Identifying clusters of active transportation using spatial scan statistics.

Statistical Research and Applications Branch, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, Maryland, USA.
American journal of preventive medicine (Impact Factor: 4.28). 09/2009; 37(2):157-66. DOI: 10.1016/j.amepre.2009.04.021
Source: PubMed

ABSTRACT There is an intense interest in the possibility that neighborhood characteristics influence active transportation such as walking or biking. The purpose of this paper is to illustrate how a spatial cluster identification method can evaluate the geographic variation of active transportation and identify neighborhoods with unusually high/low levels of active transportation.
Self-reported walking/biking prevalence, demographic characteristics, street connectivity variables, and neighborhood socioeconomic data were collected from respondents to the 2001 California Health Interview Survey (CHIS; N=10,688) in Los Angeles County (LAC) and San Diego County (SDC). Spatial scan statistics were used to identify clusters of high or low prevalence (with and without age-adjustment) and the quantity of time spent walking and biking. The data, a subset from the 2001 CHIS, were analyzed in 2007-2008.
Geographic clusters of significantly high or low prevalence of walking and biking were detected in LAC and SDC. Structural variables such as street connectivity and shorter block lengths are consistently associated with higher levels of active transportation, but associations between active transportation and socioeconomic variables at the individual and neighborhood levels are mixed. Only one cluster with less time spent walking and biking among walkers/bikers was detected in LAC, and this was of borderline significance. Age-adjustment affects the clustering pattern of walking/biking prevalence in LAC, but not in SDC.
The use of spatial scan statistics to identify significant clustering of health behaviors such as active transportation adds to the more traditional regression analysis that examines associations between behavior and environmental factors by identifying specific geographic areas with unusual levels of the behavior independent of predefined administrative units.

Download full-text


Available from: Linda Pickle, Jul 05, 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Spatial cluster analysis is a uniquely interdisciplinary endeavour, and so it is important to communicate and disseminate ideas, innovations, best practices and challenges across practitioners, applied epidemiology researchers and spatial statisticians. In this research we conducted a scoping review to systematically search peer-reviewed journal databases for research that has employed spatial cluster analysis methods on individual-level, address location, or x and y coordinate derived data. To illustrate the thematic issues raised by our results, methods were tested using a dataset where known clusters existed. Point pattern methods, spatial clustering and cluster detection tests, and a locally weighted spatial regression model were most commonly used for individual-level, address location data (n = 29). The spatial scan statistic was the most popular method for address location data (n = 19). Six themes were identified relating to the application of spatial cluster analysis methods and subsequent analyses, which we recommend researchers to consider; exploratory analysis, visualization, spatial resolution, aetiology, scale and spatial weights. It is our intention that researchers seeking direction for using spatial cluster analysis methods, consider the caveats and strengths of each approach, but also explore the numerous other methods available for this type of analysis. Applied spatial epidemiology researchers and practitioners should give special consideration to applying multiple tests to a dataset. Future research should focus on developing frameworks for selecting appropriate methods and the corresponding spatial weighting schemes.
    Geospatial health 05/2013; 7(2):183-98. DOI:10.4081/gh.2013.79 · 1.00 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper offers a comprehensive analysis of physical recreational activity engagement by simultaneously examining the location, time of day, day of week, and social context of these activities. Data from the 2007 American Time Use Survey is used to estimate a mixed multiple discrete continuous extreme value (MDCEV) model that simultaneously captures the multiple activity categories that people may choose and the time that they allocate to each of the activity categories. It is found that a host of socio-economic, demographic, household, employment-related, and environmental variables affect the choice of physical recreational activity engagement with respect to location, time of day, day of week, and social context.
    Transportation Letters The International Journal of Transportation Research 08/2009; 3(3). DOI:10.3328/TL.2011.03.03.201-217 · 0.41 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Temporal, spatial and space-time scan statistics are commonly used to detect and evaluate the statistical significance of temporal and/or geographical disease clusters, without any prior assumptions on the location, time period or size of those clusters. Scan statistics are mostly used for count data, such as disease incidence or mortality. Sometimes there is an interest in looking for clusters with respect to a continuous variable, such as lead levels in children or low birth weight. For such continuous data, we present a scan statistic where the likelihood is calculated using the the normal probability model. It may also be used for other distributions, while still maintaining the correct alpha level. In an application of the new method, we look for geographical clusters of low birth weight in New York City.
    International Journal of Health Geographics 10/2009; 8:58. DOI:10.1186/1476-072X-8-58 · 2.62 Impact Factor