 
|
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
07-May-2004, Vol 117 No 1193
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Frequent attenders at Christchurch Hospital’s Emergency
Department: a 4-year study of attendance patterns
Debra Kennedy and Michael Ardagh
Frequent attenders are common in Emergency Departments
(EDs). They make a disproportionate number of
visits1–3 and a previous study at
Christchurch Hospital suggests their numbers are
increasing.4 Despite this, there has been
little research on this population, with neither a standardised term for them
(they are varyingly described as regulars, repeaters, frequent flyers, and
frequent attenders), nor a standardised definition (study definitions vary from
more than 4 to 12 or more visits per
year).2,5–9
Most previous studies have focused on snapshot demographics
and agree that this population is a vulnerable one: predominantly male,
single—with a high incidence of multiple medical problems, psychiatric
illness, alcohol abuse, and poor social support
networks.1,2,5,7
A commonly held belief is that those patients who are
identified as frequent attenders persist with these habits over long periods of
time, and that those patients who stop presenting must be using another ED or
have died. Kne et al suggested that the frequent-attender population is not a
constant one and that most frequent attenders do not remain so over
time.9 However, Kne’s study is the only
one to examine patterns of use over time, and it was undertaken in a large
American city where patients might attend multiple EDs.
Our study aims to describe the pattern of ED-use among a
cohort of frequent attenders to Christchurch Hospital’s ED over a 4-year
period.
MethodsA descriptive retrospective
cohort study was performed. For this study, a frequent attender was defined as a
patient who presented 10or more times in a designated 1-year period.
Christchurch ED has a computer system that is able to identify patients
attending the ED 10 or more times per year, and this was used to identify
frequent attenders who presented between 1 January 1997 and 31 December 1997.
Paediatric patients (aged less than 15 on 1 January 1997) were then excluded.
This cohort was described as Year 1.
The ED at Christchurch Hospital has approximately
65,000 patient visits per year, and serves a population of
330,000.10 Christchurch Hospital is a tertiary
referral centre and, except for obstetrics and gynaecology, provides all major
specialties on site. Its ED is the only ED in Christchurch City, and therefore
all ED visits in the city are captured on the Christchurch ED computer
system.
Demographic information was collected and a primary
diagnosis was assigned to each patient from three broad categories: medical (eg,
airways disease, ischaemic heart disease), psychiatric (including self harm), or
substance abuse (including intoxication and complications—eg, multiple
minor injuries or chronic pancreatitis). Diagnoses were assigned by a single
investigator (DK)—using information from computer records, management
plans, and case notes. When more than one illness was considered a significant
factor in the reason for consultations, then entries were made in more than one
category. This categorisation is consistent with the methodology used in
previous studies.4,9
All ED visits made by this cohort were then documented
for the next three consecutive years: Year 2 (January 1998 – December
1998), Year 3 (January 1999 – December 1999) and Year 4 (January 2000
– December 2000).
Patients were then classified on a yearly basis, either
as: continued frequent attenders (10 or more visits), making less than 10
visits, making no visits, or deceased. The total number of frequent attenders
for each of the study years was also obtained, together with their total number
of attendances.
Efforts were made to establish the continued presence
in Christchurch of patients in the cohort who were no longer frequenting the ED.
This was done using computer records, which document any other contact with the
hospital (including outpatient clinics, inpatient admissions, and admissions to
hospitals outside of the Christchurch area). A documented contact with
Christchurch Hospital (after the conclusion of the study period) was accepted as
confirmation of a continued residence within the area. Patients who could not be
accounted for in this way were followed up via telephone contact with their
general practice. Continued contact with the GP after the conclusion of the
study was accepted as confirmation of presence in Christchurch throughout the
study period.
ResultsA total of 80 patients were
identified; and of these, 3 paediatric patients were excluded, thus providing us
with a cohort of 77 adults.
These 77 patients made a total of 1127 visits to the ED in
the index year (making up 1.7% of the total ED presentations
[65,007]).
The median age of the cohort was 41 years (with a range of
17 to 95 years), and there were 44 males (57%) and 33 females (43%).
The most common primary diagnosis was of a medical problem
(45%)—followed by psychiatric (29%), and substance-related conditions
(26%). Thirty-five percent of the patients had problems in more than one
category. (See Table 1.)
Table 1. Diagnoses assigned
The total number of presentations made by the cohort fell
substantially in each of the subsequent years, from 1127 to 384, as shown in
Table 2.
Table 2: Presentations to Christchurch Hospital’s
Emergency Department (ED)
The median number of visits made to ED by those who were
still attending also declined each year from 12 to 4.
The numbers of patients from the cohort presenting as
frequent attenders declined each year—from 77 in the index year, 28 (36%)
in year 2, 18 (23%) in year 3, to 13 (17%) in year 4. Forty-six (60%) of the
original 77 patients were defined as frequent attenders in the index year only,
and not in any of the subsequent years. Thirteen (17%) of the original cohort
that were still presenting as frequent attenders did, however, continue to
present with great frequency, with a median of 18 (range 10–46)
presentations in Year 4.
Of these 13 patients, only 2 had isolated medical
problems—the other 11 having either psychiatric or substance abuse as a
primary or secondary diagnosis. The male-to-female ratio was 6:7.
Seven (9%) of the original 77 were frequent attenders in all
4 consecutive years. Of these 7, 6 had a primary or secondary psychiatric
diagnosis, and the remaining 1 patient had a diagnosis of substance abuse. The
male to female ratio was 2:5.
A few fluctuations were noted; for example, some patients
fell below the threshold definition in one year only to be redefined in
subsequent years.
Of note during Year 4, there were a total of 76 frequent
attenders to the ED; of these, only 13 (17%) were from the original cohort. (See
Figure 1.)
Figure 1. Number of patients presenting to the ED as
frequent attenders
 In Year 4, 50 (65%) patients from the original cohort
continued to make ED visits; median 4 (1–46)—of these, 37 (48%) made
less than 10 visits with a median of 3 presentations.
Each year, an increasing number of patients made no further
ED visits. In Year 4, 27 (35%) of the cohort did not make any ED visits. Of
these, 10 had died, 5 were identified as having left the area, and the remaining
12 were confirmed as still in the Christchurch area at the end of the study
period. All other patients were confirmed as being present in Christchurch at
the end of the study period.
Only 1 patient was lost to follow-up after having presented
in each of the 4 study years; his whereabouts could not be confirmed at the
completion of the study period.
Of those 10 (13%) patients who died, 4 died in Year 1, 1 in
Year 2, 2 in Year 3, and 3 in Year 4. The mean age of death was 56.5 years
(range 33–95 years). All died from medical illnesses or the medical
complications of substance abuse. There were no documented deaths from trauma or
suicide.
DiscussionStudies have consistently shown the
frequent-attender population to be predominantly male, single—with
multiple medical problems, often complicated by high rates of psychiatric
illness and substance abuse.1,2,5,7 The
patients in this study showed a slight male predominance, and although the
majority of patients presented with medical problems (45%), this population also
had high rates of psychiatric illness (22%) and substance abuse (26%). Many
patients had multiple pathologies, with 35% having illnesses classified in more
than one category. These results are comparable to a previous study in
Christchurch,4 although (at that time) the
frequent-attender population showed an equal male-to-female
distribution.
Frequent attenders have been shown in several studies to
make a disproportionate number of ED visits.1–4
The study cohort of 77 patients made (1127) 1.7 % of the total ED
presentations (65,007) in the index year—compared to the average
population of Christchurch who generate 20 visits per 100 head of population per
year.10 Indeed, frequent attenders place a
large financial burden upon EDs,2 they have
higher admission rates to inpatient beds,2,6
and they have increased mortality,3,11
especially from suicide or violence. The Christchurch cohort had a high
mortality over 4 years, however all of the deaths were from medical causes, with
none attributed to suicide or violence.
There has been only one attendance cohort study previously
published. Kne et al9 (in Rochester, USA)
studied a similarly sized cohort of 76 frequent attenders (defined as more than
10 visits—differing slightly to the definition in this study [of 10 or
more visits]) from a department of similar size receiving 59,000 visits per
annum. Demographic similarities have already been demonstrated between frequent
attenders to this ED and frequent attenders to the ED in
Christchurch.4
Both cohorts accounted for a disproportionately large number
of ED attendances: 1.9% of all ED visits in the defining year in Rochester,
compared to 1.7 % of Christchurch. They also showed a decrease in the number of
patients who continued to be defined as frequent attenders over the study
period.
Forty-six percent of the American cohort were frequent
attenders in the defining year only, and not in any of the subsequent years (as
compared to 60% in Christchurch)—and 17% remained frequent users in all 4
of their study years (9% in Christchurch). The majority of frequent attenders in
both studies had medical problems as a primary diagnosis.
Both studies also found that the patients who continue to
present as frequent attenders over several years tend to have psychiatric
illness and substance-abuse problems either as a primary diagnosis or as
complicating a medical illness.
The Rochester study required a more complex follow-up of
attendances, to 5 city hospital EDs, and had the potential to miss visits made
to other EDs. In Christchurch, we are ideally placed to study the attendance
patterns of our patients, as there is only one emergency department in the
city.
Therefore, we can be confident of capturing all ED visits
made by our cohort. None of the patients (who stopped presenting to the ED) were
lost to follow-up, and only one patient was lost after presenting in all 4 of
the study years.
This study demonstrated that the population of frequent
attenders in Christchurch ED is not constant, but continually changing, thus
supporting the findings of the Rochester study. The majority of these patients
have medical problems and reduce their frequency of attendance over time,
presumably because there has been an improvement in their medical condition or
the resolution of a social stressor.
This attrition pattern is similar to that shown in frequent
attenders presenting to general
practicioners,12 although the population itself
is somewhat different, with the majority of these patients being female and
married.
Although most of the cohort decrease their rates of
presentation, a small number of patients continue to make multiple ED visits
over several years, with no apparent decline in the number of presentations they
make each year. Furthermore, this group of patients were shown in both studies
to have a high incidence of psychiatric problems or substance abuse as either a
primary or secondary
diagnosis.
To instigate management interventions, new frequent
attenders must be regularly identified given the dynamic nature of this
population. Since the majority of these patients have medical problems and are
short-term frequent attenders, they may be more easily managed. By identifying
this group early, and maximising input into their care, it may be possible to
improve their health, and reduce the number of visits they make. The habitual
frequent attenders tend to have substance abuse and psychiatric problems and
they may require intensive management to reduce their reliance on the
ED.
The aim of intervention (in the management of frequent
attenders) should be to improve the quality of care delivered as well as to
reduce the number of ED visits. Given the complex medical and psychosocial
issues, this would seem most likely to be achieved using a multidisciplinary
team to produce an individualised and consistent management plan for each
patient.
To date, very few studies have been performed to assess the
effectiveness of interventions.
Andren and Rosenqvist concluded that social work
intervention reduced the number of ED visits.13
A small study by Pope et al14 showed a
reduction in the number of visits made after implementation of a management
plan; however, Spillane et al 8 (in a
randomised trial) showed no difference in numbers of visits to ED after
implementation of multidisciplinary care plans when compared to a control group,
although improvements of the patients’ health were not
addressed.
In Christchurch Hospital, frequent attenders are currently
identified on a monthly basis and are flagged after making 6 visits in the
preceding 12 months. An individualised management plan is then produced by a
multidisciplinary team (including ED staff, family practitioner, appropriate
medical specialist, psychiatric service, and social worker).
Attempts are made to involve the patient in the production
of this plan by allowing them a free GP visit to discuss their plan with their
GP. Other initiated interventions include a voucher system to enable free GP
visits. Studies are currently underway to assess the effectiveness of these
interventions in Christchurch.
ConclusionThis study has demonstrated that
frequent attenders are a constantly changing population. Although most continue
to attend at a reduced rate, only a small proportion remain habitual frequent
attenders, and these patients tend to have psychiatric illnesses and substance
abuse problems. To initiate any management interventions, ED attendances must be
regularly monitored to identify the constant stream of new frequent attenders,
and also to flag habitual frequent attenders. Further research into the
effectiveness of intervention is also required, both in terms of reduction of
numbers of visits to the ED and the quality of care delivered to this
population.
Author information:
Debra Kennedy, Senior Registrar in Emergency Medicine, Christchurch
Hospital, Christchurch; Professor Michael Ardagh, Emergency Department
Christchurch Public Hospital and Christchurch School of Medicine,
Christchurch
Correspondence:
Professor Michael Ardagh, Emergency Department, Christchurch Hospital,
Private Bag 4710, Christchurch. Fax: (03) 3640286; email: michael.ardagh@cdhb.govt.nz
References:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|