Data Awareness for Sending Help (DASH)
Lead Research Organisation:
King's College London
Department Name: Informatics
Abstract
This project explores integration of new and emerging data sources for potential impact on emergency response. In an emergency medical situation, ambulances must get to those in need as quickly as possible in order to provide care and, ultimately, to save lives. Decisions about which ambulance should respond to each incident need to be made rapidly. However, making such decisions is complicated: incidents can occur simultaneously or in short succession over a wide area; the locations of ambulances are constantly changing; and there are many environmental factors that can affect response times, such as traffic and weather conditions. In such a complex and dynamic environment, a form of automated decision support, known as computer assisted dispatch (CAD), is often installed to help staff make these decisions.
In recent years, there has been a sharp expansion in the volume and types of data sources that might potentially be linked to CAD, presenting new possibilities for improving current decision-support systems. Doing so could enable ambulances to respond faster as emergency situations develop, improving emergency care, lowering costs, increasing efficiency and improving health outcomes for patients. Potentially useful data might come from any of the following sources: the general population (via social media and other mobile Apps); specific user segments (via in-home/wearable sensors, particularly for high-risk patient groups); urban infrastructure (via public transport monitors, embedded road sensors or weather stations); and other public sector actors (such as collaborating emergency response agencies or healthcare providers).
This proposed Policy Demonstrator Project, entitled "Data Awareness for Sending Help" (DASH), aims to explore the potential of these "new data" sources for improving ambulance response times. The project builds on a new research collaboration between King's College London (KCL) and the London Ambulance Service (LAS), which is evaluating novel methods for ambulance dispatch by simulating ambulance call-outs based on historical LAS system logs. DASH will lay the groundwork for extending this preliminary study in important new directions, by predicting changes in response times due to integration of additional data sources.
There are a number of challenges that DASH aims to address, both technical, in terms of how feasible it is to access and use new data sources reliably, and social or ethical, in terms of how acceptable and appropriate it is to use data in this way, particularly data about individuals. DASH asks three specific research questions:
(1) What are the benefits and risks for emergency service agencies, healthcare providers and the public related to linking new and emerging data sources to emergency response? A comprehensive literature review and targeted focus groups will highlight which new data sources could be tapped and will weigh benefits, costs and risks associated with data-enhanced emergency response, across various sectors.
(2) What are the technical challenges involved in linking new and emerging data sources to CAD technologies to provide the most important benefits? A technical investigation will consider practical aspects of linking new data sources to CAD and will explore innovative modelling methods that could be applied in a low-cost but high-impact and secure manner.
(3) How can practitioners and policymakers learn from our study of linking new and emerging data sources to the London Ambulance Service? DASH will produce a set of outputs designed to inform practitioners and policymakers, including: a Policy Brief outlining our findings; a Case Study that assesses linking new data and associated methodologies to LAS's CAD system; a Software Prototype, built on previous work, that demonstrates how a data-enhanced CAD system might work; and a report on the broader applicability of the findings to other emergency response agencies in the UK.
In recent years, there has been a sharp expansion in the volume and types of data sources that might potentially be linked to CAD, presenting new possibilities for improving current decision-support systems. Doing so could enable ambulances to respond faster as emergency situations develop, improving emergency care, lowering costs, increasing efficiency and improving health outcomes for patients. Potentially useful data might come from any of the following sources: the general population (via social media and other mobile Apps); specific user segments (via in-home/wearable sensors, particularly for high-risk patient groups); urban infrastructure (via public transport monitors, embedded road sensors or weather stations); and other public sector actors (such as collaborating emergency response agencies or healthcare providers).
This proposed Policy Demonstrator Project, entitled "Data Awareness for Sending Help" (DASH), aims to explore the potential of these "new data" sources for improving ambulance response times. The project builds on a new research collaboration between King's College London (KCL) and the London Ambulance Service (LAS), which is evaluating novel methods for ambulance dispatch by simulating ambulance call-outs based on historical LAS system logs. DASH will lay the groundwork for extending this preliminary study in important new directions, by predicting changes in response times due to integration of additional data sources.
There are a number of challenges that DASH aims to address, both technical, in terms of how feasible it is to access and use new data sources reliably, and social or ethical, in terms of how acceptable and appropriate it is to use data in this way, particularly data about individuals. DASH asks three specific research questions:
(1) What are the benefits and risks for emergency service agencies, healthcare providers and the public related to linking new and emerging data sources to emergency response? A comprehensive literature review and targeted focus groups will highlight which new data sources could be tapped and will weigh benefits, costs and risks associated with data-enhanced emergency response, across various sectors.
(2) What are the technical challenges involved in linking new and emerging data sources to CAD technologies to provide the most important benefits? A technical investigation will consider practical aspects of linking new data sources to CAD and will explore innovative modelling methods that could be applied in a low-cost but high-impact and secure manner.
(3) How can practitioners and policymakers learn from our study of linking new and emerging data sources to the London Ambulance Service? DASH will produce a set of outputs designed to inform practitioners and policymakers, including: a Policy Brief outlining our findings; a Case Study that assesses linking new data and associated methodologies to LAS's CAD system; a Software Prototype, built on previous work, that demonstrates how a data-enhanced CAD system might work; and a report on the broader applicability of the findings to other emergency response agencies in the UK.
Planned Impact
Timely and efficient ambulance response is a major challenge for the UK. The BBC News recently reported that only 1 in 13 ambulance services are meeting their targets for response times (http://www.bbc.co.uk/news/health-38077409). One of the reasons for slowdowns occurs in A&Es, where ambulance crews have to wait before handing over patients to under-staffed hospital personnel. Wales is the only region not lagging behind, and that is because they changed how incident calls are classified so that fewer are required to be completed in the shortest response time of 8 minutes. With an ageing population nationwide and over-stressed National Health Service (NHS), the demands on ambulances and crews will only increase.
This proposed "Data Awareness for Sending Help" (DASH) project investigates how new and emerging forms of data could help improve efficiency of emergency response services. The interdisciplinary nature of the project means that it has the potential to benefit a wide-ranging set of stakeholders:
* the general public, who are the ultimate beneficiaries of the research proposed here, the people who will experience first-hand more efficient operational models of emergency response services;
* the providers of emergency services and health care, and partner organisations such as Clinical Commissioning Groups (CCGs), as well as government officials and general policy makers; and
* researchers who study policy, healthcare and technology, from emergency response and medical specialists to general policy researchers, to computer science, artificial intelligence and operational research experts.
DASH will produce a Case Study around the London Ambulance Service (LAS), which will not only benefit LAS directly, but also its working partners and peers: other emergency services agencies and healthcare providers. DASH will make a valuable contribution to future strategic and operational development of LAS by identifying and assessing specific benefits and risks of linking new and emerging sources of data to LAS computer assisted dispatch (CAD), quantified through operational efficiency indicators. Findings will be extrapolated to other emergency services agencies nationwide. The impact of producing the Case Study will be measured in terms of stakeholder feedback in the short term and changes in practice in the long term.
DASH will build a Software Prototype, extending prior work of project investigators, to simulate dispatch conditions and allow researchers to visualise historical scenarios and ask "what if" questions about future scenarios. Researchers from operational research, computer science and artificial intelligence, as well as applied researchers in policy and public health, will benefit from the opportunity to consider the effects of linking new and emerging data and modelling methodologies to CAD, within a specific real-world context (LAS), as well as more broadly across the UK. The impact of producing the Software Prototype will be measured directly in terms of usage statistics, and indirectly in terms of citations.
DASH will publish a Policy Brief describing the potential for new and emerging forms of data to have a transformative effect on the effectiveness and efficiency of emergency services delivery. Thus, specialist practitioners, general policymakers and the general public will benefit from insights into the policy potential of new and emerging forms of data. The impact of the Policy Brief will be measured through direct feedback from those who read the brief, as well as other, indirect measures such as social media and press monitoring.
This proposed "Data Awareness for Sending Help" (DASH) project investigates how new and emerging forms of data could help improve efficiency of emergency response services. The interdisciplinary nature of the project means that it has the potential to benefit a wide-ranging set of stakeholders:
* the general public, who are the ultimate beneficiaries of the research proposed here, the people who will experience first-hand more efficient operational models of emergency response services;
* the providers of emergency services and health care, and partner organisations such as Clinical Commissioning Groups (CCGs), as well as government officials and general policy makers; and
* researchers who study policy, healthcare and technology, from emergency response and medical specialists to general policy researchers, to computer science, artificial intelligence and operational research experts.
DASH will produce a Case Study around the London Ambulance Service (LAS), which will not only benefit LAS directly, but also its working partners and peers: other emergency services agencies and healthcare providers. DASH will make a valuable contribution to future strategic and operational development of LAS by identifying and assessing specific benefits and risks of linking new and emerging sources of data to LAS computer assisted dispatch (CAD), quantified through operational efficiency indicators. Findings will be extrapolated to other emergency services agencies nationwide. The impact of producing the Case Study will be measured in terms of stakeholder feedback in the short term and changes in practice in the long term.
DASH will build a Software Prototype, extending prior work of project investigators, to simulate dispatch conditions and allow researchers to visualise historical scenarios and ask "what if" questions about future scenarios. Researchers from operational research, computer science and artificial intelligence, as well as applied researchers in policy and public health, will benefit from the opportunity to consider the effects of linking new and emerging data and modelling methodologies to CAD, within a specific real-world context (LAS), as well as more broadly across the UK. The impact of producing the Software Prototype will be measured directly in terms of usage statistics, and indirectly in terms of citations.
DASH will publish a Policy Brief describing the potential for new and emerging forms of data to have a transformative effect on the effectiveness and efficiency of emergency services delivery. Thus, specialist practitioners, general policymakers and the general public will benefit from insights into the policy potential of new and emerging forms of data. The impact of the Policy Brief will be measured through direct feedback from those who read the brief, as well as other, indirect measures such as social media and press monitoring.
Publications
Drake A
(2019)
Data awareness for sending help (DASH): policy opportunities & challenges
in EMERGENCY MEDICINE JOURNAL
| Description | The "Data Awareness for Sending Help" (DASH) project was funded as a one-year Policy Demonstrator project which focussed on identifying and assessing new and emerging data sources and technologies for ambulance dispatch. The project followed three key lines of inquiry: (1) exploring POLICY issues surrounding the acquisition of and continued access to timely data sources in order to improve emergency response; (2) investigating TECHNOLOGY issues related to the introduction and integration of new data sources and tools to the tactical and operational functions of an emergency response service; and (3) considering both types of issues in the context of a CASE STUDY around the London Ambulance Service (LAS). The most significant achievement of the project was the recommendation of six directions for future data-centric initiatives for LAS, each of which were identified and explored as part of the DASH project: (1) to encourage broader, pan-London connections around the integration of HEALTH AND SOCIAL CARE DATA, to improve evidence on what works; (2) to partner with Transport for London (TfL) to allow ambulance services to navigate traffic more intelligently through real-time integration of TRANSPORT DATA; (3) to engage with the London Air Quality Network to help predict demand for ambulance services for those with breathing problems by linking AIR QUALITY DATA to planning models; (4) to use mobile network providers' data and insight to support service effectiveness by tracking POPULATION MOBILITY DATA that indicates where people are in large numbers; (5) to extend the use of VIDEO COMMUNICATION TECHNOLOGY to improve triage and remote treatment where appropriate; and (6) to encourage agencies and challenge researchers to explore specific ways to facilitate access to and uses of WEATHER DATA that can improve planning at the tactical and operational levels. In May 2018, DASH produced an 80-page report published by The Policy Institute at King's College London entitled "Data for Ambulance Dispatch: New and emerging forms of data to support the London Ambulance Service". This report provides details of the project investigations and is organised around the six directions listed above. In addition, two specific Artificial Intelligence (AI) technologies were applied to the domain. First, an innovative methodology that employs market-based mechanisms to allocate ambulances to incidents was tested with historic LAS data and results demonstrated that the market-based methodology could improve on response times significantly. Second, a state-of-the-art technology for decision support based on computational argumentation was demonstrated for emergency response decisions at the planning (tactical) and execution (operational) levels and received favourable reviews from representatives of multiple emergency response agencies nationwide. |
| Exploitation Route | The DASH investigation was conducted from two perspectives, policy and technology, and revolved around three key research questions. As a one-year Policy Demonstrator, the project has served its purpose with respect to identifying a number of future avenues of investigation, from both policy and technology perspectives. These are described below in the context of the three research questions around which the project activities were structured. (1) What are the benefits and risks for emergency service agencies, healthcare providers and the public related to linking new and emerging data sources to emergency response? The study conducted within the scope of the DASH project was comprehensive, though necessarily not exhaustive given the timeframe and resources behind the one-year project. The focus was specifically on the London Ambulance Service and, as mentioned, six key areas for future integration of new data sources and technologies were identified, as well as two state-of-the-art AI tools. A range of benefits and risks associated with each of these data sources and technologies were identified and assessed from the policy perspective. The primary benefits can be summarised by projecting improved performance and efficiency with respect to daily response times; communication with patients, healthcare facilities and resources during an incident; and tracking of results after an incident for future planning and evaluation. The risks are largely centred around issues facing all health-related data processing in today's technology-rich society: protecting the privacy and security of patient data. The challenges around the acquisition of and continued access to data include managing data to mitigate the risks around privacy and security, assuring timeliness and accuracy, and providing easy-to-use tools and appropriate training for emergency services personnel who would benefit from access to the new data sources. (2) What are the technical challenges involved in linking new and emerging data sources to CAD technologies to provide the most important benefits? A range of technical challenges were identified by the DASH project. These include: access to accurate data, which may come from multiple sources; intelligent merging of data from multiple sources; real-time updates of relevant and appropriate data; security of data during transmission, storage while a case is "live" and after a case has completed; privacy of patient information, as well as information about healthcare facilities and resources; and careful attention to and respect for personnel operating in high-stress conditions, particularly in terms of not overwhelming personnel with too much data and new tools. Developing practical solutions to these challenges will require additional research resources and collaboration with LAS and other agencies (e.g., Transport for London and the Met Office). (3) How can practitioners and policymakers learn from our study of linking new and emerging data sources to the London Ambulance Service? The London Ambulance Service is one of 11 regional ambulance services operating in England, with a further 3 national services covering Wales, Northern Ireland and Scotland. One of the DASH activities was to consider the implications of the LAS case study to these other 13 agencies across the UK. There is a wide range of access to and integration of new technologies at each of these agencies: some are more advanced than LAS in some aspects, while others are further behind. Overall, the implications of the DASH study are relevant for all the agencies. There is significant potential for impact across the UK in addressing the challenges (listed above) that are associated with integrating the new and emerging data sources and technologies identified by the project. |
| Sectors | Communities and Social Services/Policy,Digital/Communication/Information Technologies (including Software),Environment,Healthcare,Government, Democracy and Justice,Transport |
| URL | https://dash.kcl.ac.uk |
| Description | The findings from the DASH Policy report have been cited in the London Assembly Health Committee Report of December 2018. The report includes three recommendations for the "wider urban environment", one of which is: "Recommendation 7: The Mayor should instruct his Chief Digital Officer to explore partnership working with the LAS on the Smart Cities data programme, with particular reference to the six new data initiatives identified by the King's College DASH project. As part of this, the Mayor should set out to the committee how he will encourage a conversation with Londoners about the use of health data and analytics, using the London Ambulance Service as a case study" (p13). The specific findings of DASH, as reported in our Policy Report "Data for Ambulance Dispatch" are cited in full on page 42 of the London Assembly Health Committee Report. This citation helps solidify the impact of DASH within the city of London and will strengthen our case as we move forward in pursuit of future funding to support exploration of the technologies underlying the "six new data initiatives" identified by the DASH project: (1) better integration of health and social care data; (2) partnership with TfL for intelligent and timely use of transport information; (3) engagement with the London Air Quality Network to improve response to patients with chronic breathing problems; (4) exploration of access to mobile network providers' data to better map population mobility; (5) investigate use of video communications technologies to support interaction with patients at the scene of an incident; and (6) facilitate use of weather data models to help maintain service levels as weather patterns change. |
| First Year Of Impact | 2018 |
| Sector | Healthcare,Government, Democracy and Justice |
| Impact Types | Policy & public services |
| Description | DASH Findings mentioned in the London Assembly Health Committee Report, December 2018: The report includes three recommendations for the "wider urban environment", including "Recommendation 7: The Mayor should instruct his Chief Digital Officer to explore partnership working with the LAS on the Smart Cities data programme, with particular reference to the six new data initiatives identified by the King's College DASH project. As part of this, the Mayor should set out to the committee how he will encourage a conversation with Londoners about the use of health data and analytics, using the London Ambulance Service as a case study" (p13). |
| Geographic Reach | Local/Municipal/Regional |
| Policy Influence Type | Citation in other policy documents |
| URL | https://www.london.gov.uk/sites/default/files/london_ambulance_report_final.pdf |
| Description | Impact Acceleration Award |
| Amount | £15,000 (GBP) |
| Organisation | ESRC Impact Acceleration Account Cambridge |
| Sector | Academic/University |
| Country | United Kingdom |
| Start | 01/2019 |
| End | 07/2019 |
| Title | Operational Decision Making for Ambulance Dispatch |
| Description | Note that none of the "type of research tools" listed on the dropdown list fits this tool. We have been developing a methodology to provide decision support at the operational level for ambulance dispatch. This builds on prior work of ours and collaborates with members of a research team at Birkbeck University. |
| Type Of Material | Improvements to research infrastructure |
| Year Produced | 2017 |
| Provided To Others? | No |
| Impact | Results are preliminary at this stage. |
| Title | Tactical/Strategic Decision Making for Ambulance Dispatch |
| Description | Note that the type of tool/method from the list is not a good fit. We have been developing a methodology for aiding in resource allocation for ambulance dispatch at the strategic/tactical levels, building on a tool called ArgTrust, which was developed by two of the project investigators previously. We have been adapting ArgTrust to the LAS situation and aim to demonstrate its potential by the end of the project funding period. |
| Type Of Material | Improvements to research infrastructure |
| Year Produced | 2018 |
| Provided To Others? | No |
| Impact | Work is ongoing and results in the LAS domain are anticipated in 2018. |
| Title | LAS Incidences & Responses Database |
| Description | We have developed a structured database of incidences and responses at LAS, built on data that LAS provided to us. The data is sanitised, but is still sensitive and not publicly available. |
| Type Of Material | Database/Collection of data |
| Year Produced | 2017 |
| Provided To Others? | No |
| Impact | The database feeds the software models reported in the "Tools and Methods" section here. Results are forthcoming. |
| Description | CUSP London - Centre for Urban Science and Progress |
| Organisation | King's College London |
| Country | United Kingdom |
| Sector | Academic/University |
| PI Contribution | The DASH project has brought about a collaboration with the new CUSP London - Centre for Urban Science and Progress, which is a partnership led by King's College London, with University of Warwick and New York University (US). Through DASH workshops, members of the DASH team and LAS were introduced to members of CUSP. This has established a collaboration which is supporting the CUSP Hackathon 2019 and several student research projects in 2018-19. |
| Collaborator Contribution | The annual CUSP Hackathon, to be held 18-21 March 2019 at King's College London, with participation from King's College London, University of Warwick, New York University and University College London students, working on data and problems provided by LAS. |
| Impact | The formal aspects of this collaboration have just begun and it is too soon to have any outputs. |
| Start Year | 2019 |
| Description | London Ambulance Service |
| Organisation | London Ambulance Service NHS Trust |
| Country | United Kingdom |
| Sector | Public |
| PI Contribution | We have been analysing data (below) provided to us by LAS. |
| Collaborator Contribution | London Ambulance Service (LAS) are a partner on the grant and they have provided data for us to work with on the project, as well as introductions to people throughout the organisation who have provided information about how they operate and their challenges. |
| Impact | Outputs are listed in the publications section of this report. Collaboration is multi-disciplinary: computer science (artificial intelligence), engineering (robotics) and social science (policy). |
| Start Year | 2017 |
| Description | DASH Workshop 8 May 2018 |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Professional Practitioners |
| Results and Impact | 61 people attended the closing workshop of the DASH project in which project results were summarised, a panel discussion ensued, and plans for the future were reviewed. |
| Year(s) Of Engagement Activity | 2018 |
| URL | https://www.kcl.ac.uk/sspp/policy-institute/publications/DASH-policy-findings.pdf |
| Description | DASH Workshop, 12th September 2017 |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Professional Practitioners |
| Results and Impact | This workshop shared and tested ideas being pursued by the DASH project, as well as discussion around open policy making principles by concentrating on participants' ideas and problems. The objective was to strengthen the networks which support both researchers and LAS in achieving impact through their work. We focussed on three particular data sources from the wide range DASH is considering: 1. Health and care system data 2. Environmental data (including climate and air quality) 3. Mobility data (including transport system and location) |
| Year(s) Of Engagement Activity | 2017 |
| URL | https://dash.kcl.ac.uk/2017/10/23/report-from-a-workshop-on-12-september-2017/ |
| Description | DASH Workshop, 9th January 2018 |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | Local |
| Primary Audience | Professional Practitioners |
| Results and Impact | This workshop provided an opportunity for LAS personnal not directly involved with DASH to interact with members of the DASH team and other interested researchers from King's College London to discuss work-in-progress with DASH and, more importantly, future directions for progressing the research questions after the current DASH funding ends (May 2018). |
| Year(s) Of Engagement Activity | 2018 |