Elsevier

Preventive Medicine

Volume 120, March 2019, Pages 126-130
Preventive Medicine

Associations of local-area walkability with disparities in residents' walking and car use

https://doi.org/10.1016/j.ypmed.2019.01.017Get rights and content

Highlights

  • We examined associations of walkability with disparity of walking and car use.

  • Higher walkability was associated with lower levels of disparity in walking.

  • Lower walkability was associated with lower levels of disparity in car use.

  • Improving walkability has the potential to reduce inequities in physical activity.

Abstract

Research has examined spatial distribution of physical activity, mostly focusing on between-area differences by examining associations of area-level walkability measures with physical activity. Within-area distribution is also relevant, since larger disparities in physical activity within an area can contribute to greater inequalities in health. However, associations of within-area disparity in walking and walkability have been examined only at a large geographical scale (city level). This cross-sectional study examined associations of local-area walkability measures with within-area disparities in residents' walking and car use, using data collected in the 2009 South-East Queensland Travel Survey in Australia. For each Statistical Area 2 (SA2), we calculated disparity indices of the duration of walking and car use among participants aged 18–84 years, using Gini coefficients. Linear regression examined associations of the disparity measures with population density, street connectivity, and Walk Score. Analyses were conducted for 196 SA2s, which contained 15,895 participants. Higher walkability was associated with lower levels of disparity in walking and higher levels of disparity in car use, regardless of the measures used. Each one-SD increment in Walk Score was associated with a 0.64 lower SD in walking disparity and a 0.50 higher SD in car-use disparity, after adjusting for covariates. The associations remained significant after further adjusting for car ownership. Higher walkability is known to be associated with more walking and less car use. This study extends previous knowledge by showing that higher local-area walkability can be associated with less inequality in residents' walking and higher diversity in their car use.

Introduction

Lack of physical activity is a major health risk and a leading cause of chronic diseases and premature death (Lee et al., 2012). Environmental attributes related to walkability (population density, street connectivity, and availability of utilitarian and recreational destinations), which differ between areas, are associated with residents' overall physical activity and with their walking (Christiansen et al., 2016; Cole et al., 2015; Koohsari et al., 2018; Thielman et al., 2015). A recent study using data from 14 cities worldwide found several built-environment attributes including higher residential density, well-connected streets, better access to public transport, and higher number of parks to be associated with higher levels of adults' physical activity (Sallis et al., 2016).

There are complex patterns in the distribution of physical activity. Disparities in physical activity can contribute to widening socio-economic inequalities in health (Petrovic et al., 2018). Socio-demographic disparities have been documented for self-reported leisure-time physical activity (Blackwell et al., 2014), self-reported walking for transportation (Paul et al., 2015), and accelerometer-measured physical activity (Troiano et al., 2008). However, geographic disparities in physical activity are less understood. There is a need to better understand spatial disparities in physical activity to help develop the place-based interventions that address contextual factors contributing to health problems and inequalities based on understanding of local socioeconomic and environmental characteristics (Smedley and Amaro, 2016).

There are studies examining how physical activity is spatially distributed, but they typically focus on between-area differences in physical activity. For instance, an Australian study found that a measure of walkability (consisting of dwelling density, intersection density, and land use diversity) constructed at a postal-area level accounted for between-area variation in those engaging in walking and moderate-to-vigorous physical activity sufficient for health benefits (Mayne et al., 2017). Similarly, a study conducted in the US showed that a measure of urban sprawl constructed at the level of county (a large administrative unit, which can include an entire metropolitan area) partially explained spatial variations in physical inactivity (Congdon, 2016). Investigating between-area differences in physical activity is an important step to understand its spatial distribution. However, such studies can assume that each unit area is homogeneous in environmental characteristics and activity levels, which is unlikely to be a tenable assumption across large areas, such as counties or cities. Investigating the within-area distribution of physical activity is therefore relevant, since physical activity can vary within an area, and greater within-area disparities may contribute to greater inequalities in health (Petrovic et al., 2018).

A recent study in the US examined within-city disparities of walking steps using the Gini coefficient, and found that cities with higher walkability (measured by Walk Score®) were more likely to have lower disparities in walking (Althoff et al., 2017). Higher city-level walkability was associated not only with higher levels of “mean” walking among residents but also with lower “dispersion” in walking within the city. These findings suggest that improving walkability may contribute to increasing community-level physical activity as well as reducing disparities in physical activity, which may mitigate health inequalities.

In this context, it will be informative to understand to what extent local-area walkability may be associated with within-area disparities in physical activity. Examining such associations at the level of local area, where destinations providing goods and services necessary for everyday life exist, is relevant, as urban design/planning decisions that can affect residents' travel mode choice (e.g., residential density, land use, public open space, sidewalk) are usually made at the local level. In addition to the adverse health consequences of lack of physical activity, there are also detrimental associations of prolonged car use with health outcomes (McCormack and Virk, 2014; Sugiyama et al., 2016). It can be postulated that car use is likely to be more common (less dispersion) in low-walkable areas, while it may be more variable in high-walkable local areas.

We examined the associations of local-area walkability measures with within-area disparities in residents' walking and car use, using household travel survey data collected in a socially and geographically heterogeneous region around an Australian capital city.

Section snippets

Data source

Data were drawn from the 2009 South-East Queensland Travel Survey (SEQTS), a cross-sectional household travel survey administered by the Queensland (Australia) State Government in Brisbane (the state capital city), and the adjacent Sunshine Coast and Gold Coast Statistical Divisions. This region covers a mix of urban, suburban, and regional areas, with a geographic size of 10,946 km2. Its population was approximately 2.9 million in 2009. The SEQTS used a multi-stage random sampling in which

Outcome measures

All members in the selected households on the night before the specified “survey day” were asked to complete a 24-hour travel diary. A short-span activity diary has been found to be more valid and reliable in measuring physical activity and sedentary behavior than conventional self-report questionnaires (van der Ploeg et al., 2010). They reported details of their instances of travel on the survey day, including origin, destination, start time, end time, mode, and purpose. The variables employed

Results

Participants who reported no trip (by any mode) on the survey day were excluded from analysis (n = 4204). In addition, 15 SA2s where none of the participants reported home-based walking were excluded, as the study was designed to examine disparity in walking. After the exclusions, 15,895 participants living in 196 SA2s were retained for analysis. Table 1, Table 2 show the characteristics of participants and SA2s, respectively. The median number of participants in SA2s was 51. The lowest number

Discussion

We examined the associations of disparities in the duration of walking and car use within a local area was explained by walkability-related measures, using household travel survey data collected in Australia. We found the mean disparity in walking duration to be high, due to a large proportion of participants reporting 0 min of walking on the survey day. The disparity in car use was lower, indicating that residents were relatively more homogeneous in the duration of their car use.

Higher

Conclusion

Our study found there was more equity in walking in high-walkable areas. By contrast, low-walkable areas, where people tend to rely on cars for daily travels, showed a greater disparity in walking. Thus, neighborhoods designed to be walkable have the potential to reduce inequities in health-promoting physical activity, while reducing exposure to harmful effects of prolonged car use. Improving walkability of existing neighborhoods is not an easy process. However, planning initiatives such as

Acknowledgments

Koohsari was supported by Postdoctoral Fellowship for Research in Japan from the Japan Society for the Promotion of Science (#17716). Owen was supported by National Health and Medical Research Council (NHMRC) Centre of Research Excellence Grant (#1057608), NHMRC Senior Principal Research Fellowship (#1118225), and the Victorian Government's Operational Infrastructure Support Program.

Conflict of interest statement

None of the authors has any financial interest in walkscore.com.

References (35)

  • J.F. Sallis et al.

    Physical activity in relation to urban environments in 14 cities worldwide: a cross-sectional study

    Lancet

    (2016)
  • T. Sugiyama et al.

    Adverse associations of car time with markers of cardio-metabolic risk

    Prev. Med.

    (2016)
  • J. Thielman et al.

    Neighborhood walkability: differential associations with self-reported transport walking and leisure-time physical activity in Canadian towns and cities of all sizes

    Prev. Med.

    (2015)
  • P.D. Allison

    Measures of inequality

    Am. Sociol. Rev.

    (1978)
  • T. Althoff et al.

    Large-scale physical activity data reveal worldwide activity inequality

    Nature

    (2017)
  • Australian Bureau of Statistics

    Statistical geography

  • Australian Bureau of Statistics

    Socio-Economic Indexes for Areas (SEIFA) - Technical Paper 2006

    (2008)
  • Cited by (13)

    • Built environment attributes and their influence on walkability

      2022, International Journal of Sustainable Transportation
    View all citing articles on Scopus
    View full text