Background
Injury is the most common cause of death in people under the age of 40 worldwide, largely driven by road traffic injury (RTI). The United Nations previously designated 2011-2020 the Decade for Action on Road Safety, with the aim of halving such injury. But progress has been severely limited. Data on road collisions and injury remains frequently incomplete, fragmented and slow in becoming available for analysis. We need innovative solutions to accelerate progress over the decade to come.
Concept
There has been a global surge in social media use, citizen-reporting of incidents and other forms of mobile data, web data and traffic navigation apps. This presents a data-mining opportunity that could offer a different means of identifying and predicting areas of high risk. This ‘next-generation’ surveillance using crowdsourced data, natural language programming and machine learning, has been successfully used to map and predict spread of infectious diseases including influenza, dengue and Zika.
We’re exploring how these methods could be applied to road traffic injury to boost the quality of public health data and give healthcare workers, public health planners and policy-makers the opportunity to access high volumes of useful data, which could overcome previous difficulties with traditional data sources.
Aims
We’re exploring these novel data sources and methods in the city of London, where high-quality transport and health datasets are available, to validate their use across London and beyond. Such a strategy could facilitate better planning for post-crash emergency care and divert resources to reduce risk, potentially saving lives.
Through this work, we seek to:
- Collect and pool novel digital data sources relating to road traffic collisions, such as online media, news, mobile navigation apps and vehicle monitoring systems;
- Assess the ability of these sources to identify the real-world location, timing and severity of road traffic collisions and injury;
- Compare the accuracy of these sources for road collision and injury surveillance to well-established high-quality data from London’s transport for London, ambulance, health services; and
- Explore how these sources could be used in other global contexts to overcome inadequate infrastructure, raise awareness and improve outcomes.
Project Team
Miss Seema Yalamanchili – Clinical Research Fellow, Project Lead (s.yalamanchili@imperial.ac.uk)
Ana Sofia Bras Pinto – Data Scientist
Mr James Kinross – Senior Clinical Lecturer in Surgery
Dr Matthew Harris – Senior Clinical Lecturer in Public Health
Prof. Ara Darzi – Co-Director, Institute of Global Health Innovation
This project is funded by the RAC Foundation and the FIA via their Road Safety Grants Programme.
Mechanisms for dissent
As part of our analysis, we will be using historic health and transport data relating road traffic collisions in London over the past 10 years (1st January 2009 – 31st December 2019). This includes data from Transport for London, the Department for Transport, London Ambulance Service, Hospital Episode Statistics (NHS Digital) and the Trauma Audit Research Network.
To link across the various data sets, NHS Digital will use personal data such as date of birth, gender and date and time of injury/incident. Thereafter records are pseudonymised before being sent to us for analysis. No identifiable data will be made available to us, and any data we publish will be in aggregated (grouped) form to ensure no individual identification is possible
All research is compliant with Health Research Authority Guidance and with a legal basis under data protection laws (General Data Protection Regulation Article 6 (1) (e) and General Data Protection Regulation Article 9 (2) (j)). However if you would like to withdraw consent to any of your records being used, please contact the Project Lead using this link.