Seth White graduated from ScHARR MSc in Clinical Research in 2016. His Masters' dissertation project was supported by Rotherham Emergency Department, and looked into reasons underlying increasing demand on the Emergency Department, particularly for less severe and urgent health problems.
In collaboration with his supervisor Jo Coster, and Jeremy Reynard from Rotherham District General Hospital, Seth's research adresses problems facing Emergency Departments throughout the UK and beyond.
Authors and Affiliations
Seth White MB ChB 1,3; Joanne Coster BSc 1; Jeremy Reynard MBBS FCEM 2
1 School of Health and Related Research, University of Sheffield, Sheffield, S1 4DA
2 Emergency Department, Rotherham District General Hospital, Rotherham, S60 2UD
3 Corresponding author: seth.t.white@gmail.com, MD Anderson Cancer Center, 1515 Holcombe, Houston TX, USA 77030, T: +17137455253
KEYWORDS: Emergency demand, qualitative, interview study, non-urgent attenders, inappropriate attenders, ambulatory attendance, Accident and Emergency, managing demand
Abstract
Introduction: A&E demand is rising inexorably across developed countries. Demographic factors and population health do not fully account for this rapid rise. Considerable demand comes from relatively low-urgency patients, who are likely to be discharged without investigation or treatment. Understanding why such patients choose A&E is important when configuring current and future services.
Methods: Semi-structured interviews undertaken at a local district A&E department situated in an urban deprived area. We included A&E attenders triaged to category 4 or 5 on the Manchester Triage Scale that did not present with an acute traumatic injury or were identified as vulnerable. Transcripts were analysed using framework analysis.
Results: Fifteen patients were recruited and all were interviewed. None described their health problem as an emergency. Patients responded along three main themes. 1) Access to and confidence in primary care; participants lacked confidence in their GP and had difficulty getting appointments. There was a belief that GP consultations were futile as they would inevitably be referred to A&E. 2) Pull factors to A&E; relates to how perceptions of need and positive views of A&E (availability of facilities for investigation, speed and perceived quality) influence health seeking behaviour. 3) Individual and external factors; many patients were referred to A&E and NHS 111 was a common source of referral. Individual factors mostly related to anxiety and pain or using A&E as an informed decision.
Conclusion: These findings demonstrate that patients will select A&E in lieu of other available options due to perceived benefits of A&E care and perceived barriers and limitations to primary care. Previous health care experiences (positive and negative) have a strong influence on future health care choices. The NHS 111 number appears to routinely refer low-urgency patients to A&E and this warrants further investigation.
What
this paper adds
What is already known on this subject:
A&E demand is rising rapidly. Some of this demand is from patients who
may not require emergency or urgent care. Why such patients choose A&E is
unclear.
What this paper adds:
Patients attend A&E because they believe it is a
superior service with better access to investigations, better clinicians and
greater ease of access. These patients
are unlikely to choose primary care if given the option. This has implications for service planning.
|
Introduction
Rising demand for Accident and Emergency
(A&E) care is a well-known and much politicized issue. Demand has risen 32% over the past ten years
alone [1] and
is a common problem across developed countries.[2] The factors driving this
increase are not fully understood – population increase and ageing alone do not
account for the magnitude of rises seen. Several interrelated factors have been suggested to explain this increase;
increasingly litigation-averse practices, organizational changes within healthcare
systems and increased community expectations are often cited.[3]
One significant factor, which may be
difficult to assess, is a shift in patient behavior. There is a view that an increasing number of
patients bypass primary care and attend A&E ‘inappropriately’.[4] Many studies have assessed
the ‘appropriateness’ of patient attendance in A&E, asking whether patients
may have been more appropriately treated elsewhere. Such studies (using the judgement of physicians)
have estimated various inappropriate or avoidable attendance rates of between
6-80%.[5]
The concept itself is poorly
defined in the literature, with individual studies often using independently
crafted definitions with little agreement between emergency physicians
themselves.[6] Although the term “inappropriate attender” is difficult
to systematically operationalise and may be unhelpful, it is clear that a
significant subpopulation present with low-acuity problems to A&E departments. Such patients may
contribute to overcrowding and increase strain on health services.[3] Attempts to divert such
patients to other urgent or immediate healthcare services , such as walk-in
centers or other services, have not been shown to reduce A&E demand or influence
patients’ health seeking behaviour.[4 7]
This research was therefore undertaken to
identify and understand motivating factors in health care decision making for low-acuity
patients who choose to attend A&E. We
sought to explore the underlying reasons behind patient decision making and how
previous health care experiences and patient views of health services impact
and shape future decision making.
Methods
Setting
The
research was undertaken at Rotherham District General A&E department
located in England. This A&E department serves the local area
of Rotherham (population: 257,000), which is a relatively ethnically homogenous
area [8] with a high index of social deprivation, being ranked in the bottom
quintile of local authorities nationally.[9] The A&E department is currently
undergoing refurbishment with plans to integrate a primary care service with
the emergency department to form the ‘Emergency Centre’,[10] expected to be completed in 2017.
Participant Selection
Ambulatory adult patients who were triaged
to category four or five on the Manchester Triage Scale (the two lowest
acuities) were eligible for the study.
Patients with recent onset of trauma (<24 hours), were in vulnerable
patient groups or acutely distressed were excluded from the study. These criteria were designed to capture a
less urgent population able to participate in an interview who could feasibly have
accessed treatment elsewhere. Formal prospective
assessment of suitability was not made as this was felt to be too subjective to
be useful. Ethical approval was gained
from Leicester Central research ethics committee, and study site approval was
obtained from NHS Rotherham and HRA prior to the commencement of the study (REC
reference: 16/EM/0205).
Exclusion Criteria
-
Pregnancy
-
Acutely
distressed
-
Mental
health issue
-
Trauma (if
<24hrs or injury visible)
-
Formal
referral (with letter)
-
Inadequate
English
-
Cognitive
impairment
Data Collection
All
data and information was collected within the A&E department. Patients were
screened during triage by the triage nurse, who held a copy of the study eligibility
criteria. Potential participants were
then identified to the principle investigator, and were provided with
participant information. After having time to read the information, participants
were given the opportunity to participate and if they consented, were invited
to a quiet side room where consent was taken in writing. Interviews were semi-structured in nature and
recorded on digital Dictaphone. A topic guide
was used to inform the interview and this was further developed and added to as
the research progressed. Initial topics were
derived from previous research in the UK and abroad and included a brief
chronology of events, perceptions of A&E and primary healthcare services,
past experiences and perceived health needs for this care problem.
A
second stage exclusion criteria was applied as recruitment only occurred when
the wait between triage and being seen was at least one hour, so as not to delay
care to participants. The study was restricted
to regular office hours (9-5 Mon-Fri) to avoid factors around out-of-hours
access, as the study was primary concerned with patients who could feasibly
have chosen to attend a primary care
service. Interviews were carried out during
summer 2016.
Analysis
Interviews were transcribed verbatim by the
lead researcher (SW). Transcripts were analyzed within NVIVO 11 software [11] using
the method of framework analysis, which involves a series of clearly defined,
systematic and sequential steps.[12] Following familiarization
with the interview transcripts, initial themes were coded around basic concepts
from interview responses such as “anxiety” or “perceived clinician quality” to
develop a working analytic framework.
These were then discussed and revised with a second researcher (JC) and
applied to subsequent interview transcripts.
Revisions were undertaken over several meetings. Continuing the same process, interview data
was gradually interpreted and collated into three superordinate themes: “access
to and confidence in primary care”, “pull factors to A&E”, and a third,
looser superordinate theme “individual and external factors”.
Results
Participant characteristics
Fifteen participants were recruited; participants ranged in age from 18 to 63 and included seven women and eight men with a broad variety of presenting complaints (see Table 1). Six participants gave home addresses outside of the catchment area of Rotherham A&E. Disposal outcome was only available for thirteen participants – one participant’s outcomes could not be traced and another was recorded only as ‘seen by GP’. Four participants brought a relative into the interview (no carers were interviewed) but only three contributed to the interviews.Out of the thirteen participants with disposal outcomes available, five received advice only, and three received advice and a prescription. Only one participant received a specialty referral and no participants were admitted. One participant did not wait to be seen by a clinician after the interview. Two received wound care in the department, both judged to be candidates for wound care in general practice by the researcher. Participants came from a geographically diverse area and nearly all participants resided in areas with high levels of deprivation. Six participants had home addresses listed with an A&E in closer geographical proximity (although location of incident/onset of illness was not recorded). Three participants reported travelling to Rotherham in preference of another A&E during the interview (with one being advised to go to specifically Rotherham via 111 due to presence of maxillofacial surgeons – though they were not seen by the maxillofacial surgeons).
Table 1 - participant characteristics
ID#
|
Age
|
Sex
|
Complaint
|
Outcome
|
Deprivation quintile
|
Closest ED
|
1
|
63
|
M
|
Growth on nose
|
Seen by GP
|
5th
|
Rotherham
|
2
|
57
|
M
|
R leg problem
|
Advice
|
5th
|
Mexborough (MIC)
|
3
|
41
|
M
|
L side numbness
|
Advice, prescription
|
4th
|
Rotherham
|
4
|
18
|
M
|
Injury to R foot
|
Dressing, local anaesthetic, advice
|
3rd
|
Rotherham
|
5
|
54
|
M
|
Pain L knee
|
Prescription, advice
|
4th
|
Rotherham
|
6
|
27
|
F
|
Dental problem
|
Prescription, advice
|
5th
|
Barnsley
|
7
|
41
|
F
|
Knee prob
|
Outcome unavailable
|
||
8
|
52
|
F
|
Left leg pain
|
Did not wait
|
Sheffield
|
|
9
|
32
|
M
|
R foot injury
|
Advice
|
5th
|
Rotherham
|
10
|
32
|
M
|
R leg injury
|
Advice
|
5th
|
Rotherham
|
11
|
21
|
F
|
Left knee injury
|
Observation, specialty referral, advice
|
3rd
|
Mexborough (MIC)
|
12
|
30
|
F
|
Groin pain
|
Observation, advice
|
3rd
|
Doncaster
|
13
|
26
|
F
|
R foot wound prob
|
Wound cleaning, advice
|
3rd
|
Bassetlaw
|
14
|
51
|
M
|
RTC/neck pain
|
Advice
|
2nd
|
Rotherham
|
15
|
55
|
F
|
R wrist injury
|
Advice
|
5th
|
Rotherham
|
MIC = minor injury
centre. Closest ED – closest emergency department or nurse led minor injuries
unit (walk-in centres not included). DQ
= deprivation quintile
Thematic analysis
After applying several revisions to the
coding structure, a definitive thematic framework emerged describing the
factors behind A&E attendance. Twenty-one
individual subthemes were identified during the initial coding and many of
these were closely related or interlinked. All respondents contributed to at
least one main theme and often more. Three overarching themes were identified
and each contained subthemes: access to and confidence in primary care, pull
factors to A&E, and individual and external factors.
Bypassing Primary Care
Access
to and confidence in primary care related very
closely to pull factors to A&E.
A&E was almost universally better regarded than primary care, even in
participants who voiced positive opinions of their general practitioner. Many participants dismissed the idea of
attending primary care as ‘futile’ and were attracted to A&E as able to
provide more definitive care. Many
participants also believed that if they attended their GP it was merely
delaying the inevitable A&E attendance and referred to ‘cutting out the
middle man’ when describing their reasoning behind this.
“Just because it was something that I didn’t know what it was, I thought I’d cut the middle man out and go straight to A&E” – Participant 2
“I just thought that I'd just cut the middle man out by coming here for X-ray because I thought I would need one” – Participant 5
"Whenever I've gone to the walk in or phoned 111 they tell me A&E straight away, or that you need to be seen within the next 4 hours…….so now when I get the pains I jump in a taxi and I come here.” – Participant 3
“It’s just that every time we’ve been to see the GP it’s that they’ve more or less told him that they don’t specialize in that kind of thing so we go to A&E and you’ll get better treatment.” – Participant 1
This closely related to three subthemes: 1)
Perceived need for diagnostic equipment, 2) Difficulty in accessing timely
primary care, and 3) Lack of confidence in primary care.
Perceived need for diagnostic services was
an important factor in many decisions and a very prominent theme throughout the
interviews. Radiographs were commonly
thought to be required. No other
specific investigations were listed.
“So I was going to ring the doctor and then I thought they might have to do an x-ray so I thought I’d come here and get it checked out as soon as possible” – Participant 9
“I just thought that by coming here I'd get an X-ray” – Participant 5
Difficulty in accessing the general
practitioner was also very common theme in the research – some participants
spoke of long waits, and first-come-first-serve telephone appointments as
barriers to GP attendance. It is worth
noting that this was still a prominent theme despite all the research being
conducted during normal weekday working hours.
“The thing is that it’s really hard to get an appointment with them they’ve got a system where you need to ring at bang on eight in the morning and it’s either one or half one in the afternoon to get an afternoon appointment. But if you try to ring at eight… I’ve had it before where I’ve tried to get an appointment and it’s three [minutes] past eight and they’ve all gone” – Participant 11
Confidence in general practice was mixed
overall, with several participants speaking of previous issues or disagreements
that led them to seek alternate care in A&E.
“I’ll tell you when I got there I couldn’t breathe, I’ve got COPD, and the thing he said to me that the best thing to do was to go up to the moors, park car, and take a few deep breaths. And that night, ended up in hospital with a blood clot on the lung and pneumonia. That’s how good he [GP] were.” – Participant 1
Although opinions
of GP ability were mixed, there was generally a greater confidence in emergency
physicians.
“I do like hospital because of all the equipment they have, everything right here. But when you come here it's like they know what they're doing more than other places” – Participant 4
Choice and External Influences on Agency
Individual
and external factors is a looser association of
themes. This represents factors in the more immediate decision making process.
Anxiety and perceived severity of illness was cited by several participants,
and was an expected finding based on previous research. It was however, less prominent than
expected. Some patients expressed
uncertainty.
“I’m a little bit worried – do I need seeing to straight away?” – Participant 2
“Well I’m not quite sure… I don’t know 100% whether or not I should be in A&E. I don’t know if I qualify to be in A&E with my injury. Well I don’t know it’s not really an emergency. It is an accident though! But not an emergency. So I don’t know whether it was right to come to A&E” – Participant 13
A number had been referred to go to A&E
by 111 and a similar number by friends, colleagues or informally by other
medical professionals. Interestingly a small but significant minority made an
informed choice based on convenience related factors, resulting in travelling not
insignificant distances past other A&E departments to reach their preferred
A&E department.
“Well yeah, I do actually live in Sheffield. I don't like the northern general it's always too busy… and it's a nightmare parking at the northern general isn't it?” – Participant 7
One participant stated that Rotherham
A&E was their usual place of medical care. Participants were aware of
pressure on A&E departments and often had relatively detailed knowledge of locally
available services.
Discussion
Implications for service delivery
This research builds on previous research
into patient decision-making in seeking A&E care and explores why patients in
a deprived population setting attend A&E in lieu of other health care
services for low urgency problems in-hours.
Patients were unanimous in their perception that A&E is the best
place to attend for ease of access and availability of diagnostic
equipment. Patients perceived A&E as
a more ‘definitive’ option for care, implying that general practice was unable
to decisively attend to their complaints.
This relates to the concept of convenience, which has strongly occurred
throughout the literature.[3 13] What was distinct in this
research was the concept of futility
associated with GP attendance; many patients believed that they would have
ultimately been referred to hospital to either A&E or for investigations
available in A&E.
Anxiety and perceived severity were less
frequent complaints than expected; participants did not believe their problems were
emergencies. External sources of advice
appear to have played a part; some patients cited discussions with healthcare
professionals informally, or the 111 service.
It is clear from the responses within the
study that A&E attendance in low-urgency patients is largely a logical and
considered choice, as it meets their perceived health needs with convenience. Although reasons for attending A&E varied
between participants, it was evident that most respondents had considered their
options before attending.
In this sense A&E may be perceived to
be ‘a victim of its own success’.[14] Confidence and positive
perceptions of the service may be increasing footfall through the service and
inadvertently decreasing use of other, more conventional means of non-urgent
healthcare for reasons that go beyond simple indecision/anxiety and access
issues. Public satisfaction with General
Practice, although high, has been on a downward trend for several years.[15] It is possible that changing
perceptions between services are playing a role in demand.
Given that, from a patient perspective,
A&E is a logical and convenient choice, the very concept of “appropriateness”
around attendance is unhelpful as it suggests
patients are expected to make a decision on their care which physicians themselves
sometimes find difficult to explain, as discussed previously.[6] Furthermore, the very logic
of approaching the problem as one of “appropriateness” is fundamentally flawed. It is unreasonable to place the burden of
decision of choice on patients with no medical expertise and can lead to inequity
within the healthcare system. Such
opinions have been raised before,[14] emphasizing the need for a patient-centered approach to health
service design.
This research also helps explain why the extension
of primary care options has not demonstrably impacted demand.[7] Availability of primary care
may have some influence, but a majority appear to be deliberately seeking
A&E care for the clear benefits it brings.
The role of 111 in referring patients unnecessarily
may need to be investigated further as 111 was a common source of referral of
low urgency patients to A&E. As 111
was introduced to provide advice to self-care and encourage alternatives to
A&E, this is a counterproductive outcome.
This is consistent with available evidence which does not demonstrate
any reduction in A&E demand with 111 rollout.[16]
As many A&E departments integrate
primary care services into emergency departments this research demonstrates the
logic of this, from a patient perspective.
A unified triage service between primary and emergency services may be necessary
as patients may, for various reasons, select emergency care in preference if
offered a choice. Future research into the
integration of primary care and emergency care services and research exploring
the organizational responsibilities and remit of each service within a shared
system is required.
Limitations
Though the study appears to have captured a
largely low-urgency group of patients, it is not certain how generalizable
these answers may be to the broader population.
Certain populations, specifically vulnerable groups such as pregnant
women and children, were excluded on ethical grounds and may have important
differences in their presentation factors which were not assessed here. Children may be an especially important source
of low-urgency demand, as has been suggested by others.[17] Furthermore, this is a
single-centre study and the population captured here may not be generalizable
to the wider population, given the particular demographic features of this setting.
Finally, the findings must be interpreted
in the context where they were taken.
The principle investigator is a physician who has previously worked in
the department. Participants were made
aware of this fact and may have spoken of A&E in a more positive light than
they would have otherwise. The authority
of the investigator may also have placed additional pressure on participants to
give socially desirable answers.
Conclusion
Low urgency patients often choose A&E
due to perceived benefits in clinician quality, access and available diagnostic
equipment as well as other external influences and patient factors. Such patients may actively be choosing to
come to A&E for its perceived benefits and may not be dissuaded by the
offer of parallel services. This
research supports the logic of shared triage in the context of a joint primary
care-emergency service. The research
also implies difficulty in applying the Keogh report, as low urgency patients
may be less likely to seek care outside of an A&E setting.
Disclaimer
This
study has received no funding and was undertaken in the context of a master’s
dissertation by Seth White supervised by Jo Coster at the University of Sheffield. There are no conflicts of interest.
References
1. HSCIC. Accident and Emergency Attendances in England. 2001-2015
2. Pines JM, Hilton JA, Weber EJ, et al. International perspectives on emergency department crowding. Academic Emergency Medicine 2011;18(12):1358-70
3. Lowthian JA, Curtis AJ, Cameron PA, Stoelwinder JU, Cooke MW, McNeil JJ. Systematic review of trends in emergency department attendances: an Australian perspective. Emergency Medicine Journal 2010:emj. 2010.099226
4. Benger J, Jones V. Why are we here? A study of patient actions prior to emergency hospital admission. Emergency Medicine Journal 2008;25(7):424-27
5. Murphy AW. 'Inappropriate'attenders at accident and emergency departments I: definition, incidence and reasons for attendance. Family Practice 1998;15(1):23-32
6. Richardson S, Ardagh M, Hider P. New Zealand health professionals do not agree about what defines appropriate attendance at an emergency department. The New Zealand Medical Journal (Online) 2006;119(1232)
7. Ismail SA, Gibbons DC, Gnani S. Reducing inappropriate accident and emergency department attendances. Br J Gen Pract 2013;63(617):e813-e20
8. ONS. National Census, 2011.
9. Department for Communities and Local Government. Indicies of Multiple Deprivation, 2015.
10. Rotherham NHS Foundation Trust. Rotherham Emergency Centre - About Us. Secondary Rotherham Emergency Centre - About Us 2016.
11. NVIVO 11 for Windows [program], 2016.
12. Ritchie J, Spencer L. Qualitative data analysis for applied policy research. The qualitative researcher’s companion 2002;573:305-29
13. Agarwal S, Banerjee J, Baker R, et al. Potentially avoidable emergency department attendance: interview study of patients' reasons for attendance. Emergency Medicine Journal 2012;29(12):e3-e3
14. Khaldi AWA. The NHS gets the A&E demand it deserves. Secondary The NHS gets the A&E demand it deserves 2014. https://www.hsj.co.uk/sectors/commissioning/the-nhs-gets-the-ae-demand-it-deserves/5070615.article.
15. Fund TKs. Public Satisfaction with the NHS in 2015: Results and trends from the British Social Attitudes survey: The King's fund, 2015.
16. Turner J, O'Cathain A, Knowles E, Nicholl J. Impact of the urgent care telephone service NHS 111 pilot sites: a controlled before and after study. BMJ open 2013;3(11):e003451
17. Prince M, Worth C. A study of ‘inappropriate’attendances to a paediatric accident and emergency department. Journal of Public Health 1992;14(2):177-82
