Emergency-Department-Initiated Palliative Care Consults: A Descriptive Analysis

Introduction

The impact of palliative medicine (PM) might be increased by moving palliative care (PC) services upstream to the emergency department (ED), because the majority of seriously ill patients start their in-hospital course in this acute setting.^1–3 In the ED, determinative decisions about aggressive life-support measures are often made and these early hours of care set the trajectory for future in-patient care.^1,4–5 The ED setting may be ideal for initiating early goals-of-care discussions for patients who suffer from advanced, end-stage life-limiting illnesses so that treatment pathways can be closely aligned to meet patient and family values and preferences.^1,4,6 Although an increased PM and emergency medicine (EM) collaboration may be optimal for the seriously ill/injured, this practice varies depending on the institution and ED culture. The major focus of traditional ED care/training is on caring for the acutely ill or injured.^5,7 Resuscitation, stabilization, and procedural skills are prioritized and ED physicians may not consider PC skills as part of the ED mission.^5,7–8 Therefore, an initial step may be to describe the current PC services provided in ED so both disciplines may build upon this common foundation.

Evidence is accumulating on ED patients as studies address the questions of “who has unmet PC needs in the ED” and “how do we best identify such patients early.”^3,9 The patients who use the ED safety net are more likely to have ill-defined disease trajectories as compared with the traditional in-patient model. ¹⁰ To provide optimal PC in this acute setting, it may be beneficial for the consult team to delineate the most commonly requested ED-PC services and understand why ED clinicians currently request palliative care consults (PCC). We studied patterns of ED-initiated PCC (EDI-PCC) and describe here the use of PC services in an urban tertiary-care-center ED. We then compare these with PC services provided in the traditional in-patient consult setting.

Methods

Descriptive analyses were performed on all PC team consults conducted from March 2008 to June 2009. The study was a retrospective review of data gathered by the PC team on services and consults provided. Data were reviewed in electronic medical records for accuracy. This study was approved by the hospital's Institutional Review Board.

Results

A total of 1481 PC team consults were identified (320 entries were duplicates/multiple consults for same patient). Of the 1261 unique encounters, 108 (8.6%) originated in the ED. Of these, 19 were re-assigned as floor-initiated because the PCC was requested by the in-patient team on an admitted ED patient awaiting a room; 89 true EDI-PCC were available for analysis. During the study period, the PC team consulted on 578 patients in the intensive care unit (ICU-initiated PCC) and 583 patients on the general medical/surgical floor (floor initiated) (Table 1).

The majority of the EDI-PCC patients were African American (60/89, 67%) and male (58% males; 42% females) with an average age of 47 years. EDI-PCC patients had a high mortality rate (55/89, 62%) and the majority (84%, 46/55) died in the ED.

EDI-PCC patient dispositions are shown in Table 2. The average hospital and ICU lengths of stay were 2 days and 0.5 days, respectively, in hospitalized EDI-PCC patients (14/89, 16%).

Services and communication provided by the PC team to the EDI-PCC as compared with other traditional consult settings are shown in Table 2. No significant differences existed in PC provision for the EDI-PCC patients who died, but there was an expected increase in bereavement support to family survivors (43/55, 78%).

When compared with PC services provided in the traditional in-patient setting (Table 2), the EDI-PCC service was called more frequently for emotional/bereavement support and was involved in more family meetings. Nurses in the ED were involved in the majority of communications with PC team members. As anticipated, the ICU and floor settings had a higher involvement of PC team in goals-of-care discussions, whereas services related to withdrawal of life-support were most needed in the ICU.

In those ICU-initiated and floor-initiated PCC patients who died in-hospital, the average time to PCC was from 7 to 11 days (Table 1). This is significant because the patients died soon after (average 12-day hospital length of stay).

Discussion

The major findings of our analyses are: 1) ED patients who have PCC are young (compared with traditional in-patient consults); 2) a PCC is most often requested for imminently dying ED patients, who often die in the ED setting; 3) time-sensitive PC services are needed due to imminent deterioration; 4) higher frequency of emotional/bereavement support is usually required; and 5) communication with family is a major ED PC need.

Care for chronically ill patients suffering with advanced, end-stage illnesses is optimized by a healthy relationship between both EM and PM specialties, based on a mutual understanding of needs and scope of practice.^1,3–5,7,11 PM, in most hospitals, is based on the traditional consultation model; the managing provider initiates a consult to address specific needs. The “best consultant” (the one whose recommendations are likely to be accepted and who is most likely to be used again) is described as one who practices good consultation etiquette in that he/she “determines the question to be answered, addresses it, and doesn't reach beyond that remit unless requested by the managing team.” ¹² Often, when such etiquette is routinely breached or the needs of managing providers are routinely unmet, PM consultants may risk being ‘‘shut out.’’ ¹² We retrospectively analyzed PCC initiated by the ED clinicians in an effort to identify what questions they ask of and assistance they seek from PC consultations.

ED physicians, often challenged by end-of-life issues, struggle to “do the right thing.” ⁸ In a procedure-based culture, it is frustrating not to be “aggressive” in the terminally ill, when the training (and reimbursement) favors an emphasis and priority on resuscitation. Communication skills have a lower priority and value in training than technical skills. ⁸ Time constraints, lack of prior relationships with family, and often the unexpected nature of the terminal event add to the complexity of managing ED patients at end of life.^3,5,7,8 It is therefore not surprising that we find most PCC are requested for the imminently dying ED patients. This also highlights that ongoing communication with family around complex decision making at end of life (goals of care, advance directives, hospice), as well as family emotional/bereavement support are key services ED clinicians seek from the PCC team. We also identified that most patients have a rapid downhill trajectory and die in the ED or soon after, so ED consults should receive the highest priority. ¹

The ED patients with PCC are often young due to traumatic or sudden critical events (e.g., children/young adults in motor vehicle accidents). These occurrences render families unprepared and confused as they struggle with the rapidly changing status information overload and often an inability to cope with their loved one's terminal event. PCC teams should therefore anticipate the unexpected and be prepared to assist distraught, often angry and in-denial family members, in this emotionally charged setting of uncertain prognosis and chaos.

Our study suggests some benefits to the early EDI-PCC. We found that it took more than a week to initiate a PCC for an in-patient, even in those who were dying, often too late for meaningful contribution because death followed soon after. ¹³ A possible explanation for this PCC delay may be a perceived “ownership” conflict among primary providers. We propose that an EDI-PCC may address this issue because ED clinicians routinely transfer patient care to other service providers and are hence expected to experience less “ownership” conflict. An additional benefit in those EDI-PCC patients who survive may be an earlier consensus around goals of care/care plans resulting in early disposition to home/hospice, as shown by our short hospital length of stays. ¹⁴

Conclusion

EDI-PCC patients are young, likely secondary to traumatic and critical, sudden events. The in-hospital mortality rate for EDI-PCC patients is very high (most die early and in the ED setting), signifying a trend for ED clinicians to request PC consults in those who are imminently dying. PC consult teams called to the ED should expect to provide high-priority, time-sensitive services and anticipate the need for a high level of bereavement/emotional support for distraught and unprepared families, with major discussions around end-of-life care.

Further prospective, large studies are needed to establish current ED physician practices regarding PCC and as a group identify their expectations from a PC team consult.