Addressing Patient-Reported Barriers to Advance Care Planning in Family Practice

Background Although patient-centred care has become increasingly important across all medical specialties, when it comes to end of life care, research has shown that treatments ordered are not often concordant with people’s expressed preferences. Patient and family engagement in Advance Care Planning (ACP) in the primary care setting could improve the concordance between patients’ wishes and the healthcare received when patients cannot speak for themselves. The aim of this study was to better understand the barriers faced by older patients regarding talking to their family members and family physicians about ACP. Methods In this multi-site cross-sectional study, three free text questions regarding reasons patients found it dicult to discuss ACP with their families or their family physicians were part of a self-administered questionnaire about patients’ knowledge of and engagement in ACP. The questionnaire, which included closed ended questions followed by three probing open ended questions, was distributed in 20 family practices across 3 provinces in Canada. The free text responses were analyzed using thematic content analysis and form the basis of this paper. Results 102 participants provided an analyzable response to the survey when asked why they haven’t talked to someone about ACP. 254 answered the question about talking to their physician and 340 answered the question about talking to family members. Eight distinct themes emerged from the free text response analysis: 1. They were too young for ACP; 2. The topic is too emotional; 3. The MD should be responsible for bringing up ACP 4. A fear of negatively impacting the patient-physician relationship; 5. Not enough time in appointments; 6. Concern about family dynamics; 7. It’s not a priority; and 8. A lack of knowledge about ACP. Conclusions Patients in our sample described many barriers to ACP discussions, including concerns about the effect these discussions may have on relationships with both family members and family physicians, and issues relating to patients’ knowledge and interpretation of the importance, responsibility for, or relevance of ACP itself. Family physicians may be uniquely placed to leverage the longitudinal, person- centred relationship they have with patients to mitigate some of these barriers.

may not be able to communicate these preferences at the time decisions are needed. [11,12] If they have not engaged in the process of advanced care planning (ACP) with family members and health care providers, these preferences cannot be known.
ACP is a communication process wherein people plan for a time when they cannot make decisions for themselves. It includes re ection, deliberation, and determination of a person's values and wishes or preferences for future treatments and at the end-of-life. ACP should include communication between an individual, their family/loved ones, future substitute decision-maker(s) (SDM), and health care provider(s) about these values and wishes. [13] Prior engagement in ACP has been shown to improve patient and family experiences with healthcare near the end of life, resulting in greater concordance between patient wishes and the healthcare they receive, as well as lower stress and depression among family members. [14,15] The public tends to think of ACP as only the process of creating a written advance directive for future health care [16]. These advance directives tend not to be used by clinicians, [17][18][19] and patients creating them tend to lack important knowledge relating to the treatment options described. [20] There is recognition that multiple behaviors are involved in successful ACP, notably conveying values and wishes to family members or friends even beyond the substitute decision-maker, and discussing wishes and future plans with health care professionals [21][22][23].
Patient and family engagement in ACP in the primary care setting could better prepare patients and families to make future healthcare decisions. In our previous research surveying patients in primary care about their ACP engagement, half of patients were aware of ACP and had talked to someone about it, however only 18% had talked to a health care professional. [24] Yet the process of ACP requires patients to integrate information about health conditions with their own values and preferences to ultimately be prepared for future decisions. The majority of Canadians have longitudinal relationships with their family physicians who provide care across the life cycle. [25] In addition, family physicians are well acquainted with their patients' health conditions and needs and they care for their patients in the context of family units and support systems.
If we are to implement interventions to increase the uptake of ACP discussions as part of routine primary care, we need to understand the barriers patients face in having these discussions with their family members and family physicians. [26] As part of multi-site prospective audit of ACP in primary care settings, we conducted a content analysis of open-ended responses to a survey of patients regarding their engagement in ACP to gain a deeper understanding to the question: "What are the barriers faced by older adult patients regarding talking to their family members and family physicians about ACP?"

Setting and Design
This multi-site cross-sectional study, conducted from October 2014 to March 2015 involved a selfadministered questionnaire about patients' knowledge of and engagement in ACP. The survey was conducted in a convenience sample of 20 family practices (i.e. with de ned patient populations, not walk-in clinics), recruited by several of the investigators: thirteen from Ontario, ve from Alberta, and two from British Columbia. Research assistants were present in the practices to provide participants with the questionnaires and to answer questions if they arose.

Participants
Full details of the survey methods have been published previously . [24] Brie y, staff or clinicians were asked to invite consecutive eligible patients to participate on days the research assistant was present.
Eligible patients were 50 years of age and older, could read and speak/understand English, and did not have a cognitive impairment. We chose the age of 50 based on literature suggesting the need for early engagement in ACP [27] and alignment with routine primary care delivery in this age group (e.g. screening). [28] Questionnaire Details of questionnaire development and administration have previously been described. [24] The questionnaire asked whether patients had heard of ACP previously (a de nition was provided),whether they had engaged in four key ACP behaviors including: 1) thought about the kinds of medical treatments they would want or not want if they were to get very sick and be in hospital, 2) talked to anyone about their wishes (and to whom), 3) written down their wishes, and 4) formally nominated a substitute decision maker, and nished questions about barriers to discussions.
Questions also included sociodemographic characteristics and a clinical frailty scale. [29] Patients selfcompleted the questionnaire. We report on the free-text responses to three questions: 1. If the respondent indicated they had not spoken with anyone about ACP: "Why haven't you talked with someone?" 2. "What is the one thing that makes it very hard for you to talk to your family doctor about medical treatments at the end of life?" 3. "What is the one thing that makes it very hard for you to talk to your family members about medical treatments at the end of life?" Analysis Thematic content analysis was used to analyze free text responses. Five members of the research team (CB AT MS NS GA) independently reviewed the free text responses line by line and then categorized the emerging themes into groups in an iterative manner. The group then inductively developed a preliminary coding framework over multiple readings and discussions. These initial ndings were discussed with the senior author (MH) to reach consensus on key themes and sub themes. These themes were reviewed by the research assistants (DE, NA) who had experience addressing any questions posed by the research participants while completing the survey. In a nal meeting of the rst two authors, the analysis was further re ned through consensus and this analysis was presented to the remaining authors for review. The codes, themes, and subthemes were created and modi ed throughout the entire research process with full participation of 4 of the authors, ensuring that the rigor and trustworthiness of the data were maintained through a transparent and iterative process.
The study was approved by the research ethics boards of each participating academic institution. Research assistants received informed written consent at the time of administration

Participants
Most of the 20 participating family practices were group clinics and most employed allied health professionals as part of their team. Of the 878 patients who agreed to meet with the research assistant, 810 (92.2%) completed the survey. The mean age of patients was 66 years old (range 50-95 years) and 56% were female (440/809) ( Table 1).
[Insert Table 1] Most patients identi ed as White/Caucasian (88%; 713/810) and the majority of patients (88%; 711/810) self-reported being 'very t', 'well', or 'managing well' on the frailty scale. 439 respondents provided answers to the open ended questions. A total of 696 comments were analyzed. Thematic Analysis 102 participants provided an analyzable response when asked why they haven't talked to someone about ACP. 254 provided an analyzable response to the question about talking to their physician and 340 provided an analyzable response to the question about talking to family members Eight distinct themes emerged from the free text response analysis (See Table2).
[Insert Table 2 ] Themes which emerged across all questions A belief that the participant is too young/too healthy for ACP Participants described feeling too young or too healthy to be thinking about ACP. Indeed participants seemed to believe that ACP should occur only after they were diagnosed with a life-limiting illness and that it was inappropriate to consider aspects of ACP when they perceived themselves to be 'young'. Confusion around the age at which starting ACP might be appropriate was evident, with one participant writing, "Did not see this as necessary, it's a bit soon. I am only 80!" The topic is too emotional for discussions The fact that discussing ACP has the potential to bring up strong emotions was described as a deterrent to having the discussion. The topic was described as sad and depressing, and there was fear associated with discussing the end-of-life. It emerged that these conversations were perceived as being too emotional not only for the respondent ("it's a little frightening to talk about dying") but also for the family, as participants reported being worried about "causing emotional pain to family members".
Believing that ACP is the MD's responsibility (not the patient's, not the family's, the MD should bring it up) Participants noted that someone else, other than the participant, should be primarily responsible for initiating ACP discussions. In response to the question about why participants did not discuss ACP with family members there was a clear theme that ACP should be the doctor's responsibility, not the family's. Participants indicated that although they felt comfortable discussing ACP with their MD, they saw it as the MD's role to initiate and lead the discussion: "If a doctor wanted to talk to me about the end of life, it would not be hard for me to do".
Themes which emerged relating to discussions with family physicians Fear of a potential negative impact on the relationship with the physician It emerged that participants were concerned that having an ACP discussion might adversely affect the relationship with the doctor. One concern was that the doctor's values may not align with the patient's values, making the conversation di cult: "Not knowing if his beliefs and values are the same as mine". Other's worried that the doctor may try to in uence their decisions in a way that did not align with their wishes: "I may be in uenced to sway my thoughts". Many respondents indicated they were concerned that it might be inappropriate to bring up such personal issues in a doctor's appointment "it's a personal family matter".

Not enough time in the appointments
Insu cient time in appointments with family physicians emerged as a barrier to ACP discussions. Participants reported that there were signi cant time constraints on appointments which "do not allow for an in depth conversation about something that is not happening now or may not ever happen. Appointments are too short". Further, they seemed to worry about whether this type of appointment may take time away from others: "Doctor is always too busy. I feel I will impose on his time to talk about this when he has other patients waiting".
Themes which emerged relating only to discussions with family members Concern about family dynamics Participants expressed concern about how an ACP discussion may affect family dynamics. Participants described a worry that such a discussion would cause con ict within the family and they wanted to avoid such con ict: "I don't want any arguments arising". Analysis brought forth concerns about how discussions might affect one's spouse, or one's children. Some participants did not want to discuss ACP because of dynamics and lack of trust among family members, which may affect the ACP discussion: "I don't trust them very much to leave this kind of decision to them".
Themes which emerged relating to the general question about ACP discussions with anyone It's not a priority A key theme that ACP was not a priority for participants arose. This is exempli ed by quotes such as: "[I'm] too preoccupied" and "[I have] thought about it but just haven't acted on it yet". I don't know enough about ACP When participants explained that they did not know about ACP, many of them went on to add that they felt that someone else would take care of it anyway "Did not know about [advance] care planning; I think my family will know what to do".

Conclusions
Through qualitative analysis of free text responses to open ended questions about what makes it di cult to have ACP discussions, we found several overall themes relating to patients' ACP discussions with physicians and family members. The emergent themes included barriers as perceived by the patients to ACP discussions, concerns patients have about the effect these discussions may have on relationships and dynamics with both family members and family physicians, and issues relating to patients' knowledge and interpretation of the importance, responsibility for, or relevance of ACP itself.
The centrality of the unique patient-family physician relationship has the potential to mitigate some of the challenges to advance care planning we uncovered in our study. Our study indicates that even when patients are indeed ready to have a conversation about ACP, they often expect their family doctor to let them know when to broach the topic. Patients who believe that they are "too young" or "too healthy" likely do not understand that ACP is necessary to prepare their SDM to speak for them at any point when they may not be able to speak for themselves; this may not be in the context of end of life care but applies to acute serious reversible conditions. Thus, it seems that it is essential that family doctors take responsibility for initiating these discussions and for educating patients about the importance of ACP in many different circumstances.
Although in our study patients expressed an expectation that their physician would let them know when to discuss ACP, in a previous study some patients in primary care felt they, themselves, should initiate the discussion. [30] Even when this is the case, patients who believe that they are "too young" or "too healthy" for ACP may not initiate ACP conversations early enough to be of bene t. Given that family physicians regularly engage in proactive medicine and bring up issues which patients do not bring up themselves as a usual part of care, including ACP as a routine topic in preventive care and screening discussions would be an appropriate place to start. [31,32] Family physicians also care for their patients within the context of their families, sometimes caring for several family members in one family, and attend to patients' physical and emotional needs. Helping a patient to navigate through the fears and worries identi ed in our study regarding discussing ACP, either by working with the patient in the context of their family or by engaging other members of the health care team is a meaningful contribution that most family physicians could ful ll. As noted above, by educating patients on the misconception that ACP is only related to end-of-life decision-making, family physicians can help quell the heightened emotions that patients may fear will arise in ACP discussion. Further, helping patients navigate through the purposeful re ection of their values and wishes with regards to care can be empowering for patients.
Making ACP discussions a regular part of periodic health preventive visits has the potential to change cultural perceptions about ACP and, help patients to see this as routine preventive care. Perhaps our ndings will empower family physicians who believe that patients or families should initiate ACP discussions [33] to instead make the rst move and introduce ACP to their patients. Education on effective resources already developed, such as the "Just Ask" tool [34] and "Speak Up" [34,35] as part of the Canadian Advance Care Planning website (http://www.advancecareplanning.ca/resource/just-ask/), and the PlanWell© website (https://planwellguide.com/) are tools that can both remind and facilitate family physicians in their efforts to initiate these conversations with their patients.
The nding that patients were concerned about how an ACP discussion might affect their relationship with their family physician suggests that efforts are needed to normalize such discussions as part of routine clinical care for the patient and for the practitioner. We found that patients worry about disagreeing with the family doctor, being 'unduly in uenced' by an ACP discussion, and taking too much of the 'doctor's time'. This stands in stark contrast to the commitment to patient-centredness which rests at the core of the doctor-patient relationship in family medicine. [4,6] Family physicians may need to be attuned to these concerns expressed by patients and reassure patients of the advisory role that the family physician plays in facilitating patients to make fully informed decisions that are in keeping with the patient's own values and goals.
Both patients and family physicians are aware of how current clinical care and relationships are in uenced by time pressures and con icting priorities. Family doctors are aware that more time is needed to allow for ACP discussions. [33,[36][37][38] Indeed time has been described as the "ultimate barrier" to creating the best possible circumstances for a "good death" in one study of family physicians' experiences of con ict with substitute decision-makers. [39] Given the ongoing longitudinal relationships that exist between family physicians and their patients, seeing ACP as a process as opposed to a onetime conversation could help to manage the time-related pressures perceived in the context of individual appointments. Strategies which include engaging patients before the discussion, use of clinical decision support tools (e.g. https://planwellguide.com/), and involving other primary care team members can also help to address this issue. [40][41][42][43] Patients can be engaged outside the health care setting to be better prepared for discussions with their physicians. For example, when diverse older adults were randomized to self-directed use of an interactive website to engage in ACP, they were more likely to raise the issue of ACP with their primary care provider compared to people who were given an advance directive form only. [44] Given the evidence that patients can be activated to engage in ACP, funding models need to be adjusted to enable family physicians to spend the time needed for ACP discussions [39] and to create innovative initiatives to engage patients prior to family physician appointments.
Previous work revealed that patients did not perceive ACP as relevant to them because they were healthy [21] and that the topic brings up di cult emotions among people with whom they have relationships such as family. [21,45,46] In addition, the belief that patients' family members will know the patient's wishes is not uncommon. [16,45] Yet the presence of decisional con ict and regret is known to be high in family members of critically ill patients [47] and negative emotional burden [48] and burden of decisionmaking is lower when ACP has been undertaken [49]. Our study shows similar nding but provides new insights into the role r family physicians in engaging and educating patients about the importance of ACP as it may pertain to the involvement of others in decisions about their health care. Having these conversations in the context of preventive visits may help to change culture so that patients see choosing an SDM and sharing goals and wishes as something usual and necessary to prepare for any time where a patient may lose the capacity to participate in decision-making and not related only to end of life care.

Strengths and limitations
A strength of this study is that it was conducted across multiple family practices and geographies. The free text nature of our questions allowed for additional information to be gathered to elaborate on categorical responses to questions about ACP behaviours. The themes derived from the data could guide the creation of closed-ended response options in future questionnaires assessing ACP engagement and barriers.
Our study also has limitations. Not all participants provided responses to the open-ended questions and we cannot be certain that the results would represent the views of all participants. In addition, participants self-completed the survey and we were unable to ask for clari cation of ideas expressed and most responses were short. The participating practices and patients were not randomly selected, which may reduce generalizability. In addition, the patients were required to speak and understand English, and only 12% identi ed themselves as other than white/Caucasian, which is not representative of the contemporary Canadian population. [50] Finally, the de nition of ACP which was provided to patients "re ection, deliberation, and determination of a person's values and wishes or preferences for future treatments at the end-of-life" may have directed their attention towards considering ACP as something one only considers near the end of life. The newer consensus de nition from 2017: "Advance care planning is a process that supports adults at any age or stage of health in understanding and sharing their personal values, life goals, and preferences regarding future medical care" [22] may have yielded different responses Implications Our ndings indicate that lack of awareness about ACP in all contexts and not seeing ACP as a priority were key challenges among patients who reported a barrier to ACP discussions. They further indicate that, even when aware of the importance of ACP, patients may not feel that it is their role to initiate the discussion with their physician or family members.
It may be possible to address these nding through the unique longitudinal relationship that family physicians have with their patients and through a focused effort to bring the ACP conversation into the forefront of normal preventive care Patients clearly need education about the appropriateness of ACP discussions for adults at any stage of health so that their SDM's are well prepared to speak on their behalf should it be necessary at any point in time. Public health campaigns highlighting the 2017 consensus de nition [22] could help the population as a whole to better understand the importance ACP for all adults. Further, materials aimed at engaging patients should be available in family medicine o ces, encouraging patients to be ready for such discussions with family physicians as a usual aspect of care for all adults.
Family physicians need to see initiating ACP discussions as a part of the routine preventive care they provide to their patients. Although these discussions may seem daunting in terms of the time they may require, seeing ACP as a process that happens over many appointments and engages others including the patient's family and the overall health care team can help address these time-related concerns.
Further, providing family physicians with appropriate resources they can provide to patients, such as "Speak Up" handbooks, could help structure the ACP process, allowing patients to spend time in between appointments re ecting on their goals and values.
Although ACP discussions should unfold over multiple visits, it is important to ensure that patients feel that there is adequate time in appointments to work through the sometimes di cult discussions relating to prognosis and treatment choices. Given that there is evidence that patients can be activated to participate in ACP discussions, it is essential that funding models be adjusted to allow adequate time for these discussions to unfold [39].
With better patient education, a preventive focus situated in the ongoing longitudinal family physicianpatient relationship, and appropriate funding models, ACP discussions could happen at appropriate times and could help patients receive the care that respects their goals and values in their future care or if they are not able to speak for themselves.

Declarations
Ethics approval and consent to participate: All participants were provided with research ethics boardapproved study information.