Introduction
The US population is growing older and more ethnically and culturally diverse (Jordon, Reference Jordon2020). With increasing life expectancy and advances in medical technology, proactive approaches to advance care planning (ACP) are needed to achieve informed, goal-concordant medical care. Yet, nationally, ACP is underutilized (Yadav et al., Reference Yadav, Gabler and Cooney2017).
ACP is a process in which adults reflect upon and share personal values and goals for future care. There are multiple benefits of ACP, including improved quality of care, decreased rates of decision burden for surrogates, and a higher alignment of care with goals (Brinkman-Stoppelenburg et al., Reference Brinkman-Stoppelenburg, Rietjens and Van der Heide2014). Despite these benefits, physicians report they are not sufficiently prepared to engage patients in this critical process (Furman et al., Reference Furman, Head and Lazor2006). Many feel uncomfortable, report limited training in conducting end-of-life (EOL) conversations, and have limited knowledge about advance directives (ADs), an important component of ACP (Morrison et al., Reference Morrison, Morrison and Glickman1994; Scholz et al., Reference Scholz, Goncharov and Emmerich2020). Provider-level barriers to effective ACP discussions include a lack of communication skills, and concerns about managing patient emotions and their own emotions (Meier et al., Reference Meier, Back and Morrison2001; Friedenberg et al., Reference Friedenberg, Levy and Ross2012; Brighton and Bristowe, Reference Brighton and Bristowe2016). Operational barriers include lack of support, space, time, and effective training in interdisciplinary communication (Scholz et al., Reference Scholz, Goncharov and Emmerich2020). Difficulties in interdisciplinary communication have also been identified as a key barrier to effective EOL care, as each member of the healthcare team may play a vital role in supporting ACP (Grimshaw et al., Reference Grimshaw, Eccles and Lavis2012). Together, these barriers may partially explain why fewer than half of Americans have completed an AD (Yadav et al., Reference Yadav, Gabler and Cooney2017).
The gaps in effective ACP and EOL communication are even more pronounced for members of racial and ethnic minority groups (Yadav et al., Reference Yadav, Gabler and Cooney2017). Communication addressing cultural identity and spirituality has been shown to benefit ACP acceptance among minority group members (McDermott and Selman, Reference McDermott and Selman2018). There are limited ACP educational interventions that incorporate cultural competency using an interdisciplinary approach.
Researchers have identified key issues to be addressed in interventions focused on professional training in EOL care. There is a need for more research on interdisciplinary training of professionals in EOL care (Grimshaw et al., Reference Grimshaw, Eccles and Lavis2012). Training models should be able to be rapidly disseminated to frontline clinicians, potentially building on available training curricula. Incorporating opportunities to practice communication skills is valuable given the benefits of experiential learning (Xu et al., Reference Xu, Su and Deboer2019). Training would benefit from building competence in addressing cultural/spiritual concerns, given disparities in EOL care (Xu et al., Reference Xu, Su and Deboer2019; Ferrell and Rosa, Reference Ferrell and Rosa2021). Advance Directives-Live Action Simulation Training (AD-LAST) builds on the strengths of existing programs (Browne et al., Reference Browne, Braun and Mokuau2002; Doorenbos et al., Reference Doorenbos, Briller and Chapleski2003; Schim et al., Reference Schim, Doorenbos and Borse2006; Cruz-Oliver et al., Reference Cruz-Oliver, Malmstrom and Roegner2017; Ferrell et al., Reference Ferrell, Buller and Paice2019; Li et al., Reference Li, Smothers and Fang2019; Davis et al., Reference Davis, Lippe and Burduli2020). AD-LAST integrates each of these approaches by developing effective EOL communication using interdisciplinary teamwork, a culturally sensitive approach, using existing curricular tools and experiential learning.
Method
AD-LAST was developed to bridge the educational gaps identified in prior research. AD-LAST includes didactic teaching combined with active clinical simulation.
The program is novel in three ways. First, the AD-LAST program builds on five existing and widely disseminated curricular tools, including SPIKES (Setting, Perception, Invitation, Knowledge, Empathize, Summarize, and Strategize) (Buckman, Reference Buckman2005), NURSE (Naming, Understanding, Respect, Support, Explore) (Back et al., Reference Back, Arnold and Baile2005), ask-tell-ask (The Regents of the University of California, 2014), FICA (Faith, Influence, Community, Address) (Puchalski and Romer, Reference Puchalski and Romer2000), and FRIENDS (Familiarize with state forms, Routinize conversation, Introduce topic/Inquire past experiences, Elicit preference/Expect Emotions/Explore, Name proxy, Document, Summarize) (Pan et al., Reference Pan, Palumbo and Palathra2020). Additionally, participants were taught to use the phrase “tell me more about X,” an open-ended question to foster patient engagement and rapport building. These tools provide essential didactic information in a consistent format.
In combination, the first three tools (i.e., SPIKES, NURSE, and ask-tell-ask) were employed to improve clinicians’ competencies in three aspects of communication: to supportively deliver serious news to patients and families; to provide emotional support to patients and families; and to ask open-ended questions while providing therapeutic listening. The SPIKES tool equips clinicians with a communication roadmap to facilitate disclosure of difficult news to patients and/or surrogates. The NURSE mnemonic highlights various methods to name emotions, empathically support patients, and build a trusting, therapeutic relationship. The FICA assessment tool teaches clinicians how to assess faith orientation, community resources, and spiritual needs of patients. The FICA tool, in our setting, was modified to permit a structured assessment of aspects of cultural influences in patients’ lives relevant to medical decision-making. Specifically, AD-LAST participants received didactic teaching on the importance of cultural influences on medical decision-making, including the ways in which cultural background may affect strategies for disclosing serious news. The cultural assessment tool is based on the ELNEC curriculum (Matzo et al., Reference Matzo, Sherman and Mazanec2002). Lastly, the FRIENDS tool was used to summarize major learning points about the ACP process, including familiarizing oneself with state statutes and ACP forms, routinizing the ACP conversation, attending to emotional responses, naming proxy decision-makers, and documenting effectively.
Second, the program is delivered in an interdisciplinary context, providing opportunities for cross-discipline communication and problem-solving. This approach addresses difficulties in interdisciplinary relationships. Improving interdisciplinary relationships has been identified as a critical goal of effective EOL care training (Grimshaw et al., Reference Grimshaw, Eccles and Lavis2012).
Third, as recommended by researchers (e.g., Doorenbos et al., Reference Doorenbos, Briller and Chapleski2003) and necessitated by the diverse patient population in Queens, NY, AD-LAST focuses on cultural/spiritual issues in EOL care communication across diverse groups. This paper presents outcomes of the AD-LAST program, which integrates cultural competency skills and teamwork development into advance care communication skills training.
Participants
This study was regarded as exempt by the Institutional Review Board of the hospital. Clinicians from a wide range of disciplines were recruited throughout the hospital for the small group, one-day intervention. Participants were recruited via email, flyers, and personal contacts. While we strived to achieve interdisciplinary small groups, scheduling and staffing dictated group composition and presented a barrier to including multiple disciplines in all intervention groups.
Workshop format
AD-LAST workshops were small group seminars. Ideally, groups had no more than 8–10 participants, and 85.7% included fewer than 10. Each group was led by one to two experienced facilitators. The facilitators were palliative care physicians with at least 5 years of palliative care experience. These physicians received facilitation training through the GeriTalk Program (Gelfman et al., Reference Gelfman, Lindenberger and Fernandez2014). As shown in Figure 1, the first half (Part 1) of the one-day workshop focused on teaching communication skills, including the importance of introductions and forming a rapport with patients/families, medical decision-making, capacity assessment, ADs and state statutes, cultural/spiritual variations in decision-making, interdisciplinary teamwork process, and conflict resolution. Participants broke out into small groups to practice their learned skills in real-time. The second half (Part 2) of the day focused on applying communication curricular tools. The format for this component is similar to GeriTalk (Gelfman et al., Reference Gelfman, Lindenberger and Fernandez2014), but differs in the following ways: There was a presimulation clarification of team roles and a discussion on cultural/spiritual influences on ACP. Furthermore, during simulations, the actor was allowed to respond in organic ways to the participants. Participants set goals for learning, received instruction on providing effective feedback to peers, and practicing communication in real-time with standardized patient-actors. Participants had the opportunity to practice goal-setting, role negotiation, and medical case review in an interdisciplinary manner, a key skill that is often neglected in training. This allowed participants to develop a greater appreciation for the skills and expertise which colleagues from different disciplines bring to patient care and family support. Peer feedback and facilitated discussions highlighted the ways different disciplines can collaborate to optimize communication with patients/families.
Fig. 1. AD-LAST format.
Standardized patient cases
AD-LAST faculty created two standardized patient cases for the workshop. Case #1 centered around healthcare proxy discussions during an appointment with a patient recovering from cancer. This case permitted the application of skills from ask-tell-ask, FICA, FRIENDS, and NURSE. Case #2 involved an elderly man with Parkinson's disease who had sustained a fall at home which resulted in a hip fracture. This patient was now being treated in the surgical intensive care unit and had developed acute renal failure, sepsis, and delirium. Participants were asked to meet with this patient's wife (and later, the daughter) regarding ADs and goals of care discussions. This case allowed participants to practice skills drawn from SPIKES as well as ask-tell-ask, FICA, FRIENDS, and NURSE.
Measures
Both qualitative and quantitative evaluations of outcomes are performed.
Before and after the AD-LAST workshop, participants were asked to complete paper and pencil self-report measures of experience with aspects of EOL care and ACP knowledge, and self-efficacy. Two domains of knowledge were assessed. The first domain, “ACP Knowledge”, consisted of questions on informational aspects of ACP. The second domain, “Communication/Relationships Knowledge,” included case vignettes in which participants were asked to identify appropriate communication strategies. Self-efficacy questions covered self-evaluations of confidence conducting ACP and other EOL communications. Participants were asked to complete a program evaluation to measure satisfaction.
Analytic plan
Preliminary factor analyses (Proc Factor, SAS 9.4) were performed using iterated principal factors analyses with varimax rotation to examine if there were distinguishable dimensions of self-efficacy (e.g., confidence in communicating with patients vs. confidence in working in teams or confidence in discussing ADs). Repeated measures ANOVAs (Proc GLM, SAS Institute 9.4) were used to measure changes from pretest to posttest in knowledge and self-efficacy using average scores on each measure. To compare differences between physicians and nurses on each self-efficacy item, we employed a multivariate analysis of variance (MANOVA). Using MANOVA reduces the risk of type I error when conducting analyses on multiple outcomes (i.e., multiple items on the self-efficacy scale). Descriptive statistics were employed to describe participant satisfaction.
Results
A total of 163 clinical staff from multiple disciplines and specialties volunteered to participate in 21 AD-LAST training sessions offered between August 2015 and January 2019. Table 1 summarizes the characteristics of participants, who were diverse in terms of demographics and professional disciplines. Of the 21 intervention groups, 19% included one discipline, 9% included two disciplines, and in 72% there were three or more disciplines present for each workshop, despite scheduling and staffing challenges. All participants serve an ethnically and socioeconomically diverse patient population.
Table 1. Characteristics of sample: age, gender, race, and discipline
Note: For some questions, there are missing data.
a Race/Ethnicity data exceed 100%, due to some participants reporting more than one category.
* Although N = 163, some demographics do not total 163 due to missing information, or exceed 163 (for example race, where some participants identified with more than one).
Preliminary analyses
Factor analyses revealed the self-efficacy questions comprized a single factor. Therefore, we used the average of all items to create a self-efficacy scale with good internal consistency (α = 0.89).
Analyses of differences among professional groups were limited to comparisons between nurses (n = 23) and physicians (n = 102), the groups with the largest number of participants. At baseline, there were no significant differences between nurses and doctors in total knowledge (p = 0.80) or in self-efficacy (p = 0.77).
Changes in knowledge from pretest to posttest
In the full sample, repeated measures ANOVA indicated there was a significant increase from pretest to posttest in total knowledge (p < 0.001), ACP knowledge (p < 0.001), and communication/relationships knowledge (p < 0.001) (Table 2). Results on knowledge questions revealed that participants had an average of 77% correct responses prior to participating in the workshop and 86.5% correct responses post-workshop. There were no differences between nurses and physicians in changes from pretest to posttest in total knowledge acquisition (p = 0.76).
Table 2. Improvements in knowledge and self-efficacy
Notes: Self-efficacy means out of a maximum score of 4. Knowledge means out of a maximum score of 15.
* p < 0.001.
Self-efficacy
Repeated measures ANOVA showed a significant increase in pre to posttest self-efficacy overall (p < 0.001). The mean self-efficacy score across clinicians improved from pretest to posttest (Table 2). Univariate ANOVAs indicated there were significant improvements on every self-efficacy item, including those concerned with cultural issues (i.e., “identifying cultural issues affecting decision-making”) and interdisciplinary communication (i.e., “working in interdisciplinary teams”).
MANOVA with the Professional Group serving as a predictor (contrasting physicians to nurses) and the difference between pretest and posttest on each self-efficacy item serving as outcomes indicated a significant interaction of Professional Group × Self-Efficacy items (p = 0.012). Follow-up univariate analyses indicated four self-efficacy items for which nurses demonstrated greater improvements than physicians: “discussing bad news with a patient or family member” (p < 0.001), “discussing DNRs” (p < 0.01), “conducting patient/family goal-setting meetings” (p < 0.02), and “managing conflicts over medical decisions” (p < 0.04) (Table 3).
Table 3. Comparisons between nurses and physicians in components of self-efficacy
Note. N values may not always add up to total sample N due to missing data.
* p < 0.05.
** p < 0.01.
*** p < 0.001.
Relationship between changes in knowledge and changes in self-efficacy
Despite improvements in both knowledge and self-efficacy, Pearson's correlational analyses indicated that knowledge and self-efficacy were not related and represent independent dimensions of response to the intervention. Knowledge at either baseline or posttest did not predict self-efficacy at baseline or improvements in self-efficacy at posttest (r: 0.00–0.13, p: 0.10–0.98).
Program evaluation and participants’ comments
More than 96% of participants were highly satisfied with the workshop, agreeing it accomplished stated objectives, provided them with new information, improved their knowledge on ACP, and was pertinent to improving their practice. Participants agreed the course will change how they manage their patients.
In addition to nurses and physicians, participants from other disciplines also found AD-LAST to be an effective workshop.
Sample comments about the experiential training component included:
“[The training] mimicked ‘real-life- experiences that we all had - and perhaps have not known the most appropriate way to steer the conversation or break bad news.” — Speech Pathologist
“[AD-LAST] helped me practice dealing with spontaneous interactions with patients/family members.” — Social Worker
“I know how to initiate these discussions now, something I struggled with before.” — Physician Assistant
Sample comments about the benefits of the interdisciplinary teamwork component included:
“Instant feedback … allowed me to explore what other people would do in a situation.” — Resident Physician
“Getting perspectives from coworkers …” — Social Worker
“Everybody has the opportunity to practice and learn from each other.” — Patient Advocate
Discussion
AD-LAST was a didactic and experiential educational group intervention designed to teach effective clinical communication skills around ACP. The program targeted and successfully recruited interdisciplinary health professionals. To improve ease of dissemination, the program deployed existing packaged curricular tools. The components focused on enhancing skills in recognizing cultural issues in EOL care to address the needs of an increasingly diverse patient population. The AD-LAST workshop yielded a significant increase in ACP knowledge and self-efficacy, including in key areas concerning cultural communication and interprofessional teamwork.
Exposure to interdisciplinary learning early in ones’ professional progression shapes attitudes toward collaboration and promotes teamwork (Fineberg, Reference Fineberg2005; Grey et al., Reference Grey, Constantine and Baugh2017). A recent systematic review provides evidence that interdisciplinary teams improve ACP and quality of life for patients in palliative care (Ahluwalia et al., Reference Ahluwalia, Chen and Raaen2018). Yet, many healthcare professionals have not received training on how to collaborate with others in an interdisciplinary team. Some interdisciplinary training programs are limited to relatively few professional groups [e.g., medical and social work students (Fineberg, Reference Fineberg2005); medical and nursing students (Grey et al., Reference Grey, Constantine and Baugh2017)], and recent research suggests that outcomes may be improved by increasing the diversity of the professional disciplines (Ahluwalia et al., Reference Ahluwalia, Chen and Raaen2018). Another approach to interdisciplinary training included the ELNEC program, which is a comprehensive, intensive, and widely disseminated program to develop EOL communication skills among healthcare providers. The program was originally developed for nurses, and the bulk of evaluation focuses on nurses (Ferrell et al., Reference Ferrell, Buller and Paice2019; Li et al., Reference Li, Smothers and Fang2019; Davis et al., Reference Davis, Lippe and Burduli2020).
AD-LAST expanded on existing interventions and demonstrated the capacity to include multiple professional roles in training sessions. It met the goal of giving healthcare professionals from various disciplines the opportunity to train together with the joint goal of becoming better communicators. Almost three-quarters of the groups had members from three or more professions.
Experiential training can enhance interprofessional training, as role-play exercises can provide healthcare professionals with crucial experience prior to attempting EOL conversations with real patients (Fineberg, Reference Fineberg2005; Grey et al., Reference Grey, Constantine and Baugh2017; Smith et al., Reference Smith, Macieira and Bumbach2018; Ferrell et al., Reference Ferrell, Buller and Paice2019). AD-LAST provided meaningful experiential training. Partner dyads were formed during the training sessions to permit members to conduct small group communication exercises and simulated interdisciplinary family meetings. This practice allowed for interdisciplinary role negotiation and an opportunity to learn about colleagues’ areas of healthcare expertise. During the role plays, learners obtained feedback from members of each discipline. Qualitative responses indicated this was a meaningful component of the program.
Relatively few educational interventions have focused on decreasing health disparities in ACP within an interdisciplinary palliative care approach (Browne et al., Reference Browne, Braun and Mokuau2002; Doorenbos et al., Reference Doorenbos, Briller and Chapleski2003; Cruz-Oliver et al., Reference Cruz-Oliver, Malmstrom and Roegner2017). The Browne et al. (Reference Browne, Braun and Mokuau2002) curriculum aimed at improving cultural competency in EOL communication, but targeted social work students and not interdisciplinary healthcare professionals. Schim et al. (Reference Schim, Doorenbos and Borse2006) designed a novel educational intervention to enhance cultural competency amongst interdisciplinary team members. The program was limited to hospice providers, who tend to have prior experience practicing within a team-based setting. Cruz-Oliver et al. (Reference Cruz-Oliver, Malmstrom and Roegner2017) developed a novel telenovela program to improve EOL communication. This program was focused on Latino/a EOL patients. Although nurses and other professionals participated, the curriculum was not directed at building interdisciplinary collaboration to improve EOL care for diverse populations. AD-LAST included components focused on cultural competency, and the benefits were enhanced by the participants’ experiences with diverse patient groups and their own cultural diversity. Participants demonstrated significant increases in self-efficacy in communicating about cultural issues in ACP.
Although the AD-LAST workshop proved to be a successful tool in improving interdisciplinary professionals’ knowledge and self-efficacy regarding ACP, there was no relationship of gains in knowledge to improvements in self-efficacy. Knowledge alone may not produce the motivation and confidence needed to change behavior.
Research on the relationship of knowledge acquisition to self-efficacy and behavior change in clinical settings is limited. The Geritalk program, a communication skills acquisition training for geriatrics and palliative care fellows, has investigated the link between knowledge and behavior change. This program combines experiential training with a commitment to the deliberate practice of specific communication skills. Investigators demonstrated behavioral gains through pre/post family meeting assessments, which indicated that participants acquired greater than 20% skill improvement (Gelfman et al., Reference Gelfman, Lindenberger and Fernandez2014). Future research would benefit from the inclusion of behavioral observation in the assessment process.
Communication training has been compared to procedural training, commonly employed in surgical and other medical training (Nakagawa, Reference Nakagawa2015). However, unlike in surgical training, real-time feedback/coaching from a supervisor is often unavailable during difficult discussions with patients/families. With less immediate feedback in a real clinical setting, it may take longer for individuals to feel efficacious in conducting EOL conversations. It may be useful to consider standardized assessments for communication, in which communication procedures are logged, reviewed, and certified for competency for unsupervised-independent practice. Peer coaching may also be an effective way to promote interdisciplinary communication, transfer acquired knowledge into practice, and support peer colleagues (Ladyshewsky, Reference Ladyshewsky2010).
Limitations
Outcome measures are limited. Assessments of the effects of EOL conversations on the patient, family, and provider satisfaction levels are needed. ACP education at both levels (the provider and patient) is important to promote optimal patient care. Efforts to combine patient education plus provider-directed education are needed. Assessments of AD completion rates among patients whose clinical staff has completed the course would be important. Multiple disciplines were included despite scheduling challenges, but greater participation from a broader range of disciplines and specialties would be optimal. A pre-post design was employed; a stepped wedge approach may permit more controlled comparisons (Hemming et al., Reference Hemming, Haines and Chilton2015).
Conclusion
The AD-LAST workshop has promising outcomes and is feasible and effective, even in a busy hospital environment. The application of existing curricular tools suggests that wide dissemination is feasible. AD-LAST fills a training gap in interdisciplinary collaboration and culturally sensitive approaches to the ACP process.
Acknowledgments
The authors would like to acknowledge that the two principal facilitators for AD-LAST, Drs. Evgenia Litrivis & Gabrielle Goldberg, received training through the GeriTalk program, whose guiding principles are based on VitalTalk. VitalTalk is an interactive course that aims to improve clinicians’ communication skills and faculty facilitative teaching skills. We would like to thank The Collaborative Health Integration Research Program (CHIRP) at St. John's University for their support in research methods and data analysis. We would like to thank Dr. Blanca Sckell, who was pivotal in recruiting and assisting the facilitation of internal medicine residents’ participation in AD-LAST. We would also like to thank Elizabeth Schlesinger for her administrative coordination of the AD-LAST program. Lastly, this research was supported by a generous grant provided by the Y.C. Ho/Helen and Michael Chiang Foundation.
Conflict of interest
There are no conflicts of interest.
