Keywords

Introduction

Lung disease is a common cause of morbidity and mortality with care trajectories and symptom burden comparable to those associated with cancer [1, 2]. Despite these similarities, studies have shown that patients with lung disease, such as chronic obstructive pulmonary disease (COPD), lack an awareness of death and dying compared to those with cancer. Data indicates that frequency, timing, and quality of palliative care communication between clinicians and this patient population are poor. This is problematic because palliative care communication done well improves patients’ symptoms and quality of life, leads to better concordance between patient goals and care delivered, and improves family psychosocial outcomes.

Communication surrounding palliative care rarely occurs in care of patients with advanced lung diseases [3]. As an example, a systemic review showed that most studies exploring palliative communication report rates of discussion regarding treatment preferences and end-of-life issues in less than 30% of patients with COPD [4]. Only a small proportion of patients with moderate to severe COPD have discussed treatment preferences and end-of-life care issues with their physicians, and the vast majority of these patients believe their physicians do not understand their preferences for end-of-life care [5]. When discussions do occur, the quality is often rated by patients as low [4] due to the absence of important elements such as prognosis, anticipatory guidance, and spirituality [6]. Patients with COPD are more likely than those with HIV/AIDS and cancer to express concern about the lack of education that they receive about their disease, treatment, prognosis, and advance care planning [7]. Clinicians acknowledge that discussion of goals of care and end-of-life preferences should occur early in the illness and when the patient is stable [8]. However, when discussions do occur, they often take place at advanced stages of the illness or hospitalization when patients may be unable to participate in decision-making, a situation that is burdensome to their surrogate decision-makers.

Herein we will review the existing literature on why effective palliative care communication is important for patients with advanced lung diseases, their families, and the healthcare system, discuss common barriers encountered by clinicians, and consider potential solutions to those barriers.

Why communication is important for patients, their families, and the healthcare system:

Effective palliative care communication includes both a [1] discussion of the values and preferences that shape end-of-life care and [2] exploration of factors contributing to patient suffering and symptom management.

For patients with lung disease, discussions regarding end-of-life care frequently occur at advanced stages of illness when patients are often unable to participate in decision-making. In the absence of early goals of care discussions, patients are at risk to receive unwanted medical care resulting in prolongation of the dying process and undue burden on surrogate decision-makers. When patients are decisionally incapacitated, surrogate decision-makers are asked to collaborate with clinicians to assist in making emotionally and cognitively difficult decisions. This surrogate role has been associated with negative emotional outcomes, such as stress, guilt, or doubt, in at least one third of surrogate decision-makers [9]. Anxiety, depression, and post-traumatic stress disorder have all been described in surrogate decision-makers [10,11,12,13]. Families of patients who died in the ICU were found to have higher levels of both PTSD and depression when they experienced discordance between their preferred and actual decision-making roles [14]. Furthermore, a cross-sectional study of bereaved family members found that perceptions on quality of end-of-life care are associated with complicated grief [15] which is known to be associated with poor physical health due to diseases, such as cancer, heart disease, and hypertension, and poor psychosocial outcomes, including anxiety, depression, and suicide [16, 17].

These poor outcomes for surrogate decision-makers are largely due to the burden of making high stakes decisions without knowledge of patients’ wishes. This burden may be alleviated by clarifying patients’ preferences early in their disease course. One systematic review found that when surrogates knew which treatments were consistent with the patient’s preferences, it reduced negative outcomes for surrogate decision-makers such as feelings of guilt and stress. In this study, surrogates who were confident in their loved one’s treatment preferences felt as though they were simply reporting the patient’s preferences as opposed to deciding on behalf of their loved one [9]. Other studies have demonstrated that end-of-life discussions were associated with better caregiver quality of life and bereavement adjustment at follow-up [18]. Additionally, terminally ill patient’s prognostic awareness has been found to be associated with a higher quality of death reported by their caretakers who were also more physically and mentally healthy 6 months post-bereavement [19].

Physicians’ ability to engage in goals of care discussions also contributes to outcomes for healthcare providers and the healthcare system. Caring for patients with advanced lung diseases who are seriously ill, suffer repeated exacerbations, and have limited treatment options makes clinicians vulnerable to moral distress and burnout. One study demonstrated that delivering bad news to patients can contribute to burnout among physicians who feel inadequately trained in communication skills [20]. Training interventions to improve end-of-life communication skills have shown significant improvements in confidence in communicating, attitudes toward psychosocial care, and sense of personal accomplishment [21]. Furthermore, the perception of providing harmful or futile care leads to moral distress and loss of empathy [22] which serves as a common cause of clinician burnout [23, 24]. Physician burnout has significant consequences on physician health with increased rates of depression [25], substance dependence [26], and suicidal ideation [27]. In addition to a moral and ethical cost, there is an economic cost to burnout. Cross-sectional studies have uncovered associations between physician burnout with decreased productivity, increased turnover, and increased medical errors resulting in poorer quality of care [28,29,30]. Each of these issues presents financial implications for healthcare organizations that subsequently incur significant losses [31, 32].

Early goals of care conversations have the potential to mitigate these patient, surrogate, and clinician outcomes. A substantial field of research has established that engaging patients in conversations about end-of-life care is associated with improvement in patient-centered outcomes. This is because a patient’s values and preferences help to frame their medical decisions. When patients elect a surrogate decision-maker or complete an advanced directive, they are more likely to have their wishes known and followed [33]. One study found that end-of-life discussions were associated with less aggressive medical care near death and earlier referral to hospice, which in turn were associated with not only better patient but also caregiver quality of life and bereavement adjustment at follow-up [18]. As an example, in a multisite longitudinal study, patients with advanced cancer who reported awareness of their terminal prognosis had lower rates of psychological distress and higher rates of advance care planning [19]. Other studies show that patients with cancer are more likely to receive end-of-life care that is consistent with their preferences when they have had the opportunity to discuss their wishes with a physician. Patients aware of their terminal prognosis were more likely to desire symptom-focused care [34]. While most existing data focuses on patients with advanced cancer and COPD, patients with other advanced lung diseases and their families are likely to benefit as well.

Patients with advanced lung disease experience debilitating physical and emotional symptoms leading to loss of independence and the ability to fulfill social roles. Palliative medicine specialists approach symptoms as multifactorial, influenced not only by physical but also psychological, social, and spiritual suffering [35]. These symptoms are often best addressed by a multidisciplinary Palliative Medicine Team. However, nonspecialists with experience in the traditional biologic model for symptom assessment and treatment can begin the process of identifying symptoms and attempting treatment with both non-pharmacologic and pharmacologic therapies. When patients’ symptoms are not assessed, distressing experiences such as breathlessness, fatigue, anorexia, pain, and mood disorders often go untreated with detrimental consequences on quality of life. A systematic review found that distressing symptoms present at end of life are equally prevalent for patients with end stage COPD as among advanced cancer patients. Specifically, among patients with COPD, 90–95% experience breathlessness, 68–80% fatigue, and 34–77% pain [36]. Early palliative care integration for seriously ill patients with lung cancer has been shown to be associated with improvements in both mood and quality of life [37]. In a meta-analysis, palliative care interventions delivered by a range of palliative care and non-palliative care specialists were found to be associated with improvements in symptom burden and quality of life [38].

The healthcare system also benefits from early, skillful palliative care communication. Clinicians who feel insufficiently trained in communication skills have been shown to have a higher prevalence of depersonalization and low personal accomplishment than those who perceived themselves to be sufficiently trained [39]. A qualitative study exploring oncology physicians’ approach to end-of-life care found that those who viewed end-of-life communication and care as an important role reported increased job satisfaction and decreased burnout [40]. Communication skills training is associated with less burnout and work-related stress [20, 41]. Studies directly examining cost at end of life associated with early involvement of palliative care have shown mixed results. One study found that among patients with advanced cancer, those that reported discussing their end-of-life wishes with their physician had better quality of death and significantly lower healthcare costs in their final week of life [42]. A landmark study published in 2010 showed that among patients with metastatic non-small cell lung cancer, early palliative care referral led to less aggressive care at end of life and longer survival [37]. While the study did not measure costs, these findings suggest that timely introduction of palliative care has the potential to mitigate unnecessary societal costs [37].

In response to accumulating data, medical societies, such as the American Thoracic Society, recommend that clinicians who care for patients with chronic or advanced respiratory diseases should be trained in, and capable of, providing recommended basic competencies in palliative care which include communication in goals of care and symptom management [43].

Barriers and Solutions

Palliative care communication including discussions about goals of care and symptom management has important consequences for patients, families, and healthcare providers. However, there are significant barriers to providing consistent, skillful communication in these areas. Herein we will present these barriers and discuss potential solutions.

Prognostication

Clinicians view prognostication as one of the most difficult parts of their profession [44], and many are reluctant to provide patients and their families with prognostic information due to fear of erring or losing credibility [45]. Uncertainty regarding prognostication and, therefore, identification of which patients are most likely to benefit from end-of-life conversations are common challenges in patients with advanced lung diseases.

Many forms of advanced lung diseases follow the disease trajectory prior to death commonly seen in organ failure [46, 47]. Similarly to patients with heart failure, the course is one of overall functional declines punctuated by intermittent decompensation (Fig. 9.1). During each episode of decompensation, patients deteriorate significantly and either succumb to their disease or recover to a lower level of functioning. It is difficult to predict the outcome of each deterioration. Further complicating the clinical picture, chronic organ failure may represent a comorbidity that patients die with and not of. Physicians frequently cite difficulty in prognostication as a barrier to communication about palliative care and end of life in patients with organ failure [4, 48,49,50,51].

Fig. 9.1
figure 1

Functional decline prior to death due to organ failure. (Lunney et al. J Am Geriatr Soc) Acknowledgment wording: reproduced with permission of the © ERS 2020: Curtis [50]

Full size image

Findings from the SUPPORT study demonstrated the difficulty in prognostication of patients with chronic organ failure due to COPD compared to those with lung cancer. Using the Acute Physiology and Chronic Health Evaluation (APACHE) II model, they found that at 5 days prior to death, patients with lung cancer were predicted to have <10% chance of surviving 6 months and those with COPD were predicted to have >50% chance [52]. Despite the development of prognostic models specific to lung disease, prognostic accuracy remains challenging. One such prognostic model is the BODE (body mass index, airflow obstruction, dyspnea, exercise capacity) index. Of patients receiving the highest score on the BODE index, 62.3% were still alive at 3 years [53]. While this represents an improvement from broad prognostic models, such as APACHE II and FEV1 alone, it is still not predictive of 6-month prognosis.

One potential solution to the challenge of accurate prognostication is shifting focus away from survival estimates and toward identifying patients who are relatively well and may benefit from end-of-life conversations. Universal factors associated with ≥50% 6-month mortality in maximally treated disease include poor performance status, advanced age, malnutrition, comorbidities, organ dysfunction, and hospitalization for acute decompensation. Societies have proposed various options for identifying which patients are likely to benefit, all emphasizing the importance of early discussion. One such example was suggested by Curtis et al. to prompt discussion about end-of-life care for patients with COPD (Table 9.1) [50]. More broadly, clinicians can ask themselves “Would I be surprised if this patient died within the year?” An answer of “no” should prompt a discussion. This question has been shown to be moderately sensitive in predicting 1-year mortality, with a pooled sensitivity of 67% in meta-analysis [54]. Medicare eligibility for hospice is shown in Table 9.2. These criteria were found to have both poor sensitivity and specificity in identifying patients who would die within 6 months [55].

Table 9.1 Characteristics that should prompt discussion of end-of-life preferences for patients with COPD
Full size table
Table 9.2 Disease-specific hospice eligibility criteria for patients with COPD (each criteria required)
Full size table

Each change in clinical status represents an opportunity to reassess the patient’s quality of life, determine whether it remains acceptable, and consider with the patient what brings value to their life. In the absence of definitive prognostic information, which is nearly uniformly elusive in clinical medicine, it is important to acknowledge the uncertainty of each patient’s future when communicating about end of life.

Perception or Fear of Harming the Patient or Provoking an Emotional Response

Clinicians caring for patients with advanced illness report fear of harming the patient by sharing upsetting prognostic information or discussing end-of-life care. In one study, 23% of physicians caring for patients with COPD cited concerns that discussing end-of-life care will take away their patient’s hope, and 21% felt the patient was not ready to talk about the care they would wish for if they got sicker [56]. Some patients do not wish to receive their physician’s prognostic information or discuss advance care planning and end-of-life preferences. In some cultures, it is believed that communicating unfavorable prognoses to the patient becomes a self-fulfilling prophecy. However, existing data suggests that most patients wish to receive this information and expect their physician to initiate discussions [57]. Absence of communication about prognosis has been associated with prolongation of the dying process, undesired hospitalizations, and patient mistrust in the healthcare system [34]. Research has also demonstrated that clinicians’ fear of removing patients’ hope is largely unfounded. In fact, nondisclosure is associated with decreased quality of life and increased rate of depression in survivors [58]. Patients and families with overly optimistic prognostic estimates were more likely to die following receipt of aggressive medical interventions, such as intubation or cardiopulmonary resuscitation, with no difference in outcomes [59].

Prognostic information provides a framework for patients to make informed treatment decisions and consider important aspects of life care planning, such as legacy planning, addressing personal relationships, and saying goodbye. By balancing hope and reality, clinicians can potentially allow their patients to shift focus toward other, perhaps more attainable, goals while continuing to hold the hope for a better outcome. Clinicians can mitigate the potential for inflicting harm by eliciting patients’ preferences for prognostic information. A proposed strategy is detailed in Table 9.3 [60].

Table 9.3 Discussing prognostic information
Full size table

Emotional distress associated with receipt of bad news can further be lessened by ensuring discussions take place in an appropriate setting, delivering the information tactfully, allowing the appropriate amount of time free of interruptions, and providing emotional support through verbal and nonverbal expressions of empathy. (See Section “Lack of Comfort with Palliative Care Topics”)

Lack of Comfort with Palliative Care Topics

Medical education has long been deficient in end-of-life care and improving physicians’ competencies has increasingly become a topic of focus. In a national survey, both medical students and residents perceived preparation for providing care at end of life was worse than other common clinical tasks [61]. Additionally, a national survey of internists revealed that 56.8% reported inadequate training in prognostication [44]. Teaching of communication skills should include focused skills practice and be integrated with biomedical training. Most clinicians have not received the skills training required to provide high-quality palliative care communication or teach others how to do so.

Symptom Management

Lung disease is a common cause of morbidity with similar symptom burden compared to cancer [1, 2]. However, patients with lung disease are less likely to have their symptoms elicited and, therefore, appropriately managed. Dyspnea is the most common symptom experienced by patients with advanced lung disease; however pain and mood disorders are also frequently encountered. The experience of dyspnea increases in frequency and severity as death approaches. The primary goal in managing dyspnea is optimizing treatment of the patient’s underlying disease. For patients with refractory dyspnea, there are options for non-pharmacologic and pharmacologic therapies. When dyspnea is refractory to non-pharmacologic management, systemic low-dose opiates are the first-line medication, serving as a safe and effective pharmacologic treatment for refractory dyspnea.

Despite numerous studies documenting the safety of opiates for treatment of dyspnea in patients with advanced lung disease, there remains stigma associated with prescribing opiates, and patients suffer from poor control of their dyspnea [62,63,64]. Commonly cited source of discomfort in prescribing includes insufficient knowledge, lack of experience, and fear of adverse effects [62, 63]. In one of these studies, physicians with palliative care experience were more comfortable prescribing opiates [63]. Interestingly, a study from France showed that despite high rates of refractory breathlessness among COPD patients and reported comfort among the majority of clinicians in prescribing opiates, the rates of prescribing opiates within this population were low [64]. These findings may suggest assessment and treatment of dyspnea lies in a failure to assess breathlessness and provide treatment. Based on these findings, clinicians may benefit from a trigger during outpatient visits to assess common distressing symptoms in patients with advanced lung disease.

Empiric studies have demonstrated that short-term use of opiates reduces breathlessness in patients with a variety of conditions, including COPD and interstitial lung disease [65,66,67]. A randomized controlled trial showed that sustained-released morphine had a positive effect on disease-specific health status in patients with moderate to very severe breathlessness [68]. Importantly, opioids have not been shown to significantly reduce oxygen saturation, raise arterial carbon dioxide, or reduced respiratory rate; however patients retaining carbon dioxide were excluded in most of these studies [69]. In 2011, the American College of Chest Physicians released a consensus statement recommending the use of opiates for relief of dyspnea in patients with advanced cardiopulmonary disease in consideration of comorbidities and titrated to individual effect [70]. Knowledge about this literature may help clinicians feel more comfortable engaging their patients in conversations about symptom management. Additionally, providers may consider referral to specialty palliative care for patients with advanced lung disease with multiple comorbidities or a baseline elevation in carbon dioxide experiencing refractory dyspnea.

Goals of Care Communication Skills

Specialized clinicians from a variety of fields that frequently care for patients with life-limiting illnesses report feeling inadequately trained to conduct end-of-life conversations [71,72,73,74]. Physicians specifically report difficulties with communicating prognoses, facilitating end-of-life decision-making, and managing the emotional responses these conversations naturally uncover [75, 76]. In response, national leaders and experts in communicating with seriously ill patients have developed multiple frameworks for approaching these core communication skills.

Each of these communication tasks is recommended to take place in a comfortable and private place. Every attempt should be made to minimize interruptions and allow an appropriate amount of time. Prior to engaging in the conversations, clinicians should ask whether the patient would like to have support persons present and assess preferences regarding receipt of prognostic information.

Responding to Emotion

The experience of advanced illness and receipt of medical information challenging patient’s hopes and expectations for the future evokes intense negative emotions. While clinicians may be unable to restore a patient’s health, research demonstrates that providing support lessens the experience of emotional distress. Additionally, when patients are experiencing emotions, they are unable to process medical information effectively. Attending emotions will also bring them into the person’s awareness and allow them to shift away from them toward a cognitive state capable of planning complex behaviors and decision-making. Lastly, studies demonstrate that patients have higher trust for the clinicians when they respond to emotion which leads to greater shared decision-making [77].

Clinicians can show emotional support by listening and expressing empathy both verbally and nonverbally. The acronym N-U-R-S-E [78] summarizes how to respond verbally to emotion (Table 9.4).

Table 9.4 Verbal responses to emotion
Full size table
Giving Serious News

Sharing serious news is difficult for clinicians who struggle with feelings of apprehension at the prospect of causing distress, fear of provoking an emotional response, and professional failure when unable to protect their patients from disease. Despite these difficulties, patients require an understanding of the medical situation to make value-based treatment decisions and plan for the future.

Evidence-based recommendations for giving serious news are organized into the Ask-Tell-Ask framework summarized in Table 9.5 [79]. The first “Ask” refers to eliciting the patient’s understanding of the medical situation. This is followed by “Tell,” consisting of labeling the news to give them a chance to emotionally prepare for the news. When the news or headline is then shared, clinicians should be concise, use simple language, and refer to outcomes that are meaningful to the patients. The last “Ask” is to check understanding or check in with the patient.

Table 9.5 Delivering serious news
Full size table
Engaging in Decision-Making

VitalTalk designed REMAP as a road map for goals of care conversations to increase the quality and efficiency of conversations and make communication skills easier to learn [80]. The acronym serves a conversation guide toward reaching a patient-centered and shared decision. Table 9.6 details each step with accompanying example statements. The conversation begins with clinicians ensuring the patient or their surrogates are on the same page in terms of medical information. Sometimes, clinicians will have additional news to communicate (see sharing serious news above.) Throughout the conversation, clinicians will need to recognize and respond to emotion, particularly after communicating difficult news. If the patient or their surrogate is on the same page as clinicians and is prepared to talk further, the next step is to explore values and preferences in light of the medical situation. Throughout the conversation, it is also recommended to reflect what you are hearing to ensure you are aligned with the patient and family. Lastly, after gathering sufficient information, VitalTalk recommends at least three mapping questions, you can offer a patient-centered recommendation based off the information provided.

Table 9.6 Goals of care discussions
Full size table

Palliative Care Communication in a Fragmented Healthcare System

Over the past decade, patient care has become increasingly complex. An individual’s exposure to healthcare is likely to be spread across multiple providers, practice settings, and even healthcare systems. The resulting diffusion of responsibility and difficulty accessing prior medical documentation make iterative communication about goals of care and symptom management challenging and compound the barriers to communicating about palliative care (Table 9.7).

Table 9.7 Summary of barriers and potential solutions for clinicians when communicating about palliative care issues
Full size table

We know that patients often expect their providers to initiate discussions regarding prognosis and goals of care [51]. However, prior studies have demonstrated that a physician’s decision to share prognosis is often based on whether the patient made a specific request [57]. A survey of general practitioners caring for patients with COPD in the United Kingdom found that 41% reported discussing prognosis often or always and 15% rarely or never and 30% left it for patients or their relatives to raise the subject [81]. There is also ambiguity among providers about who is responsible, specialists or general practitioners. A systematic review and narrative synthesis found that no single group of healthcare providers felt that their roles, relationships with patients, or work setting made them the most appropriate to have goals of care conversations with their patients with COPD [48]. Primary care physicians report uncertainty regarding their role in goals of care discussions when specialty providers are involved [51].

Further challenges arise with the documentation of goals of care conversations. A variety of clinicians may engage patients with advanced lung diseases in conversations about goals of care without communication or documentation to guide further discussions as the patient continues along the trajectory of their illness. Even if a clinician documents goals of care conversations, it may not be easily retrieved or accessible to providers from other healthcare systems. Among patients who had previously completed an advance directive, only 26% had it recognized during a hospitalization [82]. There is no standardization for the location of advance care planning or goals of care documentation [51]. A study exploring the location of advance care planning documentation within an electronic health record showed only 33.5% of patients with documentation had a scanned document and the remainder were within a progress note or problem list [83]. Lastly, the situation may be further complicated when visits with outpatient providers and hospital exposures occur across different healthcare systems with distinct EHR that do not communicate.

When multiple clinicians are caring for a patient, the oncologic literature recommends collaboration among providers to decide who will be primarily responsible for discussing prognosis and end-of-life care then ensuring the other providers are aware of the outcome of discussions [84]. Healthcare systems would benefit from a designated site within the EHR where an iterative account of the patient’s wishes, values, and preferences in light of the current medical situation can be located. Ideally, this site would be readily accessible across healthcare systems.