By Archie Drake
This is an exciting time for ambulance services: technological developments over the last decade or so suggest a range of possible changes and improvements. Working out what can be done better in practice will help them save lives.
Since the 1980s, ambulance services have made use of ‘computer assisted dispatch’ (CAD) systems to become more sophisticated in the way they allocate ambulances to respond to incidents. The way in which CAD systems can draw in outside information to support decisions is now changing fast. Just like users of apps like Google Maps, they can make use of more reliable and up-to-date routing information. Like users of WhatsApp, they might use video calling to communicate with people rather than audio-only. Improved system awareness and user responsiveness has the potential to enhance ambulance CAD, in some ways like Uber has challenged traditional taxi firms or Amazon has transformed warehouse management practices.
These comparisons are unfair; ambulances are certainly not taxis, and ambulance service staff handle information which is very different to that involved in these other applications. But are we using new technology primarily for our convenience rather than to save lives? This raises uncomfortable questions about where our priorities lie and why.
The temptation is to ask these questions exclusively of ambulance services themselves. Many would say that they simply need to improve and work harder to deliver better services. Some might even say they just need to become more like the Ubers or Googles of this world.
That’s not a constructive approach. Yes, clearly the services themselves are responsible for providing urgent and emergency care in the best possible way. Yes, improvements can always be made. But the issues here are beyond the control of ambulance services acting alone. Technological change is complex and conditioned by context, not the simple ‘flick a switch’ leap imagined by solutionists. We need to look at the wider picture to work out how ambulance services can make the best of the opportunities that technology seems to be offering them.
In the first place, public services are finding it difficult to cope with significant social change under the tight fiscal policy of recent years. An incisive National Audit Office report on NHS Ambulance Services earlier this year summarised the great pressures with which UK ambulance services are coping. Demand is rising sharply, for reasons which are poorly understood but seem likely to include our ageing population, wider pressures on the NHS and social care system (not least on mental health services), public health challenges like pervasive alcohol abuse, and – at a more fundamental level – the changing ways that we seek help with our problems.
Meanwhile, resources have been squeezed. Funding for urgent and emergency activity increased by 16 per cent between 2011-12 and 2015-16, as against the 30 per cent increase in activity over that period. Ambulance services are struggling to retain their staff as their working environment has become more dangerous, threatening, pressured and stressful. Is it any wonder, then, that we have seen ambulance performance tend to deteriorate in recent years?
This is not just a question of financial and human resources, momentous though those issues are. Information resources are the primary driver for modern technology companies. Are ambulance services obtaining all the information necessary to optimise their response to demand from the community?
Working from a clinical perspective with the University of Sheffield, the NHS Ambulance Response Programme (ARP) has found ways to improve how services gather information on incidents, focusing the correct response on priority needs (‘Dispatch on Disposition’). Ambulance services all over England are now moving to implement the ARP’s recommendations.
As computer science and policy researchers working here at King’s, we think that in today’s increasingly networked, digitally enabled world, ambulance services can also look further afield for information to support their work. Our local London Ambulance Service (LAS) agrees, and together we’ve started work on a one-year demonstrator project, ‘Data Awareness for Sending Help’ (DASH). The DASH project explores the potential impacts of integrating new and emerging data sources on emergency response and wider policy. Although we’re only in the early stages of our work, we’re already developing some interesting perspectives.
There are a wide range of new and emergent data sources relevant to LAS service delivery, ranging from types which are more relevant to strategic planning (like wider health and care system data), to types with tactical relevance (like weather forecasts), right through to types which relate to real-time decision-making (like caller locations). LAS processes already integrate a wide range of data into planning and dispatch decisions. Data mining techniques can be applied to data that LAS is already collecting to enhance their decision-making; our early investigations have shown that there are significant insights to be gained by taking better advantage of data already in hand.
There are significant benefits available here. Research can help clarify what advantages might result from uptake of emergent data, as well as how best to go about it. Extending to wider or improved data sources will further enhance response team efficiency, helping ambulance services deliver the best possible service within tight budgets. The benefit should be felt by patients themselves as ambulance services are supported in their ongoing effort to become increasingly responsive (including reaching people with life-threatening conditions quicker) and effective (delivering the best possible care in each individual situation). Ultimately, ambulance services will be enabled to save lives which would otherwise be lost.
We have been encouraged to find that there is a great deal of fresh activity in this area already. GoodSAM is an app, developed in partnership with LAS, which notifies a community of nearby medically qualified responders of an incident at the same time as the emergency services. Geotracker is a system developed for LAS by researchers at Birkbeck, which provides a platform for improved demand prediction and route management. From next year, all new cars will be obliged to include an eCall device, which calls the emergency services automatically when there is a crash (a comparable app called RealRider is already available for motorcyclists). A system called 999Eye is being trialled, which gives emergency service control rooms access to livestream video of incidents. It seems likely that new and emergent sources of data will have a transformative effect on ambulance dispatch.
National leadership has done little to advance these changes. Although it is technically overseen by the Department of Culture, Media and Sport, as one major stakeholder observes, ‘no-one really “owns” 999’. Within the field of health and social care, the National Data Guardian, Dame Fiona Caldicott, established as far back as 2013 that ‘professionals should have the confidence to share information in the best interests of their patients’. But the NHS has struggled to deliver on that vision for the system overall. The Digital Economy Act 2017 provides for wider data sharing between public bodies. That could help with certain public data sources if ambulance services make the case, but not with privately held data and not (apparently) with initiating the public conversations which are so important for success.
Conditions may be more productive at the city level. Intelligent transport system data is one of the most exciting types to emerge recently. Transport for London has set up an integrated Surface Transport and Traffic Operations Centre and draws on various sources of data, including from Waze, to produce real-time traffic information that LAS can call on. That said, LAS still faces a very difficult task working with London’s 32 borough councils (plus the City) and its 32 Clinical Commissioning Groups whose work includes commissioning LAS. In the absence of national progress, it is hard for local actors like LAS to make any headway, although they press on energetically with local initiatives like the Pre-Hospital Emergency Department Data Sharing project, which has tracked patient outcomes and will feed those back to improve their experiences in future.
This is a field of public activity in which private actors wield a surprising amount of effective control. Some choose to act responsibly, others less so. Take the example of mobility data. The Advanced Mobile Location function is a good example of a private initiative which supports ambulance services. But aggregate location data is a different story. As one researcher has pointed out, when it comes to anticipating ambulance demand, ‘the most relevant data is how the city’s population moves hourly’. If ambulance services can better understand where groups of people are at any one time, they can better understand where people are most likely to need an ambulance. Mobility data is one of the most remarkable sources of data to emerge in recent years, largely driven by the ubiquity of mobile phones and controlled by network and application providers. But while the effective controllers of that data are happy to sell it to marketing businesses, they don’t currently make it available for safe use by ambulance services.
Many people involved in this topic have deep-seated concerns about the ethical and legal implications of data sharing. Some of the new and emergent data which is or could be relevant for ambulance services is personal data, meaning it could be used to identify specific individuals and should therefore only be used with their proper consent. It is of course right to protect data of this kind, but privacy concerns are too often used as an excuse not to share data or use it for the public good in forms which no longer reveal personal information. It does appear, to borrow a term used by the Royal Statistical Society and Ipsos MORI in a 2014 study, that the public have a ‘data trust deficit’. But there is evidence that, in terms of support for the sharing of data, this relates much more to private companies acting for profit than it does to the NHS acting to save lives.
So why do we find that the practical effect of the rules and regulations governing data sharing are at odds with public opinion, and what can be done? A ‘lost generation’ has handed over information casually to private companies in exchange for services. Public sector organisations are barely coping with the pressures placed on them. As the Royal Society and British Academy jointly observed last month:
‘[data-enabled technologies] can make a great contribution to human flourishing but to realise these benefits, societies must navigate significant choices and dilemmas: they must consider who reaps the most benefit from capturing, analysing and acting on different types of data, and who bears the most risk.’
One way to increase the benefits and mitigate risks to the public – and boost trust – would be for appropriate levels of our government to pursue policies to safely improve the information available to ambulance services. Our main challenge here is to show how this can be done effectively.
Recent events here in London have highlighted the great service which LAS delivers for us all under the most difficult of circumstances. We are proud to be involved in a small way in supporting LAS with our investigation into how its information resourcing can be improved, and look forward to developing our initial observations into fully developed recommendations both for LAS and for other services facing similar demands and constraints.
Archie Drake is a Research Fellow at the Policy Institute.
DASH is funded by the Economic & Social Research Council (ESRC) grant ref. ES/P011160/1 (April 2017 – March 2018). Principal Investigator: Dr Elizabeth I Sklar.
For suggestions about this post, please contact Archie Drake: firstname.lastname@example.org.