Yan Kestens

GPS tracking in neighborhood and health studies: A step forward for environmental exposure assessment, a step backward for causal inference?

Recent studies have relied on GPS tracking to assess exposure to environmental characteristics over daily life schedules. Combining GPS and GIS allows for advances in environmental exposure assessment. However, biases related to selective daily mobility preclude assessment of environmental effects, to the extent that these studies may represent a step backward in terms of assessment of causal effects. A solution may be to integrate the Public health / Nutrition approach and the Transportation approach to GPS studies, so as to combine a GPS and accelerometer data collection with an electronic mobility survey. Correcting exposure measures and improving study designs with this approach may permit mitigating biases related to selective daily mobility.

Using experienced activity spaces to measure foodscape exposure

Researchers are increasingly interested in understanding how food environments influence eating behavior and weight-related health outcomes. Little is known about the dose-response relationship between foodscapes and behavior or weight, with measures of food exposure having mainly focused on fixed anchor points including residential neighborhoods, schools, or workplaces. Recent calls have been made to extend the consideration of environmental influences beyond local neighborhoods and also to shift away from place-based, to people-based, measures of exposure. This report presents analyses of novel activity-space measures of exposure to foodscapes, combining travel survey data with food store locations in Montreal and Quebec City, Canada. The resulting individual activity-space experienced foodscape exposure measures differ from traditional residential-based measures, and show variations by age and income levels. Furthermore, these activity-space exposure measures once modeled, can be used as predictors of health outcomes. Hence, travel surveys can be used to estimate environmental exposure for health survey participants.

Social capital and core network ties: A validation study of individual-level social capital measures and their association with extra- and intra-neighborhood ties, and self-rated health

Research on social capital and health has assumed that measures of trust, participation, and perceived cohesion capture aspects of peoples neighborhood social connections. This study uses data on the personal networks of 2707 Montreal adults in 300 different neighborhoods to examine the association of socio-demographic and social capital variables with the likelihood of having core ties, core neighborhood ties, and high self-rated health (SRH). Persons with higher household income were more likely to have core ties, but less likely to have core neighborhood ties. Persons with greater diversity in extra-neighborhood network capital were more likely to have core ties, and persons with greater diversity in intra-neighborhood network capital were more likely to have core neighborhood ties. Generalized trust, perceived neighborhood cohesion, and extra-neighborhood network diversity were shown associated with high SRH. Conventional measures of social capital may not capture network mechanisms. Findings suggest a critical appraisal of the mechanisms linking social capital and health, and the further delineation of network and psychosocial mechanisms in understanding these links.

Field validation of listings of food stores and commercial physical activity establishments from secondary data.

BackgroundFood- and activity-related establishments are increasingly viewed as neighbourhood resources that potentially condition health-related behaviour. The primary objective of the current study was to establish, using ground truthing (on-site verification), the validity of measures of availability of food stores and physical activity establishments that were obtained from commercial database and Internet searches. A secondary objective was to examine differences in validity results according to neighbourhood characteristics and commercial establishment categories.MethodsLists of food stores and physical activity-related establishments in 12 census tracts within the Montreal metropolitan region were compiled using a commercial database (n = 171 establishments) and Internet search engines (n = 123 establishments). Ground truthing through field observations was performed to assess the presence of listed establishments and identify those absent. Percentage agreement, sensitivity (proportion of establishments found in the field that were listed), and positive predictive value (proportion of listed establishments found in the field) were calculated and contrasted according to data sources, census tracts characteristics, and establishment categories.ResultsAgreement with field observations was good (0.73) for the commercial list, and moderate (0.60) for the Internet-based list. The commercial list was superior to the Internet-based list for correctly listing establishments present in the field (sensitivity), but slightly inferior in terms of the likelihood that a listed establishment was present in the field (positive predictive value). Agreement was higher for food stores than for activity-related establishments.ConclusionCommercial data sources may provide a valid alternative to field observations and could prove a valuable tool in the evaluation of commercial environments relevant to eating behaviour. In contrast, this study did not find strong evidence in support of commercial and Internet data sources to represent neighbourhood opportunities for active lifestyle.

Social Inequalities in Food Exposure Around Schools in an Urban Area

BACKGROUND The obesity epidemic among children and youth, and the social gradient in this relationship, could be related to differential exposure to food sources in primary environments. Although the positive association between area-level deprivation and fast-food outlets offering high-calorie foods has been well documented, few studies have evaluated food sources around school settings. PURPOSE This study evaluated the relationships among food sources around schools, neighborhood income, and commercial density. METHODS A GIS was used to derive measures of exposure to fast-food outlets, fruit and vegetable stores, and full-service restaurants near primary and secondary schools in Montreal, Canada, in 2005. Food source availability was analyzed in 2009 in relation to neighborhood income for the area around schools, accounting for commercial density. RESULTS For the 1168 schools identified, strong neighborhood income gradients were observed in relation to food sources. Relative to the highest income-quartile schools, the odds of a fast-food outlet being located within 750 m of a low income-quartile school was 30.9 (95% CI=19.6, 48.9). Similar relationships were observed for full-service restaurants (OR=77, 95% CI=35, 169.3) and fruit and vegetable stores (OR=29.6, 95% CI=18.8, 46.7). These associations were reduced, but remained significant in models accounting for commercial density. CONCLUSIONS Food source exposure around schools is inversely associated with neighborhood income, but commercial density partly accounts for this association. Further research is necessary to document food consumption among youth attending schools in relation to nearby food source opportunities.

Heterogeneity in hedonic modelling of house prices: looking at buyers’ household profiles

This paper introduces household-level data into hedonic models in order to measure the heterogeneity of implicit prices regarding household type, age, educational attainment, income, and the previous tenure status of the buyers. Two methods are used for this purpose: a first series of models uses expansion terms, whereas a second series applies Geographically Weighted Regressions. Both methods yield conclusive results, showing that the marginal value given to certain property specifics and location attributes do vary regarding the characteristics of the buyer’s household. Particularly, major findings concern the significant effect of income on the location rent as well as the premium paid by highly-educated households in order to fulfil social homogeneity.

The Impact of Surrounding Land Use and Vegetation on Single-Family House Prices

The aim of this paper is to assess the marginal effect of land-use locational externalities on the sale price of single-family houses, considering various spatial scales—in accordance with perception theories—and trade-off with accessibility to the city centre. From land-use and vegetation data derived from aerial photographs and Landsat TM satellite images, two sets of hedonic models, using OLS regression, are built from two samples of single-family properties sold in Quebec City. A standard model integrates property-specific factors, census factors, accessibility, and location attributes. In a second model, land-use and vegetation variables are considered on various spatial scales; a third step introduces the interaction effect of the surrounding land use with location, with car-time distance to the main activity centres being used as the main indicator. This allows for an analysis of the spatial variation of the environmental impact throughout the city considering relative proximity to the centre. The successful integration of environmental variables concerning location enhances our understanding of the local land-use and vegetation effects. It also improves the overall performance of the model while virtually removing spatial autocorrelation among residuals. Such models could be used in order to assess the fiscal impacts of various land zoning by law policies, thereby providing planning administrations with a useful decisionmaking tool.

Association between Activity Space Exposure to Food Establishments and Individual Risk of Overweight

Objective Environmental exposure to food sources may underpin area level differences in individual risk for overweight. Place of residence is generally used to assess neighbourhood exposure. Yet, because people are mobile, multiple exposures should be accounted for to assess the relation between food environments and overweight. Unfortunately, mobility data is often missing from health surveys. We hereby test the feasibility of linking travel survey data with food listings to derive food store exposure predictors of overweight among health survey participants. Methods Food environment exposure measures accounting for non-residential activity places (activity spaces) were computed and modelled in Montreal and Quebec City, Canada, using travel surveys and food store listings. Models were then used to predict activity space food exposures for 5,578 participants of the Canadian Community Health Survey. These food exposure estimates, accounting for daily mobility, were used to model self-reported overweight in a multilevel framework. Median Odd Ratios were used to assess the proportion of between-neighborhood variance explained by such food exposure predictors. Results Estimates of food environment exposure accounting for both residential and non-residential destinations were significantly and more strongly associated with overweight than residential-only measures of exposure for men. For women, residential exposures were more strongly associated with overweight than non-residential exposures. In Montreal, adjusted models showed men in the highest quartile of exposure to food stores were at lesser risk of being overweight considering exposure to restaurants (OR = 0.36 [0.21–0.62]), fast food outlets (0.48 [0.30–0.79]), or corner stores (0.52 [0.35–0.78]). Conversely, men experiencing the highest proportion of restaurants being fast-food outlets were at higher risk of being overweight (2.07 [1.25–3.42]). Women experiencing higher residential exposures were at lower risk of overweight. Conclusion Using residential neighbourhood food exposure measures may underestimate true exposure and observed associations. Using mobility data offers potential for deriving activity space exposure estimates in epidemiological models.

Use of a new public bicycle share program in Montreal, Canada.

BACKGROUND Cycling contributes to physical activity and health. Public bicycle share programs (PBSPs) increase population access to bicycles by deploying bicycles at docking stations throughout a city. Minimal research has systematically examined the prevalence and correlates of PBSP use. PURPOSE To determine the prevalence and correlates of use of a new public bicycle share program called BIXI (name merges the word BIcycle and taXI) implemented in May 2009 in Montreal, Canada. METHODS A total of 2502 adults were recruited to a telephone survey in autumn 2009 via random-digit dialing according to a stratified random sampling design. The prevalence of BIXI bicycle use was estimated. Multivariate logistic regression allowed for identification of correlates of use. Data analysis was conducted in spring and summer 2010. RESULTS The unweighted mean age of respondents was 47.4 (SD=16.8) years and 61.4% were female. The weighted prevalence for use of BIXI bicycles at least once was 8.2%. Significant correlates of BIXI bicycle use were having a BIXI docking station within 250 m of home, being aged 18-24 years, being university educated, being on work leave, and using cycling as the primary mode of transportation to work. CONCLUSIONS A newly implemented public bicycle share program attracts a substantial fraction of the population and is more likely to attract younger and more educated people who currently use cycling as a primary transportation mode.

Modelling interactions of location with specific value of housing attributes

This paper presents a procedure for considering interactions of neighbourhood quality and property specifics within hedonic models of housing price. It handles interactions between geographical factors and the marginal contribution of each property attribute for enhancing values assessment. Making use of simulation procedures, it is combining GIS technology and spatial statistics to define principal components of accessibility and socio‐economic census related to transaction prices of single‐family homes. An application to the housing market of the Quebec Urban Community (more than 3,600 bungalows transacted in 1990 and 1991) illustrates its usefulness for building spatial hedonic models, while controlling for multicollinearity, spatial autocorrelation and heteroskedasticity. Distance‐weighted averages of each property attribute in the neighbourhood and interactions of property attributes with each principal component are used to detect any spatial effect on sale price variations. This first‐stage spatial hedonic model approximates market prices, which are then used in order to compare “expected” and actual property tax amounts, which are added to obtain a second‐stage model incorporating fiscal effects on house values. Interactions between geographical factors and property specifics are computed using formulae avoiding multicollinearity problems, while considering several processes responsible for spatial variability. For each property attribute, they define sub‐models which can be used to map variations, across the city, of its marginal value, assessing the cross‐effect of geographical location (in terms of neighbourhood profiles and accessibility to services) and its own valuation parameters. Moreover, this procedure distinguishes property attributes, exerting a stable contribution to value (constant over the entire region) from those whose implicit price significantly varies over space.