Unplanned, urgent and emergency care: what are the roles that EMS plays in providing for older people with dementia? An integrative review of policy, professional recommendations and evidence

Objective To synthesise the existing literature on the roles that emergency medical services (EMS) play in unplanned, urgent and emergency care for older people with dementia (OPWD), to define these roles, understand the strength of current research and to identify where the focus of future research should lie. Design An integrative review of the synthesised reports, briefings, professional recommendations and evidence. English-language articles were included if they made any reference to the role of EMS in the urgent or emergency care of OPWD. Preparatory scoping and qualitative work with frontline ambulance and primary care staff and carers of OPWD informed our review question and subsequent synthesis. Results Seventeen literature sources were included. Over half were from the grey literature. There was no research that directly addressed the review question. There was evidence in reports, briefings and professional recommendations of EMS addressing some of the issues they face in caring for OPWD. Three roles of EMS could be drawn out of the literature: emergency transport, assess and manage and a ‘last resort’ or safety net role. Conclusions The use of EMS by OPWD is not well understood, although the literature reviewed demonstrated a concern for this group and awareness that services are not optimum. Research in dementia care should consider the role that EMS plays, particularly if considering crises, urgent care responses and transitions between care settings. EMS research into new ways of working, training or extended paramedical roles should consider specific needs and challenges of responding to people with dementia.


INTRODUCTION
It is reported that emergency ambulance crews frequently encounter older people with dementia (OPWD) and it can be difficult to take history, assess pain and access suitable alternatives to the emergency department (ED), especially out of hours. In 2009-2010, the cost of urgent and emergency ambulance services in England was £1.5 billion. 1 Although 999 emergency call-outs are associated with only 2% of unscheduled care activity, they are thought to be responsible for about 22% of commissioning costs. 2 With the present landscape of financial austerity, the recent NHS England Urgent and Emergency Care Review 3 and the national focus on dementia care 4 it is pertinent to ask what role ambulance services or emergency medical services (EMS) play in the urgent and emergency care of OPWD.
Currently, around 670 000 people in England are estimated to be living with dementia. 5 Research in the last 5 years has begun to consider the impact of having a dementia on the emergency use of hospital services: A US study found that patients arriving in the ED by EMS have higher prevalence of cognitive impairment (CI) than those arriving by other means (13% vs 8% p<0.01). 6 The presence of a dementia diagnosis also means that an older person is more likely to be admitted to hospital as an emergency than those who do not have a dementia diagnosis. 7 Other studies have shown that OPWD admitted to hospital have longer length of stay and higher mortality than patients without dementia. 8 9 Analysis of one UK hospital's emergency admissions data found that people with a dementia diagnosis were more likely to arrive at the ED via ambulance than those without a dementia diagnosis: 82% versus 54% OR 3.73 (95% CI 3.49 to 3.99) (Unpublished data, Okereke U, CLAHRC-CP Fellowship Report, 2011).
There is a body of research that explores extended roles for paramedics/EMS personnel that go beyond the traditional 'emergency' care roles to urgent and even primary care roles. 10 11 There is another body of EMS research around falls, for example, the SAFER 2 study. 12 OPWD are at high risk of falling 13 14 and have a similar number of chronic medical conditions as older people without dementia. 15 However, to date, none of this research has specifically looked at the needs of OPWD.
This review seeks to synthesise the existing literature on the roles EMS play in unplanned, urgent and emergency care for OPWD, to define these roles, to understand the strength of current research and to identify where the focus of future research should lie.

METHODS
We predicted that the literature would be sparse and diverse and used integrative review methodology which critiques and synthesises the literature on a topic, allowing for a combination of diverse methodologies and literature types. 16 17 Our aims were to bring an evidence-informed commentary on how EMS respond to OPWD and derive a research agenda that logically flows from our synthesis.

Preparatory phase
To inform our review question and subsequent synthesis, MB and JF held a workshop with frontline ambulance and primary care staff and carers of OPWD. MB held discussions with an Alzheimer's Society carers' group on use of 999 ambulance services. In both cases a summary was written up and sent to group participants and any disagreements or comments invited. Participants responded that the summaries represented their views and a few additional comments were received.
Workshop participants were asked to put forward, from their experience, reasons for a 999 ambulance call-out to OPWD. Group discussion agreed that there were three overlapping categories: clinical reasons, lack of information and carers not coping.
The visit to the Alzheimer's Society carers' group gave a very different perspective. Most of the participants had not yet had the experience of calling 999 and the focus of discussion was largely on information needs: carers' sense that they were illprepared but wanted to prepare for this eventuality, surprise or frustration at the barriers to healthcare professionals having access to the right information, and the resultant worry about the additional onus of responsibility this placed on carers; the need to brief professionals about the OPWD's medical history.
This engagement with members of the public early in the research process shaped the review question. Findings from the review were then fed back to patient and public representatives to further ensure our review findings had resonance with the experience of frontline staff and carers. 18

Literature search
We used systematic and iterative search techniques. All databases available via NHS Evidence were searched (MEDLINE, EMBASE, CINAL, Psych INFO, AMED and HMIC) using terms to encompass the context-prehospital urgent or emergency care-and the condition-dementia. End of life (EoL) was used as a search term alongside dementia. We knew from recent publications and discussions with paramedical colleagues that there might be relevant observations about EoL care for OPWD in prehospital EoL literature (table 1). Iterative search techniques included recommendations from colleagues, forward citation searching, searching references of selected papers and relevant internet sites or email newsletters.
Articles were included if their scope covered the prehospital urgent or emergency care of OPWD where EMS could have a role. We excluded any roles in patient hospital discharge. No time limit was imposed, but the earliest source was from 1992. Only studies that had, at least, an abstract in English were reviewed. We did not restrict the type of literature included, thus included national reports, briefings, guidance and internal reports alongside articles and studies from peer-reviewed journals. We wanted to understand the evidence and narrative around EMS roles.
We identified 5005 documents from the systematic search, screened 714 abstracts and included 108 for fuller review. Of these 108, 19 sources were included for analysis. Each document source (title or abstract or full paper) was independently reviewed by two reviewers at each stage in the systematic search. At abstract review, a third experienced systematic reviewer reviewed disagreements and made a decision. A further 14 documents were identified from iterative techniques and at least two authors reviewed for inclusion. These 33 sources were included for an initial analysis and 17 documents were included in our final synthesis (figure 1).

Information extraction and initial analysis
We extracted information from the initial 33 sources into a data extraction form (DEF); in addition to recording standard descriptive details we included information under four headings that formed the start of our analysis: lens, ambulance role descriptor, ambulance role components and 'nuggets'. 19 Lens reflected the professional background or assumptions of the authors, for example, EMS or palliative care research. Ambulance role descriptor captured the role of EMS described and ambulance role component was more specific about what EMS were actually doing. 'Nuggets' captured particular ideas or quotes from the sources that seemed most relevant.
After this initial analysis, 16 sources were excluded because a second review by the team clarified that, despite being of related interest, they did not strictly meet our review inclusion criteria. It became clear we had included some sources inappropriately, for instance, only because they resonated with paramedical experience of dealing with people with dementia in their work, or gave weight to the argument that ambulance services are 'forgotten' in research, even though the source itself did not specifically deal with dementia or could not be expected to include EMS. Reasons for initial interest and subsequent exclusion were noted in the DEF.
Emerging roles from the descriptor and component fields were distilled into a smaller set of roles, and the 'Lens' field was further honed into four broad perspectives (box 1).
Subsequent thematic analysis of the key points from each source defined and refined the key issues to emerge from the literature. After cross-checking these with themes that emerged from our preparatory consultation and examination by the full review team, the validity of our findings was discussed and confirmed at a postreview stakeholder meeting of carers and relevant professionals.

Quality appraisal
We applied an informal critical appraisal, influenced by the Critical Appraisal Skills Programme 20 approach. Many of the sources were reports, briefings and guidance, which were not expected to meet any of the agreed quality appraisal criteria. Given the paucity of any information on our review topic, we aimed for inclusivity and agreed not to exclude any documents Table 1 Search terms used in the systematic search CONTEXT: prehospital, emergency or urgent care, that is, where we would expect the ambulance service to have a role CONDITION: OPWD/cognitive impairment/End-of-life urgent adj3 care OR 'urgent-care' OR urgent care .ti,ab emergenc*.ti,ab prehospital* OR pre-hospital.ti,ab ambulance*.ti,ab ambulance OR "emergency medical" adj3 service* ((patient* adj3 transport*)).ti,ab paramedic*.ti,ab ((((emergency OR "emergency care" OR "emergency medical" OR ambulanc*) adj3 (crew* OR staff OR practitioner OR team* OR technician)))).ti,ab care adj3 pathway*.ti,ab dement*.ti,ab alzheimer*.ti,ab ((Cognit* AND (disord* OR impair* OR declin* OR function*))).ti,ab ((memory adj2 (complaint* OR impair* OR problem* OR loss*))).ti, ab ("End of Life" OR end-of-life).ti.ab OPWD, older people with dementia. on quality appraisal grounds. A field in the DEF captured this. (The DEF is available, on request, from the authors).

RESULTS AND SYNTHESIS Description of evidence
The 17 literature sources included are summarised in tables 2 and 3. Over half (n=10) were from policy and professional literature (table 2) of which only three were found via the systematic search. The evidence base in this grey literature included case studies, internal service evaluations, workshop reports, best practice guidance and briefing statements. Work, based on one small-scale staff survey and frontline experience, undertaken by Great Western Ambulance Service aimed at improving prehospital care of OPWD and their carers, is the basis of the evidence in three sources: the original report, 21 a workshop 22 and a briefing 23 by the UK Ambulance Service Network (ASN i ). There was one non-UK grey literature source, a dementia training resource for Australian EMS personnel. 24 The remaining seven documents were from journals (five of which were from the systematic search) and included a literature review, an information/training article, three US evaluative studies around training, knowledge, triage of patients ( psychiatric, distressed senior, Alzheimer's) and two English qualitative studies of EoL care for people with dementia (table 3).
Following analysis, the literature could be divided into four role categories: emergency transport, assess and manage, last resort/safety net and EMS role not considered. In the synthesis, 'emergency transport' and 'assess and manage' roles are discussed together as 'emergency transport' was not found as a stand-alone role in any of the literature (box 2).

Synthesis 'Emergency transport' and 'assess and manage' roles
Emergency transport is considered in two of the documents and the 'assess and manage' role considered in 14. The majority of the literature for this review is from 2009 to 2012 and linked to the developing policy focus on considering and planning for the impact of dementia on modern society. 4 However, there was one much earlier source from 1992, which was a small US-based evaluation of a training programme to teach prehospital personnel management of patients with Alzheimer's disease. 25 Three issues were identified from research around aspects of care carried out by EMS practitioners that relate to their role in the care of OPWD: ▸ Ambulance crews do not have access to validated pain assessment tools for OPWD. 26 The review author noted, 'Reference to cognitive impairment and the consequent impact this condition has on the paramedic's ability to assess pain is rarely mentioned in the paramedic literature.' ▸ There is limited evidence of under triage for patients with dementia by emergency medical technicians using a triage protocol. This study highlighted the difficulty of assessing patients with dementia and that the, then current (2001), protocols do not specifically address dementia. 27 ▸ Anxiety is the most common diagnosis in the 'distressed senior' and most transported to hospital. 28 Work undertaken by English ambulance services and a dementia training resource for Australian ambulance workers indicate that there is recognition within EMS for the need to provide improved care to OPWD. However, the majority of the literature is setting out probable problems and there is a lack of any evidence and tested interventions to address them.
Two sources with an EMS perspective address wider roles that ambulance services can have, beyond emergency transport, consider OPWD as a particular patient group for whom the EMS may have a significant role: 'Seeing ambulance services in a different light: More than a patient transport service' 23 and 'The route to success in end of life care: achieving quality in ambulance services'. 29 The latter report highlights the ongoing role EMS may have at the EoL, particularly for those with chronic obstructive pulmonary disease or dementia; 'Ambulance services and staff have a more significant role and contribution during this period [end-of-life] than might be first thought. People at the end of life may have contact with ambulance services on several occasions, for example when a complication occurs which creates a sudden health crisis, or for an unrelated event such as a fall.' 29 Two English 'Best Practice Guidance' documents that focus on the urgent and emergency care of older people draw out ambulance service roles around appropriate assessment and management of OPWD. Neither source directly explores the role of EMS in the care of OPWD because dementia, delirium and ambulance roles are discussed in isolation. However, both do highlight the 'on-scene' role EMS have in attending the patient at home or in the community. The 2007 UK Department of Health guidance 30 recommends that ambulance clinicians should routinely assess for confusion using a cognitive assessment instrument. They also note the 'unique position' ambulance clinicians have in attending patients in their own home and state, 'Often no other healthcare professional is made aware that the ambulance service has attended this group of patients [here referring to patients who have fallen, a significant proportion of whom may have dementia] and there is therefore a missed opportunity to use the valuable information collected about the person and the fall.' 30 The 2011 'Silver Book' 31 suggests that there 'is scope for delivering a more specialist programme in geriatric medicine for ambulance clinicians.' The EMS role as a 'last resort' or 'safety net' is considered in seven of the documents; however, it is mentioned in passing rather than with substantial exploration of the role and the evidence base for the statements made is not clear. For example: ▸ The briefing paper from the ASN states that 'Patients find the current system for accessing NHS services confusing and frustrating…Many people therefore dial 999 for non-emergencies because they know that they will receive advice or a response'. 23 In the specific dementia-related case study the authors describe patients with dementia being taken to hospital when a care home does not have the capacity to make a proper assessment-even if the person does not have an urgent or emergency care need. ▸ In presentations from the ASN dementia care workshop, 22     To review and convey an awareness of the initial assessment of seniors with diverse psychiatric emergencies Quantified the psychiatric emergencies for a cohort of patients (dementia being a factor in some of these emergencies). Identifies that care givers (of senior patients) often rely on EMS as initial responder to meet their crisis. Highlights role of appropriate assessment by EMS and that they are often the first call in a crisis (a safety net).
▸ Two US papers make mention of the safety net or last resort role; 'EMS frequently provide the initial evaluation of the "distressed senior"' 28 and 'Calling 911 is the first option or last resort for families [of OPWD] in crisis'. 32 ▸ The report considering the role of EMS in EoL care 29 refers to anxious carers and family members calling 999 for emergency support if there are worsening symptoms or a sudden crisis. It does not frame this in an 'inappropriate call' or 'last resort' context as some of the other sources do but rather considers how best to improve the response. An English qualitative study investigating barriers to providing EoL care for OPWD provides a stronger evidence base for this last resort/safety net role: Dening et al 33 highlighted the issue of ambulance services responding 'out-of-hours' when other, perhaps more appropriate services, are unavailable, leading to hospital admission due to lack of alternatives. Another English study dealing with EoL care for OPWD highlighted an incident at a care home where 999 would have been used as a safety net had the general practitioner not been on site in that particular case. 34 However, this is one case study and an interpretation by the author of what may have happened.

Role of EMS in urgent and emergency care of OPWD not considered
The role of EMS was not considered in one particular document and we found the absence notable. The report from a discussion workshop 'Understanding out of Hospital Dementia Care' describes a system 'not fit for purpose' where 'Patients and carers are left to navigate their way through a complex and often baffling health and social care system. This can lead to the all-too familiar outcome of patients ending up in inappropriate care settings…'. 35 This resonates with the last resort/safety net role of other documents, but it was striking that a document about out-of-hospital care did not once consider the role of EMS.

LIMITATIONS
This international review identified literature sources only from the UK, US and Australian EMS. All the grey literature retrieved, with the exception of the dementia training resource for ambulance workers, 24 is UK based and the unpublished reports are all from the East of England, reflecting the authors' local access. As there was a small number of different research settings identified in the literature, we are uncertain how generalisations can be made to other contexts. However, there are sufficient similarities between the UK, US and Australian EMS to provide a useful contribution to understanding the role of EMS in dementia care in these countries. As we did not include any non-English-language literature we cannot be sure how useful it may be to EMS systems in other countries, though the roles identified may serve as a useful starting point for research.
Systematic searching may have excluded articles where the role of EMS is only mentioned in the full text. We tried to guard against this by including papers at abstract that looked like there may be an ambulance service role; however, some papers may not have been identified by the prehospital search terms. It is highly unlikely that this excluded any significantly different or new roles from those already found.
There was a large degree of heterogeneity between the included studies and a general lack of research investigating the role of EMS in dementia care, so some residual publication bias may be present in the review.
The paucity of research made it difficult to make judgements on what the basis of the evidence was for many of the statements made. Formal quality assessment was not carried out as this review does not attempt to synthesise the empirical findings but rather provides an overview of the current narratives and research.

DISCUSSION
The use of EMS by OPWD is not well understood. The literature reviewed demonstrated a concern for this group, need for training for EMS staff and awareness that current services are not optimal. There were notable omissions in the literature reviewed initially, for example, in 'Understanding out of hospital dementia care' the role of EMS was overlooked. This omission was also clear in some of the literature we removed in the last round of exclusions. The 2011 'Joint statement on the emergency care of older people' 36 asserts, 'It is essential to provide the best care for frail older people at the first opportunity [our emphasis] to ensure optimal efficiency of the health and social care system'. Although speaking about older people in general, not just those with dementia, the statement does not mention ambulance services; instead, the 'first opportunity' appears to be at the hospital door. A review of the definition of crisis in dementia care 37 did not select papers that reported crisis and use of EMS, though conversely EMS work describes crisis in dementia care as a trigger for calling EMS. This is very likely due to the inclusion criteria for the review which required the paper to define crisis. There was a common narrative (in both UK-based and US-based literature) describing EMS taking on a last resort or safety net-type role, but this is not well defined and we found no empirical evidence to support this view.
The 'emergency transport' role was not as evident in the literature as we were initially led to believe: the strong narrative for this came from the briefing 'Seeing ambulance services in a different light: more than a patient transport service' 23 and from paramedical reviewers. However, we excluded research that only included EMS calls as outcome measures 38 or co-variables. 39 For example, a paper exploring the prevalence of depression and CI in older adult ED patients that examined differences between patients arriving by EMS and by other means.
This scoping work has highlighted a populist narrative around 'inappropriate' calls (the last resort/safety net role) and identified recurrent themes about crews needing better training in assessing OPWD, the absence of an alternative to hospital, the lack of integration in healthcare and that 999 (or 911 or equivalent) is the default option due to lack of knowledge or access to more appropriate services. Few of these assumptions have been tested empirically. The review revealed literature of varying quality, focus and depth. There is an emerging consensus about the lack of information, information sharing, service integration and missed opportunities of ambulance crews in attending to people at home. Yet, these issues have not been studied with respect to OPWD who have additional care needs, not least co-morbidities that may exacerbate the current challenges. 40 There are high levels of patient satisfaction and trust in NHS ambulance staff; 41 they have a duty to respond and they offer one-to-one patient care. In the absence of evidence, emergency services and EMS personnel will respond to the perceived needs with training, pathways development, use of nonvalidated but 'best fit' tools (such as the Abbey Pain Scale for pain management). There is an increasing interest internationally in extended paramedical roles and skills. 3 10 11 A recent review looking at which extended paramedical skills are making an impact 10 found that the strongest evidence base for paramedical skill development was in assessing and managing acute minor conditions in older patients and found the most promising means of reducing demand on EDs to be interaction of paramedics with other community services. Although Evans, McGovern et al concluded that the evidence was not strong enough to guide policy, their findings point towards redesign or reconfiguration of paramedical practice and the role of EMS to encompass both 'assess and manage' and 'last resort/safety net' roles. Our review would add that when doing this specifically considering the additional needs of OPWD is essential.

CONCLUSIONS AND RECOMMENDATIONS
There is no common framework to understand or interpret how emergency care for OPWD should be provided or how effectiveness can be judged. There is an urgent need to address these issues, which will become more acute, given forecast demographics. In clinical practice, there are ever-increasing numbers of service development initiatives that tend to focus on the 'assess and manage roles' attempting to go beyond, or avoid, the 'emergency transport' role and robust evaluation of their effectiveness should be a priority. Our review found insufficient evidence to quantify the extent to which EMS acts in the different roles identified and was unable to establish how ambulance responses meet the needs of OPWD-potential questions for future research. If the last resort/safety net role of EMS is reframed as rapid response to urgent care needs for OPWD, perhaps, this is an appropriate EMS role and interventions to address these needs could be tested.
Research in dementia care should consider the role EMS play, particularly if considering crises, urgent care responses and transitions between care settings. EMS research into new ways of working, training or extended paramedical roles should consider specifically the challenges and needs of responding to people with dementia. A mixed methods project combing qualitative interviews and surveys could be conducted to further develop our understanding the role of EMS in dementia care. Findings of this research could contribute to the development of a framework to improve emergency care and adjudicate on EMS effectiveness.

Emergency Transport and Assess and Manage
When the EMS, emergency medical services (EMS) role is described within their expected roles, either as a means of getting a person to medical help as quickly as possible, usually hospital, or as making an assessment of a patient, deciding on what immediate management is appropriate and providing treatment, onward referral or conveyance to hospital.

Last Resort/Safety Net
When the role is described as an EMS crew responding outside of their expected or appropriate remit to a situation that is not a medical emergency, though it may be a crisis or urgent care need.

EMS role not considered
When the EMS role is not described and in that context the omission is notable.