Towards an Integrated Global Transport and Health Assessment Tool (TIGTHAT)

Lead Research Organisation: University of Cambridge
Department Name: MRC Epidemiology Unit

Abstract

Globally, how people and goods move is important for urban health. Transport causes ill-health through road traffic injuries, noise, and air pollution, but is an opportunity for benefiting health through active travel. Although there are good examples, problems of poor safety, pollution, and environments unconducive to walking and cycling are common in many lower and middle income cities (LMICs). These cities are often changing fast, many are growing rapidly, and motorbike and then car use is increasing. The problem and the opportunity are large.

A number of studies have used computer based models to estimate the health impacts of more active travel in high income Western countries. However, we are aware of only three studies in LMICs (Delhi, Sao Paulo and Kuala Lumpur), all involving the applicants.

The high income studies found more walking and cycling and fewer cars could bring large health benefits. However, our work shows things are more complicated in LMICs. For example, how much people benefit from increasing their physical activity depends how active they are to start with. It also depends on how polluted the air is, because pedestrians and cyclists breath in faster. Road injuries numbers may go up or down as pedestrians and cyclists are at higher risk than drivers, but fewer cars may make the roads safer.

One way to support decision makers is to provide tools that help them to compare the health impacts of their choices. The challenge is to be both practically useful and scientifically accurate. In this project we lay the basis for creating a model that can be readily applied to cities around the world. A key challenge is data comparability and quality. Our international team of health and transport researchers in the UK, India, Brazil, USA, and Peru will assess what data are available and work to develop rules for drawing the best conclusions from it.

We will build a model for two Indian cities (Bengaluru and Visakhapatnam) and do further work with the models we have recently created for Sao Paulo and Delhi. Running the model will not just produce new results but we also use statistical techniques to work out where getting better data is most valuable e.g. do our results change more if we have better information on injuries or on walking.

To estimate health impacts we need scenarios about how travel might change. We will do this in two ways. Firstly asking what would happen if people were as likely to walk, cycle (including electric-assist bikes) a trip of a given distance as people are in more active cities. Secondly, we will look at how more compact cities might affect how people travel.

We will look at the data available in a further four Latin America and sub-Saharan African cities in depth to model them in the future. We will also look across a wide range of settings to see what data is available on how people travel and useful it is for modelling.

We will use a computer programme to get information from Google Street View for different cities and use it to estimate the mix of traffic across the city. We will see if we can use this data to predict how much time people spend travelling by each mode, and using other data about the city how active people are in total.

The project team has worked previously on using a range of data to estimate the number of road traffic victims by mode in different countries. We will now apply this approach to the urban level and work with estimating the kind of data we need for the health impact model. For calculating air pollution risk we will develop a simple method for estimating the contribution of transport to AP using global databases and we will compare this against more detailed analysis in cities for which we have better data.

In summary this project lays the foundation for a scientifically robust webtool that can help policy makers tackle a major determinant of population health.

Technical Summary

Urban land transport has positive (physical activity) and negative (road traffic injuries, noise and air pollutants) side effects. Studies in high income cities found substantial population health benefits from mode shift to active travel, with physical activity (PA) dominating. However, our work in Brazil, India, and Malaysia identifies a more varied picture, with a larger burden and trickier trade-offs.

In this project we lay the basis for a globally applicable stochastic synthesis engine with a user friendly interface, to support evidence based decision making on transport and health. A key challenge is data comparability and quality. Our international, multidisciplinary team will assess what data is available and develop approaches for mapping (calibrating) available to desired data, while representing propagated uncertainty.

We will undertake integrated health impact simulation modelling for 3 Indian cities and Sao Paulo, Brazil, using sensitivity analysis to inform future empirical work. We will test counterfactual scenarios assuming changes to mode choice and land use. We will evaluate travel behaviour data for future modelling studies in further Latin American, India, and sub-Saharan settings.

We will query the Google Street View (GSV) API to estimate ratio(s) of traffic modes in several cities. We will build an ecological model to predict travel time mode split and self-report PA from GSV data, and explore mapping to objective PA. The applicants have previously combined and imputed data to estimate RTIs by mode at the national level. We will now do this at the urban level. For AP we will develop a simple method for estimating the transport's contribution to AP using global databases and compare this to detailed analysis in several cities.

This project lays the foundation for an analytic modelling framework to estimate multiple outcomes of interest under counterfactual scenarios with respect to a major determinant of population health.

Planned Impact

This project will develop methods and tools to improve health and well-being in urban environments in Low and Middle Income Countries (LMICs). The methods we will develop in this project will allow to predict how changes in urban transport would change the air pollution, injuries and physical activity, and how these changes would improve health and well-being, and link to greenhouse gas emission targets.

The way people and goods move around is a public health issue. Impacts are at their starkest in LMICs with rapidly increasing urbanisation and motorisation. The burden of road traffic injuries (RTIs) is predicted to rise globally from the 9th to 7th leading cause of death. In India there are an estimated 200,000RTI deaths per year, while Africa has the highest per capita burden. Urban outdoor air pollution (AP) also poses a large burden, greatest in Asia. At the same time LMICs face a growing burden of chronic disease associated with lack of physical activity (PA). For example around 30% of the adult population are inactive in Brazil.

In longer term our plan is to develop a tool that can be applied in all larger urban areas in LMICs, as well as other cities in the world. A web interface will be provided so that stakeholders (e.g. public health professional, transport professionals, other interested parties) can assess the evidence on likely health consequences of transport interventions and scenarios, and in this way promote policies that improve health. We aim that in five years we will have Version 1 of an open source, freely available health impact model developed and disseminated for India, Brazil and Peru, with other Latin American and Caribbean countries, and then Africa and East Asia to follow.

In the short term this project will bring evidenced understanding of the availability and usefulness of data for modelling the health impacts of transport. This will be strongly advanced in India, advanced in Latin America, and started in Africa. We will have appraised and tested the suitability of multiple data types for inclusion in a health impact model. We will have started data mapping of Google Street View to travel time mode share and total physical activity. We will be able to inform science and surveillance about where enhanced travel, PA, RTI, and AP data collection would most reduce decision uncertainty. We will have worked with a range of potential model users. This project will also increase our project partners' capacity in LMICs to develop and conduct health impact assessment studies. On a larger scale, availability of the model code through an open-source portal will enable scientific reproducibility.

The challenge for the study is to disseminate knowledge from these methods to potential users around the world. For details on how we aim to do this, see Pathways to Impact, and for academic dissemination Communications Plan.
ODA justification

This research will bring long-term benefits to LMICs by providing public health, transport, and planners the means to assess health risks and benefits of transport choices. The vast majority of heath impact fo transport modelling studies are for high income settings. Few tools integrate risks due to PA, RTI and AP. The World Health Organization (WHO) Health Economic Assessment Tool (HEAT) is mainly used in Europeans setting with good data, and, focusing on walking and cycling, does not provide integrated assessment of motorised modes. On the other side, travel demand models, where available, contain little or no data on active travel.

In the short-term, our project will benefit the cities in India, Brazil and Peru. We will develop and test the methods in these cities and disseminate the preliminary results to local stakeholders through our local collaborators. We will also disseminate best practice through our partners. However, the main impact will be achieved after the project when we have developed and disseminated our impact assessment model.

Publications

10 25 50

publication icon
Aldred R (2019) Contextualising Safety in Numbers: a longitudinal investigation into change in cycling safety in Britain, 1991-2001 and 2001-2011. in Injury prevention : journal of the International Society for Child and Adolescent Injury Prevention

publication icon
Goel R (2018) Modelling of road traffic fatalities in India in Accident Analysis & Prevention

 
Description Global & local health impact assessment of transport
Amount € 1,661,804 (EUR)
Organisation European Research Council (ERC) 
Sector Public
Country European Union (EU)
Start 01/2019 
End 12/2023
 
Title ITHIM 
Description ITHIM refers to a range of related models and tools developed at CEDAR to perform integrated assessment of the health effects of transport scenarios and policies at the urban and national level. The health effects of transport policies are modelled through the changes in physical activity, road traffic injury risk, and exposure to fine particulate matter (PM2.5) air pollution. Some versions of ITHIM also predict changes in CO2 emissions. ITHIM is being used in research and by health and transport professionals to estimate the health impacts of scenarios, compare the impact of travel patterns in different places, and model the impact of interventions. ITHIM works either as a stand-alone model, or it can be linked with other models (e.g. transport, health, economic). 
Type Of Material Computer model/algorithm 
Year Produced 2012 
Provided To Others? Yes  
Impact Fed into transport plans in London, California and Portland. ITHIM has is being used in research projects in Brazil, India, and Malaysia. It has been sent to research & practitioner groups in Ireland, Taiwan, Singapore, Canada, and multiple groups in the US. It has been used to generate results for reports (for British Cycling and CTC) that have attracted media and parliamentary attention. 
URL http://www.cedar.iph.cam.ac.uk/research/modelling/ithim/
 
Title ITHIM R 
Description The Integrated Transport and Health Impact Modelling (ITHIM) tool has been considerably upgraded and implemented into R. https://shiny.mrc-epid.cam.ac.uk/ithim/ 
Type Of Material Computer model/algorithm 
Year Produced 2018 
Provided To Others? Yes  
Impact A version of the model was developed with inputs from local and WHO stakeholders in Accra, Ghana. Impacts and outputs have also been realised from spin off versions of the model (and earlier versions) including citation under objective 6 of the California "Health in All Policies Task Force 2014-2018 Active Transportation Action Report" with recent papers here https://ajph.aphapublications.org/doi/full/10.2105/AJPH.2018.304879 https://www.nzma.org.nz/journal/read-the-journal/all-issues/2010-2019/2018/vol-131-no-1472-23-march-2018/7529 https://www.sciencedirect.com/science/article/pii/S1361920918309052 https://www.mdpi.com/1660-4601/15/5/962 
URL https://shiny.mrc-epid.cam.ac.uk/ithim/
 
Description 4th Safer City Streets Network Meeting (LG) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Dr Leandro Garcia presented at the 4th Safer City Streets Network Meeting held on 20 November 2018 in London.
Year(s) Of Engagement Activity 2018
 
Description 7th International Society for Physical Activity and Health Congress (ISPAH) (MT) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Dr Marko Tainio both chaired a session and presented at the 7th International Society for Physical Activity and Health Congress (ISPAH), held in London on 15 October 2018.
Year(s) Of Engagement Activity 2018
 
Description 7th International Society for Physical Activity and Health Congress (LG) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Dr Leandro Garcia presented at the 7th International Society for Physical Activity and Health Congress held in London on 15 October 2018.
Year(s) Of Engagement Activity 2018
 
Description All party Parliamentary Cycling Committee (JW) 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Policymakers/politicians
Results and Impact Dr James Woodcock was on the panel at the All party Parliamentary Cycling Committee held in the Houses of Parliament on 26 February 2018.
Year(s) Of Engagement Activity 2018
URL https://allpartycycling.org/2019/03/01/meeting-with-public-health-england/
 
Description High-level Roundtable on Cities and sustainable infrastructure in South Asia (JW) 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Policymakers/politicians
Results and Impact Dr James Woodcock was invited to speak at the High-level Roundtable on Cities and sustainable infrastructure in South Asia meeting, held in Delhi, India.
Year(s) Of Engagement Activity 2018
 
Description International Society of Exposure Science (ISES)-International Society for Environmental Epidemiology (ISEE) Joint Annual Meeting (MT) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Other audiences
Results and Impact Dr Marko Tainio chaired and spoke at the International Society of Exposure Science (ISES)-International Society for Environmental Epidemiology (ISEE) Joint Annual Meeting, held in Ottawa, Canada, on 26 August 2018.
Year(s) Of Engagement Activity 2018
 
Description MIT Technology Review 'Goodbye, census-hello, Street View' (RG, JW) 
Form Of Engagement Activity A magazine, newsletter or online publication
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact News article on MIT Technology Review 'Goodbye, census-hello, Street View' https://www.technologyreview.com/s/610293/goodbye-census-hello-street-view/ on Google Street View paper.
Year(s) Of Engagement Activity 2018
URL https://www.technologyreview.com/s/610293/goodbye-census-hello-street-view/
 
Description Modelling World Predicting mode share with Google Street View (JW) 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach National
Primary Audience Professional Practitioners
Results and Impact Dr James Woodcock presented a talk on Predicting mode share with Google Street View at the Modelling World conference held in Birmingham in June 2018.
Year(s) Of Engagement Activity 2018
URL http://landor.co.uk/modellingworld/2018/home.php
 
Description Press Release - Google Street View travel patterns (RG, JW) 
Form Of Engagement Activity A press release, press conference or response to a media enquiry/interview
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Media (as a channel to the public)
Results and Impact Article in Futurism https://futurism.com/gsv-google-street-view-public-health/ Indian Express http://indianexpress.com/article/technology/social/google-street-view-can-map-travel-patterns-in-cities-study-5163589/ and six other online news outlets. Authors were subsequently invited on 7 May 2018 to attend a Google Street View summit on 30/31 May 2018 to participate in dscussions on how to use view imagery to index ground-level observations that enable socioeconomic insights.
Year(s) Of Engagement Activity 2018
URL https://eurekalert.org/pub_releases/2018-05/uoc-ugs050118.php
 
Description Travel to work in India: Future concerns (JW) 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Dr James Woodcock provided information for the "Travel to work in India: Current patterns and Future Concerns" report from the Transportation research and Injury Prevention Programme (TRIPP) at the Indian Institute of Technology Delhi.

Urban structure is very closely linked to the mobility patterns of city residents. An understanding of city mobility patterns has a major impact on residents well being both directly and indirectly. Access to employment, education and health and recreation facilities have to be enabled by safe, green and inclusive transport systems. Reliable data on mobility patterns is a pre requisite to understanding many complex issues that are faced by cities today. Census 2011 has for the first time in India included two questions on travel:
We have summarized the responses to this questions in a brief report.
A round table discussion on "Travel to work in India: Future concerns" was organized on 15th May, 2018 at IIT Delhi.
The participants discussed the following themes:
I Travel and city Structure with reference to Pedestrians, and bicyclists. Broadly focusing on the following questions:
What lessons do we have from census travel patterns?
Are we designing cities for pedestrians and bicyclists?
II City Structure, planning and regulatory measures with reference to trips by para transit systems, bus and metro systems. Broadly focusing on the following questions:
What lessons do we have from census travel patterns?
Are we planning cities and governance structures for public transport systems?
List of participants is enclosed.
Year(s) Of Engagement Activity 2018
URL http://tripp.iitd.ernet.in/assets/publication/WorkTravelReport.pdf
 
Description Urban Health Initiative workshop in Accra, Ghana (JW) 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Policymakers/politicians
Results and Impact Dr James Woodcock was invited to present the model at the workshop for stakeholders on urban health models in Accra, Ghana, on 26 June 2018.
Year(s) Of Engagement Activity 2018