap in the literature (Ogilvie et al., 2007). Such an opportunity arose in Australia, where capital cities are experiencing average annual growth of around 1.6% (ABS, 2012), and providing sufficient affordable housing is a major policy challenge for government (Department of Infrastructure and Transport, 2011). The current study was conducted in the west coast city of Perth, which is one of Australia's fastest growing capital cities (2.5% annually) (ABS, 2012). Nevertheless, outer suburban metropolitan regions of other Australian capital cities are also experiencing even more rapid growth (e.g., some areas in the North-West metropolitan region of Melbourne grew almost 8% in the 12 months to June 2011) (ABS, 2012). Thus, the findings in this study are relevant to other greenfield outer suburban areas in Australia and elsewhere.
This paper uses data from the RESIDential Environment Project (RESIDE), a longitudinal natural experiment of people building houses and relocating to new neighbourhoods, to examine the impact of the built environment on walking for transport and recreation following relocation. We hypothesized that people relocating to neighbourhoods with infrastructure supportive of transport or recreational walking, would walk more. To understand the independent effects of the built environment, the study adopted an ecological framework (Giles-Corti et al., 2008), which allowed for adjustment for changes in demographic, intrapersonal and interpersonal characteristics as well as the changes in the built environment. In addition, measures of residential location preference were collected to adjust for one aspect of self-selection.
Section snippets
Methods
The University of Western Australia's Human Research Ethics Committee (#RA/4/1/479) provided ethics approval. RESIDE involved participants who moved into 73 new housing developments across metropolitan Perth. Intrapersonal, interpersonal and perceived built environmental data were collected using questionnaires completed by participants, while objective built environmental data were derived from a Geographic Information System (GIS).
Statistical analysis
Generalized Linear Mixed Models (that included a random cluster effect to allow for clustering by (new) developments) in SAS v9.2 were used to examine associations with changes in neighbourhood recreational and transport-related walking. All models were adjusted for baseline age, gender, education level, marital status, having children <18 years at home, and baseline total minutes of recreational or transport-related walking. Socio-demographic change variables (p ≤ 0.20) were included in all
Results
Table 3 shows the socio-demographic characteristics of study participants who completed both surveys (n = 1420) and those who completed T1 only. Those remaining in the study at T2 were significantly more likely to be female, slightly older, and more likely to be married or in a de facto relationship and to have children under 18 years at home (p < 0.05). They were also significantly less likely to work and were engaged in significantly fewer hours of paid or unpaid work (p < 0.05).
Discussion
The physical characteristics of neighbourhoods can facilitate health-enhancing behaviours, thus having the potential to reduce social inequalities (Marmott, 2011). Nevertheless, to date much of the evidence on the impact of the built environment on physical activity is cross-sectional. There has been a small number of quasi-experimental studies (Berry et al., 2010; Boone-Heinonen et al., 2009, 2010; Hou et al., 2010; Krizek, 2000; Lee et al., 2009; Ludwig et al., 2011; Wells & Yang, 2008), but
Limitations
Our findings are limited to new home buyers moving into new housing developments, most of which were greenfield developments located on the urban fringe. This was unavoidable given RESIDE aimed to study the impact of the built environment on behaviour, controlling for self-selection. This required that participants be surveyed prior to relocation and new home buyers were more readily identified than other groups, for example, those who rent accommodation. This is a major limitation and future
Conclusion
This study provides longitudinal evidence that both transport and recreational-walking behaviours change in response to changes to access and the diversity of local transport- and recreational destinations, highlighting the importance of local facilities to enable and promote active living. Together these will help reduce social inequality by providing a supportive environment. However, longer-term follow-up is paramount to explore the full impact on behaviour, as some of the features designed
References (63)
- et al.
The built environment and obesity: a systematic review of the epidemiologic evidence
Health & Place
(2010) - et al.
Stepping towards causation: do built environments or neighborhood and travel preferences explain physical activity, driving, and obesity?
Social Science & Medicine
(2007) - et al.
Increasing walking: how important is distance to, attractiveness, and size of public open space?
American Journal of Preventive Medicine
(2005) - et al.
The relative influence of individual, social and physical environmental determinants of physical activity
Social Science & Medicine
(2002) - et al.
Socioeconomic status differences in recreational physical activity levels and real and perceived access to a supportive environment
Preventive Medicine
(2002) - 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
Preventive Medicine
(2008) - et al.
Development of a reliable measure of walking within and outside the local neighborhood: RESIDE's Neighborhood Physical Activity Questionnaire
Preventive Medicine
(2006) - et al.
Health and climate change. Public health benefits of strategies to reduce greenhouse-gas emissions: overview and implications for policy makers
Lancet
(2009) - et al.
Correlation or causality between the built environment and travel behavior? Evidence from Northern California
Transportation Research Part D-Transport and Environment
(2005) - et al.
Longitudinal associations between neighborhood-level street network with walking, bicycling, and jogging: the CARDIA study
Health & Place
(2010)