Background Community emergency medicine (CEM) aims to bring highly skilled, expert medical care to the patient outside of the traditional ED setting. Currently, there are several different CEM models in existence within the UK and Ireland which confer multiple benefits including provision of a senior clinical decision-maker early in the patient’s journey, frontloading of time-critical interventions, easing pressure on busy EDs and reducing inpatient bed days. This is achieved through increased community-based management supplemented by utilisation of alternative care pathways. This study aimed to undertake a national comparison of CEM services currently in operation.
Method A data collection tool was distributed to CEM services by the Pre-Hospital trainee Operated Research Network in October 2020 which aimed to establish current practice among services in the UK and Ireland. It focused on six key sections: service aims; staffing and training; job tasking and patient selection; funding and vehicles used; equipment and medication; data collection, governance and research activity.
Results Seven services responded from across England, Wales and Ireland. Similarities were found with the aims of each service, staffing structures and operational times. There were large differences in equipment carried, categories of patient targeted and with governance and research activity.
Conclusion While some national variations in services are explained by funding and geographical location, this review process revealed several differences in practice under the umbrella term of CEM. A national definition of CEM and its aim, with guidance on scope of practice and measurable outcomes, should be generated to ensure high standard and cost-effective emergency care is delivered in the community.
- prehospital care
- doctors in PHC
- emergency ambulance systems
- emergency care systems
Data availability statement
All data relevant to the study are included in the article or uploaded as supplemental information.
Statistics from Altmetric.com
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.
What is already known on this subject
Community emergency medicine (CEM) services aim to bring the ED to the patient, providing effective emergency care in the community.
This model of emergency care has been evidenced in several international countries, which use doctors as part of their prehospital emergency medical response, reporting lower ED attendances and better utilisation of community and non-emergency hospital services.
There is no study that reviews and compares services currently delivering CEM in the UK and Ireland.
What this study adds
Despite similar service goals and values, there were notable differences identified between services in terms of operational activity, equipment and governance structure.
CEM services have overlap with other community services but crucially are unique in their collaboration between the ED and ambulance service and offer an alternative proven approach to prehospital care.
Future collaboration between services and national bodies is needed in order to define CEM, develop national clinical and operational standards, and to evaluate services.
Community emergency medicine (CEM) is a relatively new and evolving aspect of emergency medicine. Operating in the prehospital setting, services aim to provide the knowledge, skill and expertise of clinicians in order to ‘bring the ED to the patient’.1 One aim of this model of care is to reduce ED attendances for patients who would otherwise have been transported to hospital through the utilisation of advanced on-scene diagnostics, provision of a senior clinical decision-maker early in the patient journey and facilitating access to alternative care pathways.
The CEM model has proven successful within inner city environments, such as the London-based Physician Response Unit (PRU), which has been shown to deliver safe and effective care to reduce hospital attendances.1 Their recently published cost analysis demonstrated significant savings as a consequence of reduced ED conveyances in addition to a reduction in inpatient bed days. There are also other theoretical benefits of replicating services such as this by improving relationships with the ambulance service, staff education and improving patient satisfaction by providing hospital-level care in the community.1 2 Several other models exist within both urban and rural settings. The Newport PRU, in operation since 2014 and serving southeast Wales, has suggested that their PRU service has the potential to significantly reduce local ED attendances by up to 10%, helping ease patient flow within the local ED.3
The burden on emergency medicine nationally is growing year on year, with over 24.9 million A&E attendances in England in 2018–2019.4 Patients are spending longer in the ED, with the number of patients spending more than 4 hours in the department rising to 12% of cases.4 As a result, national guidance and health policies have indicated an urgent need to identify alternative methods of delivering emergency healthcare with a move towards increased community-based care. The NHS Long Term Plan aims to ‘dissolve the historic divide between primary and community health services’ and in doing so reduce the pressure on hospital services.5 The Royal College of Emergency Medicine CARES guidance, focusing on Crowding, Access, Retention, Experience and Safety (CARES), was relaunched during the COVID-19 pandemic. It describes several key areas for improvement: adapting other patient pathways as alternatives to the traditional ED route, reducing waiting times to see a clinician and improving the accuracy of subsequent triage to the most appropriate service.6
It should be noted that there exist a number of community-based emergency care services undertaking operational activities which overlap with some of the remits of CEM services. These include prehospital critical care services, community medical frailty services and general practice enhanced and out-of-hours schemes in addition to enhanced services delivered by several ambulance services such as falls teams and advanced paramedic practitioner models. One such example is the South East Coast Ambulance Service that has developed a Community Paramedic Project,7 where specially trained paramedic practitioners respond to selected 999 calls coupled with improved integration with other primary healthcare providers and full access to patient records. Since the introduction of this scheme, they have reported a 15% reduction in conveyance to local EDs.
The PRU model of CEM is unique in that it incorporates a prehospital practitioner with a team member (doctor or other suitable hospital personnel) who is familiar with hospital processes and pathways; to provide a team with a skill set and knowledge that compliments each other. There does remain some debate regarding CEM services in terms of their effectiveness in reducing strain on the healthcare system and the cost–benefit in doing so.8 However, it has been shown at a local level that a CEM service provides a viable and safe solution for departments to employ.1
There has not yet been national comparison of CEM or PRU services. With several organisations planning the introduction of similar services imminently, it is felt that an evaluation of current national practice is a priority. The objectives of this study were to review CEM and to outline its current provision in the UK and Ireland with identification of priorities for future work in the field of CEM.
The Pre-Hospital trainee Operated Research Network (PHOTON) is an independent trainee-led research group supported by the Faculty of Prehospital Care, with an established national network of doctors and other prehospital emergency care practitioners interested in the delivery of prehospital research. To date, there exists no national directory of CEM services. Therefore, PHOTON identified operational CEM services using a snowball sampling technique.9 Services were initially identified through word of mouth, before expanding service identification through review of pre-existing published literature, social media posts and use of an invitational letter circulated to all members of PHOTON to establish what CEM services existed. Clinical leads for each CEM service were then contacted individually by email and invited to participate and suggest any other known services that should be included.
Pertinent data categories which would allow full appraisal of currently operating CEM services were identified and then further expanded by the study authors in order to develop the data collection tool. This covered six key sections: service aims; staffing and training; job tasking and patient selection; funding and vehicles used; equipment and medication; governance and research activity. This was translated onto an online data collection form in order to facilitate ease of completion (see online supplemental appendix A). A full breakdown of the data points included is detailed in table 1. The questionnaire was circulated to the clinical leads for each service. The data collection tool was live from 17 October 2020 to 17 July 2021 to allow the service leads to obtain the relevant data. Reminders were sent to invited participants in order to maximise response rate.
Within the data collection period for this study, responses were received from all seven known CEM services at that time. Four representing England: London’s PRU, Leicester’s PRU, Lincolnshire’s CEM Service (CEMS) and Northumberland’s X-ray Response Team; two services from Wales: Cardiff’s Cardiff & Vale PRU and Newport’s Gwent PRU and Cork’s Alternative Prehospital Pathway (APP) team from Ireland.
CEM service overview
The first service to come into operation was London’s PRU, which was set up in 2001 to cover the northeast area of inner London. Newport’s PRU was then formed in 2014, serving southeast Wales. The Cork APP team and the Lincolnshire CEMS were both set up in 2019, covering Cork city and the county of Greater Lincolnshire, respectively. Finally, Leicester’s PRU, Northumberland X-ray Response Team and Cardiff & Vale PRU were initiated in 2021 (Table 2).
All services serve a varied population covering both urban and rural communities over an almost full range of deprivation indices.
Service leads were asked to identify the main aims of the care provided. All seven services reported aims that focused around providing enhanced patient care in the prehospital setting. In doing so, they aspire to reduce ambulance conveyances, avoid hospital admission, use alternative patient pathways or refer to other community services.
Five services were able to provide hospital conveyance rates based on internal service evaluations of varying sample sizes which were related to the length of time each service had been in operation. One service was not able to provide these data. ED conveyance rates ranged from 17% to 60%. Two services detailed the additional use of community follow-up services in order to facilitate community-based care (table 3).
Where ED attendance is warranted, two services aim to improve the level of prehospital care given by providing enhanced or critical care interventions such as advanced airway management, trauma management and early delivery of intravenous antibiotics in patients with suspected sepsis. Several services also described aims of improving patient satisfaction and overall patient experience when accessing emergency care. One service indicated an additional operational objective of assisting the ED through provision of community care or follow-up to aid early discharge from the ED.
Staffing and training
All services indicated that their operational teams were staffed with a dual crew of varying background but typically consisting of one hospital clinician and one member of staff from the ambulance service. One service included a radiographer as part of their clinical team. There was a reported range of seniority within these clinical teams, with some describing a consultant/paramedic model, and others deploying a registrar alongside an ambulance technician with senior support available remotely.
In terms of training and supervision, a wide variation in the degree of training and supervision provided to team members prior to commencing work with each service was described. One service detailed providing new recruits with a single induction day and approximately five supervised shifts prior to being allowed to operate independently, while another reported having a 3-month period of supervised shifts and a formal sign-off process (table 4).
Job tasking and patient selection
Tasking was described as being achieved in a variety of ways: all services are tasked by their local Ambulance Control Centre (ACC) based on either predetermined criteria or at dispatcher discretion based on knowledge of the services operations. All but one service accept crew requests and requests from other community teams such as oncology or frailty services. Three services self-task from the ambulance stack of current calls following some form of interaction with the ACC, with one listing this as its predominant method of dispatch. One service also reported accepting referrals from the ED or acute medicine in order to facilitate earlier hospital discharge into the community.
Four of the seven services that responded described attending all categories of emergency call (category 1: life-threatening, category 2: emergency, category 3: urgent and category 4: non-urgent). One service reported only targeting category 3 or 4 calls. Six of the services responded as attending patients from the full spectrum of patient types, with one service not targeting significant trauma or cardiac arrests and one service focusing on frailty and low mechanism falls. Box 1 outlines several case vignettes which act as examples of enhancements to standard patient care in those attended by a CEM team. An anonymised example of a CEM dispatch aid is included in online supplemental file 2. Notably, several services reported avoiding interaction with mental health patients requiring specialist psychiatric input, therefore necessitating an attendance at an ED or Mental Health Unit. Teams would however attend if they could provide enhanced care such as in the context of self-harm or overdose. Teams were asked to indicate whether specific service users were targeted; several mentioned key groups which included frequent service users, palliative care patients and paediatric patients.
Examples of patient interactions*
Example 1: blunt chest trauma older patient
A 77-year-old man; fallen while in garden and injured chest wall.
Assessment performed including point-of-care ultrasound (POCUS) to aid exclusion of pneumothorax.
Blunt chest wall risk score performed and patient given appropriate analgesia, exercises and safety netting.
Example 2: frequent caller—chest pain
A 29-year-old woman identified as a frequent service user presenting with chest pain.
CEM team were able to access hospital records and identify agreed multidisciplinary ED action plan for investigations.
Patient assessed, appropriate investigations performed, patient reassured and discharged at scene.
Example 3: fallen and confused
An 88-year-old woman. Ambulance was rung as carers found patient fallen at lunchtime and seemed more confused over the last few weeks. Patient assessed by CEM team for injuries. Point-of-care renal function/creatine kinase checked and further investigations sent including infection/confusion screen. Fall hazards identified.
Referred onto frailty team for ongoing investigation into long-term confusion deterioration. Falls team to complete review of patient within their own home.
Example 4: acute oncology patient
A 51-year-old woman with a background of breast cancer on second cycle of chemotherapy called the chemotherapy hotline reporting a fever and dysuria.
Referred to the CEM team for review. Found to be unwell with tachycardia, fever and evidence of urinary sepsis. Resuscitative treatment initiated at home with intravenous antibiotics, fluid and paracetamol.
Identification that patient had been neutropenic on recent blood tests. The patient was conveyed directly to a side room on the acute oncology ward, thus bypassing the ED. The patient stabilised clinically en-route to hospital.
Example 5: a good death
An 82-year-old man with a background 6 cm abdominal aortic aneurysm (AAA) known to vascular surgery team and multi-disciplinary team (MDT) decision made that patient is not fit for surgery.
Cardiac arrest in public place. CEM team attended as part of prehospital team. Return of spontaneous circulation (ROSC) achieved and ultrasound revealed large AAA and free fluid in the abdomen.
Following discussion with patient’s family, a decision was made to convey patient to home. Patient taken to own bed, kept comfortable with anticipatory medications, a do-not-resuscitate order in place and an emergency referral for community palliative care follow-up for support.
Example 6: maintaining patient comfort in palliative care
A 79-year-old woman with advanced malignancy presented with acute pain in right arm. Family and carers unable to move patient onto provided hospital bed due to lack of equipment, frailty and severe pain in arm.
CEM team diagnosed pathological humeral head fracture with ultrasound and acute on chronic renal impairment due to reduced oral intake using point-of-care bloods.
Improved patient comfort at home, transferring to hospital bed, providing sling for fracture, pain relief, ensure for nutrition and arranging palliative care follow-up at home.
*The patient care examples have been modified to ensure patient anonymity.
CEM, community emergency medicine.
Vehicles and funding
All seven services receive funding from their corresponding health boards or clinical commissioning groups. Three services receive additional funding, predominantly for the vehicles and ambulance staff, from their local ambulance service. One service also reported receiving charitable funds from their corresponding air ambulance charity.
All seven currently active services operate a vehicle with blue light capability. Depending on the services’ funding structure, the vehicle is provided either directly by themselves, the local ambulance service or is charity funded. Two services reported using their own service vehicle, whereas the other three services reported using a vehicle provided by their local ambulance service. Two services reported the use of multiple vehicles in order to increase operational hours and maximise job capacity, with one service bringing a third vehicle into operation as part of their response to the COVID-19 pandemic.
Medication and equipment carried
Services were asked to provide details of equipment carried in addition to the standard medications and equipment carried by an ambulance vehicle within their local ambulance service. There was notable variation in terms of oral medications carried, such as anti-arrhythmics, anti-emetics, antibiotics and allergy medications (table 5). On the other hand, among the four services that included major trauma and cardiac arrest within their targeted demographics, there was greater consistency in terms of carriage of controlled drugs, with all reporting carrying fentanyl, ketamine, midazolam and rocuronium.
Equipment to facilitate management of major trauma and haemorrhage was almost identical between all services. However, equipment carried which related to provision of advanced cardiac arrest care, including delivery of advanced anaesthesia, was varied. This included mechanical cardiopulmonary resuscitation (CPR) devices, ventilators, analgesic and anaesthetic drugs.
Another key difference observed between the various services related to the carriage of additional electromedical equipment such as point-of-care (POC) testing. Some services reported having access to POC blood analysers (allowing on-scene biochemistry and blood gas analysis), blood ketometers and coagulation testing, whereas other services carry none. Only one service carried a portable X-ray device.
Governance and research activity
Services use either paper worksheets or electronic patient care records to record operational data during shifts. All services reported collecting data which was broadly similar (table 6); however, there are differences both in terms of timing of data collection (typically either immediately following each patient encounter or following completion of the clinical shift) and how the data are then entered and stored within each service’s local database. No service actively links data collected with electronic hospital records although one service describes an attendance notification within the hospital records on a CEM service visit.
Governance structure differs significantly between services. Several services reported having their own standard operating procedures and guidelines in order to benchmark clinical standards, whereas others rely on less formalised guidance based on training delivered at initial induction. The undertaking of clinical review meetings and clinical governance days also varies greatly across the involved services, with varying frequencies of review meetings being conducted, ranging from a daily rapid case review followed by a weekly morbidity and mortality meeting to quarterly clinical governance meetings. Three services reported having named specific paramedic or consultant leads for governance areas within the service. A further three of the services reported that the duty clinical team have access to a ‘top-cover’ consultant during operational hours for clinical advice when necessitated but this may be a consultant within the service or indirectly as a consultant working in the corresponding ED.
In terms of research activity, only three services reported active research activity, with one having a record of published works in peer-reviewed journals as well as a number of national conference presentations. However, all services reported maintaining a comprehensive database of activity from which future work could be developed.
Several services that identify as delivering CEM are operational throughout the UK and Ireland. Those contacted demonstrated several common themes in their practice including their aims and objectives, staffing structures, operational times, patient numbers and non-conveyance rates. Differences identified between services, relating to job tasking and patient selection, equipment and drugs carried, and governance and research activity, may in part be attributed to the categories of call targeted, geographical location and funding structure. While the majority of services respond to all categories of call, services tend to target a predefined portion of the spectrum of acuity, with one service exclusively targeting lower acuity patients. Subsequently, each service carries medication and equipment more tailored towards the typical acuity of patients seen. In addition, services were invited to detail equipment and medications beyond that typically carried within their local ambulance service, which may further explain some of the variance identified. It is also noted that services use a ‘hospital clinician’ from a range of different specialties, with differing skill sets, which may predispose itself to further variation in the care received by patients.
Internationally, many countries use the ‘Franco-German’ model of Emergency Medical Care Systems, employing doctors as part of their ambulance service. This model reports reduced attendance at EDs and improved direction to more appropriate hospital services.10 One such example is the Service d'Aide Médicale Urgente in France, where doctors are included at all stages of the prehospital network from call take, dispatch, assessment and conveyance. They provide senior decision-making, knowledge and skills encompassing a wide range of presentations from those requiring critical care and resuscitation to direction towards more community-based services.11
A more structured definition of CEM, provided by the Faculty of Pre-Hospital Care (FPHC) or the Royal College of Emergency Medicine (RCEM), would unify the operating remit and aims of CEM services. This definition would detail scope of practice and complement the pre-existing Intercollegiate Board for Training in Pre-Hospital Emergency Medicine curriculum. This would prevent ‘mission creep’ into overlapping prehospital critical care services and community ambulance services as well as the work of general practitioners.
This definition could incorporate a more robust set of clinical standards and generation of standardised key performance indicators nationally for CEM. This should include specific outcomes of conveyance rates, further ED or CEM interaction and 30-day mortality. This would allow services to monitor their own outcomes, comparing with other similar services. To facilitate this, a register of practising CEM services brought together by the PHOTON group and its overseeing Faculty of Prehospital Care, would open communication channels for both existing and planned future CEM schemes.
There are several limitations to this study. It is feasible that there are other services, with aims and scope of practice similar to the CEM services included in this study, that were not successfully identified and included by the snowball method for data collection. Even several services that were included use different nomenclature and branding which provides a challenge when identifying other CEM services. In addition, there are other non-CEM prehospital services, such as the advanced paramedic practitioner models,12 that operate with similar aims and scope of practice, that may not identify themselves with the term PRU or CEM as they do not have the unique attribute of using a team made up of a hospital clinician and prehospital practitioner. Future work will also be required to compare CEM services with these other prehospital services to compare costs and clinical effectiveness.
London PRU has previously published their data estimating a net economic annual benefit in excess of £500 000.1 This initial scoping study did not collect data about individual service’s costs and finances. It was felt that there was likely to be a significant degree of heterogeneity between services due to variances in funding structure and geographical area covered. Therefore, a purposeful cost/benefit analysis would be extremely complex to undertake outside the remit of this study. It is acknowledged that this limits future CEM services and healthcare administrators in making decisions about the cost implications surrounding implementation of new CEM services; however, future rounds of data collection from a national register could easily address this.
This work establishes that although there are many similarities between CEM models in operation nationally, particularly centred around service aims and goals, there remains significant variation in practice. Some of the disparities described are inevitable due to variances in geographical location, patient demographics and service funding structure. However, there is a need for national collaboration between CEM services to ensure high standards of care are being delivered which is cost-effective and complements care provided within the ED and by other prehospital services. A centrally derived definition of CEM, detailing its aims, scope of practice and expected clinical standards, as well as robust measuring of agreed CEM outcomes and key performance indicators, would strengthen the work being carried out by services across the UK and Ireland.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplemental information.
Patient consent for publication
We would like to extend our gratitude to all CEM service leads for allowing their data to be included within this study.
Handling editor Mary Dawood
Twitter @LisaRamage, @leechcaroline
Collaborators Pre-Hospital trainee Operated Research Network (PHOTON): David Cookson; Scott Knapp; Nicholas Moore; Andrew Patton; Varsha Rao; Syed Masud; Matthew Bowker.
Contributors AH was involved in study design, dissemination, data interpretation, write-up and revision of the manuscript. LR was involved in study concept, study design, data interpretation, write-up and revision of the manuscript. CL was involved in study design, write-up and revision of the manuscript. The PHOTON study investigators DC, SK, NM, AP, VR, SM and MB were involved in data collation and submission.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.