Ambulance Response Programme: a socialist analysis

A socialist analysis of any health programme requires recognition of the competing tensions of resource, patient care and working conditions, and the necessity of balance to prevent a crisis. Therefore, the following analysis will reflect these tensions. Finally, I will analyse the crisis points which caused this to be necessary. During this analysis, I will exclude increased funding as a consideration, due to the low liklihood under the current government, and because increased funding would ameliorate, but not solve, some of the underlying problems.

The Ambulance Response Programme

The Ambulance Response Programme was the largest overhaul of the way the ambulance services deals with calls for 20 years, and was trialled in the West Midlands, South West and Yorkshire Ambulance Services from 2015, before now being rolled out to other services.

The major changes that the ARP brought about were:

1. Increased time to assess call requirement before resource allocation (through new ‘Nature of Call’ and ‘Dispatch on Disposition’ processes)
2. Increased number of categories with longer response times

Prior to the ARP, calls were broadly coded as follows:

After a series of trial and error, the latest iteration of the ARP codes calls as follows:

Analysis: Old vs New

Resource efficiency

Efficiency is important, no matter your political position. However, the motive for efficiency is key. Clearly, it was inefficient for 50% of ambulance calls to be responded to within 8 minutes, especially considering the 90%+ rate of false positives (University of Sheffield, 2017). This was both detrimental to patients who couldn’t get an ambulance, and workers who dealt with the stress of demand. Therefore, better ways of matching clinical need to resources were required.

There is detailed evidence-based analysis provided by the University of Sheffield (2017) (Presentation, Final Report) regarding the effect of the ARP on resource efficiency. As they make a much better argument than I can, I will only summarise their main findings:

• Increased time to correctly assess calls before resource allocation has resulted in increased operational efficiency on all measures, and subsequently better response times for seriously ill patients.
• The change in call categorisation is a large overhaul and still in its infancy, and many adjustments are yet to be made, however, the initial results reveal that resource allocation more closely matches clinical need, and reports from staff in EOC appear positive.

While it has clearly been driven in response to decreased funding to meet rising demand, the important question to be made is whether this efficiency has been beneficial to patients and workers, or whether it has been solely to meet a wider agenda.

Effect on patient care

 It is easy to take away headlines such as “Patients will wait 20 minutes instead of 8 minutes when having a heart attack”, but it is obviously more complex mechanism than that. Of course, ideal conditions and unlimited funding would lead to an ambulance on every street, but reality imposes constraints, even in a well-funded health system such as one under a Labour government.

It is important to recognise what the numbers mean. The numbers are arbitrary and represent a target (i.e. 75% of Category 1 calls responded to within 8 minutes) by which to measure performance and evaluate the effectiveness of changes. So the changes that have been made to call categorisation are only to increase the number of categories and to what delay is deemed acceptable for each category. The number of minutes can be changed at any time. That being said, studies have shown that there is no clinical benefit for response in less than 8 minutes for any condition other than out-of-hospital cardiac arrest (University of Sheffield, 2017).

Where efficiency has been gained is because of the greater time allowed to ambulance controllers to send an appropriate resource (e.g. one that may be about to come clear close to the incident), rather than forcing controllers to dispatch resources immediately, only to stand them down once half way across the city. This is especially evident when one considers the maximum response time for an emergency call was previously 19 minutes, even for very low-urgency calls (e.g. minor ailments such as a cold) – it is now 90 minutes, allowing controllers plenty of time to allocate a resource while ensuring emergency needs are met.

One unintended side-effect that has had a detrimental impact on patients arises due to the change in the ability of Rapid Response Vehicles (RRV) to ‘stop the clock’. RRVs were located in many rural towns, where being highly localised enabled them to have a good relationship with other health care providers (impossible in the urban environment), were manned by experienced paramedics who were highly skilled in the use of alternative pathways. Not only this, they were able to provide rapid response in time-critical conditions. In summary, RRVs were good for patient care.

From a performance target perspective, they were able to ‘stop the clock’, which financially justified the lack of ambulance provision to these rural areas and their presence masked the long (several hour at times) waits for a conveying ambulance to arrive (which wasn’t a performance target). This tragedy led to the removal of ability of RRVs to ‘stop the clock’ when it was apparent a conveying ambulance was required (e.g. elderly fall with hip pain). In practice, what has happened is, in organisations forced by policy makers to place meeting budget requirements above patient care, that the financial justification for RRVs disappeared and the RRVs were quickly replaced by ambulances, which spend a long time away from their locality conveying to city A&Es leaving those rural areas with little or no cover. The towns involved have rightly been incredibly angry at this change.

Clearly, a middle way is required in this circumstance, and it is imperative that NHS policy makers recognise the unintended side effect of their decision and adjust accordingly.

It is a complex picture to say whether this change has benefited patient care overall. While I would like no-one to wait a long time for an ambulance, a utilitarian approach must be taken – the greatest amount of good for the greatest amount of patients. Patients who are uncomfortable but stable, (e.g. elderly fall with no injury) might now be waiting longer for ambulances, but that means resources are available to reach seriously ill patients quicker.

The major confounding factor in the analysis of the ARP is the inability of current triage systems to differentiate between emergency and urgent conditions in some circumstances, leading to symptoms becoming groups to be dealt with together. A triage system that could sort heart attacks from other non-emergency causes of chest pain would enable, for example only, a 8 minute response to one, and a 40 minute response to the other. However, the computer triage system treats all chest pains as one common group. As a result of this, the previous categorisation system took the lowest common denominator and required an 8 minute response to all chest pains – which, along with other similar processes, proved unsustainable without massive increases in funding that’d be eye watering even under socialist governments. The current categorisation system, therefore, has had to compromise, allowing Ambulance Services a faster response than necessary to some chest pains, but a slower response than necessary for others. The weakness of computer triage systems will be examined briefly later in this analysis.

Effect on workers

 The effect of ARP on ambulance clinicians has been positive on the whole. The major concerns of Ambulance Clinicians are finishing on time, getting a break, and fatigue, including blue light driving fatigue.

Prior to the ARP, it was common to respond to a call for a low urgency condition only minutes before you were relieved by an oncoming crew.

The increased number of categories, and the increase in time allowed to respond to urgent and non-urgent calls has allowed unions to push for improved end-of-shift policies and break policies, for example, not responding to calls with a 60 or 90 minute response time within the last hour of your shift, and getting a break halfway through the shift unless there is an outstanding emergency call. Both of these policies represent major steps forward in ambulance working conditions only made possible through the ARP. No figures have been released around whether these policies have decreased late finishes and late breaks (and of course, there are many confounding factors), but I believe the positive conclusion follows logically.

The inefficiency of the previous system led quickly to crew fatigue. While, of course, fatigue is still a major issue for ambulance crews, a small amount of the burden has been improved.

Many outsiders find it hard to believe that blue lights & sirens are used to go to every call, from stubbed toes to shaving cuts and to mothers struggling to get their baby to sleep (I’ve been to all 3!).  This policy, suited to an era gone by where few calls were not life threatening emergencies, is still enforced by Trusts in the name of benefit to the patient, but it is widely believed on the frontline that the motive is purely financial (less time driving between jobs means more jobs achieved during the shift). Driving on blue lights is fatiguing, especially in rural environments where you may be driving for an hour routinely. One major shift represented within the ARP is the recognition that some responses (Category 4) don’t require a blue light response – again, reducing the burden of fatigue. In reality, few conditions benefit from blue light responses, and the correct solution would be a flexible approach: allowing the clinician to make the decision according to weather & traffic conditions and clinical need, rather than rigid blanket rules.

Fundamental tensions: is there a better way than ARP?

Now, one must ask, why was such an overhaul of response times necessary? It is only by addressing the fundamental tensions that led to such a change occurring that it would be possible to improve on ARP.

While it was certainly correct to move away from old systems of working, the driving factors arise from two points – the tension between emergency and urgent care within the Ambulance Service and the high rate of false positives as a result of computer triage systems – both of which are subjects for in-depth analysis themselves, but about which I will make only a brief assessment.

Demand for emergency pre-hospital care is what the Ambulance Service evolved to supply, and almost everything within the Ambulance Service is oriented to this demand, including previous systems of call categorisation, the medication & equipment we carry and our training. It also evolved in an evironment where 999 was infrequently used by patients for urgent healthcare. However, due to austerity in other areas, especially General Practice, Mental Health and Social Services, patients have turned to 999 (and now, 111) to meet their urgent needs. In simplest terms, the Ambulance Service is overwhelmed by urgent and primary care requests, such that when a time-critical emergency occurs, there are few or no resources to respond. The ARP is a clear shift in ambulance service policy to attempt to more efficiently bridge this tension between emergency and urgent care.

This tension is fundamental within the Ambulance Service, yet is little debated in absolute terms in the public realm. While reversing austerity in GP, MH and local authorities would reduce ambulance urgent care demand, the tension remains beneath the surface. Not only that, ambulance clinicians have shown that they have a place in supporting medicine in the community.

The reality is that no single pool of resources can meet demand for both urgent care and emergency care without one or both suffering, and therefore exclusive resources must be allocated for each purpose, probably through split emergency and non-emergency organisations (potentially into 999 & 111, Fire Service & Ambulance Service, or new organisations entirely). Implicit in this is the recognition that emergency care requires resources in reserve.

The system that made ambulance services vulnerable to the increase in urgent care calls is the use of a computer triage system which will always be designed to be near risk-free. While it brings benefits such as standardisation and ability to audit, its weakness is the removal of the human subjective element from it. The easiest example to illustrate is that of Chest Pains: A risk-free computer system will be designed to miss no heart attacks, and therefore every chest, back or jaw pain is treated as a heart attack, despite that call being for 10 year old back pain, or for dental problems. The system churns out a high number of false positives of many conditions, stretching resources such that they are unable to attend the true positives. It is due to this that over 50% of ambulance calls required 8 minute responses under the old categorisation system.

Regarding solutions, due to the non-standard presentation of medical conditions, it would near-impossible to improve the false positive rate through question refinement (e.g. at what duration of chest pain do you not treat it as a heart attack? Do they have to say “crushing chest pain”?). The job of clinicians, however, is to consider many subjective factors simultaneously and come to a conclusion as to the likelihood of a condition being present. The weakness of clinicians, on the other hand, is that they make mistakes, and their telephone clinical assessment could delay response. The return of clinicians to call-taking is expensive and introduces some risk, but would produce system wide benefits in reducing ambulance responses, and therefore improving response times to seriously ill patients.

One further question on which there has been no debate is: if it has been deemed safe for a patient to wait 60 minutes for an ambulance, but ultimately will require admission, and they live only 10 minutes from the hospital, should they receive an ambulance or be told to make their own way to A&E? In many cases, asking them to wait for an ambulance, when they could make their own way only serves to delay treatment, which is unethical.

The question that must ultimately be asked is whether they should receive an emergency ambulance at all, if their condition is so stable. I do not propose leaving such people without care at all, but I return to my previous point on the tension between emergency and urgent care. I merely propose that other forms of urgent care services could meet this demand better, and from a resource perspective, patients with urgent care conditions who can make their own way to A&E or other services should be encouraged to.

Only a brief assessment of these factors has been made here, as a true assessment must be far more in-depth than this analysis allows, and I hope to elaborate on these issues more in future.

Conclusion

The ARP, though still early on, has appeared to have improved resource efficiency, have had a broadly positive effect on patient care for the most seriously ill patients, and allowed unions greater scope to develop better working conditions. I can find no reason for socialists to oppose this change.  However, socialists must look beyond surface symptoms for deeper tensions, and by understanding and attempting to resolve these tensions, alongside others, it is possible to improve patient care, working conditions, and resource efficiency at a magnitude far greater than current attempts.

Further reading would be the in-depth report by the University of Sheffield, which provides a detailed background and evidence based analysis on its effect on resource efficiency.

James Angove is a pseudonym. The author is a socialist and a paramedic in the UK, whose identity must be hidden due to the treatment of health care professionals and other whistleblowers who talk about issues within the health service.