Characteristics of study subjects
We observed 70 cases over 450 hours. For patient characteristics, see Table 1. Some observation details are summarized in Table 2. Lengths of stay ranged from 0.75 to over 32 hours and involved 2 to 12 ED caregivers and/or consultants. Forty-eight patients (69%) were hospitalized.
Table 1
Patients’ characteristics (n = 70), mean (range) or number (%)
Age, years 75 and older | 82 (72–96) 68 (97) |
Sex (female) | 47 (67) |
Residence Community-dwelling Nursing home Rehabilitation center Unknown | 49 (70) 18 (26) 1 (1) 2 (3) |
Presence of a relative Community-dwelling Nursing home Unknown | 40 (57) 31 7 2 |
Referred by a physician Own general practitioner Another general practitioner Specialist in the institution Outside specialist | 37 (53) 24 5 6 2, including 1 from rehabilitation center |
Referred by a caregiver, with a written note Own general practitioner Another general practitioner Outside specialist Nursing home | 32 (46) 17 5 1, from rehabilitation center 9, including 7 not referred by a physician |
Falls, recent history Main reason for admission Accompanied somatic chief complaint Delirium, >= 2 risk factors Pre-existing cognitive disorders Neurological main complaints Deterioration of general status | 24 (34) 17 7 61 (87) 18 9 12 (17) |
Table 2
Hospital 1 | Hospital 2 | Hospital 3 |
34 observations | 16 observations | 20 observations |
Approximate mean length of stay: 4h | Approximate mean length of stay: 7h | Approximate mean length of stay: 6h30m |
Approximate mean number of caregivers that intervened in the process of care: 4 • ED, MD trainee: 0 • ED, MD assistant: 1/3 • ED, MD: 1 • ED, GP: 2/5 • in-hospital, MD: 1 • ED, Nurses: 2 | Approximate mean number of caregivers that intervened in the process of care: 7 • ED, MD trainee: 1 • ED, MD assistant: 2 • ED, MD: 1 • ED, GP: 0 • in-hospital, MD: 2 • ED, Nurses: 2 | Approximate mean number of caregivers that intervened in the process of care: 5 • ED, MD trainee: 2/5 • ED, MD assistant: 1 • ED, MD: 0 • ED, GP: 0 • in-hospital, MD: 1 • ED, Nurses: 2 |
ED: emergency department; MD: medical doctor; GP: general practitioner |
Main results
Almost all the patients would have benefited from screening for cognitive status, functional performance, and/or psychosocial support to highlight hidden, typically geriatric problems. Indeed, 90% (n = 63) presented with at least one of the three typical geriatric presentations, i.e. falls (n = 24, 34%), high risk of delirium (n = 61, 87%), and functional decline (n = 12, 17%) (Table 1). Two or more risk factors for functional decline were noted in 58 of the 70 (83%). These 90% of patients were taken into account in the second stage of the analysis.
No geriatric flow routine
Overall, the patients received the same care process. Their management was mainly based on a systematic biomedical approach. For example, when Dr B has an elderly patient with deterioration in general status [similar to functional decline, Ed.], she always requests a radiographic examination of the thorax and laboratory tests. (Observation notes H1-2).
ED caregivers generally had no triage routine to screen for a geriatric profile, including functional status, and identify older patients at risk of complications. One hospital had incorporated such a process, using a screening tool, into the nurse’s triage but implementation was almost never done: “I always skip it [screening]; it is in the wrong place and should be part of triage, and I forget it. Whereas charging, we’ve been drilled, we don’t forget it.” (H3-48, nurse’s interview notes); “there are more urgent things; I forget items; it’s not difficult, but going through it every time, ugh!” (H3-51, nurse’s interview notes). In particular, systematic screening for delirium and subsequent prevention measures were not part of standard procedures.
Management followed two guiding principles, prioritization and categorization, to label the problem presented. Prioritization assigned an acuteness level; categorization specified the nature of the priority. This was to exclude organic health conditions requiring rapid treatment, e.g. hip fractures, and to maintain patient flow. In some cases, procedures requiring a prescription were initiated by a nurse, “to save time because we know what the doctors want” (H3-52, nurse’s interview notes).
Categorization and prioritization at the triage stage were of paramount importance for subsequent management. For example, a patient (H1-10) admitted by ambulance was not triaged by an ED nurse but sent directly to trauma for a fall, delaying treatment of an underlying heart problem. Moreover, when falls were the main reason for admission (n = 17), they were systematically labelled as traumatic issues. Preoperative assessment in 6 of these cases ruled out any urgent underlying somatic cause. However, minor trauma, particularly uncomplicated wounds, did not automatically lead to such assessments: management focused on the primary motivating category, e.g. “wound requiring suturing”.
Categorization also allowed implementation of a treatment plan, in particular to facilitate “negotiation for a bed” with a consultant, where necessary. A bed then became the priority. One of the hospitals allocated one nurse to this. Consultants contacted during the care pathway were mainly organ specialists, corresponding to the categorization. However, categorization in a specific specialty was often a challenge for multimorbid patients. Additional technical examinations were regularly used to facilitate negotiation for a bed. Consultants often set “technical” conditions for admission; this sometimes prolonged ED consultations, with little added value for patients.
“So Dr J. phoned the pulmonologist. The doctor she spoke to wanted an echocardiogram. I was surprised and asked Dr J why. She said she didn't know why but she was doing it because she might lose out on a bed. She dressed the patient’s story up for the ultrasonographer, who agreed to conduct the examination.” (H1-8 observation and interview notes)
A geriatrician was called in for 12 of the 63 patients presenting with one of the three geriatric issues (19%), mainly in order to obtain admission to the geriatric ward. Four of these patients already had a link with that department; six others had psychosocial issues and presented with deterioration in general status. The lack of beds in geriatrics was often mentioned, which could be one reason why a geriatrician’s advice was not sought.
“There is definitely no bed in the geriatric department. By the way, there’s never any bed in the geriatric department! Many people are not treated in the right department.” (H3-47 nurse’s interview).
Risk of discontinuity of care
An average of five caregivers intervened per patient, including ED team members and others from other specialties. ED caregivers, moreover, changed during shift handovers and each was often responsible for several patients with different problems, priorities, and, possibly, very different timeliness.
The case of Mr F., aged 81, admitted from a rehabilitation center for a fall with a complicated wound (H3-55), illustrated the risks of this division of work for both care quality and ED flow. A 27-hour stay, intervention by at least 12 professionals, and unclear leadership illustrated the difficulties of passing on the information needed to bring the pathway to an end. In some cases, the observer was approached by ED caregivers, mainly with regard to continuity of information. Observation of a self-appointed case-manager illustrated the importance of coordination of the care process in EDs. A woman aged 81 (H1-12) arrived with a cardiac rhythm problem at a busy time. Her grandson, an emergency nurse assigned to triage, acted as case-manager until she returned home, while complying with the established prioritization rules. He ensured that information was passed on within the team and to the patient, identified the additional examinations needed, and the responsible physician. Eight professionals were involved in this pathway, even though she only stayed for about 3.5 hours.
Unmet basic needs and patient-centered care needs
Functional status rarely figured in the medical history or ED assessment. It was addressed in 9 of the 63 cases by an ED caregiver, including one social worker and one ambulance doctor, and spontaneously declared by the patient or a relative in two. Medico-technical care was the priority, often at the expense of traditional bedside and patient-centered care. Some ED caregivers managed to reconcile technical care with a more global approach. Nevertheless, these kinds of care were in competition, given the irregular flow of patients, which hindered efforts to meet basic needs such as comfort, pain relief, food, or hydration.
“I would like to put her in a bed because she’s in such pain (…), but I still have two infusions to set up.” (nurses’ observation notes H1-6)
Two ED caregivers were particularly attuned to these basic needs. They had experience with a different care paradigm. A social worker provided in-home coordination that took account of the patient’s functional status and preferences (H1-3, observations). One nurse – who had worked on a pediatric ward – was particularly attuned to patients’ basic needs (H1-6): she was “quite respectful and took the time to explain the steps. She put a cover over her, ensured she was comfortable, and tried to reassure her. She talked about her previous experience, which wasn’t typical for an emergency nurse… ‘I like the [observation unit] and reception. I pamper my patients. Most here have flashing lights on their heads!’ [They prefer emergency ambulance call-outs and management of critically ill and polytrauma patients, etc.]” (H1-6 observation notes and spontaneous nurse’s declaration).
Furthermore, management was carried out without consulting older people and considering their priorities. They were deemed to have agreed to the process. Although rarely involved in decisions, they seldom complained about this. But they sometimes addressed information requests and basic care concerns to the observer. A relative can be a precious resource, mainly as companionship for the patient, for communicating basic needs and transmitting information. For patients with cognitive disorders, conclusions and treatment plans were entrusted to a relative. In one case, daughters adopted a more proactive attitude, disagreeing with the doctor’s conclusion that “the assessment had turned up nothing unusual” and with “the decision to send the patient back [to the nursing home] without any explanation [of the symptoms]…. They requested a second opinion…! Dr Y…contacted the geriatric department. He was not convinced that hospitalization was justified, even if he sympathized with the daughters.” (H2-36, observation notes).
Complex older patients: unwelcome in EDs?
During certain observations, the issue of caring for elderly patients was raised with ED professionals.
Its legitimacy was almost never questioned at the front desk or at the triage step, or at least the elderly patient was not blamed. “They have often called the family doctor [unlike younger patients]. If they come to the hospital to be admitted [to a hospital care unit], it is often at the request of the family, who want to get rid of the parents.” (H1-11, interview notes) “Some people call an ambulance to jump the queue – young people, not older people. Older people seem ‘stronger’.” (H2-22, summary notes, nurse's interview) Moreover, prioritization seemed to be easier for older people “because either they [older people] arrive by ambulance and are by definition priority or they are accompanied." (H1, triage secretary interview)
However, during subsequent stages, old age was often associated with complexity of care – one caregiver even referred to troublesome cases (H3-55 observation notes) – and/or slowness, things that do not fit the current emergency department care model (H2-43 interview Dr M). “We don’t have much time but they take up a lot of time even though there are so many people [in the waiting room]” (H2-43, nurse’s interview). “They block up the emergency department” (H3-47, nurse's interview). “I don’t like geriatrics… [in the ED] you have to work quickly and nothing is simple with elderly patients.” (H1-7, interview Dr G).
The feeling that older patients are a burden was indirectly illustrated by overestimations of their proportion in EDs (H2-42, caregiver's interview; H3-43, caregiver’s interview). One caregiver described his/her distress when admitting older patients in the ED observation unit, saying [he/she] was incapable of taking care of them...and needed help “if one weeps and cannot cope” (H2-42, Dr AI).
Although the legitimacy of presenting at the department was rarely questioned directly, two doctors criticized the failure of care in nursing homes. One described a case as “prophylactic” and could not “understand why this patient has been sent in [before the weekend], especially since she has come from a nursing home” (H3-50, observation notes Dr. AC); the second observed: “when elderly people come in at night, it really is serious… or it’s a nursing home that sends them in because they can no longer cope.” (H2-24 interview notes, night-shift doctor)
These attitudes may be associated with stereotypes about elderly patients, which may explain, e.g., a caregiver mainly addressing a daughter before realizing that the patient was cognitively sound (observation notes H2-45). One professional ignored the patient and seemed uninterested in treating the elderly (observation notes H2-40). This recalls ED caregivers’ problems with consultants reluctant to admit elderly patients, as they want patients to fit their specialty’s parameters (general observation notes H1 and observations H1-2, H2-22, H2-38, H3-65).