Using remote sensing to define environmental characteristics related to physical activity and dietary behaviours: a systematic review (the SPOTLIGHT project)

Redesigning urban parks for active living in dense urban areas: a remote audit approach

The widespread use of virtual audits has transformed the assessment of urban design attributes by eliminating the necessity for on-site visits. However, there remains a lack of virtual audit tools specifically designed for evaluating urban parks in dense urban environments. This study aims to (1) adapt the Audit Tool for Activity-friendly Parks in Dense Urban Areas (TAPS) into a remote audit tool (R-TAPS), and (2) evaluate its reliability and validity. Trained auditors used R-TAPS to conduct remote audits of urban parks (n = 53) in Tokyo, with on-site audits for a subset (n = 25). Kappa statistics and percent agreement assessed inter-rater reliability, and intra-class correlation coefficient (ICC) verified convergent validity. R-TAPS showed moderate to almost perfect agreement in 89% of the items. Remote and on-site audits exhibited a high positive correlation (ICC = 0.73). R-TAPS offers a reliable tool for virtually evaluating urban parks to promote physical activity in dense urban settings.

Using Google Street View for virtual observations of neighborhoods and dwelling units: A feasibility study

In face-to-face household surveys, field interviewers are sometimes asked to make notes of characteristics of the dwelling unit on the sampled address as well as its surroundings before making contact with a household member living at the sample address. Field interviewer observations of this kind are used to improve efficiency of field data collection and to be used as nonresponse adjustment. However, field interviewer observations can be expensive and the quality of observations needs to be improved. Recently, survey organizations start to utilize Google Street View to conduct virtual observations of the dwelling unit and the neighborhood. This paper reports a feasibility study that evaluates the feasibility of using virtual observations, assesses its agreement with field interviewer observation results, and examine whether virtual observations correlate with survey response status and survey estimates. We found moderate to high agreements between virtual and interviewer observation results. We also found that some observation results are significantly related to response status and survey estimates. However, virtual observations using GSV have coverage issues, which could limit their potential use.

Integrating the Physical Environment Within a Population Neuroscience Perspective

Population neuroscience recognises the role of the environment in shaping brain, behaviour, and mental health. An overview of current evidence from neuroscientific and epidemiological studies highlights the protective effects of nature on cognitive function and stress reduction, the detrimental effects of urban living on mental health, and emerging concerns relating to extreme weather events and eco-anxiety. Despite the growing body of evidence in this area, knowledge gaps remain due to inconsistent measures of exposure and a reliance on small samples. In this chapter, attention is given to the physical environment and population-level studies as a necessary starting point for exploring the long-term impacts of environmental exposures on mental health, and for informing future research that may capture immediate emotional and neural responses to the environment. Key data sources, including remote sensing imagery, administrative, sensor, and social media data, are outlined. Appropriate measures of exposure are advocated for, recognising the value of area-level measures for estimating exposure over large study samples and spatial and temporal scales. Although integrating data from multiple sources requires consideration for data quality and completeness, deep learning and the increasing availability of high-resolution data present opportunities to build a more complete picture of physical environments. Advances in leveraging detailed locational data are discussed as a subsequent approach for building upon initial observations from population studies and improving understanding of the mechanisms underlying behaviour and human-environment interactions.

Measuring changes in neighborhood disorder using Google Street View longitudinal imagery: a feasibility study

Few studies have used longitudinal imagery of Google Street View (GSV) despite its potential for measuring changes in urban streetscapes characteristics relevant to health, such as neighborhood disorder. Neighborhood disorder has been previously associated with health outcomes. We conducted a feasibility study exploring image availability over time in the Philadelphia metropolitan region and describing changes in neighborhood disorder in this region between 2009, 2014, and 2019. Our team audited Street View images from 192 street segments in the Philadelphia Metropolitan Region. On each segment, we measured the number of images available through time, and for locations where imagery from more than one time point was available, we collected 8 neighborhood disorder indicators at 3 different times (up to 2009, up to 2014, and up to 2019). More than 70% of streets segments had at least one image. Neighborhood disorder increased between 2009 and 2019. Future studies should study the determinants of change of neighborhood disorder using longitudinal GSV imagery.

Design, calibration, and testing of a mobile sensor system for air pollution and built environment data collection: The urban scanner platform

This paper describes a mobile air pollution sampling system, the Urban Scanner, which aims at gathering dense spatiotemporal air quality data to support urban air quality and exposure science. Urban Scanner comprises custom vehicle-mounted sensors for air pollution, meteorology, and built environment data collection (low-cost sensors, wind anemometer, 360 deg camera, LIDAR, GPS) as well as a server to store, process, and map all gathered geo-referenced sensory information. Two levels of sensor calibration were implemented, both in a chamber and in the field, against reference instrumentation. Chamber tests and a set of mathematical tools were developed to correct for sensor noise (wavelet denoising), misalignment (linear and nonlinear), and hysteresis removal. Models based on chamber testing were further refined based on field co-location. While field co-location captures natural changes in air pollution and meteorology, chamber tests allow for simulating fast transitions in these variables, like the transitions experienced by a mobile sensor in an urban environment. The best suite of models achieved an R² higher than 0.9 between sensor output and reference station observations and an RMSE of 2.88 ppb for nitrogen dioxide and 4.03 ppb for ozone. A mobile sampling campaign was conducted in the city of Toronto, Canada, to further test Urban Scanner. We observe that the platform adequately captures spatial and temporal variability in urban air pollution, leading to the development of land-use regression models with high explanatory power.

Understanding perceptions of neighborhood health and non-communicable disease risk in urban contexts in Ghana

Global health surveillance reports show Africa's epidemiologic transition from one dominated by higher burdens of nutritional, maternal, and communicable diseases to one increasingly dominated by non-communicable diseases (NCDs). Debates on the increasing cases of NCDs in the African context have focused on individualistic risk factors to the neglect of other similar important determinants such as the living environment. Drawing on theoretical tenets of the protection motivation theory and using cross-sectional data, we examined neighborhood risk perceptions and self-rated risk of developing NCDs in Ghana. The dependent variable 'self-rated risk of developing NCDs' was measured as a binary outcome and the focal independent variable - perceived neighborhood health risk - as an index. We fitted multivariate multilevel regression models to a sample of 1376 individuals across 9 neighborhoods. Results show that respondents who perceived their neighborhoods as risky were more likely to rate their risk of developing NCDs high. A unit increase in neighborhood violence was associated with 8% likelihood of self-rated risk of developing NCDs. However, a unit increase in the aesthetic quality of respondent's neighborhood was associated with lower likelihood of self-rated risk of developing NCDs. Engaging in regular physical activity, and non-tobacco use were associated with a lower likelihood of perceived NCDs risk. We suggest policy agendas intended for reducing the burden of NCDs in Ghana and other LMICs could incorporate programs that target improving environmental characteristics to minimize risks and offer people the opportunity to make healthy choices.

Street View Imagery (SVI) in the Built Environment: A Theoretical and Systematic Review

Street view imagery (SVI) provides efficient access to data that can be used to research spatial quality at the human scale. The previous reviews have mainly focused on specific health findings and neighbourhood environments. There has not been a comprehensive review of this topic. In this paper, we systematically review the literature on the application of SVI in the built environment, following a formal innovation–decision framework. The main findings are as follows: (I) SVI remains an effective tool for automated research assessments. This offers a new research avenue to expand the built environment-measurement methods to include perceptions in addition to physical features. (II) Currently, SVI is functional and valuable for quantifying the built environment, spatial sentiment perception, and spatial semantic speculation. (III) The significant dilemmas concerning the adoption of this technology are related to image acquisition, the image quality, spatial and temporal distribution, and accuracy. (IV) This research provides a rapid assessment and provides researchers with guidance for the adoption and implementation of SVI. Data integration and management, proper image service provider selection, and spatial metrics measurements are the critical success factors. A notable trend is the application of SVI towards a focus on the perceptions of the built environment, which provides a more refined and effective way to depict urban forms in terms of physical and social spaces.

Geospatial Food Environment Exposure and Obesity Among Low Income Baltimore City Children: Associations Differ by Data Source and Processing Method

Due to the high prevalence of childhood obesity, it is imperative to assess the relationship children's access to food retailers and obesity. However, the influence of methodological decisions on these associations has been understudied. We examined relationships between different measures of geospatial food environment (using 4 data sources, and 2 data processing methods), and BMI in a sample of low-income children in Baltimore, Maryland. The choice of data sources and data processing methods produced large differences in estimates of children's exposures to certain store types, such as supermarket-like stores, but had less impact on associations with BMI z-scores.

Validation of desk-based audits using Google Street View® to monitor the obesogenic potential of neighbourhoods in a pediatric sample: a pilot study in the QUALITY cohort

Background

The suitability of geospatial services for auditing neighbourhood features relevant to pediatric obesity remains largely unexplored. Our objectives were to (i) establish the measurement properties of a desk-based audit instrument that uses Google Street View ^® to assess street- and neighbourhood-level features relevant to pediatric obesity (QUALITY-NHOOD tool, the test method) and (ii) comment on its capacity to detect changes in the built environment over an 8-year period. In order to do so, we compared this tool with an on-site auditing instrument (the reference method).

Methods

On-site audits of 55 street- and neighbourhood-level features were completed in 2008 in 512 neighbourhoods from the QUALITY cohort study. In 2015, both repeat on-site and desk-based audits were completed in a random sample of 30 of these neighbourhoods.

Results

Agreement between both methods was excellent for almost all street segment items (range 91.9–99.7%), except for road type (81.0%), ads/commercial billboards (81.7%), road-sidewalk buffer zone (76.1%), and road-bicycle path buffer zone (53.3%). It was fair to poor for perceived quality, safety and aesthetics items (range 59.9–87.6%), as well as for general impression items (range 40.0–86.7%). The desk-based method over-detected commercial billboards and road-sidewalk buffer zone, and generally rated neighbourhoods as less safe, requiring more effort to get around, and having less aesthetic appeal. Change detected over the 8-year period was generally similar for both methods, except that the desk-based method appeared to amplify the increase in the number of segments with signs of social disorder.

Conclusions

The QUALITY-NHOOD tool is deemed adequate for evaluating and monitoring changes in pedestrian- and traffic-related features applicable to pediatric populations. Applications for monitoring the obesogenic nature of neighbourhoods appear warranted.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12942-022-00301-8.

Examining the Implementation of Activity-Friendly Zoning and Land Use Policies Through the Use of Google Street View Measures: A Pilot Study

INTRODUCTION

Pedestrian-oriented zoning and land use policies are being used by local jurisdictions as tools to implement population-level built environmental changes to create more walkable communities. There is a paucity of evidence examining whether these policies lead to actual changes in the built environment. We used Google Street View (GSV), an established, less expensive, alternative built environment data collection method, to conduct an exploratory pilot study of 19 jurisdictions to examine associations between variations in the presence of these adopted zoning policies and their corresponding specific street-level built environment features.

METHODS

Samples of 10 large and 9 small jurisdictions (18 municipalities and 1 county) were purposively selected on the basis of the presence of activity-friendly zoning policy provisions (sidewalks, crosswalks, bike-pedestrian connectivity, street connectivity, trails/paths, bike lanes, bike parking, and other items). Corresponding activity-friendly street-level built environment measures were constructed using GSV. Street segments in these jurisdictions were sampled using ArcGIS and stratified by type (residential and arterial) and income (high, medium, and low).

RESULTS

A total of 4363 street segments were audited across the 19 sampled jurisdictions. Results show significant differences in the presence of activity-friendly street features when the corresponding zoning policy element was addressed in New Urbanist zones/districts in the site's zoning code (eg, crosswalks, 24.48% vs 16.18%; and bike lanes, 12.60% vs 7.14%). Street segments in the middle- and high-income block groups were less likely to have activity-friendly features than low-income segments, except bike lanes.

CONCLUSIONS

Results show that having activity-friendly policy provisions embedded in a jurisdiction's (municipality/county) zoning codes was associated with a greater presence of the corresponding built environmental street feature on the ground. Results suggest that the methods tested in this article may be a useful policy tool for local governments to identify high need areas that should be prioritized for built environment improvements.

A System for Monitoring the Environment of Historic Places Using Convolutional Neural Network Methodologies

This work aims to contribute to better understanding the use of public street spaces. (1) Background: In this sense, with a multidisciplinary approach, the objective of this work is to propose an experimental and reproducible method on a large scale. (2) Study area: The applied methodology uses artificial intelligence to analyze Google Street View (GSV) images at street level. (3) Method: The purpose is to validate a methodology that allows us to characterize and quantify the use (pedestrians and cars) of some squares in Rome belonging to different historical periods. (4) Results: Through the use of machine vision techniques, typical of artificial intelligence and which use convolutional neural networks, a historical reading of some selected squares is proposed, with the aim of interpreting the dynamics of use and identifying some critical issues in progress. (5) Conclusions: This work validated the usefulness of a method applied to the use of artificial intelligence for the analysis of GSV images at street level.

International evaluation of the Microscale Audit of Pedestrian Streetscapes (MAPS) Global instrument: comparative assessment between local and remote online observers

OBJECTIVES

The use of online imagery by non-local observers to conduct remote, centralized collection of streetscape audit data in international studies has the potential to enhance efficiency of collection and comparability of such data for research on built environments and health. The objectives of the study were to measure (1) the consistency in responses between local in-field observers and non-local remote online observers and (2) the reliability between in-country online observers and non-local remote online observers using the Microscale Audit of Pedestrian Streetscapes Global tool to characterize pedestrian-related features along streets in five countries.

METHODS

Consistency and inter-rater reliability were analyzed between local and non-local observers on a pooled database of 200 routes in five study regions (Melbourne, Australia; Ghent, Belgium; Curitiba, Brazil; Hong Kong, China; and Valencia, Spain) for microscale environmental feature subscales and item-level variables using the intraclass correlation coefficient (ICC).

RESULTS

A local in-field versus remote online comparison had an ICC of 0.75 (95 % CI: 0.68-0.80) for the grand total score. An ICC of 0.91 (95 % CI: 0.88-0.93) was found for the local online versus remote online comparison. Positive subscales yielded stronger results in comparison to negative subscales, except for the similarly poor-performing positive aesthetics/social characteristics.

CONCLUSIONS

This study demonstrated remote audits of microscale built environments using online imagery had good reliability with local in-field audits and excellent reliability with local online audits. Results generally supported remote online environmental audits as comparable to local online audits. This identification of low-cost and efficient data acquisition methods is important for expanding research on microscale built environments and physical activity globally.

Using Street View Imagery to Predict Street-Level Particulate Air Pollution

Land-use regression (LUR) models are frequently applied to estimate spatial patterns of air pollution. Traditional LUR often relies on fixed-site measurements and GIS-derived variables with limited spatial resolution. We present an approach that leverages Google Street View (GSV) imagery to predict street-level particulate air pollution (i.e., black carbon [BC] and particle number [PN] concentrations). We developed empirical models based on mobile monitoring data and features extracted from ∼52 500 GSV images using a deep learning model. We tested theory- and data-driven feature selection methods as well as models using images within varying buffer sizes (50-2000 m). Compared to LUR models with traditional variables, our models achieved similar model performance using the street-level predictors while also identifying additional potential hotspots. Adjusted R² (10-fold CV R²) with integrated feature selection was 0.57-0.64 (0.50-0.57) and 0.65-0.73 (0.61-0.66) for BC and PN models, respectively. Models using only features near the measurement locations (i.e., GSV images within 250 m) explained ∼50% of air pollution variability, indicating PN and BC are strongly affected by the street-level built environment. Our results suggest that GSV imagery, processed with computer vision techniques, is a promising data source to develop LUR models with high spatial resolution and consistent predictor variables across administrative boundaries.

Adapting the SPOTLIGHT Virtual Audit Tool to assess food and activity environments relevant for adolescents: a validity and reliability study

Background

Physical inactivity and unhealthy diet are key behavioral determinants underlying obesity. The neighborhood environment represents an important arena for modifying these behaviors, and hence reliable and valid tools to measure it are needed. Most existing virtual audit tools have been designed to assess either food or activity environments deemed relevant for adults. Thus, there is a need for a tool that combines the assessment of food and activity environments, and which focuses on aspects of the environment relevant for youth.

Objective

The aims of the present study were: (a) to adapt the SPOTLIGHT Virtual Audit Tool (S-VAT) developed to assess characteristics of the built environment deemed relevant for adults for use in an adolescent population, (b) to assess the tool’s inter- and intra-rater reliability, and (c) to assess its criterion validity by comparing the virtual audit to a field audit.

Methods

The tool adaptation was based on literature review and on results of a qualitative survey investigating how adolescents perceived the influence of the environment on dietary and physical activity behaviors. Sixty streets (148 street segments) in six neighborhoods were randomly selected as the study sample. Two raters assessed the inter- and intra-rater reliability and criterion validity, comparing the virtual audit tool to a field audit. The results were presented as percentage agreement and Cohen’s kappa (κ).

Results

Intra-rater agreement was found to be moderate to almost perfect (κ = 0.44–0.96) in all categories, except in the category aesthetics (κ = 0.40). Inter-rater agreement between auditors ranged from fair to substantial for all categories (κ = 0.24–0.80). Criterion validity was found to be moderate to almost perfect (κ = 0.56–0.82) for most categories, except aesthetics and grocery stores (κ = 0.26–0.35).

Conclusion

The adapted version of the S-VAT can be used to provide reliable and valid data on built environment characteristics deemed relevant for physical activity and dietary behavior among adolescents.

A dual-method approach toward measuring the built environment - sampling optimization, validity, and efficiency of using GIS and virtual auditing

In recent years, GIS and virtual auditing have been widely used to measure the built environment, and each method carries its strengths and weaknesses. To generate higher quality, more cost-effective, and less time-consuming measures, it is necessary to explore dual- or multi-method strategy toward sampling optimization, improvement of measurement, and enhancement of efficiency. To justify the proposed dual-method approach, the study has three major objectives. First, it examines the uncertainties associated with different sample sizes by using GIS to generate scenarios that contrast the validity of measurements to aid sampling optimization in auditing. Second, it compares the validity of GIS measures with those generated through Google Street View Auditing (GSVA) by human raters. Third, it further examines the efficiency of the proposed dual-method approach in comparison to the two individual methods. Such investigation generates several novel findings. First, the study presents important evidence to support that GIS measures can offer sampling guidance applicable to the GSVA method. It leads to a recommendation of sampling sizes (5%-20%) for cases in settings with a mixture of affluent and disadvantaged neighborhoods. Results further indicate that different communities and certain individual features and characteristics may demand different sampling practices. Second, the study found that while GSVA is trustworthy for most characteristic variables, especially those that required subjective input, GIS provides well-validated measures for certain objective environmental attributes. Furthermore, the study reports that a dual-method approach of GIS and GSVA had a lower financial and time burden than using GSVA alone and is thus recommended as a comprehensive solution for optimal measurement of an objective built environment in mixed urban neighborhoods.

Predicting Traffic-Related Air Pollution Using Feature Extraction from Built Environment Images

This study develops a set of algorithms to extract built environment features from Google aerial and street view images, reflecting the microcharacteristics of an urban location as well as the different functions of buildings. These features were used to train a Bayesian regularized artificial neural network (BRANN) model to predict near-road air quality based on measurements of ultrafine particles (UFPs) and black carbon (BC) in Toronto, Canada. The resulting models [adjusted R² of 75.87 and 79.10% for UFP and BC and root mean squared error (RMSE) of 21,800 part/cm³ and 1300 ng/m³ for UFP and BC] were compared with similar ANN models developed using the same predictors, but extracted from traditional geographic information system (GIS) databases [adjusted R² of 58.74 and 64.21% for UFP and BC and RMSE values of 23,000 part/cm³ and 1600 ng/m³ for UFP and BC]. The models based on feature extraction exhibited higher predictive power, thus highlighting the greater accuracy of the proposed methods compared to GIS layers that are solely based on aerial images. A comparison with other neural network approaches as well as with a traditional land-use regression model demonstrates the strength of the BRANN model for spatial interpolation of air quality.

Interrater Reliability of Historical Virtual Audits Using Archived Google Street View Imagery

Impaired mobility occurs in up to half of community-dwelling older adults and is associated with poor health outcomes and high health care costs. Although the built environment impacts mobility, most studies of older adults lack information about environmental-level factors. In-person observational audits can be utilized but cannot assess the historical environment. We applied a 78-item checklist to archived Google Street View imagery to assess historical residence access and neighborhood characteristics. Interrater reliability between two raters was tested on 50 addresses using prevalence-adjusted and bias-adjusted kappa (PABAK). The mean PABAK for all items was .75, with 81% of the items having substantial (PABAK ≥ .61) or almost perfect (PABAK ≥ .81) agreement. Environmental assessment using archived virtual imagery has excellent reliability for factors related to residence access and many neighborhood characteristics. Archived imagery can assess past neighborhood characteristics, facilitating the use of historical environment data within existing cohorts.

Bus Stops Near Schools Advertising Junk Food and Sugary Drinks

Children rarely understand the full extent of the persuasive purpose of advertising on their eating behaviours. Addressing the obesogenic environments in which children live, through a quantification of outdoor advertising, is essential in informing policy changes and enforcing stricter regulations. This research explores the proportion of bus stop advertisements promoting non-core food and beverages within walking distance (500 m) from schools in Auckland, New Zealand while using Google Street View. Information was collected on: school type, decile, address, Walk Score^®, and Transit Score for all 573 schools in the Auckland region. Ground-truthing was conducted on 10% of schools and showed an alignment of 87.8%. The majority of advertisements on bus shelters were for non-food items or services (n = 541, 64.3%). Of the advertisements that were for food and/or beverages, the majority were for non-core foods (n = 108, 50.2%). There was no statistically significant difference between the variables core and non-core food and beverages and School decile (tertiles), Walk Score (quintiles), and Transit Score (quintiles). 12.8% of all bus stop advertisements in this study promoted non-core dietary options; highlighting an opportunity for implementing stricter regulations and policies preventing advertising unhealthy food and drink to children in New Zealand.

Field Validation of Commercially Available Food Retailer Data in the Netherlands

The aim of this study was to validate a Dutch commercial dataset containing information on the types and locations of food retailers against field audit data. Field validation of a commercial dataset ("Locatus") was conducted in February 2019. Data on the location and classification of food retailers were collected through field audits in 152 streets from four urban and four rural neighborhoods in the Netherlands. The classification of food retailers included eight types of grocery stores (e.g., supermarkets, bakeries) and four types of food outlets (e.g., cafés, take away restaurants). The commercial dataset in the studied area listed 322 food retailers, whereas the field audit counted 315 food retailers. Overall, the commercially available data showed "good" to "excellent" agreement statistics (>0.71) with field audit data for all three levels of analysis (i.e., location, classification and both combined) and across urban as well as rural areas. The commercial dataset under study provided an accurate description of the measured food environment. Therefore, policymakers and researchers should feel confident in using this commercial dataset as a source of secondary data.

Analyzing Associations Between Chronic Disease Prevalence and Neighborhood Quality Through Google Street View Images

Deep learning and, specifically, convoltional neural networks (CNN) represent a class of powerful models that facilitate the understanding of many problems in computer vision. When combined with a reasonable amount of data, CNNs can outperform traditional models for many tasks, including image classification. In this work, we utilize these powerful tools with imagery data collected through Google Street View images to perform virtual audits of neighborhood characteristics. We further investigate different architectures for chronic disease prevalence regression through networks that are applied to sets of images rather than single images. We show quantitative results and demonstrate that our proposed architectures outperform the traditional regression approaches.

Characterizing the Rural Opioid Use Environment in Kentucky Using Google Earth: Virtual Audit

Background

The opioid epidemic has ravaged rural communities in the United States. Despite extensive literature relating the physical environment to substance use in urban areas, little is known about the role of physical environment on the opioid epidemic in rural areas.

Objective

This study aimed to examine the reliability of Google Earth to collect data on the physical environment related to substance use in rural areas.

Methods

Systematic virtual audits were performed in 5 rural Kentucky counties using Google Earth between 2017 and 2018 to capture land use, health care facilities, entertainment venues, and businesses. In-person audits were performed for a subset of the census blocks.

Results

We captured 533 features, most of which were images taken before 2015 (71.8%, 383/533). Reliability between the virtual audits and the gold standard was high for health care facilities (>83%), entertainment venues (>95%), and businesses (>61%) but was poor for land use features (>18%). Reliability between the virtual audit and in-person audit was high for health care facilities (83%) and entertainment venues (62%) but was poor for land use (0%) and businesses (12.5%).

Conclusions

Poor reliability for land use features may reflect difficulty characterizing features that require judgment or natural changes in the environment that are not reflective of the Google Earth imagery because it was captured several years before the audit was performed. Virtual Google Earth audits were an efficient way to collect rich neighborhood data that are generally not available from other sources. However, these audits should use caution when the images in the observation area are dated.

The Association of Urban Greenness and Walking Behavior: Using Google Street View and Deep Learning Techniques to Estimate Residents' Exposure to Urban Greenness

Many studies have established that urban greenness is associated with better health outcomes. Yet most studies assess urban greenness with overhead-view measures, such as park area or tree count, which often differs from the amount of greenness perceived by a person at eye-level on the ground. Furthermore, those studies are often criticized for the limitation of residential self-selection bias. In this study, urban greenness was extracted and assessed from profile view of streetscape images by Google Street View (GSV), in conjunction with deep learning techniques. We also explored a unique research opportunity arising in a citywide residential reallocation scheme of Hong Kong to reduce residential self-selection bias. Two multilevel regression analyses were conducted to examine the relationships between urban greenness and (1) the odds of walking for 24,773 public housing residents in Hong Kong, (2) total walking time of 1994 residents, while controlling for potential confounders. The results suggested that eye-level greenness was significantly related to higher odds of walking and longer walking time in both 400 m and 800 m buffers. Distance to the closest Mass Transit Rail (MTR) station was also associated with higher odds of walking. Number of shops was related to higher odds of walking in the 800 m buffer, but not in 400 m. Eye-level greenness, assessed by GSV images and deep learning techniques, can effectively estimate residents’ daily exposure to urban greenness, which is in turn associated with their walking behavior. Our findings apply to the entire public housing residents in Hong Kong, because of the large sample size.

How has big data contributed to obesity research? A review of the literature

There has been growing interest in the potential of ‘big data’ to enhance our understanding in medicine and public health. Although there is no agreed definition of big data, accepted critical components include greater volume, complexity, coverage and speed of availability. Much of these data are ‘found’ (as opposed to ‘made’), in that they have been collected for non-research purposes, but could include valuable information for research. The aim of this paper is to review the contribution of ‘found’ data to obesity research to date, and describe the benefits and challenges encountered. A narrative review was conducted to identify and collate peer-reviewed research studies. Database searches conducted up to September 2017 found original studies using a variety of data types and sources. These included: retail sales, transport, geospatial, commercial weight management data, social media, and smartphones and wearable technologies. The narrative review highlights the variety of data uses in the literature: describing the built environment, exploring social networks, estimating nutrient purchases or assessing the impact of interventions. The examples demonstrate four significant ways in which ‘found’ data can complement conventional ‘made’ data: firstly, in moving beyond constraints in scope (coverage, size and temporality); secondly, in providing objective, quantitative measures; thirdly, in reaching hard-to-access population groups; and lastly in the potential for evaluating real-world interventions. Alongside these opportunities, ‘found’ data come with distinct challenges, such as: ethical and legal questions around access and ownership; commercial sensitivities; costs; lack of control over data acquisition; validity; representativeness; finding appropriate comparators; and complexities of data processing, management and linkage. Despite widespread recognition of the opportunities, the impact of ‘found’ data on academic obesity research has been limited. The merit of such data lies not in their novelty, but in the benefits they could add over and above, or in combination with, conventionally collected data.

Systematic review of the use of Google Street View in health research: Major themes, strengths, weaknesses and possibilities for future research

We systematically reviewed the current use of Google Street View (GSV) in health research and characterized major themes, strengths and weaknesses in order to highlight possibilities for future research. Of 54 qualifying studies, we found that most used GSV to assess the neighborhood built environment, followed by health policy compliance, study site selection, and disaster preparedness. Most studies were conducted in urban areas of North America, Europe, or New Zealand, with few studies from South America or Asia and none from Africa or rural areas. Health behaviors and outcomes of interest in these studies included injury, alcohol and tobacco use, physical activity and mental health. Major strengths of using GSV imagery included low cost, ease of use, and time saved. Identified weaknesses were image resolution and spatial and temporal availability, largely in developing regions of the world. Despite important limitations, GSV is a promising tool for automated environmental assessment for health research. Currently untapped areas of health research using GSV include identification of sources of air, soil or water pollution, park design and usage, amenity design and longitudinal research on neighborhood conditions.

Active aging - resilience and external support as modifiers of the disablement outcome: AGNES cohort study protocol

Background

Population aging increases the need for knowledge on positive aspects of aging, and contributions of older people to their own wellbeing and that of others. We defined active aging as an individual’s striving for elements of wellbeing with activities as per their goals, abilities and opportunities. This study examines associations of health, health behaviors, health literacy and functional abilities, environmental and social support with active aging and wellbeing. We will develop and validate assessment methods for physical activity and physical resilience suitable for research on older people, and examine their associations with active aging and wellbeing. We will examine cohort effects on functional phenotypes underlying active aging and disability.

Methods

For this population-based study, we plan to recruit 1000 participants aged 75, 80 or 85 years living in central Finland, by drawing personal details from the population register. Participants are interviewed on active aging, wellbeing, disability, environmental and social support, mobility, health behavior and health literacy. Physical activity and heart rate are monitored for 7 days with wearable sensors. Functional tests include hearing, vision, muscle strength, reaction time, exercise tolerance, mobility, and cognitive performance. Clinical examination by a nurse and physician includes an electrocardiogram, tests of blood pressure, orthostatic regulation, arterial stiffness, and lung function, as well as a review of chronic and acute conditions and prescribed medications. C-reactive protein, small blood count, cholesterol and vitamin D are analyzed from blood samples. Associations of factors potentially underlying active aging and wellbeing will be studied using multivariate methods. Cohort effects will be studied by comparing test results of physical and cognitive functioning with results of a cohort examined in 1989–90.

Conclusions

The current study will renew research on positive gerontology through the novel approach to active aging and by suggesting new biomarkers of resilience and active aging. Therefore, high interdisciplinary impact is expected. This cross-sectional study will not provide knowledge on temporal order of events or causality, but an innovative cross-sectional dataset provides opportunities for emergence of novel creative hypotheses and theories.

Assessing the Exposome with External Measures: Commentary on the State of the Science and Research Recommendations

The exposome comprises all environmental exposures that a person experiences from conception throughout the life course. Here we review the state of the science for assessing external exposures within the exposome. This article reviews (a) categories of exposures that can be assessed externally, (b) the current state of the science in external exposure assessment, (c) current tools available for external exposure assessment, and (d) priority research needs. We describe major scientific and technological advances that inform external assessment of the exposome, including geographic information systems; remote sensing; global positioning system and geolocation technologies; portable and personal sensing, including smartphone-based sensors and assessments; and self-reported questionnaire assessments, which increasingly rely on Internet-based platforms. We also discuss priority research needs related to methodological and technological improvement, data analysis and interpretation, data sharing, and other practical considerations, including improved assessment of exposure variability as well as exposure in multiple, critical life stages.

Reliability between online raters with varying familiarities of a region: Microscale Audit of Pedestrian Streetscapes (MAPS)

BACKGROUND

To test inter-rater reliability of the online Microscale Audit of Pedestrian Streetscapes (MAPS) tool between raters with varying familiarities of Phoenix, Arizona.

METHODS

The online MAPS tool, based on the MAPS in-field audit tool and scoring system, was used for audits. Sixty route pairs, 141 segment pairs, and 92 crossing pairs in Phoenix were included. Each route, segment or crossing was audited by two independent raters: one rater in Phoenix and the other in San Diego, California, respectively. Item, subscale scores, and total scores reliability analyses were computed using Kappa or intra-class correlation coefficient (ICC).

RESULTS

The route overall score had substantial reliability (ICC: 0.832). Of the route subscale and overall scores, sixteen out of twenty had moderate to substantial reliability (ICC: 0.616-0.906), and the four subscales had fair reliability (ICC: 0.409-0.563). Sixteen out of twenty scores in segment and crossing sections demonstrated fair to substantial reliability (ICC: 0.448-0.897), and the remaining four had slight reliability (ICC: 0.348-0.364).

CONCLUSIONS

Most of the online MAPS items, subscales, and overall scores demonstrated fair to substantial reliability between raters with varied familiarities of the Phoenix area. Results support use of online MAPS to measure microscale elements of the built environment by raters unfamiliar with a region.

Micro-scale environment and mental health in later life: Results from the Cognitive Function and Ageing Study II (CFAS II)

BACKGROUND

Poor micro-scale environmental features, such as graffiti and broken windows, have been associated with crime and signs of social disorder with a potential impact on mental health. The aim of this study is to investigate the association between micro-scale environment and mental health problems in later life, including cognitive (cognitive impairment and dementia) and common mental disorders (depressive and anxiety symptoms).

METHODS

The method of visual image audits was used to collect micro-scale environmental data for 3590 participants in the Cognitive Function and Ageing Study II, a population-based multicentre cohort of people aged 65 or above in England. Multilevel logistic regression was used to examine the associations between the quality of micro-scale environment and mental health problems taking into account urban/rural difference.

RESULTS

Poor quality of micro-scale environment was associated with nearly 20% increased odds of depressive (OR: 1.19; 95% CI: 0.99, 1.44) and anxiety symptoms (OR: 1.17; 95% CI: 0.99, 1.38) while the direction of association for cognitive disorders differed across urban and rural settings. Although higher odds of cognitive disorders were found in rural settings, living in a poor quality environment was associated with nearly twice higher odds of cognitive impairment (OR: 1.88; 95% CI: 1.18, 2.97) in urban conurbations but 20% lower odds in rural areas (OR: 0.80; 95% CI: 0.57, 1.11).

LIMITATIONS

The causal direction could not be fully determined due to the cross-sectional nature of the data. The visual nature of the environmental assessment tool means it likely does not fully capture features related to the availability of local support services, or opportunities for social participation and interaction.

CONCLUSIONS

The quality of micro-scale environment appears to be important to mental health in older people. Interventions may incorporate the environmental aspect to reduce cognitive and common mental disorders.

Online versus in-person comparison of Microscale Audit of Pedestrian Streetscapes (MAPS) assessments: reliability of alternate methods

Background

An online version of the Microscale Audit of Pedestrian Streetscapes (Abbreviated) tool was adapted to virtually audit built environment features supportive of physical activity. The current study assessed inter-rater reliability of MAPS Online between in-person raters and online raters unfamiliar with the regions.

Methods

In-person and online audits were conducted for a total of 120 quarter-mile routes (60 per site) in Phoenix, AZ and San Diego, CA. Routes in each city included 40 residential origins stratified by walkability and SES, and 20 commercial centers. In-person audits were conducted by raters residing in their region. Online audits were conducted by raters in the alternate location using Google Maps (Aerial and Street View) images. The MAPS Abbreviated Online tool consisted of four sections: overall route, street segments, crossings and cul-de-sacs. Items within each section were grouped into subscales, and inter-rater reliability (ICCs) was assessed for subscales at multiple levels of aggregation.

Results

Online and in-person audits showed excellent agreement for overall positive microscale (ICC = 0.86, 95% CI [0.80, 0.90]) and grand scores (ICC = 0.93, 95% CI [0.89, 0.95]). Substantial to near-perfect agreement was found for 21 of 30 (70%) subscales, valence, and subsection scores, with ICCs ranging from 0.62, 95% CI [0.50, 0.72] to 0.95, 95% CI [0.93, 0.97]. Lowest agreement was found for the aesthetics and social characteristics scores, with ICCs ranging from 0.07, 95% CI [−0.12, 0.24] to 0.27, 95% CI [0.10, 0.43].

Conclusions

Results support use of the MAPS Abbreviated Online tool to reliably assess microscale neighborhood features that support physical activity and may be used by raters residing in different geographic regions and unfamiliar with the audit areas.

Street Audits to Measure Neighborhood Disorder: Virtual or In-Person?

Neighborhood conditions may influence a broad range of health indicators, including obesity, injury, and psychopathology. In particular, neighborhood physical disorder-a measure of urban deterioration-is thought to encourage crime and high-risk behaviors, leading to poor mental and physical health. In studies to assess neighborhood physical disorder, investigators typically rely on time-consuming and expensive in-person systematic neighborhood audits. We compared 2 audit-based measures of neighborhood physical disorder in the city of Detroit, Michigan: One used Google Street View imagery from 2009 and the other used an in-person survey conducted in 2008. Each measure used spatial interpolation to estimate disorder at unobserved locations. In total, the virtual audit required approximately 3% of the time required by the in-person audit. However, the final physical disorder measures were significantly positively correlated at census block centroids (r = 0.52), identified the same regions as highly disordered, and displayed comparable leave-one-out cross-validation accuracy. The measures resulted in very similar convergent validity characteristics (correlation coefficients within 0.03 of each other). The virtual audit-based physical disorder measure could substitute for the in-person one with little to no loss of precision. Virtual audits appear to be a viable and much less expensive alternative to in-person audits for assessing neighborhood conditions.

Nature Contact and Human Health: A Research Agenda

BACKGROUND

At a time of increasing disconnectedness from nature, scientific interest in the potential health benefits of nature contact has grown. Research in recent decades has yielded substantial evidence, but large gaps remain in our understanding.

OBJECTIVES

We propose a research agenda on nature contact and health, identifying principal domains of research and key questions that, if answered, would provide the basis for evidence-based public health interventions.

DISCUSSION

We identify research questions in seven domains: a) mechanistic biomedical studies; b) exposure science; c) epidemiology of health benefits; d) diversity and equity considerations; e) technological nature; f) economic and policy studies; and g) implementation science.

CONCLUSIONS

Nature contact may offer a range of human health benefits. Although much evidence is already available, much remains unknown. A robust research effort, guided by a focus on key unanswered questions, has the potential to yield high-impact, consequential public health insights. https://doi.org/10.1289/EHP1663.

Optimizing Scoring and Sampling Methods for Assessing Built Neighborhood Environment Quality in Residential Areas

Optimization of existing measurement tools is necessary to explore links between aspects of the neighborhood built environment and health behaviors or outcomes. We evaluate a scoring method for virtual neighborhood audits utilizing the Active Neighborhood Checklist (the Checklist), a neighborhood audit measure, and assess street segment representativeness in low-income neighborhoods. Eighty-two home neighborhoods of Washington, D.C. Cardiovascular Health/Needs Assessment (NCT01927783) participants were audited using Google Street View imagery and the Checklist (five sections with 89 total questions). Twelve street segments per home address were assessed for (1) Land-Use Type; (2) Public Transportation Availability; (3) Street Characteristics; (4) Environment Quality and (5) Sidewalks/Walking/Biking features. Checklist items were scored 0–2 points/question. A combinations algorithm was developed to assess street segments’ representativeness. Spearman correlations were calculated between built environment quality scores and Walk Score^®, a validated neighborhood walkability measure. Street segment quality scores ranged 10–47 (Mean = 29.4 ± 6.9) and overall neighborhood quality scores, 172–475 (Mean = 352.3 ± 63.6). Walk scores^® ranged 0–91 (Mean = 46.7 ± 26.3). Street segment combinations’ correlation coefficients ranged 0.75–1.0. Significant positive correlations were found between overall neighborhood quality scores, four of the five Checklist subsection scores, and Walk Scores^® (r = 0.62, p < 0.001). This scoring method adequately captures neighborhood features in low-income, residential areas and may aid in delineating impact of specific built environment features on health behaviors and outcomes.

Variations in the Prevalence of Obesity Among European Countries, and a Consideration of Possible Causes

Over the last 10 years the prevalence of obesity across the European continent has in general been rising. With the exception of a few countries where a levelling-off can be perceived, albeit at a high level, this upward trend seems likely to continue. However, considerable country to country variation is noticeable, with the proportion of people with obesity varying by 10% or more. This variation is intriguing and suggests the existence of different profiles of risk or protection factors operating in different countries. The identification of such protection factors could indicate suitable targets for interventions to help manage the obesity epidemic in Europe. This report is the output of a 2-day workshop on the 'Diversity of Obesity in Europe'. The workshop included 14 delegates from 12 different European countries. This report contains the contributions and discussions of the materials and viewpoints provided by these 14 experts; it is not the output of a single mind. However, such is the nature of scientific analysis regarding obesity that it is possible that a different set of 14 experts may have come to a different set of conclusions. Therefore the report should not be seen as a definitive statement of a stable situation. Rather it is a focus for discussion and comment, and a vehicle to drive forward further understanding and management of obesity in Europe.

Built environmental correlates of cycling for transport across Europe

This cross-sectional study aimed to determine which objective built environmental factors, identified using a virtual neighbourhood audit, were associated with cycling for transport in adults living in five urban regions across Europe. The moderating role of age, gender, socio-economic status and country on these associations was also investigated. Overall, results showed that people living in neighbourhoods with a preponderance of speed limits below 30km/h, many bicycle lanes, with less traffic calming devices, more trees, more litter and many parked cars forming an obstacle on the road were more likely to cycle for transport than people living in areas with lower prevalence of these factors. Evidence was only found for seven out of 56 possible moderators of these associations. These results suggest that reducing speed limits for motorized vehicles and the provision of more bicycle lanes may be effective interventions to promote cycling in Europe.

A Reliable, Feasible Method to Observe Neighborhoods at High Spatial Resolution

INTRODUCTION

Systematic social observation (SSO) methods traditionally measure neighborhoods at street level and have been performed reliably using virtual applications to increase feasibility. Research indicates that collection at even higher spatial resolution may better elucidate the health impact of neighborhood factors, but whether virtual applications can reliably capture social determinants of health at the smallest geographic resolution (parcel level) remains uncertain. This paper presents a novel, parcel-level SSO methodology and assesses whether this new method can be collected reliably using Google Street View and is feasible.

METHODS

Multiple raters (N=5) observed 42 neighborhoods. In 2016, inter-rater reliability (observed agreement and kappa coefficient) was compared for four SSO methods: (1) street-level in person; (2) street-level virtual; (3) parcel-level in person; and (4) parcel-level virtual. Intra-rater reliability (observed agreement and kappa coefficient) was calculated to determine whether parcel-level methods produce results comparable to traditional street-level observation.

RESULTS

Substantial levels of inter-rater agreement were documented across all four methods; all methods had >70% of items with at least substantial agreement. Only physical decay showed higher levels of agreement (83% of items with >75% agreement) for direct versus virtual rating source. Intra-rater agreement comparing street- versus parcel-level methods resulted in observed agreement >75% for all but one item (90%).

CONCLUSIONS

Results support the use of Google Street View as a reliable, feasible tool for performing SSO at the smallest geographic resolution. Validation of a new parcel-level method collected virtually may improve the assessment of social determinants contributing to disparities in health behaviors and outcomes.

Optimising measurement of health-related characteristics of the built environment: Comparing data collected by foot-based street audits, virtual street audits and routine secondary data sources

The role of the neighbourhood environment in influencing health behaviours continues to be an important topic in public health research and policy. Foot-based street audits, virtual street audits and secondary data sources are widespread data collection methods used to objectively measure the built environment in environment-health association studies. We compared these three methods using data collected in a nationally representative epidemiological study in 17 British towns to inform future development of research tools. There was good agreement between foot-based and virtual audit tools. Foot based audits were superior for fine detail features. Secondary data sources measured very different aspects of the local environment that could be used to derive a range of environmental measures if validated properly. Future built environment research should design studies a priori using multiple approaches and varied data sources in order to best capture features that operate on different health behaviours at varying spatial scales.

•

This study compares multiple data collection methods for measuring built environment features.

•

Virtual street audits are reliable for more objective built environment measures.

•

Street-based audits are superior for collecting fine detail environmental features.

•

Routine secondary data sources need less resources but must be properly validated.

•

Appropriate methods for health studies vary depending on the research question and resources.

Using Virtual Street Audits to Understand the Walkability of Older Adults' Route Choices by Gender and Age

Walking for physical activity can bring important health benefits to older adults. In this population, walking has been related to various urban design features and street characteristics. To gain new insights into the microscale environmental details that might influence seniors’ walking, details which might be more amenable to change than neighbourhood level factors, we employed a reliable streetscape audit tool, in combination with Google Street View^™, to evaluate the ‘walkability’ of where older adults choose to walk. Analysis of the routes selected by a purposive sample of independently mobile adults aged 65 years and over living in Edinburgh, UK, revealed a preference to walk in more walkable environments, alongside a willingness to walk in less supportive settings. At times, factors commonly considered important for walking, including wayfinding and legibility, user conflict, kerb paving quality, and lighting appeared to have little impact on older adults’ decisions about where to walk. The implications for policy, practice, and the emerging technique of virtual auditing are considered.

Emerging technologies to measure neighborhood conditions in public health: implications for interventions and next steps

Adverse neighborhood conditions play an important role beyond individual characteristics. There is increasing interest in identifying specific characteristics of the social and built environments adversely affecting health outcomes. Most research has assessed aspects of such exposures via self-reported instruments or census data. Potential threats in the local environment may be subject to short-term changes that can only be measured with more nimble technology. The advent of new technologies may offer new opportunities to obtain geospatial data about neighborhoods that may circumvent the limitations of traditional data sources. This overview describes the utility, validity and reliability of selected emerging technologies to measure neighborhood conditions for public health applications. It also describes next steps for future research and opportunities for interventions. The paper presents an overview of the literature on measurement of the built and social environment in public health (Google Street View, webcams, crowdsourcing, remote sensing, social media, unmanned aerial vehicles, and lifespace) and location-based interventions. Emerging technologies such as Google Street View, social media, drones, webcams, and crowdsourcing may serve as effective and inexpensive tools to measure the ever-changing environment. Georeferenced social media responses may help identify where to target intervention activities, but also to passively evaluate their effectiveness. Future studies should measure exposure across key time points during the life-course as part of the exposome paradigm and integrate various types of data sources to measure environmental contexts. By harnessing these technologies, public health research can not only monitor populations and the environment, but intervene using novel strategies to improve the public health.

Modified ground-truthing: an accurate and cost-effective food environment validation method for town and rural areas

Background

A major concern in food environment research is the lack of accuracy in commercial business listings of food stores, which are convenient and commonly used. Accuracy concerns may be particularly pronounced in rural areas. Ground-truthing or on-site verification has been deemed the necessary standard to validate business listings, but researchers perceive this process to be costly and time-consuming. This study calculated the accuracy and cost of ground-truthing three town/rural areas in Minnesota, USA (an area of 564 miles, or 908 km), and simulated a modified validation process to increase efficiency without comprising accuracy. For traditional ground-truthing, all streets in the study area were driven, while the route and geographic coordinates of food stores were recorded.

Results

The process required 1510 miles (2430 km) of driving and 114 staff hours. The ground-truthed list of stores was compared with commercial business listings, which had an average positive predictive value (PPV) of 0.57 and sensitivity of 0.62 across the three sites. Using observations from the field, a modified process was proposed in which only the streets located within central commercial clusters (the 1/8 mile or 200 m buffer around any cluster of 2 stores) would be validated. Modified ground-truthing would have yielded an estimated PPV of 1.00 and sensitivity of 0.95, and would have resulted in a reduction in approximately 88 % of the mileage costs.

Conclusions

We conclude that ground-truthing is necessary in town/rural settings. The modified ground-truthing process, with excellent accuracy at a fraction of the costs, suggests a new standard and warrants further evaluation.

Metrics in Urban Health: Current Developments and Future Prospects

The research community has shown increasing interest in developing and using metrics to determine the relationships between urban living and health. In particular, we have seen a recent exponential increase in efforts aiming to investigate and apply metrics for urban health, especially the health impacts of the social and built environments as well as air pollution. A greater recognition of the need to investigate the impacts and trends of health inequities is also evident through more recent literature. Data availability and accuracy have improved through new affordable technologies for mapping, geographic information systems (GIS), and remote sensing. However, less research has been conducted in low- and middle-income countries where quality data are not always available, and capacity for analyzing available data may be limited. For this increased interest in research and development of metrics to be meaningful, the best available evidence must be accessible to decision makers to improve health impacts through urban policies.

Neighbourhood typology based on virtual audit of environmental obesogenic characteristics

Virtual audit (using tools such as Google Street View) can help assess multiple characteristics of the physical environment. This exposure assessment can then be associated with health outcomes such as obesity. Strengths of virtual audit include collection of large amount of data, from various geographical contexts, following standard protocols. Using data from a virtual audit of obesity-related features carried out in five urban European regions, the current study aimed to (i) describe this international virtual audit dataset and (ii) identify neighbourhood patterns that can synthesize the complexity of such data and compare patterns across regions. Data were obtained from 4,486 street segments across urban regions in Belgium, France, Hungary, the Netherlands and the UK. We used multiple factor analysis and hierarchical clustering on principal components to build a typology of neighbourhoods and to identify similar/dissimilar neighbourhoods, regardless of region. Four neighbourhood clusters emerged, which differed in terms of food environment, recreational facilities and active mobility features, i.e. the three indicators derived from factor analysis. Clusters were unequally distributed across urban regions. Neighbourhoods mostly characterized by a high level of outdoor recreational facilities were predominantly located in Greater London, whereas neighbourhoods characterized by high urban density and large amounts of food outlets were mostly located in Paris. Neighbourhoods in the Randstad conurbation, Ghent and Budapest appeared to be very similar, characterized by relatively lower residential densities, greener areas and a very low percentage of streets offering food and recreational facility items. These results provide multidimensional constructs of obesogenic characteristics that may help target at-risk neighbourhoods more effectively than isolated features.

Mismatch between perceived and objectively measured environmental obesogenic features in European neighbourhoods

Findings from research on the association between the built environment and obesity remain equivocal but may be partly explained by differences in approaches used to characterize the built environment. Findings obtained using subjective measures may differ substantially from those measured objectively. We investigated the agreement between perceived and objectively measured obesogenic environmental features to assess (1) the extent of agreement between individual perceptions and observable characteristics of the environment and (2) the agreement between aggregated perceptions and observable characteristics, and whether this varied by type of characteristic, region or neighbourhood. Cross-sectional data from the SPOTLIGHT project (n = 6037 participants from 60 neighbourhoods in five European urban regions) were used. Residents' perceptions were self-reported, and objectively measured environmental features were obtained by a virtual audit using Google Street View. Percent agreement and Kappa statistics were calculated. The mismatch was quantified at neighbourhood level by a distance metric derived from a factor map. The extent to which the mismatch metric varied by region and neighbourhood was examined using linear regression models. Overall, agreement was moderate (agreement < 82%, kappa < 0.3) and varied by obesogenic environmental feature, region and neighbourhood. Highest agreement was found for food outlets and outdoor recreational facilities, and lowest agreement was obtained for aesthetics. In general, a better match was observed in high-residential density neighbourhoods characterized by a high density of food outlets and recreational facilities. Future studies should combine perceived and objectively measured built environment qualities to better understand the potential impact of the built environment on health, particularly in low residential density neighbourhoods.

Interactions of individual perceived barriers and neighbourhood destinations with obesity-related behaviours in Europe

Perceived barriers towards physical activity and healthy eating as well as local availability of opportunities (destinations in the neighbourhood) are important determinants of obesity-related behaviours in adults. Little is known, however, about how these factors interact with the behaviours. Data were analysed from 5,205 participants of the SPOTLIGHT survey, conducted in 60 neighbourhoods in urban regions of five different countries across Europe. A virtual audit was conducted to collect data on the presence of destinations in each neighbourhood. Direct associations of, and interactions between, the number of individual perceived barriers and presence of destinations with obesity-related behaviours (physical activity and dietary behaviours) were analysed using multilevel regression analyses, adjusted for key covariates. Perceiving more individual barriers towards physical activity and healthy eating was associated with lower odds of physical activity and healthy eating. The presence of destinations such as bicycle lanes, parks and supermarkets was associated with higher levels of physical activity and healthier dietary behaviours. Analyses of additive interaction terms suggested that the interaction of destinations and barriers was competitive, such that the presence of destinations influenced obesity-related behaviours most among those perceiving more barriers. These explorative findings emphasize the interest and importance of combining objective (e.g. virtual neighbourhood audit) methods and subjective (e.g. individual perceived barriers collected in a survey) to better understand how the characteristics of the residential built environment can shape obesity-related behaviours depending on individual characteristics.

Obesity-related behaviours and BMI in five urban regions across Europe: sampling design and results from the SPOTLIGHT cross-sectional survey

OBJECTIVES

To describe the design, methods and first results of a survey on obesity-related behaviours and body mass index (BMI) in adults living in neighbourhoods from five urban regions across Europe.

DESIGN

A cross-sectional observational study in the framework of an European Union-funded project on obesogenic environments (SPOTLIGHT).

SETTING

60 urban neighbourhoods (12 per country) were randomly selected in large urban zones in Belgium, France, Hungary, the Netherlands and the UK, based on high or low values for median household income (socioeconomic status, SES) and residential area density.

PARTICIPANTS

A total of 6037 adults (mean age 52 years, 56% female) participated in the online survey.

OUTCOME MEASURES

Self-reported physical activity, sedentary behaviours, dietary habits and BMI. Other measures included general health; barriers and motivations for a healthy lifestyle, perceived social and physical environmental characteristics; the availability of transport modes and their use to specific destinations; self-defined neighbourhood boundaries and items related to residential selection.

RESULTS

Across five countries, residents from low-SES neighbourhoods ate less fruit and vegetables, drank more sugary drinks and had a consistently higher BMI. SES differences in sedentary behaviours were observed in France, with residents from higher SES neighbourhoods reporting to sit more. Residents from low-density neighbourhoods were less physically active than those from high-density neighbourhoods; during leisure time and (most pronounced) for transport (except for Belgium). BMI differences by residential density were inconsistent across all countries.

CONCLUSIONS

The SPOTLIGHT survey provides an original approach for investigating relations between environmental characteristics, obesity-related behaviours and obesity in Europe. First descriptive results indicate considerable differences in health behaviours and BMI between countries and neighbourhood types.

Assessing Walking and Cycling Environments in the Streets of Madrid: Comparing On-Field and Virtual Audits

Audit tools are useful for exploring the urban environment and its association with physical activity. Virtual auditing options are becoming increasingly available potentially reducing the resources needed to conduct these assessments. Only a few studies have explored the use of virtual audit tools. Our objective is to test if the Madrid Systematic Pedestrian and Cycling Environment Scan (M-SPACES) discriminates between areas with different urban forms and to validate virtual street auditing using M-SPACES. Three areas (N = 500 street segments) were selected for variation in population density. M-SPACES was used to audit street segments physically and virtually (Google Street View) by two researchers in 2013-2014. For both physical and virtual audits, all analyzed features score significantly different by area (p < 0.05). Most of the features showed substantial (ICC = 0.6-0.8) or almost perfect (ICC ≥ 0.8) agreement between virtual and physical audits, especially neighborhood permeability walking infrastructure, traffic safety, streetscape aesthetics, and destinations. Intra-rater agreement was generally acceptable (ICC > 0.6). Inter-rater agreement was generally poor (ICC < 0.4). Virtual auditing provides a valid and feasible way of measuring residential urban environments. Comprehensive auditor training may be needed to guarantee good inter-rater agreement.

Associations between Food Outlets around Schools and BMI among Primary Students in England: A Cross-Classified Multi-Level Analysis

INTRODUCTION

Researchers and policy-makers are interested in the influence that food retailing around schools may have on child obesity risk. Most previous research comes from North America, uses data aggregated at the school-level and focuses on associations between fast food outlets and school obesity rates. This study examines associations between food retailing and BMI among a large sample of primary school students in Berkshire, England. By controlling for individual, school and home characteristics and stratifying results across the primary school years, we aimed to identify if the food environment around schools had an effect on BMI, independent of socio-economic variables.

METHODS

We measured the densities of fast food outlets and food stores found within schoolchildren's home and school environments using Geographic Information Systems (GIS) and data from local councils. We linked these data to measures from the 2010/11 National Child Measurement Programme and used a cross-classified multi-level approach to examine associations between food retailing and BMI z-scores. Analyses were stratified among Reception (aged 4-5) and Year 6 (aged 10-11) students to measure associations across the primary school years.

RESULTS

Our multilevel model had three levels to account for individual (n = 16,956), home neighbourhood (n = 664) and school (n = 268) factors. After controlling for confounders, there were no significant associations between retailing near schools and student BMI, but significant positive associations between fast food outlets in home neighbourhood and BMI z-scores. Year 6 students living in areas with the highest density of fast food outlets had an average BMI z-score that was 0.12 (95% CI: 0.04, 0.20) higher than those living in areas with none.

DISCUSSION

We found little evidence to suggest that food retailing around schools influences student BMI. There is some evidence to suggest that fast food outlet densities in a child's home neighbourhood may have an effect on BMI, particularly among girls, but more research is needed to inform effective policies targeting the effects of the retail environment on child obesity.

Geoprocessing via google maps for assessing obesogenic built environments related to physical activity and chronic noncommunicable diseases: validity and reliability

This study analyzes the reliability and validity of obesogenic built environments related to physical activity and chronic noncommunicable diseases through Google Maps in a heterogeneous urban area (i.e., residential and commercial, very poor and very rich) in São Paulo (SP), Brazil. There are no important differences when comparing virtual measures with street audit. Based on Kappa statistic, respectively for validity and reliability, 78% and 80% of outcomes were classified as nearly perfect agreement or substantial agreement. Virtual measures of geoprocessing via Google Maps provided high validity and reliability for assessing built environments.