Refinement of an Emergency Department-Based, Advance Care Planning Intervention for Nurses

Abstract

Background:

Most older adults visit the emergency department (ED) near the end of life without advance care planning (ACP) and thus are at risk of receiving care that does not align with their wishes and values. ED GOAL is a behavioral intervention administered by ED clinicians, which is designed to engage seriously ill older adults in serious illness conversations in the ED. Seriously ill older adults found it acceptable in the ED. However, its potential to be used by nurses remains unclear.

Objective:

The aim of this study is to identify refinements to adapt an ED-based ACP intervention by eliciting the perspectives of nurses.

Design:

This is a qualitative study using semistructured interviews. Data were analyzed using axial coding methods.

Setting/Subjects:

We recruited a purposeful sample of ED nurses in one urban academic ED and one urban community ED in the northeastern region of the United States.

Results:

Twenty-five nurses were interviewed (mean age 46 years, 84% female, and mean clinical experience of 16 years). Emerging themes were identified within six domains: (1) nurses' prior experience with serious illness conversations, (2) overall impression of ED GOAL, (3) refinements to ED GOAL, (4) implementation of ED GOAL by ED nurses, (5) specially trained nursing model, and (6) use of telehealth with ED GOAL.

Conclusions:

ED nurses were generally supportive of using ED GOAL and provided insight into how to best adapt and implement it in their clinical practice. Empirical evidence for adapting ED GOAL to the nursing practice remains to be seen.

Introduction

In the last six months of life, 75% of seriously ill older adults present to the emergency department (ED) in the United States, signaling clinically significant decline in patients' illnesses.^1,2 More than 70% of them express priorities focused on quality of life rather than quantity.³ A systematic review demonstrated that 56% to 99% of elderly patients in the ED have no advance directives⁴ and may receive care unaligned with their goals.⁵

To help seriously ill, yet clinically stable, older adults formulate their goals for future care, we developed an emergency physician-led, six-minute, motivational interview intervention (ED GOAL).⁶ Our prior studies demonstrated that after ED GOAL implementation, 74% of older adults felt more ready to address goals of care with their clinicians and found ED GOAL acceptable⁶ and feasible.⁷ Yet, physicians' time constraint was a barrier to implementation.⁷ At the same time, ED nurses who observed our studies expressed interest in ED GOAL.⁸

Nurses' strength in communication skills makes them well suited to initiate serious illness conversations.^9–11 ED nurses have previously suggested that utilizing specially trained nurses who are consulted to deliver ED GOAL would result in increased efficacy and intervention fidelity,^12–14 especially because motivational interviewing is within the scope of practice for nurses.^15,16 In other settings, nurse-led, advance care planning (ACP) interventions have been shown to ease patients' decisional conflicts and increase documentation of patient care preferences.¹⁷ However, no empirical strategies to implement a nurse-led ED GOAL exist.

Using qualitative methods, we sought to understand how to refine ED GOAL to fit the scope of practice of ED nurses for effective implementation in the ED. By better understanding implementation strategies for a nurse-led ED GOAL, we hope to maximize its potential efficacy and reach to engage seriously ill, yet clinically stable, patients in serious illness conversations after leaving the ED.

Methods

Study design and setting

We conducted a qualitative study using semistructured interviews. This study was approved by our Institutional Review Board (2020P000012). ED nurses were recruited from one urban academic ED and one community hospital within the northeastern region of the United States.

Data collection and analysis

ED nursing directors provided a roster of ED nurses to identify eligible nurses: full-time (≥40 hours per week) ED nurses or specially trained nurses (wound and ostomy nurses, intravenous access nurses, and sexual assault nurse examiners/SANEs). Purposive sampling was used by soliciting participation of ED nurses who expressed preexisting interest on this topic or were specially trained. Nurses were recruited with e-mail invitations. Two follow-up e-mails were sent to eligible nurses, one week apart, to optimize recruitment. All participants provided verbal consent.

Nurses reviewed the ED GOAL script for serious illness conversations and a video demonstration of ED GOAL. A semistructured interview guide was used to assess nurses' impressions of ED GOAL. The interviewer (M.C., a research assistant) was trained in qualitative research methods in two-hour didactic and mock interviews with technique feedback by a medical anthropologist with 15 years of field experience (A.C.R). The interviews were conducted between March and May of 2020.

The interviews (∼45 minutes each) were organized into six domains to understand the following: (1) prior experience(s) in serious illness conversations, (2) overall impression of ED GOAL, (3) suggested refinements to ED GOAL, (4) potential methods to implement ED GOAL, (5) use of specially trained nurse consultation models for implementation, and (6) use of telehealth with ED GOAL. Open-ended questions followed by cognitive probes were used. After 10 interviews, we reviewed five transcripts to iteratively revise the interview guide for clarity. Transcripts of audio recordings were not returned to participants for comment. The thematic saturation was reached after 25 interviews.

Recordings were transcribed and analyzed by M.C. and B.R.R. using axial coding methods. The coding structure was collaboratively developed through open coding to identify emergent themes using the NVivo 1.3 (QSR International) qualitative analysis software. Seven transcripts (28%) were double coded to ensure inter-rater reliability, with kappa values set to ≥0.80. Each coder coded 16 transcripts. Results were reported in adherence to the Consolidated Criteria for Reporting Qualitative Research guidelines (Supplementary Appendix SA1).¹⁸

Results

We enrolled 25 nurses (average age 46 years, 84% female, and average clinical experience of 16 years). Nine nurses (36%) were specially trained in additional nursing competencies (Table 1). Major domains and emerged themes are identified in Table 2.

Table 1.

Characteristics of Study Participants

Sample size (N)	25
Age (years), mean (SD)	45.6 (12.2)
Female, n (%)	21 (84)
Years of practice in ED, mean (SD)	16.0 (11.2)
Specialized nurses, n (%)	9 (36)
Sexual assault nurse examiners	5 (20)
Intravenous access nurses	3 (12)
Wound/ostomy nurses	1 (4)
Nursing union members, n (%)	24 (96)

ED, emergency department.

Table 2.

Major Themes Identified in the Qualitative Interviews

Prespecified domain	Emerged theme	Representative quotes (participant ID no.)
Prior experience with serious illness conversations	Infrequent, mostly occurs in critical care settings	“I don't think in our ED we, very often, initiate those conversations…The only times that it really every comes up is…if the patient is in a very critical condition and is kind of meeting that point where they need comfort measures only. But with more chronic conditions where the patient is otherwise stable, I don't find too often that we are initiating those conversations in the ED…And, most often, those are led by the providers or the clinicians, not so much the nursing.” (11)
	Rapport with patient driving serious illness conversation	“I think it's very important that the rapport drives the consultation and not necessarily the comorbidities or the chronic illness.” (12)
Overall impression of ED GOAL	Positive	“My overall impression is that this is fantastic and that this needs to be pushed further and it needs to start happening more…I think that especially since we see such a … specific patient population where they're [patients are] critically ill…I think that this would be a wonderful added aspect of their care within the ER” (7) “I think having nurses do it might be a little bit more—because we spend a little more time with the patients during their visits than the physicians get a chance to. So, I think it might—being able to expand that into a nursing role might actually be able to make it happen more often.” (10)
	Neutral or negative	“I think it's very well organized. I do think it can turn into like a very long conversation, which I don't know if it's always feasible depending on how busy the ED is. But I like how it's [the ED GOAL intervention is] very structured.” (19) “It's difficult, especially in the emergency department, when someone is having a healthcare crisis, to bring this up… everything is so rushed and maybe a physician won't have the time to spend with the patient to be able to discuss everything like this. However, on the other hand, I see the need for this and I—with people who do have a primary care physician, this is, in my view, a very, very good script to use when approaching this kind of a scenario.” (20)
Refinements to ED GOAL	Add: identify additional actors central to the patient's medical care	“is there anyone who you would want kind of included in this planning in terms of like family or a caretaker included in this planning or would be helpful to reach out to … kind of discuss your overall goal planning” (19)
	Add: asking permission to disclose patients' wishes to other clinicians	“do you mind if I share some of these answers with your ED provider and we can kind of come up with a plan from there to kind of make you feel more comfortable or how to get some of your questions or your goals reached in the outside setting?” (19)
	Remove: coordinating with the patient's primary provider (second to last statement in Part 6 of ED GOAL script)	“we're almost making promises that we can't keep. Like, oh, we'll get you an appointment where your doctor can just talk about your concerns. I think, realistically, is that something that we could provide? Maybe we could put … I will send Dr. so and so an e-mail voicing your concerns and that could be something that you can follow-up with.” (6)
Implementation of ED GOAL by ED nurses: barriers	Lack of time	“It's just finding the time to actually sit down and have this and to actually make it sound genuine, without feeling rushed.” (9)
	Nurse pushback	“We have so many little assessments that we're supposed to ask every patient that comes into the ED. So I think that would be like the greatest resistance” (19)
	Varying comfort levels of ED nurses in having serious illness conversations	“I can imagine, realistically, that there are some nurses who wouldn't feel as comfortable having these conversations. Every nurse has a little bit of a different preference and background. And I, myself, would feel comfortable. But I can imagine that there are some nurses who would not.” (11)
	Difficulty of clear pass off of serious illness conversation findings to outpatient provider	“I guess more along the lines of the documentation piece, just follow-up. You've had this conversation, you've gotten—you've really gotten to a good place with the patient. Now what? Where—what happens now? And I know the follow-up in the script is relaying to their doctor. But, if these patients are being admitted to the hospital, how do we relay this information to the receiving nurses and receiving doctors, and care team, and just kind of, again, get everyone on the same page?” (11)
Implementation of ED GOAL: logistical needs of ED nurses	Need for a clear method of identifying appropriate patients for ED GOAL	“Maybe it's one of the things that's addressed in triage like, does this patient have any advanced care planning? And it's a simple yes or no. And then, if it's a yes, we confirm with the patient what that is. And then, if it's a no, it gets flagged for the primary nurse that's maybe taking care of that patient to have that conversation with them.” (9)
	Need for administrative support	“I mean, we have to have the buy-in of the administration. We have to have the buy-in of the nursing director and the medical director of the ED” (10) “I just think having the support from leadership, I think, is—if we make it a priority, it will be done. And that's been my experience with most of the kind of the assessments and interventions that we do in the ED. It requires the support of leadership. It requires someone to step forward and say, this is a priority, this is why it's a priority. People just need to understand why they're doing these things, and that it's not just checking a box.” (23)
	Need for interdisciplinary effort	“I feel like it would be a team approach. I wouldn't wanna be the sole person initiating the conversation, because I feel like the patient will have a lot of questions that the provider will be better off answering than myself. So, I think it'll be a combined team approach.” (11)
Implementation: ED GOAL training	Content of training	“I think training on the purpose of the tool, kind of when to use the tool, and then some coaching as to—to kind of get people comfortable with the language, if they're not comfortable with having this type of a conversation. How to have this conversation without being—feeling—without it feeling forced, or awkward…” (23)
	Way to evaluate nurses' competency in administering ED GOAL	“So I would go right to the nurse and say what is your comfort level with this? And do you want to have some conversations or would you like the competency related to this? And that would probably be the best way to do it—have it built into our competencies.” (2)
Specially trained nursing model	Benefits of having all ED nurses vs. specially trained nurses equipped to do ED GOAL	“They're actually sort of like your Dr. [Last Name] video there, you're actually taking away that crazy barrier where, oh, wait a minute, I'm getting beeped, oh, wait a minute, I'm getting a call, oh, wait a minute. You know what I mean? So you're actually—like the SANE nurse, she really—she dedicates time to her patients. So it'd be perfect.” (8) “I think all of the nurses should be trained on it, and then to kind of have some nurses that are more like the champions of—the ACP champions that could support the nurses. Because every nurse needs to know how to have these conversations. And then, that way, what happens if maybe that special nurse–specialty nurse isn't there that day? But it would be a nice resource for those nurses that didn't feel as comfortable. But I think everybody should be trained. But I think, at the same time, it's good to have a few of the specially trained nurses, for sure.” (9)
Use of telehealth with ED GOAL	Reasons for	“Before this COVID, I would have thought that that wouldn't be the right way to do it…But in reality, I think that this could—telehealth could have a place in this [for ED GOAL].” (28) “I think it would be a great one because it would cut down on travel time and it would be a little more convenient for the patient and the staff member because— if the patient's able to effectively receive the information through that method, then I think it would probably make things easier. If your nurse who's not directly next to the patient can just call them and have this conversation, it would be much more efficient—I would say that.” (29)
	Reasons against	“But I do think it takes away from the more personal elements of the conversation, just the ability, like I said, to sit at the bedside with the patient and hold their hand and comfort them, and provide them with a cup of water, or a blanket, or something. The telehealth model is really, really beneficial but it kind of takes away that personal affect.” (11) “I'm not sure that all of our patients, particularly some of the elderly, they can't hear it. It's kinda loud and confusing. I think a one to one—one-on-one person would probably be a better venue for this conversation.” (10)

Nurses' prior experience with serious illness conversations

Participants stated that serious illness conversations occur infrequently in the ED. They reported that these conversations only occur when patients were in a critical condition necessitating immediate decision making.

Overall impression of ED GOAL

Several nurses recognized the importance of initiating serious illness conversations with their patients with serious illnesses. Nurses reported that they were particularly well suited to initiate serious illness conversations given the amount of time they spend at the bedside and the depth of relationships with their patients. Implementation by champion nurses was endorsed. A minority expressed neutral or even negative views toward ED GOAL, stating that these conversations may not be appropriate in the busy ED environment.

Refinements to ED GOAL

Participants suggest asking whether or not the patient wants to involve family/caregivers and spiritual guides in the serious illness conversations and advocating for an interdisciplinary approach to address patients' concerns (e.g., involve physicians to contact the outpatient clinicians and communicate what patients shared in the serious illness conversations).

Implementation barriers and facilitators by ED nurses

Time constraints, lack of privacy, lack of support from some nursing colleagues, lack of comfort with serious illness conversations, and the difficulty of communicating findings to outpatient clinicians were the main barriers. Having specially trained champion nurses would improve intervention fidelity. Other emerging themes included involving emergency physicians when communicating ED GOAL findings with the outpatient teams and the use of telehealth to conduct ED GOAL.

Specially trained nursing model

Champion nurses might result in much higher utilization of ED GOAL. Another model is where every nurse is trained in the basics, while specially trained nurse champions can also be available to execute ED GOAL entirely.

Use of telehealth

Telehealth may make ED GOAL more convenient for both the patient and clinician. Others expressed neutral or negative opinions on telehealth, suspecting that it would take away the personal element of having serious illness conversations that involve emotional and sensitive topics.

Discussion

This study provides empirical evidence that some ED nurses endorse ED GOAL for seriously ill patients in the ED. These nurses expressed that ED GOAL could be feasible for ED nurses with special training. To refine and implement ED GOAL, the nurses recommended strengthening attempts to build rapport with patients up-front, creating consultation models with specially trained nurses, and utilizing an interdisciplinary approach to communicate the patient's wishes with the outpatient clinicians. These empirical findings suggested that rigorous implementation strategies are necessary for successful implementation of ED GOAL by nurses.

Strengths and implications

Champion nurses in a consultation model, such as SANE nurses, may mitigate the implementation barriers such as time constraints, lack of comfort in serious illness conversations, and need to communicate findings with patients' outpatient clinicians. Such models would allow higher quality of serious illness conversations by specially trained nurses with high intervention fidelity.¹² This approach is similar to nurse-led ACP interventions in other settings.^9,10

Furthermore, ED nurses recommended making ED GOAL an interdisciplinary effort. This recommendation is consistent with the existing international health care organizations that systematically endorse ACP support to their patients.¹⁷ In two of these international health care organizations, dedicated ACP facilitators utilize a team-based approach and work closely with physicians to disseminate ACP. This team-based interdisciplinary approach is similar to other nurse-led ACP interventions.^9,10 By having a specially trained nurse consultation model, a designated emergency physician may serve on this interdisciplinary team to communicate what patients disclosed to the outpatient clinicians.

Endorsement of telehealth to administer ED GOAL may be explained by its increased use during the COVID-19 pandemic for palliative care.¹⁹ Several ACP interventions have been successfully adapted to telehealth modalities,^20,21 which demonstrated a shift in clinicians' perspectives on the use of telehealth for serious illness conversations.^22–24

Limitations

The participants had preexisting interest in serious illness conversations. The findings are likely internally valid given that we were only interested in the perspectives of nurses who would become nurse champions for serious illness conversations. Furthermore, ED GOAL is likely administered by specially trained nurse champions only, whose perspectives are illustrated here. Social desirability bias to appease the interviewer^25,26 was minimized by designating a nonphysician interviewer (M.C.) asking questions in a nonjudgmental manner, and all data were deidentified. Last, nurses were asked to imagine administering ED GOAL when they have not been trained in serious illness communication skills. Our findings are theoretical and more empirical data are needed after the nurse champions are trained in ED GOAL.

Conclusions

This study provides empirical evidence that ED nurses believe ED GOAL would be important for seriously ill patients in the ED. The nurses expressed that the scripted serious illness conversations in the ED such as ED GOAL could be feasible for ED nurses with special training in an interdisciplinary manner involving emergency physicians as well. Further studies are needed to empirically assess the implementation in the ED.

Authors' Contributions

All authors contributed to the design, writing, and editing of the manuscript.

Funding Information

B.R.R. is supported by the Medical Student Training in Aging Research training grant through the National Institute on Aging and American Federation for Aging Research (5T35AG038027). T.F.G. is supported by the Cambia Health Foundation. K.O.'s work is supported by the Cambia Health Foundation and National Institute on Aging (K76AG064434).

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

Supplementary Appendix SA1