The ‘unnecessary’ use of emergency departments by older people: findings from hospital data, hospital staff and older people

Additional keyword: inappropriate presentation.

Demands on emergency healthcare in Australia are rising,^1–3 reflecting trends in the US, Canada and Europe.⁴ While the ageing of populations is a major contributory factor, demand in Australia is greater than expected from demographic change alone, suggesting an overuse of emergency department (ED) services.⁵ There is a view older people are more likely than other age groups to present at an ED with non-urgent conditions^4,6,7 and with some presentations reflecting a social need rather than a medical need.^6,8–10 These presentations are often classified as ‘unnecessary’ or ‘avoidable’.^11,12 As EDs may not be designed to meet the needs of older people and may place the older person at risk of further adverse outcomes,^13,14 reducing unnecessary presentations may assist with the overcrowding problem experienced by EDs and lead to more valued and appropriate alternative avenues of care and assistance for the older person.

There is no standard definition of what constitutes inappropriate attendance at EDs¹⁵ and decisions on which attendances are necessary or unnecessary depends on the criteria used. Bezzina et al.¹⁶ argue the concept is complicated by ‘differing motivations between stakeholder groups for seeking to define ‘inappropriate’ presentations.’ The Australian Institute of Health and Welfare (AIHW) recognises there are ‘potentially avoidable general practitioner (GP)-type’ presentations to EDs and consequently, these presentations could have been avoided if appropriate non-hospital services were available in the community. Data on these presentations are not considered a measure of hospital performance. The AIHW defines avoidable presentations as presentations to public hospital EDs where the patient was: 1) allocated a triage category of four (semi-urgent) or five (non-urgent); 2) did not arrive by ambulance or police or correctional vehicle; and 3) was not admitted to the hospital, was not referred to another hospital, and did not die.¹ Based on the above criteria, data for Australian hospitals indicates that in 2011–12, for the total population, 38% of all presentations in Australia could be classified as potentially avoidable GP-type presentations.¹ With no agreement on defining inappropriate attendance at EDs, there is also no consensus on what level of preventable hospital admissions in the older population, or any age group for that matter, constitutes a problem.⁶

The factors affecting the demand for emergency care are complex and multifaceted. According to the Andersen–Newman health utilisation framework, demand is related to both individual determinants (mediated by predisposing factors, enabling factors and illness level); societal determinants (norms and technology, population growth and ageing) and the health services system (through its resources and organisational structure).¹⁷ More specifically, international and national studies indicate besides the often complex health issues confronting older people, other identified factors include it is a consequence of ageing;¹² socioeconomic characteristics;^6,18 influence of family opinion;⁶ limited access to primary care,⁶ availability of services in hospital;^19,20 lack of self care;^6,21 and a lack of social supports.^12,22,23

Our understanding of the importance of these factors in terms of EDs arises from analyses of older people’s presentation to EDs regardless of triage category. Few studies have examined unnecessary presentations to EDs and much of our understanding comes from analyses of hospital and patient data, or health professionals’ viewpoints.^6,24 While professionals within the health and care systems may have particular insights into the nature and causes of potentially avoidable presentations, Gruneir et al. conclude there is limited understanding of the sociodemographic, clinical and contextual factors that result in older people attending the ED. ⁴

Only a limited number of studies have sought the older person’s view on why they attended the ED,^12,19 yet it is likely these reasons that are of most value in developing policy and practices to reduce avoidable presentations.

This paper compares and contrasts the information collected through hospital data, the perceptions of hospital staff and the viewpoints of older people themselves to gain a greater understanding of whether visits of older people to EDs are potentially avoidable.

This research was conducted at two major public hospitals in metropolitan Adelaide, one located to the north-east of the central business district and one to the west. The project used three sources of data: hospital data for the population aged 65 years and over identified as triage category four or five for the financial year 2010–11; focus groups; and individual surveys. The hospital administrative dataset includes a range of data items on each person that presents to the ED of a hospital and it is from this data that the AIHW presents annual reporting on the number of potentially avoidable ED presentations (Table 1).

Table 1.  Administrative data collected by the emergency department

As it is difficult to get access to ED staff, focus groups were deemed the most appropriate means of gaining insights into ED presentations by older people in an efficient and timely manner. Focus groups were held with nominated staff members of the ED in each hospital, including the director of each ED. These directors arranged times when the research team could conduct focus groups with the staff. The focus groups were held in late 2011 with 30 staff attending, across the two hospital sites. The focus group discussions, lasting around 1 h, centred around several questions that included patterns of attendance, reasons for attendance, the determinants of preventable hospital admissions, characteristics of the older person presenting to the ED, and the structural features of the healthcare system that influenced presentations to the ED.

The final stage of the study aimed to survey as many people as possible who qualified for the study over a 4-week period: community-dwelling older people (not those in residential care) who presented (not arriving by ambulance) to the ED of the hospitals when by triage category their health circumstances did not warrant emergency attention. Although the AIHW classifies people only triaged as category four or five as ‘non-urgent’ the focus group discussions indicated that as a matter of course, many older people who present with non-urgent conditions are often triaged as category three because of potential multiple underlying conditions (comorbidities), therefore it was decided to include older people triaged as category three in this final stage of the study.

Recruitment of people occurred between the hours of 7 am and 6 pm in late April–early May, 2012. Attendees triaged as three, four or five at the ED were approached about the project. In all, 120 people aged ≥65 years agreed to participate in the study. Arrangements were made to ring them a fortnight after our initial contact, when they were hopefully feeling able to undertake a telephone survey. During this telephone call 39 people declined to be involved in the study, 23 people were excluded from the study because they had arrived at the hospital by ambulance (which was not recorded in the hospital tracking system used to identify potential participants) and 19 people had been admitted to the hospital after our initial contact with them. A sample of 58 people was surveyed. This survey, of 40-min duration on average, collected both quantitative and qualitative data to provide not only subjective views on why older people attend the ED, but also objective insights. Data on health and wellbeing, use of health facilities and services, and social networks was collected to gain an understanding of the broader circumstances that may consciously or unconsciously influence their choice to seek out the ED as a place of treatment.

Ethics approval was granted from the Human Research Ethics Committees of The University of Adelaide, the Central Northern Adelaide Health Service and South Australia Health.

This data was only available from one hospital. Using the AIHW definition, 2688 presentations by older people were considered to be potentially avoidable GP-type presentations to the ED. The majority (87.6%) of these people were triaged as category four. Regarding demographic characteristics, there was little variation by sex and presentations decreased as age increased (Table 2). Although Australian-born people comprised the largest group, more than 70 birthplace groups were represented.

Table 2.  Administrative data 2010–11: characteristics of older people triaged as category four or five

In terms of the mode of travel to the ED, 84.6% of the people ‘walked in’. There was an even spread of people attending across the year with a slight increase in December and January when local surgeries may be closed, have reduced hours, or a preferred doctor is on leave. More than two-thirds of people arrived between the hours of 8 am and 5 pm. There were several different diagnoses of people’s conditions, with trauma, gastrointestinal tract issues and musculoskeletal and connective tissue problems the main diagnoses. Interestingly, more than 100 people, or around 5% of cases, were given a non-descript diagnosis of ‘miscellaneous, social, other’ suggesting there was no medical reason for them to be attending a place of emergency medicine. Upon assessment, most people were discharged (Table 2).

The medical and allied staff in the focus groups felt that older people were a significant group within the overall pattern of demand in the ED. It was difficult for staff to identify what proportion of the older people that they see would be triaged in the non-urgent two categories and whether this had changed over time, however they did indicate that because of potential comorbidities, older people would generally be triaged at least as a category three rather than as four or five. One medical expert stated all older people who choose the ED have a right and a need to be there. Focus group participants identified noticeable trends in terms of times of the year, days of the week and times of the day when people with less significant conditions were likely to present – peaks occurred during the winter months, first thing in the morning (6 am–9 am) and again in the evening when relatives may come home from work. It was stated ‘there can be a bit of a rush on Fridays as nursing homes know they will be short-staffed over the weekend.’ In a more negative light it was also suggested that there can be increased activity before long weekends and holidays, in what was termed ‘relative-dumping’.

The reasons identified for the presentation of older people at the ED fit within the Andersen–Newman health utilisation framework reflecting individual determinants, societal determinants and the health services system (Table 3). Focus group participants stated that underlying all presentations is a feeling of need and a medical condition that initiates the desire to seek help. For some this was a physical condition but for others it was anxiety and depression. It was felt that some people use the ED because of a lack of support at home and an inability to adequately care for themselves, or alternatively they used the hospital as a social outing. Older people commonly experience loneliness and in the ED they can have positive interactions with others. The ED can also be a convenient place for family members to take older people and sometimes families are worried and want an older relative assessed by the ED. Some older people reappear one to two days after discharge from the hospital.

Table 3.  Determinants of older people with non-urgent conditions presenting to emergency departments: views of medical and allied health staff and older people
Grey indicates reason was identified in focus groups or survey

According to the focus group participants, the ED is also an attractive option for older people because of the nature of the healthcare system. The availability of GP services appeared to be a strong influence on people’s presentation at the ED. In the western suburbs an identified lack of GP services was considered a contributing factor, however in the northern suburbs it was felt there were adequate services in the area, with many GP services open seven days per week until 10pm, as well as a locum service, (though waiting times for a locum could be considerable). Community psychiatric services for older people however were considered to be lacking and not available after hours.

Focus group participants identified that some older people feel it is easier to get service in the ED than to attend the GP and that waiting lists for treatment are long and older people (and families) mistakenly believe that by coming to the ED their case will be fast tracked. Also if a person cannot get into the GP of their choice, despite the availability of options like seeing another doctor or going to another clinic, they often go to the ED.

Focus group participants saw the availability and cost of services at the hospital as attractive features to both attendees and GPs. Treatment at the ED is at no cost to the individual, and it is a one-stop shop for all services, particularly after hours. Additionally some tasks that were common practice for GPs, such as suturing, are now diverted to hospitals as GPs have finite appointment times and such services come at a cost to the GP clinic. Changes in medical protocols were also raised as a factor in older people visiting the ED. For example, it is now recommended people go immediately to the ED or ring an ambulance for particular symptoms, and phone services staffed by qualified personnel may advise callers to go to the ED because it is difficult to diagnose a caller’s condition.

Table 4 presents the sociodemographic characteristics of the 58 people surveyed; Table 5 provides details on health status and use of health services.^25–28 As for the demographic data, there were slightly more female presentations than male and as age increased, presentation decreased. Many had limited education and were now reliant on the aged pension. Nearly one-third of the respondents felt that economically they were struggling. In terms of social connectedness in over two-thirds of households someone was able to drive and social connectedness was more reliant on family, friends and neighbours than with community activities. Overall a high level of perceived social support was present in the surveyed group, although individual answers to the questions comprising this measure indicated some people’s expectations regarding social support were not being met.

Table 4.  Sociodemographic characteristics of sample of older people presenting to the emergency department

Table 5.  Health-related data and healthcare system use by sample of older people presenting to the emergency department (ED)
GP, general practitioner

From a health perspective (Table 5) older people arrive at the ED with a range of medical complaints that varied by triage category. Overall, for 38% of people the condition was an existing problem and most of these people (91%) had sought medical help previously. Most people (81%) went to the ED because they felt that their health or the specific condition warranted immediate attention. Of interest was that the initial triaged description of presenting symptoms was not reflected in the final diagnosis upon discharge for several cases (four people). For example, a person may present with a gastrointestinal complaint but this condition is related to an underlying chronic condition of diabetes. This supports the views of medical staff reported earlier that assigning a cautious triage category prompted by the consideration of the higher likelihood of comorbidities is a prudent decision. The survey indicated that many of the people that attended the EDs had a familiarity with hospitals as a centre of care. In terms of non-hospital services, 50% visited a GP at least monthly. They followed a very traditional pattern, with 96% visiting the same doctor or clinic when seeking GP services.

Several people (72%) went to the ED because they could not access a GP, or it was the only place open. Some of the comments included: ‘have to book two days before seeing doctor’; ‘too long to wait for doctor – would take 12 h’; ‘cannot wait five days for a doctor’.

Some attended because someone else (family or friend) decided to take the person to the hospital, or the local medical clinic or doctor, for various reasons, sent them to the ED. The reasons given to the older people as to why they could not be treated at the clinic were because ‘it would be quicker’ and because ‘the doctor didn’t have the bandages required’. Older people also attended the ED because of the level of care they had previously received or expected they would receive at the hospital. For many, the condition that prompted their ED presentation was of a chronic nature and previous experiences had taught them that they would be wasting time and be confronted by frustrations if they accessed a medical clinic initially. For example, comments included: ‘I went back to the hospital where I had the operation’; ‘the ED has more facilities and is more likely to help me’; ‘problem will get sorted quickly by going to the hospital’; ‘knew it would involve X-ray so wanted to get it all done at once’; ‘more direct attention, has all the equipment’. Some people commented on the kindness and professional help they received in the hospital.

At the time of interview 50% rated their health as fair or poor, many were limited to some degree with regard to household activities, and more than 90% of the sample had some level of anxiety or depression. Overall, people recorded low quality of life scores.

This research set out to examine the validity of the AIHW classification of older people’s presentation at EDs as ‘potentially avoidable GP-type presentations’ and to examine, from the perspective of the staff attending to them and from the older people themselves, as to why they presented at the ED. Such presentations at hospitals have been perceived as a challenge for the health system in Australia, as is the case in many other countries. As EDs are a significant avenue of care for older people, there is concern about the pressures on EDs with the ageing of the population.

The AIHW classifies ‘potentially avoidable GP-type presentations’ solely on the basis of the triage category assigned to the person when they first report to the hospital. As older people may suffer from chronic complex conditions this initial triaging may not truly reflect the medical status of the person’s condition. Information received from the older people included in this study indicated that the medical condition for which they were attending the ED was often one for which they had sought help previously at the hospital or at a GP clinic or other service. While this study is limited in scope, the discussions with hospital staff and the surveying of the older people in this study indicates that to gain an insightful indication of whether older people attend EDs inappropriately, there is a need to incorporate a wider range of variables than available from administrative data alone.

As indicated in Table 3, there are many common determinants identified by both hospital staff and the older attendees for the decision to present at the ED. From the perspective of medical staff, non-urgent cases of older people living in the community were likely to arrive during the day, whereas after-hours attendees were more likely to be from residential care, due to the lack of appropriate staff in these facilities after hours. The perceptions of the medical staff as to why older people attend EDs reflect such issues as a perceived need to seek help from the ED, loneliness^12,22,23 and a lack of supports at home,^6,21 family opinions,⁶ inadequacies in the wider health system and the cost to the individual.

Two factors mentioned by the focus group participants – the health condition warranting a visit to the ED and inadequacies in the wider health system – also appear to be the most important factors from the point of view of the older person. This is reinforced by the qualitative responses from the older people in the study, where they preferred to attend at the ED due to negative or frustrating experiences when attempting to seek care at a medical clinic. Also, trust was an important component in the decision-making process of the respondents and those who could call on either their partner or children in an emergency would be recommended to contact, or indeed be taken directly to an ED. In contrast, loneliness and a lack of support at home and the affordability of care^19,20 were not issues for the group of older people included in this study.

This study has limitations in that the findings are based on only two metropolitan hospitals and 58 non-urgent ambulatory cases and therefore generalisation is, as always, questionable. However it does indicate that it is important in any development of policy and practice to gain the views of both service providers and service users at the centre of the policy. For many of the older people involved in this study, attending the ED was not an unreasonable nor avoidable event, particularly considering their perceived health status.

With the increasing number of older people in the population and the ageing of this population, the presence of older people with urgent and non-urgent conditions in the EDs of Australian hospitals is likely to accelerate. A lack of services or programs within the community to support them, increasing waiting times to see doctors^29,30 and the closing of community health centres³¹ can only exacerbate the situation. The ED of hospitals will need to continuously adapt to accommodate the needs of this older demographic and for staff to acquire necessary geriatric skills.

The present study contributes to the body of knowledge on older people and inappropriate attendances at EDs. The ED is an attractive option for older people who have a serious concern about their health. However from this study it is clear that if there was a greater availability of programs and services within the community providing person-centred care catering to the needs of older people it may be possible to ameliorate some of the demand and relieve some of the pressure on EDs. With the ageing of the population and with increasing levels of comorbidities, any practical solutions aimed at reducing ‘inappropriate’ or ‘avoidable’ ED presentations for this group must be cognisant of potential risks to a patient’s health.

None declared.