Abstract
Objectives
Long-term care (LTC) is an important setting for goals of care (GoC) discussions. Understanding clinician barriers to GoC discussions could identify opportunities for LTC-specific interventions to improve quantity and quality of GoC discussions in the context of serious illness.
Design
A multicenter, cross-sectional survey study.
Setting and Participants
1184 LTC clinicians from 34 Ontario LTC homes were invited to participate.
Measures
The questionnaire assessed (1) clinician barriers related to the LTC resident power of attorney (POA), the health care provider, and the health care system; (2) willingness to engage in GoC discussions; and (3) suggestions to address identified barriers. Responses were rated on a 7-point scale (1 = extremely unimportant/unwilling, 7 = extremely important/willing). A linear mixed-effects model determined significance between mean importance ratings for each barrier and the willingness to engage in GoC discussion between physicians and nurses. A simple content analysis was performed on written suggestions to address GoC discussion barriers.
Results
The overall response rate was 49% (581/1184). The top 3 rated barriers were (1) POA's difficulty accepting their loved one's poor prognosis, (2) POA's difficulty understanding the limitations and complications of life-sustaining therapies, and (3) lack of adequate documentation of prior discussions with LTC resident or POA. Barriers related to the health care provider, and the health care system, were deemed statistically more important by nurses. LTC physicians were more willing to exchange information, be a decision coach, and participate in the final decision than nurses. Suggestions to improve GoC discussions include a dedicated team to have these conversations in LTC, and updating policies to mandate and standardize these conversations at all family meetings.
Conclusions and Implications
This study has identified key LTC clinician–identified barriers to GoC discussions. Developing targeted interventions to these barriers could be the foundation for developing new interventions that support high-quality GoC discussions.
Keywords
Goals of care (GoC) discussions between a person, their substitute decision-makers (SDMs), and the health care team are vital as they prepare SDMs to make future health care decisions when one is deemed to not have cognitive capacity to make such decisions. GoC discussions are an integral part, but not synonymous with, advance care planning (ACP). In long-term care (LTC), GoC discussions can be difficult as SDMs attempt to make important decisions while juggling their understanding of the person's life story and identity, personal feelings of guilt, and their relationship with the person.
1
This struggle is often made even more difficult when a person has dementia, and when decisions about life-sustaining treatments may be required during periods of transition, significant change, and clinical uncertainty.2
Further compounding this difficulty are assumptions that LTC clinicians may have: the SDM intimately knows the values, wishes, and goals of those they represent; that SDMs are aware of disease trajectory; or that every SDM is ready to engage in a GoC discussion.1
However, it has been established that engaging in conversations about GoC at the end of life (EOL) is important to improving satisfaction with care and the provision of value-congruent care in LTC.3
, 4
, 5
In the context of serious illness at EOL in LTC, the lack of high-quality GoC discussions to facilitate decision making, and the poor documentation of these discussions, can lead to poor care coordination. In 1 study, Cwinn et al reported that information gaps occur in 86% of all hospital transfers from LTC; most striking is that almost half of these gaps related to missing information about GoC.
6
This lack of continuity may result in LTC residents receiving care that is not consistent with their goals and potentially overly aggressive (eg, hospital admission and treatment escalation).7
, 8
, 9
Despite these factors, GoC discussions are feasible and acceptable in the LTC setting.
10
To increase the quantity and quality of GoC discussions in LTC, interventions guided by knowledge of the specific barriers and facilitators in this setting are needed. A search for ACP implementation literature in LTC noted few reports articulating barriers specific to this setting. When reports were available, applicability of results was limited because of methodology (eg, clinician barriers were not actively identified during the study or small sample sizes). Current barriers identified relate to the reluctance among staff and family members to discuss GoC, lack of physician availability, and legal uncertainties.11
Therefore, there is a need to address the paucity of evidence in order to support high-quality GoC discussions, and to facilitate the development of LTC home-level or system-level interventions. First, there is a need to better understand and quantify the context-specific barriers in LTC from the perspective of all health care professional (HCP) groups. More specifically, it would be important to describe profession-specific barriers, as each HCP group within LTC has a different clinical role. Lastly, it is important to understand whether different HCP groups had different levels of willingness to participate in GoC discussions. Therefore, we conducted a study to (1) elicit overall LTC clinician barriers to GoC discussions, (2) identify specific barriers that were significantly different between nurses vs physicians, and (3) compare the willingness of nurses vs physicians to participate in the GoC discussion process.Methods
Study Design and Setting
This was a multicenter survey of LTC clinicians working in LTC homes from 1 LTC corporation in Ontario, Canada. The LTC corporation policy is for each LTC resident to have a legally designated power of attorney for health care (POA) as their SDM; therefore, “POA” is used synonymously with “substitute-decision makers” or “SDMs” in the questionnaire and results.
Recruitment
All physicians and registered staff [eg, nursing, allied health professionals (AHPs), and administrators] working in participating LTC homes were eligible to participate. Forty LTC homes were invited to a study information webinar. The corporation also made an announcement supporting this project. An opt-in approach was used; 34 homes expressed interest to participate. Each participating LTC home was awarded a pro-rated honorarium based on site questionnaire completion rate. The honorarium was designated for ongoing LTC staff education purposes. The corresponding author's institutional research ethics board approved this study (project no. 2757). LTC administration did not have access to individual staff responses nor which staff participated in the study.
The DECIDE-LTC (Decision-making About Goals of Care for Long-Term Care Patients) Tool
The study questionnaire was adapted from a published hospital-based questionnaire on GoC discussion barriers.
12
The overall structure (eg, format, instructions, response scales, and conceptual categorical grouping of barriers) of the hospital-based questionnaire was maintained. The adaptation process sought to demonstrate questionnaire face and content validity. This involved an item generation phase through literature review and expert opinion; a cognitive interviewing phase with 10 active LTC clinicians (2 physicians, 2 family medicine resident physicians, 4 nurses, and 2 social workers); and a pilot study with 55 practicing LTC clinicians. Each adaptation phase refined the questionnaire leading to the final version of the DECIDE-LTC questionnaire used in this multicenter study (unpublished data). For respondents, GoC discussions were defined as a conversation between a capable LTC resident or a POA (for LTC residents without decisional capacity), and the health care team to establish the goals of treatment (eg, cure, prolongation of life, and comfort) and agree on the types of life-sustaining treatments that will (or will not) be used to achieve those goals (eg, CPR, mechanical ventilation, dialysis, intensive care unit admission, feeding tubes, and intravenous hydration).The questionnaire is divided into 4 sections. The first section assesses the importance of various barriers to GoC discussions as they pertain to the LTC resident or POA (10 items), the HCP (7 items), and the health care system (8 items). Space was provided for respondents to add additional barriers. Respondents rated the importance of each barrier on a 7-point scale (1 = extremely unimportant, 7 = extremely important). The second section is a free text section that requested respondents to provide their suggestions for solutions to address the barriers listed in the first section of the DECIDE-LTC tool. The third section assesses respondent willingness to participate in the 4 actions of the GoC discussion process (to initiate GoC discussions; to exchange information; to be a decision coach; and to participate in making a final GoC decision) using a 7-point scale (1 = extremely unwilling, 7 = extremely willing). The last section includes demographic questions including level of formal training and skill and priority for skill acquisition in performing GoC discussions.
The primary outcome is the identified barriers to GoC discussions in LTC as they relate to the LTC resident or POA, the HCP, and the health care system. Key secondary outcomes include the willingness to participate in GoC discussions and potential enablers to GoC discussions in LTC.
Questionnaire Administration
Participating LTC homes use an online learning portal as a cross-site platform for distributing education and announcements. Registered staff would log into their personal accounts and receive messages regarding mandatory online education modules and local reminders. This portal was used to distribute an invitation letter to all registered staff at participating LTC homes. The platform was then used to distribute the online link to the DECIDE-LTC tool, which was hosted by Lime Survey. For those LTC homes not using the portal, and for those staff (eg, physicians, other AHPs) without portal accounts, paper questionnaires with a prepaid return envelope were mailed to these LTC homes. The voluntary submission of a completed online or paper questionnaire to the research team was considered to be implied consent for participation.
A modified-Dillman method
13
,14
was used to maximize response rates. A study invitation letter was sent 2 weeks prior to the start of the study, then the questionnaire (online link or paper questionnaire), followed by 2 reminders. The online questionnaire remained active for 8 weeks (July 2017–August 2017); for paper questionnaires, 16 weeks (July 2017–November 2017) was given to account for postage delays.Data Collection and Analysis
Data from online and paper questionnaires were combined and analyzed with SPSS Statistics 25.0 and SAS 9.4. Comparison analyses were performed between physicians and nursing (including advanced nurse practitioners, registered nurses, registered and practical nurses) respondents to determine significant differences in importance ratings of each barrier, and willingness to participate in the different actions of the GoC discussion process. AHP respondents were excluded from this analysis because less than 50% of these respondents indicated their specific allied health profession. A linear mixed-effects model with a P value < .05 was used to determine significance; the LTC site was determined to be a random effect in order to control for the dependence between participants within the same LTC home. Categorical variables are described as counts, percentages, and medians; continuous variables are described as means with standard deviations (SDs).
Written responses in section 2 were collated, and responses that were not suggestions to address barriers to GoC discussions were removed. A simple content analysis was performed on the remaining responses.
15
Two authors (D.E., H.S.) independently coded each response into themes. The barriers listed in the DECIDE-LTC tool served as the initial themes; new themes identified were iteratively added. During the independent coding stage, there was 60%, 40%, and 20% consensus for system-level, LTC or POA resident, and HCP barriers, respectively. The 2 coding authors resolved coding disagreements through discussion without the need of the assigned third author (M.H.). The resulting final list of themes was then grouped by overarching category.Results
The overall response rate was 49% (581/1184); 374 of the 1184 surveys distributed were paper. One hundred forty-one questionnaires were excluded because fewer than 10 of the items were completed. Overall, 339 nurses, 30 physicians, and 20 AHPs responded; 51 respondents did not identify their profession. Half of the respondents (225/389) rated their GoC discussion skills as average or poorer, and 74% (266/358) indicated that improving their GoC discussion skills was an important priority. Table 1 presents the respondent cohort demographics.
Table 1Demographics of DECIDE-LTC Respondents (N = 440)
| Category | |
|---|---|
| Recruited participants, n (%) | |
| Male | 44 (10) |
| Female | 346 (79) |
| Did not answer or missing | 50 (11) |
| By profession, n (%) | |
| Physician | 30 (7) |
| Nursing (eg, advance nurse practitioner) | 8 (2) |
| Nursing (eg, registered nurse, registered practical nurse) | 331 (75) |
| Other allied health professional | 20 (5) |
| Missing or incomplete | 51 (11) |
| Age, y, mean ± SD | 44.9 ± 13.4 |
| Physician | 54.7 ± 14.6 |
| Nursing (eg, advance nurse practitioner) | 41.1 ± 7.8 |
| Nursing (eg, registered nurse, registered practical nurse) | 44.3 ± 13.1 |
| Other allied health professional | 41.6 ± 11.9 |
| Years in practice, y, mean ± SD | 12.8 ± 11.0 |
| Physician | 20.3 ± 14.6 |
| Nursing (eg, advance nurse practitioner) | 6.5 ± 6.3 |
| Nursing (eg, registered nurse, registered practical nurse) | 12.1 ± 10.4 |
| Other allied health professional | 15.6 ± 10.5 |
| Have received formal training in communicating with patients and families about goals of care, n (%) | 175/389 (45) |
| Physician | 15/30 (50) |
| Nursing (eg, advance nurse practitioner) | 2/8 (25) |
| Nursing (eg, registered nurse, registered practical nurse) | 148/331 (45) |
| Other allied health professional | 10/20 (50.0) |
| Self-rated skills in goals of care discussions | 389 (100) |
| Limited | 30 (8) |
| Fair | 31 (8) |
| Average | 164 (42) |
| Very good | 147 (38) |
| Expert | 17 (4) |
SD, standard deviation.
∗ Percentage of respondents by profession.
Table 2 displays the mean importance rating of each barrier by profession. The overall top 5 barriers, regardless of profession, were (mean, SD) (1) POA's difficulty accepting their loved one's poor prognosis (6.2, 1.0), (2) POA's difficulty understanding the limitations and complications of life-sustaining therapies (6.1, 1.0), (3) LTC resident not having any form of advance care planning (eg, advance directive) (6.1, 1.3), (4) not having adequate time to have conversations with LTC residents or POA (5.7, 1.4), and (5) lack of adequate documentation of prior discussions with LTC resident or POA (5.5, 1.5).
Table 2The Mean Importance Rating, by Health Care Profession, of Each Potential Barrier to Goals of Care Discussions in the Long-term Care
| Barrier Related To | Total (n = 389) | Physician (n = 30) | APN, RN, or RPN (n = 339) | Other AHP (n = 20) | P Value | ||||
|---|---|---|---|---|---|---|---|---|---|
| Mean | SD | Mean | SD | Mean | SD | Mean | SD | ||
| LTC Resident or POA | |||||||||
| Patient does not have any form of advance care planning (eg, advance directive) | 6.05 | 1.34 | 5.37 | 1.77 | 6.14 | 1.25 | 5.55 | 1.70 | .008 |
| Patient has an advance care plan but the plan cannot be applied to the current situation because it lacks sufficient detail to inform a goals of care discussion | 5.83 | 1.32 | 4.80 | 1.61 | 5.93 | 1.26 | 5.60 | 1.27 | <.001 |
| Lack of patient capacity to make decisions about goals of care | 5.91 | 1.18 | 6.00 | 1.02 | 5.93 | 1.17 | 5.50 | 1.50 | .73 |
| POA's difficulty accepting their loved one's poor prognosis | 6.21 | 1.02 | 6.30 | 1.06 | 6.22 | 0.99 | 5.85 | 1.35 | >.99 |
| POA's difficulty understanding the limitations and complications of life-sustaining therapies | 6.14 | 1.04 | 6.07 | 1.23 | 6.16 | 1.00 | 5.80 | 1.36 | .58 |
| Personal, religious, or cultural values of patients/family about life-sustaining therapies | 5.78 | 1.23 | 5.20 | 1.38 | 5.87 | 1.18 | 5.15 | 1.46 | .007 |
| Language barriers of POAs that make goals of care conversations difficult to conduct | 5.76 | 1.41 | 4.83 | 1.32 | 5.88 | 1.35 | 5.20 | 1.88 | <.001 |
| Lack of agreement among the POA and other family members about goals of care | 6.02 | 1.17 | 5.87 | 1.20 | 6.06 | 1.13 | 5.60 | 1.70 | .29 |
| Conflicts or disagreements between the POA and the health care team about goals of care | 5.92 | 1.28 | 5.77 | 1.41 | 5.96 | 1.23 | 5.40 | 1.78 | .41 |
| Lack of timely availability of POAs for discussions around goals of care | 5.79 | 1.25 | 5.43 | 1.04 | 5.88 | 1.22 | 4.95 | 1.76 | .17 |
| Health care practitioner | |||||||||
| My uncertainty in estimating the patient's prognosis | 5.61 | 1.54 | 4.73 | 1.31 | 5.73 | 1.51 | 4.75 | 1.68 | .005 |
| My lack of training to have these conversations (eg, not sure what to say, cultural sensitivity, etc) | 5.68 | 1.73 | 3.77 | 2.00 | 5.91 | 1.56 | 4.70 | 1.98 | <.001 |
| I feel too stressed or burnt out to engage in goals of care discussions | 5.36 | 2.01 | 3.30 | 1.99 | 5.59 | 1.88 | 4.55 | 2.31 | <.001 |
| My uncertainty about who or which health care provider is (most) responsible for having the goals of care discussion | 5.52 | 1.74 | 4.13 | 1.78 | 5.70 | 1.63 | 4.40 | 2.23 | <.001 |
| My long-term relationship with patient makes it difficult to change from intervention to palliation | 4.96 | 2.06 | 2.93 | 1.64 | 5.19 | 1.98 | 4.20 | 2.14 | <.001 |
| My fear that goals of care discussions will cause POAs to perceive that the health care team is “giving up” | 5.29 | 1.87 | 3.83 | 1.86 | 5.46 | 1.79 | 4.65 | 2.13 | <.001 |
| My desire to avoid repeating a discussion about goals of care already performed by another colleague | 4.98 | 1.94 | 3.27 | 1.34 | 5.20 | 1.89 | 3.85 | 1.98 | <.001 |
| Health system or external factors | |||||||||
| Lack of access to documentation of prior discussions with hospital or community-based physicians | 5.50 | 1.47 | 5.70 | 1.18 | 6.09 | 1.36 | 5.90 | 1.37 | .18 |
| Lack of adequate documentation of prior discussions with patient or POA | 5.45 | 1.32 | 5.47 | 1.22 | 6.14 | 1.25 | 5.70 | 1.06 | .007 |
| Not having adequate time to have conversations with patients or POA | 5.70 | 1.54 | 4.77 | 1.98 | 6.21 | 1.27 | 6.00 | 1.70 | <.001 |
| Insufficient remuneration for this activity | 5.10 | 2.15 | 4.20 | 1.94 | 5.57 | 1.85 | 4.60 | 2.50 | .003 |
| Current regulations from my professional association limits my role in having these discussions | 4.95 | 1.96 | 3.00 | 2.51 | 5.70 | 1.92 | 5.00 | 2.16 | <.001 |
| Current LTC policies limit my role in having these discussions | 4.65 | 1.74 | 3.22 | 2.42 | 5.58 | 1.92 | 4.70 | 2.00 | <.001 |
| My concern that goals of care designation forms will be incorrectly interpreted by some health care providers as consent for use or nonuse of life-sustaining treatments | 5.60 | 1.90 | 4.27 | 2.03 | 5.71 | 1.76 | 5.60 | 2.12 | <.001 |
| Lack of institutional tools or guidelines for having discussions about goals of care | 4.95 | 1.57 | 4.07 | 2.07 | 5.88 | 1.62 | 4.90 | 1.79 | <.001 |
AHP, allied health professional; ANP, advance nurse practitioner; RN, registered nurse; RPN, registered practical nurse; SD, standard deviation.
Responses are rated on a 7-point Likert scale (1 = extremely unimportant, 7 = extremely important).
∗ The ratings for the 51 respondents who did not indicate profession were excluded in the calculation of the overall mean importance of the barrier.
† Only the mean barrier ratings between physicians and nursing respondent groups were compared and reported in this table.
Nursing respondents provided higher mean importance ratings for all survey items except for “POA's difficulty accepting a LTC resident's poor prognosis.” The mean importance ratings for the top 5 physician barriers were not rated significantly different by nursing respondents. However, a significant difference was noted between physician and nurse ratings for barriers related to HCPs and the health care system. Nurses provided higher importance ratings to these barriers than did physicians. Specifically, barriers such as lack of any form of ACP, time constraints, and poor documentation of prior discussions were among the top 5 barriers identified uniquely by nurses.
One hundred forty-eight questionnaires included potential suggestions to GoC discussions. Table 3 presents the overarching categories and examples of suggestions to improve GoC discussions in LTC. Specific examples include (1) developing a standardized protocol and documentation strategy to ensure proper communication of previous ACP during transition periods; (2) using teleconferencing to ensure that all relevant HCPs, the POA, and the LTC resident are present for GoC discussions; (3) providing high-quality POA educational materials around prognosis and life-sustaining treatments; (4) advocating for administrative scheduling changes to protect time for LTC nurses; and (5) creating a dedicated team to routinely review GoC with the LTC resident and their POA outside of providing clinical care.
Table 3A Summary of 148 Respondent Suggestions to Address Barriers to Goals of Care Discussions in Long-term Care
| Overarching Category (Number of Suggestions) | Included Themes | Example(s) |
|---|---|---|
| Supporting values-congruent care provision across health care sectors (44) | Care planning; documentation and tools; patient-centered; administrative policies; pre-existing advance care plans | “[ACP] Must be communicated at hospital discharge and LTC admission as part of… [a] discharge/admission process…[need a] standardized protocol… Must not be accepted into LTC without prior documentation [of ACP]…” “One specific form for all sectors (Long-term care, hospitals…retirement residences…etc) to be transferred with the resident at all times…[it] is to be initiated by the first service provider.” “Having…accessible staff members in the care team to discuss resident's specific needs… to gain a holistic image of the resident…and implement plans specific to that resident.” “clearer…forms to document conversations and discussions held prior to admittance to [long-term care] and expressed wishes prior to…dementia progression.” |
| Supporting communication with long-term care resident or POA (36) | Communication; interpersonal relationships; team approach; resident factors | “Have translator available 100% of the time” “[ensure] all staff...including physicians…are talking the same talk. This is most important in end stage palliation to ensure family and friends…receive the same answers…” “ideally, the health care team and the family/POA/SDM would meet in person to discuss goals of care…where this is not always possible…for long period of time…videoconference…may be able to accommodate all parties…” “[Goals of care discussions] could be required…with annual medicals/quarterly med reviews…” |
| Supporting POA education (24 | POA knowledge resources (ie, prognosis, end-of-life treatments); POA role understanding | “Educating the family/POA/SDM about the patient/resident's prognosis…it may help to give examples of similar situations so they have a better understanding…about the…care that there loved one would most likely choose for themselves” “To help POA…it may be helpful to have pamphlets available to better understand treatments…they can look through and return and ask questions if needed...” |
| Supporting staff competence and capacity (44) | Training; consultants; funding; workload; timing of conversations | “Staff education about how to communicate with families…so that conversation is consistent, covers all necessary information…for other team members that may not have been involved in the discussion” “Provide a form or procedure…that has prompting questions in order to help guide the conversation. By having prompts, the conversation will be sure to cover important topics.…It would be helpful to have phrases that are known to be effective and sensitive for approaching such a delicate topic.” “Have a…care team available with a plan to support family and patient in advance so that support is available when required.” “Palliative training and courses for all staff as a team” “…if a [nurse] had a smaller group of residents that they were totally responsible for, it would lead to more personalized care and communication with families and POAs... impossible to know everything in the care plans of 50 residents.” “[With] a full shift… there is no time for anything more—have a specific day for…staff members to meet with each family to talk freely (without interruption…) about goals of care…every 6 months or every 3 months...” |
Table 4 presents the mean willingness of each health care profession to participate in the GoC discussion process. All respondents indicated a similar willingness to participate in initiating GoC discussions. Nurses were less willing than physicians to participate in exchanging information, being a decision coach, and in the final decision making about GoC. Some nurses indicated that they were neutral or unwilling to participate in being a decision coach or in making a final GoC decision, respectively [ie, 15% (50/339) and 14% (48/339)]. A small proportion of nurses felt that it was outside their scope of practice to exchange information (6%, 21/339), be a decision coach (3%, 10/339), or participate in a final decision-making process (4%, 14/339).
Table 4The Mean Willingness Rating of Each Health Care Profession to Participate in Each of the 4 Steps in the Goals of Care Discussion Process in Long-term Care
| Goals of Care Discussion Process | Overall (n = 389) | Physician (n = 30) | ANP/RN/RPN (n = 339) | Other Allied Health Professional (n = 20) | P Value | ||||
|---|---|---|---|---|---|---|---|---|---|
| Willing, n (%) | Mean (SD) | Willing, n (%) | Mean (SD) | Willing, n (%) | Mean (SD) | Willing, n (%) | Mean (SD) | ||
| To initiate goals of care discussions | 353 (91) | 6.0 (1.2) | 28 (93) | 6.3 (1.0) | 309 (91) | 6.0 (1.1) | 16 (84) | 5.8 (1.6) | .15 |
| To exchange information | 321 (83) | 5.9 (1.4) | 29 (97) | 6.7 (0.5) | 281 (84) | 5.8 (1.4) | 11 (58) | 5.6 (2.0) | <.001 |
| To be a decision coach | 324 (83) | 5.8 (1.3) | 30 (100) | 6.4 (0.7) | 279 (82) | 5.7 (1.3) | 15 (75) | 5.6 (0.9) | .01 |
| To participate in making a final decision about goals of care | 320 (82) | 5.8 (1.3) | 30 (100) | 6.5 (0.7) | 277 (82) | 5.7 (1.3) | 13 (68) | 5.6 (1.8) | .002 |
ANP, advance nurse practitioner; RN, registered nurse; RPN, registered practical nurse; SD, standard deviation.
∗ The results for the 51 respondents who did not indicate a profession were excluded from this table.
† The statistical significance reported in this table refers to the mean willingness ratings between physicians and nursing respondents.
Discussion
In this multicenter survey of 440 LTC clinicians, the top-rated barriers to GoC discussions in LTC were (1) POA's difficulty accepting their loved one's poor prognosis, (2) POA's difficulty understanding the limitations and complications of life-sustaining therapies, (3) LTC resident not having any form of advance care planning (eg, advance directive), (4) lack of adequate documentation of prior discussions with LTC resident or POA, and (5) not having adequate time to have conversations with LTC residents or POA.
Our results reinforce previously reported barriers to GoC discussions (eg, reluctance among staff and relatives to discuss GoC, lack of clinician training, and lack of proper documentation).
11
,16
, 17
, 18
, 19
, 20
, 21
, 22
Small sample sizes and lack of LTC experience in study participants limit the applicability of these previous findings to the LTC setting. The current study addresses these methodologic issues and has identified common barriers between different HCP groups; specifically, respondents commonly identified that issues with LTC resident or POA understanding of prognosis, and the options, benefits, and complications with treatments at the EOL, were the most significant barriers to GoC discussion.Interestingly, nursing staff identified the lack of time for GoC discussions as their top barrier, whereas physicians provided a significantly lower importance rating. This may be because the day-to-day responsibilities of LTC nurses prevent in-depth GoC discussions with families, whereas LTC physicians have the ability to more freely structure their clinical time and could consider GoC discussions as a routine part of their role. Further, nurses felt that the lack of remuneration was a more significant barrier than did physicians. Because nursing remuneration in Ontario is dependent on time spent providing clinical care and physician remuneration is less so, it would be consistent that nurses identified both time and remuneration as more important barriers than physicians. Therefore, providing additional remuneration to physicians may not necessarily incentivize them to engage in more discussions, whereas providing additional clinical time, and thereby remuneration, for nurses could incentivize increased engagement in GoC discussions.
Nurses also indicated that regulatory limitations and LTC policies were significant barriers to GoC discussions. Our nurse respondents included advance nurse practitioners, with the broadest scope of practice, to registered practice nurses, with the narrowest scope of practice. Registered nurses and registered practice nurses may wish to participate more in GoC discussions but could restrict participation as they feel that certain aspects of GoC discussions exceed their defined scope of practice (eg, being a decision coach or participating in making a final decision about GoC). LTC physicians are not similarly restricted; this difference in regulatory limitations could explain why, for physicians, relational and communication-based issues with the LTC resident or POA are the most important barriers.
A recent systematic review reported that decision-making tools such as educational videos and pamphlets could improve knowledge, ease decisional conflict, and improve clinical communication in a variety of palliative conditions.
23
These tools should be tailored for use in the LTC context with residents with multimorbidity, extreme frailty, and dementia, and with SDMs who could be involved in decision making. Therefore, LTC homes should be implementing LTC resident- or POA-facing decision aids as an evidence-based approach to address the barriers identified in this study around POA knowledge and preparedness.24
An example could be to adopt a standardized structured GoC discussion template; this template could provide educational opportunities to improve LTC resident or POA knowledge (eg, non–health care professional description of pathophysiology, prognosis, intervention and treatment options, and POA role and responsibilities). This template would also necessitate clinician knowledge and communication skills training, which would build capacity and competence.The successful use and implementation of LTC resident- or POA-facing tools would require local LTC policy and regulatory changes. Examples of LTC policy changes provided by our respondents include protecting time for LTC nurses, or by developing a local multidisciplinary LTC clinician team. This team could help LTC residents and POA work through and use decision aids, provide care continuity, streamline ongoing communication, and serve to advocate for care that is congruent to LTC resident goals, wishes, and values.
LTC clinicians overwhelmingly indicated a desire to improve their skills in GoC discussions. One specific example of improving GoC discussion skills could be to eliminate mutually exclusive language when proposing treatment options. For example, “We can treat issue X medically, which we hope may work, or we don't treat and your loved one will decline” could lead POAs to choose some form of treatment because they think they must do something. Meanwhile, their choice could lead physicians to think POAs lack understanding around prognosis and treatment options.
Addressing the quality and content of LTC GoC discussions by training with existing communication frameworks (eg, Serious Illness Conversation Guide
25
or VitalTalkSerious Illness Care: Ariadne Labs.
https://www.ariadnelabs.org/areas-of-work/serious-illness-care/
Date accessed: April 10, 2019
26
) could also lead to more effective discussions about important topics such as resident goals and values, prognoses, and the benefits and risks of life-sustaining therapies. In addition, providing clarity around practice scope in GoC discussions for LTC nurses could mitigate anxiety about overstepping boundaries; empowering nurses to practice to the fullest extent of their scope in GoC discussions could improve the rate of GoC discussions.VitalTalk.
https://www.vitaltalk.org
Date accessed: April 10, 2019
However, improving clinician skills would not address the top barriers reported in this study. The health care system needs to encourage primary care clinicians and patients to start engaging in GoC and ACP discussions well upstream of LTC admission. Empowering older adults to speak with their potential future POAs while they have decisional capacity is important, and would require implementing tools that improve older adults and their family's knowledge and preparedness to make health care decisions in LTC. Doing so could enable POAs to better advocate for value-congruent care when the LTC resident can no longer speak for themselves. This consideration would be consistent with best practice recommendations in patients with serious illness in more acute care settings
27
and could help to address the identified barriers and improve GoC discussions in LTC.Strengths and Limitations
Our study is unique as all respondents were LTC clinicians from multiple homes across a wide geographical area. Several limitations are that questionnaire responses are specific to Ontario, Canada. Therefore, health care systems in other provinces or countries may have other context-specific barriers resulting from government-level administrative policies. Second, respondents were predominantly nurses; as nurses are the most prevalent front-line LTC clinicians, our respondent demographics reflect the clinician distribution in LTC homes. A future study could plan to administer this questionnaire in a larger group of LTC physicians and AHP; this may shed more light onto professional and scope-based differences in important GoC discussion barriers. Lastly, personal support workers (PSWs) were excluded from this study; the investigator group decided that because of their scope of practice, they would not have the experience necessary to identify clinical barriers to GoC discussions. However, PSWs spend the most time with LTC residents and may be the first to notice clinical decline in LTC residents. Therefore, ensuring that GoC and symptom management plans are effectively and appropriately communicated to PSWs, and acknowledging their observations of clinical change would be important to ensuring high-quality care for LTC residents.
Conclusions and Implications
In this study, the highest-rated clinician barriers to GoC discussions related to the lack of preparedness of LTC residents and POAs to have GoC discussions, specifically, the POA's difficulty accepting the LTC resident's poor prognosis, and their difficulty understanding the limitations and complications of life sustaining therapies. System-level barriers identified by nursing respondents related to poor documentation of previous conversations and the lack of time to engage in proper GoC discussions with residents and their POAs. Developing interventions that target the most important common barriers could form the foundation of local quality improvement initiatives. The results of these initiatives could impact policies at the local and corporate LTC level, and other policy makers to support high-quality GoC discussions in LTC. Ultimately, these initiatives will increase the quality and quantity of GoC discussions and, subsequently, the quality of EOL care provided to LTC residents.
Acknowledgments
The authors acknowledge Dr John J. You for his insights and contribution during the initial stages of this study.
References
- End of life care: The experiences of advance care planning amongst family caregivers of people with advanced dementia—a qualitative study.Dementia (London). 2016; 15: 958-975
- Advance care planning and end-of-life care for hospitalized nursing home residents.J Am Geriatri Soc. 2002; 50: 829-835
- Vital conversations with family in the nursing home: Preparation for end-stage dementia care.J Soc Work End Life Palliat Care. 2014; 10: 112-126
- The positive association of end-of-life treatment discussions and care satisfaction in the nursing home.J Soc Work End Life Palliat Care. 2015; 11: 307-322
- End-of-life conversations and hospice placement: Association with less aggressive care desired in the nursing home.J Soc Work End Life Palliat Care. 2017; 13: 61-81
- Prevalence of information gaps for seniors transferred from nursing homes to the emergency department.CJEM Can J Emerg Med Care. 2009; 11: 462-471
- Decision making in the treatment of acute illness in nursing homes: Framing the decision problem, treatment plan, and outcome.Med Anthropol Q. 1995; 9: 236-256
- Approaching Death: Improving Care At the End of Life.National Academy Press, Washington, DC1997
- Consequences of Hospital Restraint Use for Older Nursing Home Residents.(The Gerontological Society of America 50th Annual Scientific Meeting) Gerontologist, Cincinnati, Ohio1997: 252
- Conversations about goals and values are feasible and acceptable in long-term acute care hospitals: A pilot study.J Palliat Med. 2017; 20: 710-715
- A review of the implementation and research strategies of advance care planning in nursing homes.BMC Geriatr. 2016; 16: 24
- Barriers to goals of care discussions with seriously ill hospitalized patients and their families: A multicenter survey of clinicians.JAMA Intern Med. 2015; 175: 549-556
- Mail and Internet Surveys: The Tailored Design Method.Wiley, New York, NY2000
- Development of a standard e-mail methodology: Results of an experiment.Public Opin Q. 1998; 62
- Three approaches to qualitative content analysis.Qual Health Res. 2005; 15: 1277-1288
- Advance care planning: Understanding clinical routines and experiences of interprofessional team members in diverse health care settings.Am J Hosp Palliat Care. 2017; 34: 946-953
- Nurses' knowledge of advance directives and perceived confidence in end-of-life care: A cross-sectional study in five countries.Int J Nurs Pract. 2016; 22: 247-257
- Advance care planning in the skilled nursing facility: What do we need for success?.Hosp Pract (1995). 2011; 39: 85-90
- Advance care planning in care homes: The experience of staff, residents, and family members.Int J Palliat Nurs. 2013; 19: 550-557
- Advanced care planning in care homes for older people: A qualitative study of the views of care staff and families.Age Ageing. 2011; 40: 330-335
- Nurses' views regarding implementing advance care planning for older people: A systematic review and synthesis of qualitative studies.J Clin Nurs. 2015; 24: 2057-2073
- Advance care planning in care homes for older people: An English perspective.Palliat Med. 2009; 23: 332-338
- Tools to promote shared decision making in serious illness: A systematic review.JAMA Intern Med. 2015; 175: 1213-1221
- “Now I don't have to guess”: Using pamphlets to encourage residents and families/friends to engage in advance care planning in long-term care.Gerontol Geriatr Med. 2017; 3 (2333721417747323)
- Serious Illness Care: Ariadne Labs.(Available at:)https://www.ariadnelabs.org/areas-of-work/serious-illness-care/Date accessed: April 10, 2019
- VitalTalk.(Available at:)https://www.vitaltalk.orgDate accessed: April 10, 2019
- American College of Physicians High Value Care Task Force. Communication about serious illness care goals: A review and synthesis of best practices.JAMA Intern Med. 2014; 174: 1994-2003
Article info
Publication history
Published online: October 28, 2019
Footnotes
This work was supported by the Canadian Institutes of Health Research (CIHR; Grant PHE-135930). The CIHR was not involved in the study design, data collection, data analysis, interpretation, or decision to submit this manuscript for publication. A.V. was an employee of Extendicare Assist at the time of the study; however, Extendicare Assist did not provide any financial support for this study. Extendicare Assist was not involved in study design, data collection, data analysis, interpretation, or decision to submit this manuscript for publication. All other authors declare no conflicts of interest.
Identification
Copyright
© 2020 The Authors. Published by Elsevier Inc. on behalf of AMDA - The Society for Post-Acute and Long-Term Care Medicine.
User license
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
