The recent journal article on Paramedic Pathfinder (Newton et al, 2013) was clearly of great interest to a lot of paramedics— here was a new tool that was being proposed to support paramedics in non-conveyance decisions and to help them divert patients away from the emergency department. However, on closer inspection of the article, and with some critical examination, it becomes apparent that this is not necessarily the step forward that the article claims it to be, and instead is more an indication that there needs to be more rigorous research before a change in current methods can be recommended.
The article justifies the work based on the premise that paramedics and emergency medical technicians are unable to consistently identify those patients who would benefit from non-emergency department (ED) care, and provides a number of references to evidence this. On closer examination of the references, a number are based on American Emergency Medical Services (EMS) provision and there is no justification or evidence that this could be transferred to UK EMS provision given the significant variation between the two systems.
Once the potentially relevant references have been identified, it is evident that not one of them provides any evidence on whether paramedics can identify those patients that require the ED, only on whether paramedics can identify if patients will require hospital admission. These are two fundamentally different concepts as it is perfectly possible that a patient may require ED treatment or assessment but not actually require hospital admission, especially with the increasing use of ambulatory care schemes.
The references include a number of articles—such as Mason et al (2007) and Gray and Walker (2008)—that have identified that paramedics, when given extended training in assessment, treatment and diagnostic skills, are able to safely reduce attendance at the ED. Snooks et al (2004) investigated a protocol-based approach to see if this caused a higher proportion of patients to be left at home and found that this was not the case.
Finally, research by Clesham et al (2008) has shown that, ultimately, ambulance staff are able to correctly identify most patients who could be diverted away from the ED. This would suggest that the original reasoning for the article is flawed.
Having established that the assumption that paramedics require this tool in the first place might not be valid, it is important to consider if the evidence is strong enough to suggest that the tool is a good thing to implement.
The first concern is that the study on ambulance clinicians using the Pathfinder tool is compared to that of North West Ambulance Service NHS Trust (NWAS) clinicians and then generalised to include all ambulance staff. Latest performance data shows NWAS are the poorest ambulance service in the country in terms of the percentage of patients not conveyed to an ED, as well as the poorest in closing calls at telephone triage level. This is coupled with only mid-level performance on re-contact figures compared with other Trusts. This would suggest strongly that more work needs to be done locally to find out why other services are able to achieve much better results, before assuming that the convenience sample of clinicians is representative of all paramedics (Association of Ambulance Chief Executives, 2014).
The second concern is that the tool specifically excludes patients who are difficult to assess in the pre-hospital environment. To then compare this fairly to normal paramedic practice, it would seem prudent to first demonstrate how good paramedics are at determining destination with the same caveat. What this paper appears to compare is a triage tool that is only being used on the more straightforward patients against paramedics trying to correctly establish appropriate destination on the full range of presenting complaints.
Another area that has not really been explained is how the study compares numerous grades of clinicians but makes no comment as to whether higher grades are more successful in correctly establishing the appropriate destination. As has already been evidenced above by Mason et (2007) and Gray and Walker (2008), better training allows clinicians to avoid theED safely and more often, so it would be important to establish if this triage tool actually performs better than an average EMT, paramedic, paramedic practitioner or emergency care practitioner individually. By grouping all the clinicians together, the study is actually comparing the tool to an average of multiple levels of clinician and no information as to the makeup of the group is provided. Presumably the sensitivity and specific occurrence of a clinician making the same decision without the tool should increase as the clinical grade becomes more senior. This comes from research showing that advanced grades of paramedic, such as emergency care practitioners, are able to use their extended training and clinical reasoning to safely avoid ED attendance in a greater number of patients than paramedics (Dixon and Gaisford, 2014).
The tool makes use of the Pre-Hospital Early Warning Score (PHEWS), yet does not provide any evidence that this is an effective method of establishing patients at risk of deterioration. The tool also gives no consideration to patients with abnormal physiology normally, nor does it provide for the altered normal values expected in paediatric patients, despite Paramedic Pathfinder being recommended for use in anyone over the age of five years. There is no justification as to why this score was used over a better evidenced and more widely used one such as the National Early Warning Score (NEWS).
The study is based around a panel of three providing the gold standard answer, using not only the information provided at the time of assessment to the clinician, but also using the benefit of extra information provided by the receiving department. When the panel did not agree, it was assumed to be gold standard if a majority verdict was reached. The study does not disclose how often the panel were unable to provide a unanimous verdict, and in those cases, which of the three disagreed or why.
Finally, where the patient has not been conveyed or referred to another agency there was no follow-up to determine if this was a safe and appropriate thing to do. As a matter of importance it would seem vital to ensure that the patients that the tool recommends for non-conveyance are not being left behind in error.
Fundamentally, and ethically, this tool fails to treat patients as the individual they are with their own thoughts, feelings and concerns. While paramedic care has progressed towards a holistic approach and has rightly recognised the importance of involving patients in their own care, this tool takes a retrograde step in reducing every patient to a set of numbers with a pre-determined outcome. Even the gold standard decision of the panel is one made based on pieces of data written on paper in front of them with no involvement of the person that is receiving the treatment themselves.
In summary, this paper has been unable to provide any fair comparison of the Paramedic Pathfinder tool with that of current UK paramedics as a generalised group. Moving forward, research would need to be undertaken on the paramedic's ability to determine specifically the correct destination with the same exclusions as the tool and crucially on a more representative sample of paramedics from all Trusts rather than the one with the poorest non-conveyance rate. It is methodologically unsound to currently compare the triage tool to the ability of paramedics to predict if a patient requires admitting to hospital because they are not answering the same question. Indeed, it would be perfectly possible for a patient to attend an urgent care centre and then be admitted, and this to be a completely appropriate pathway for the patient to follow.
The Paramedic Pathfinder tool ultimately takes the human element of the paramedic profession away from patient assessment and reduces them to a flowchart that does not take into account their holistic needs or wishes. Providing increased training and skills to front-line paramedics has been shown to reduce ED attendance safely, and this paper has provided no evidence that increased reliance on rigid and inflexible protocols instead will achieve a better outcome for patients in the pre-hospital arena. Until further research is completed, there seems to be no compelling evidence that this tool is an improvement over current paramedic practice.