Background English Ambulance Services are faced with annual increases in emergency demand. Addressing the demand for low acuity emergency calls relies upon the ability of ambulance clinicians to accurately identify the most appropriate destination or referral pathway. Given the risk of undertriage, the challenge is to develop processes that can safely determine patient dispositions, thereby increasing the number of patients receiving care closer to home.
Aims The aim of the study was to evaluate the clinical utility and safety of triage support tools (Paramedic Pathfinders).
Methods Two triage filters (Pathfinders) were developed (one medical, one trauma). These were applied by ambulance clinicians to 481 patients who had been transported to emergency departments (EDs). Preferred (gold standard) patient dispositions were established by senior medical practitioners using both ambulance and ED clinical records. The clinical utility of ambulance clinicians using Pathfinders was evaluated against this gold standard.
Results The Medical Pathfinder was applied to 367 patients (76.3%) and the Trauma Pathfinder to 114 (23.7%). Agreement between ambulance clinician and gold standard was achieved in 387 cases (80.5%) giving the tools a combined sensitivity of 94.83% and specificity of 57.9%. 20.9% of medical patients and 30.7% of trauma patients who had been transported to hospital could have been safely cared for elsewhere.
Conclusions Ambulance clinicians using Pathfinders have demonstrated acceptable levels of sensitivity in identifying patients who require ED care. The actual impact of the tools in clinical practice will be dependent on the provision of suitable alternatives to ED.
- Clinical Assessment
Statistics from Altmetric.com
English Ambulances Services are facing increases in demand every year. Total emergency calls have risen by 83.2% (from 4.41 to 8.08 m) since the turn of the millennium. Of those patients conveyed, the majority of patients (88%) are still transported to emergency departments (EDs).1 While those with critical and emergent conditions will continue to require conveyance to EDs, the present challenge is to safely identify alternative treatment and referral options for non-critical patients at the point of first contact.
Studies suggest that Paramedics and Emergency Medical Technicians are not able to accurately or consistently identify those patients who would benefit from non-ED care.1–3 It has been suggested that a protocol-based system will be required pending the development of reflective ambulance practitioners through academic education.4 Further, a recent summary of the evidence shows that most UK studies have been based not on identifying alternative pathways, but on focused non-conveyance initiatives.1 Further, these studies consistently show challenges in safety and reliability in the identification of patients who are suitable for non-conveyance. Two studies2 ,3 have explored referral to alternative providers, one to a Minor Injuries Unit in the UK, the other to a Primary Care Centre in the USA. Both studies found that while the clinical risk associated with these strategies was low, it was dependent on clear assessment protocols and adequate training of those using them. The studies recommend alternative strategies that optimise the potential advantages of identifying all suitable patients, not just those falling in certain triage categories or conditions.
Finally, and perhaps most relevant to this study, no evidence was found that assessed the use of non-condition specific guidelines. There is a clear evidence gap requiring an integrated emergency/urgent care systems approach that considers response and clinical management/referral options across all categories of emergency.
Development of Paramedic Pathfinder
A multidisciplinary clinical working group was established to design the tools, using symptom-based discriminators derived from the Manchester Triage System.5 It was decided to develop two generic protocols (named Pathfinders)—one for patients who had suffered trauma and the other for patients with medical conditions. This generic approach obviated the need for multiple guidelines (considered cumbersome in an operational environment).
The Pathfinders were designed to safely categorise patients into four groups (or pathways): those requiring emergency care, those requiring urgent care, those whose care could be safely delivered in the community with appropriate support and those who could self-care. Pathfinder was designed to be used by ambulance clinicians with a skillset of Institute of Healthcare Development (IHCD) Technician (EMT2) and above.
The aim of the study was to evaluate the clinical utility and safety of these triage support tools (Pathfinders) using a mixed clinician sample.
During the evaluation process, the Pathfinder tools were in printed sheet format available in hardcopy to all clinicians who participated in the evaluation. The original Pathfinder tools themselves are shown in figures 1 and 2 below.
Eight evaluation sites were identified based on the existing or intended provision of Urgent Care Centres (UCC). All patients were ‘live’ 999 activity, but no change was made in conveyance or destination.
Staffs of EMT2 grade and above were asked to participate. This was to ensure that the sample was a fair and accurate representation of clinical staff and their range of capabilities. The identity of the attending practitioners was not recorded, although their job title was recorded for comparison and evaluation purposes. Informed consent was sought from the lead clinician on each conveying ambulance. No one refused.
Clinicians were asked a single, simple question: Using the ‘Paramedic Pathfinder’ what is the most appropriate destination for this patient?
Community Care Pathway
The evaluations were completed sequentially in each unit over a period of 2–3 days. Patients for evaluation were identified by convenience on arrival at the ED and therefore patients who had been managed without transportation to an ED were excluded from the study. At the time of evaluation, the non-conveyance rate was around 18.5% and this encompassed patients who were documented as refusing transport, those who were absent at scene, and those who met existing self-care guidelines. Patients with the following conditions or characteristics were excluded from the evaluation as they are specifically excluded from the tool itself:
Face Arm Speech Time (FAST) positive
Cardiac chest pain
Obstetric and gynaecological presentations
Lack of mental capacity
Less than 5 years of age
Early Warning Score > 4
Patients transported to hospital after evaluation by another healthcare professional were also excluded from the evaluation as the patient outcome was predetermined.
The gold standard disposition for each patient was determined by a group of three senior clinicians who had full access to both the ambulance and the ED notes. This group included one senior clinician from each: the ED, the Ambulance Service and the UCC (or a nominated clinician if the UCC was not yet open). Each clinician made an independent assessment and the decisions were then collated to establish the outcome. A majority decision was accepted. The gold standard clinical review was conducted using a set of strict evaluation guidelines, which clearly set out a minimum requirement for UCC level of care.
Formal ethical approval was not required as this study is a service evaluation. However, the study was approved through the Ambulance Trust Research and Development process and complied with the principles of good ethical practice.
538 data forms were collected during the data gathering process. Data were ‘cleaned’ by removing all patients admitted by a Healthcare Professional (n=25) in cases where no ED notes could be traced (n=26), or where no expert consensus could be established (n=6). This produced a final data sample of 481 cases. The Medical Pathfinder was applied to 367 patients (76.3%) and the Trauma Pathfinder to 114 (23.7%).
The Medical Pathfinder was applied to 367 patients. In 298 cases (81.2%) the lead ambulance clinician and the expert panel agreed about disposition. In 56 cases (15.26%) patients determined to need primary ED assessment by the lead ambulance clinician, were deemed suitable for an Urgent Care, alternative pathway, or non-conveyance by the expert panel. Thirteen patients (3.54%) thought suitable for an Urgent Care by the lead ambulance clinician were deemed to need ED assessment on review of the clinical notes. Table 1 indicates the use of ambulance clinicians at determining disposition of patients with medical presentations: this means that if alternatives are available, and in eligible medical patients, ambulance clinicians using the tool can deflect 20.9% of cases away from the ED with a sensitivity of 94.4% (90.7–96.7%) and a specificity of 57.9% (49.4–65.9%).
Thirteen (3.54%) patients would have been incorrectly deflected from the ED by ambulance clinicians. Where Pathfinder was found to be correctly applied, cases were subject to a senior clinical review by three ambulance medical directors. Themes were identified in ability to differentiate between cardiac-related chest pain and chest pain of musculoskeletal origin resulting in change to the discriminator to ‘non-traumatic chest pain’. This reduced specificity but clearly improved sensitivity in this respect. Patients presenting with haematuria, first episode retention and history of loss of consciousness were also increased in outcome to ED from Urgent Care.
The Trauma Pathfinder was applied to 114 patients. In 89 cases (78.1%), the lead ambulance clinician and the expert panel agreed about disposition. In 23 cases (20.2%), patients determined to need primary ED assessment by the lead ambulance clinician, were deemed suitable for an Urgent Care, alternative pathway or non-conveyance by the expert panel. Two patients (1.7%) thought suitable for an Urgent Care by the lead ambulance clinician were deemed to need ED assessment on review of the clinical notes. Table 2 indicates the use of ambulance clinicians at determining disposition of patients with traumatic injury: this means that for eligible trauma patients, ambulance clinicians using the tool can deflect 30.7% of cases away from the ED with a sensitivity of 96.4% (87.9–99%) and a specificity of 60.3% (47.5–71.9%).
The two patients who were incorrectly deflected from the ED were both looked at in detail. One was a patient incorrectly thought to have musculoskeletal chest pain (there was significant cardiac history), while the second was a weightbearing elderly patient with unresolved hip pain after a fall who had a fractured neck of femur.
This study has sought to investigate the utility and safety of the ‘Paramedic Pathfinder’ in patients currently transported to EDs by ambulance. The study group included those patients arriving at EDs by ambulance after existing opportunities for non-dispatch (hear and treat) and non-conveyance (see and treat) had already been taken. Patients with serious conditions (stroke, cardiac chest pain), patients with conditions not amenable to non-medical assessment (mental illness, obstetrics and gynaecology), patients with deranged physiology (Early Warning Score (EWS) >4), those lacking capacity and those under 5 years old were excluded from Pathfinder application. Patients previously assessed by a healthcare professional were also excluded. As the sample group explored only those patients who were conveyed, the Pathfinder was not retrospectively applied to patients who remained at home or referred to alternative services. The eligible group therefore represents approximately 79% of the total patients dealt with by the Paramedic Emergency Service in the North West of England. It is of course possible that application of Pathfinder to those patients previously not conveyed would have resulted in increased conveyance. However, this will be subject to further review in due course.
Both Pathfinders pre-exclude patients with Pre Hospital Early Warning Score (PHEW) of >4. This score, a combination of physiological indicators allocating a higher score to observations further from physiological norms, is designed to identify patients who are deteriorating at an earlier stage than pure presenting symptoms. The PHEW used as shown in figure 3 is a adaption version of the modified EWS.
The study identified 38 (7.9%) patients with a PHEW score of >4. Of these patients, 35 were medical presentations, the remaining three, from the trauma cohort. All but three of the 38 patients were deemed in need of A&E attendance, two from the trauma group and one medical presentation were UCC or pathway suitable.
Analysis of whether the discriminators would have identified these patients without the PHEW exclusion is difficult as paramedics did not generally continue to apply the algorithm once pre-excluded. None of the patients incorrectly thought suitable for non-ED destinations by ambulance clinicians had PHEW scores >4. Indeed five had a PHEW of 0. The PHEW score needs further evaluation, especially against the recent Royal College of Physicians proposal for a National Early Warning Score,6 although at this stage, given the overall specificity of the Pathfinders is not significantly affected by PHEW inclusion, the project group recommended that the PHEW remains a general exclusion within both Pathfinders.
Statement of principal findings
Analysis of the 481 cases considered in this study suggests that given the provision of a kitemarked UCC and appropriate community care provision, only 57.7% of eligible patients arriving by ambulance at EDs actually required assessment there. Some 23.3% of patients would have been correctly deflected from the ED if the Pathfinder tool were in use. A further 16.4% of eligible patients were considered by the expert reviewers to be suitable for deflection from the ED if suitable services are in place while 3.1% of patients were incorrectly selected as suitable for alternative provision. Of those patients considered by the expert panel to be suitable for deflection (39.6%), 5.8% were considered appropriate for referral to specialist community services, 33.8% into Urgent Care. Further work is ongoing to establish referral options into primary care in areas with no UCC provision.
Limitations and strengths of the study
The sample size was lower than originally anticipated due to the time constraints on collection of data and availability of expert clinicians. However, further evaluation of the Pathfinder including physiological scoring and impact on sensitivity and specificity of the recommended changes is being undertaken in a live environment. The inclusion of additional outcomes, such as community and primary care referral will also be subject to further evaluation. This Trust had undertaken internal evaluation on the ability of clinicians to determine destination without use of the Pathfinder tools. The findings were supportive of the results of previous international service evaluation, producing an undertriage rate of 11.5%.
A particular strength of this study is that it has looked at patients actually delivered by Paramedic Emergency Service to EDs.
Unanswered questions and future research
Application of the Pathfinder may increase on-scene times but this will be offset by reduced hospital/UCC turnaround times. Reduced conveyance distances may also impact on job cycles times although further evaluation will be required to assess the impact in those geographically challenged areas.
Under-triage (taking patients to a facility unable to deal with the acuity of their problem) represents the greatest risk clinically. A detailed review of under-triage attributable to the tool will allow changes to be made.
Both Pathfinders have demonstrated acceptable levels of sensitivity and specificity, even without the adoption of the recommended changes, post data analysis. The improved levels of sensitivity compared with existing evidence, suggests that the use of reductive triage methodology based on symptom recognition, is preferable to diagnostic-based decision making. The safety and potential benefits of implementation will be entirely dependent on the provision of suitable community and primary care alternatives to traditional ED.
Contributors MN is the lead contributor and has been the programme manager for development of the Pathfinder triage tools. He was directly responsible for coordinating and developing the Pathfinder concept and development, and remained so throughout the entire process. MN developed the evaluation methodology, coordinated the evaluation and drafted the initial paper prior to submission to EMJ. MN was responsible for collecting and analysing the data and producing the evaluation report for the Trust Board. MN is responsible for the overall content. KM-J is the Medical Director of the NWAS NHS Trust, and as such, was the responsible officer, overseeing the development and evaluation of the Pathfinder concept. He has reviewed and approved the draft algorithms for pilot and approved evaluation methodology. He assisted in the drafting of the submitted article. ET and DR are both Assistant Medical Directors of the employing NWAS NHS Trust. They were fully involved in the conceptual planning, design and evaluation of the Pathfinder Triage filters. They contributed to the drafting and review of the submitted article. IM is Head of Service Modernisation and contributed directly to the development and evaluation of the Pathfinder triage algorithms. He was jointly responsible for developing the Trust's Urgent Care strategy, which was the primary driver for creation of Pathfinder. IM contributed to authorship of the submitted article and the development of the paramedic Pathfinder Tool. All authors credited in the paper made substantial contributions to conception and design of the Pathfinder Triage filters, acquisition of evaluation data and analysis and interpretation of data. Each was involved in either drafting the article or revising it critically for important intellectual content, as well as final approval of the version to be published.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Additional data are available in the form of full statistical analysis of each decision made by both the senior medical team and the clinicians involved. This has been broken down into specific area identifiable data. The original Trust report is an open report and is available on request from NWAS NHS Trust.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.