Debate on the paper by Diez Roux

Saldiva, Paulo Hilário Nascimento; Alves, Nilmara de Oliveira; Veras, Mariana Matera

doi:10.1590/0102-311xco02s115

Cited by 1 publication

(1 citation statement)

References 9 publications

(2 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…City design can contribute to the risk of type 2 diabetes mellitus, cardiovascular disease, chronic respiratory disease, some cancers, mental illness (commonly existing as multi-morbidities due to shared risk factors and interacting biological mechanisms [11]), in addition to road injury. Despite their relative wealth, Australia and the UK are not immune, having experienced among the greatest global increases in rates of overweight and obesity leading to the development of NCDs and metabolic diseases since the 1980s [12,13]. The combined effect of air pollution, poor diet, high sugar intake, physical inactivity, smoking, and other modifiable lifestyle factors, driven in part by the reinforcing influence of low-density, car and fossil-fuel dependent cities, pose serious barriers to change.…”

Section: Introductionmentioning

confidence: 99%

Computer vision, causal inference and public health modelling approaches to generate evidence on the impacts of urban planning in non-communicable disease and health inequalities in UK and Australian cities: A proposed collaborative approach

Hunter

Garcia

Stevenson

et al. 2023

Preprint

View full text Add to dashboard Cite

Background: Given that the majority of the worlds population live in cities, it is essential to global health efforts that we design them in ways that both reduce non-communicable diseases (NCDs) risk and that facilitate adoption and maintenance of healthy lifestyles. Current approaches tend to focus on the relationship between urban design-related factors that affect health at the local or neighbourhood level but few studies have explored this relationship both within and across entire cities, nor explored the causal pathways between urban-designed related factors and NCDs. The aim of this research program is to use computer vision, causal inference, and public health modelling methods for understanding the causal relationship between urban design and health at the neighbourhood level, and to explore intervention approaches at the city scale. Methods: Phase 1 will use machine learning and computer vision techniques to analyse gridded, local-level aerial images (with an optical resolution of <20cm), of all UK and Australian cities with populations over 100,000. It will identify a variety of urban features within these images and derive associations between them and NCD incidence and risk factors identified through location-based health surveys. Phase 2, using data from prospective health cohorts and linked objective built environment data, will apply Bayesian networks to investigate the possible causal pathways between built environment, lifestyle factors, and NCD incidence. Phase 3 will estimate the health impacts of actionable changes in urban design. Using health impact assessment modelling, we will calculate the NCD burden that could be prevented if cities were to adopt urban features of healthier counterparts. A similar approach will be applied on finer-grained scale within all case study cities, enabling assessment of health impacts of changes in individual locations. Phase 4 will develop an interactive web-based toolkit to enable urban designers, planners and policymakers to inform the decision-making cycle, co-designed with intended users involving participatory workshops. Discussion: We use state-of-the-art approaches to: (i) generate evidence on the impacts of urban planning and design in NCDs and health inequalities in UK and Australian cities, and (ii) provide stakeholders with tools for advocacy and designing healthier cities.

show abstract

Section: Introductionmentioning

confidence: 99%