Introduction
A life-limiting diagnosis is devastating for the patient and their family to receive. Yet the advantage of timely truth disclosure regarding a life-limiting illness enables a patient and their family to be involved in informed decision-making and physically, mentally, and spiritually prepare for the end-of-life (Kogan et al., Reference Kogan, Enguidanos and Penido2013; Becker et al., Reference Becker, Jors and Block2015). The health professional needs to explain bad news in language that the patient and family can understand while considering the impact of cultural and spiritual preferences (Mostafazadeh-Bora and Zarghami, Reference Mostafazadeh-Bora and Zarghami2017). Such a conversation requires empathy, tact, intuition, and skill (Bousquet et al., Reference Bousquet, Orri and Winterman2015) and can be one of the most challenging tasks for health professionals to undertake (Berkey et al., Reference Berkey, Wiedemer and Vithalani2018).
Difficulty in breaking bad news may result from the doctor's fear of emotional responses from patients and families, a lack of time for detailed dialogue, inadequate training in end-of-life discussions, or not being comfortable with prognostication (Berkey et al., Reference Berkey, Wiedemer and Vithalani2018; Johnston and Beckman, Reference Johnston and Beckman2019). Patients’ or family members’ reactions to bad news may be intensified by various factors such as age and family responsibilities (Ptacek and Eberhardt, Reference Ptacek and Eberhardt1996), faith, culture, past experiences, and symptoms of the disease (Mostafazadeh-Bora and Zarghami, Reference Mostafazadeh-Bora and Zarghami2017).
Receiving bad news in a healthcare context refers to any information which will negatively alter a person's view of their future, yet the definition of “bad” is proportionate to the patient's understanding of the life-limiting illness and may differ from the reality as understood by the health professional (Buckman, Reference Buckman2005).
The shift away from a paternalistic medical culture to patient autonomy in which a patient has the right to know about their diagnosis and prognosis has become a legal and ethical requirement in Western countries (Goldberg, Reference Goldberg1984; Bousquet et al., Reference Bousquet, Orri and Winterman2015), yet in practice, the truth can be withheld, distorted, and manipulated by health professionals for various reasons (Fallowfield et al., Reference Fallowfield, Jenkins and Beveridge2002; Sarafis et al., Reference Sarafis, Tsounis and Malliarou2014). Health professionals are faced with ethical dilemmas of “do no harm” at the same time as respecting patients’ autonomy (Beste, Reference Beste2005) especially when patients are unable to contribute to decision-making due to their physical or emotional state (Sarafis et al., Reference Sarafis, Tsounis and Malliarou2014).
Although there is literature discussing the delivery of bad news and various communication models to assist health professionals, examining truth disclosure within bad news is not as prolific. Therefore, this qualitative meta-synthesis aimed to explore and examine how health professionals, patients, and families experience truth disclosure when delivering or receiving bad news in the inpatient/outpatient palliative care setting.
Methods
Like the meta-ethnography technique described by Noblit and Hare (Reference Noblit and Hare1988), the meta-synthesis aims to use an inductive and interpretive methodology to synthesize and combine the included papers with going beyond the original interpretations (Britten et al., Reference Britten, Campbell and Pope2002; Thomas and Harden, Reference Thomas and Harden2008). The meta-synthesis was undertaken in three phases: a systemized search, quality appraisal, and synthesis.
Search strategy
A systemized search was undertaken in September 2020 following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines (Liberati et al., Reference Liberati, Altman and Tetzlaff2009). With the aid of a librarian, a scoping search was undertaken using (life limit* OR life threatening OR fatal OR terminal) diagnosis OR prognosis AND disclosure of information and repeated using (communication). Of the papers that matched the selection criteria and the paper's aim, the keyword “truth disclosure” appeared as a commonality among the papers. With that in mind, the search process was recommenced using a combination of keywords and MeSH terms using the CINAHL, MEDLINE, and PsycInfo databases. MeSH terms of (“truth disclosure”) were used with CINAHL and Medline and (“truth”) with PsycInfo. Keywords included (“palliative care or end-of-life care or terminal care or dying”) NOT (“children or adolescents or youth or child or teenager”) NOT (“ICU or intensive care unit”) NOT (“emergency department or emergency room or accident and emergency or a&e or a & e”). Restrictors were set to retrieve peer-reviewed articles written in English and published between 2013 and 2020. An age restrictor for “adults” was also added. The search was repeated, combining the three databases and the terms (“bad news”) with the keywords and restrictors. In addition to searching the databases, a comprehensive hand search of the available literature was conducted by looking at reference lists. A total of 212 articles were retrieved, including two hand-searched articles, and once duplicates were excluded (n = 156), all three authors screened 56 articles for relevance by reading titles and abstracts and excluded 12 articles. The remaining 44 articles were then read in full and assessed against the inclusion/exclusion criteria (see Table 1), with the final eight papers being eligible for a quality assessment. Figure 1 outlines this process. The rationale for limiting the search to the above keywords, period, cohort, paradigm, and hospital department was to examine the latest papers discussing truth disclosure during a bad news conversation within a restricted environment, looking at a specific behavior.
Fig. 1. PRISMA flow diagram.
Table 1. Inclusion and exclusion criteria
Quality appraisal
The Consolidated Criteria for Reporting Qualitative Research (COREQ): 32-item Checklist for Interviews and Focus Groups (Tong et al., Reference Tong, Sainsbury and Craig2007) was used to assist the authors in examining the papers’ rigor and methodological designs. Within the Research Team and Reflexivity domain, there was limited information regarding the researchers’ characteristics, and none of the eight papers clearly disclosed the researchers’ relationship to the participants. This was also reflected for each paper when assessed using the Critical Appraisal Skills Programme (CASP, 2020), and as a result, each paper scored ≥90%. While some inadequate methodological reporting was evident, no paper was discarded on this basis as all papers yielded important findings and provided in-depth analysis and discussion of their findings (Dixon-Woods et al., Reference Dixon-Woods, Shaw and Agarwal2004). Table 2 details each paper's characteristics.
Table 2. Characteristics of included papers
Synthesis
The purpose of a meta-synthesis is to develop new concepts by going beyond the findings of the original papers (Britten et al., Reference Britten, Campbell and Pope2002; Thorne et al., Reference Thorne, Jensen and Kearney2004; Thomas and Harden, Reference Thomas and Harden2008). The papers were synthesized based on Noblit and Hare's (Reference Noblit and Hare1988) phases of conducting a meta-ethnography. The first author circled pertinent words, drew links between them, and annotated phrases throughout the Findings and Discussion sections of each paper. This became an iterant process in which concepts or third-order constructs were developed, as per the adaptation for health research by Britten et al. (Reference Britten, Campbell and Pope2002) and detailed in Table 3. Following the terminology of meta-ethnography (Noblit and Hare, Reference Noblit and Hare1988), first-order constructs are the participants’ quotes, second-order constructs are the themes generated by the original authors, and third-order constructs are the development of new concepts as a result of translating the papers into one another. The second and third authors examined the development of the third-order constructs and pursuant concepts until agreement and consensus were reached.
Table 3. Summary of emerging concepts
Results
Description of included papers
Eight papers were included in the meta-synthesis and were represented by five Western countries being Sweden (n = 2), UK (n = 2), USA (n = 2), Canada (n = 1), and The Netherlands (n = 1). Data collection arose from semi-structured interviews with patients (n = 2), nurses (n = 2), doctors (n = 1), recorded consultations between doctors and patients (n = 2) in which only one included the patients’ voices in the data, open-ended questions from a survey with nurses (n = 1), and focus groups (FGs) followed by semi-structured interviews with doctors, nurses, and dieticians (n = 1). The first FG consisted of (n = 1) dietician, (n = 2) nurses, and (n = 1) specialist palliative care nurse. The second FG consisted of (n = 2) nurses, (n = 1) dietician, (n = 1) specialist palliative care nurse, and (n = 1) doctor. Twenty-five individual semi-structured interviews then followed with (n = 10) nurses, (n = 5) specialist palliative care nurses, (n = 3) dieticians, and (n = 7) doctors. Settings included outpatient chemotherapy clinics, general medical wards, stroke units, and oncology centers across 11 sites. Five of the eight studies were either follow-up or part of wider studies. See Table 2.
Participant characteristics
Overall, (n = 77) patient voices were heard (n = 11) with non-cancer diagnoses. The first group of patient participants included (n = 20) women aged between 40 and 80 years with incurable breast cancer, undergoing outpatient palliative chemotherapy. These participants preferred only to hear good news, handed over decision-making to others, and used treatment as a form of denial. A second group of patient participants were aged between 55 and over 89 years and included (n = 5) patients with cancers and (n = 11) with CHF, severe COPD, liver cirrhosis, or other medical issues and were inpatients in medical wards. As most of these participants had chronic disease, they had built therapeutic relationships, valued being known, and understood the importance of open diagnostic and prognostic conversations when they were ready. The final group of (n = 41) cancer patient participants were undergoing palliative chemotherapy and had an estimated <1 year to live. They expressed their desire for prognostic information overtly or covertly but only 30% of the time this was addressed.
Of the (n = 235) health professionals whose voices were heard, the nursing cohort consisted of (n = 167) oncology nurses, (n = 11) stroke unit nurses, and (n = 12) palliative care specialists. Of the physician cohort (n = 19) were oncologists, (n = 23) were generalists, (n = 4) were stroke specialists, and (n = 2) palliative consultants. Within the dietician cohort, (n = 3) were in oncology and (n = 2) in palliative care.
Specialist palliative care nurses and consultants were confident and experienced in conducting diagnostic, prognostic, and end-of-life conversations. They highlighted the need to establish relationships to develop trust before delivering bad news and the importance of shared decision-making. On the other hand, the majority of health professional participants in the synthesis were generalists, oncologists, stroke experts, or medical/oncology nurses who were not comfortable delivering diagnostic or prognostic bad news. Most waited for prompting or acknowledgment by the patient, used vague terminology or overly optimistic forecasting while others avoided it through fear, to protect the patient/family, or believed it was not their job. The majority believed they needed further education about how to deliver bad news and some nurses wanted better collaboration among the team.
Description of the conceptual model
Two concepts were identified following the synthesis process: “Enablers in breaking bad news” and “Truth avoidance.” Several elements formed the concept of Enablers for breaking bad news, such as the therapeutic relationship, reading cues, acknowledgment, language/delivery, time/place, and qualities that will be discussed separately. Figure 2 illustrates how the Enablers create an environment within which to deliver bad news, leading to a crossroad, even if momentarily and unconsciously, of whether bad news will be shared or avoided. At the crossroads, a decision is also made as to the level of truth disclosure which is affected by personal barriers which maybe fear, and anxiety felt by the health professional and is discussed in Truth avoidance. The process is circular as conversations need to be repeated. There are consequences for both courses of action.
Fig. 2. Enabling truth disclosure model.
Enablers in breaking bad news
Several elements assisted in the sensitive delivery of bad news, such as the development of a therapeutic relationship, reading cues, language/delivery, appropriate time/place, and qualities.
Therapeutic relationship
The importance of a therapeutic relationship between the health professional and the patient/family unit emerged as an important factor in creating a conducive environment for the delivery of bad news in five of the eight papers (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Millar et al., Reference Millar, Reid and Porter2013; Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016; Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017). In the palliative care setting, health professionals were able to understand the patient's preferences, history, culture, religion, and health literacy and learn how to phrase conversations tailored to their preferences (Millar et al., Reference Millar, Reid and Porter2013; Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). A relationship fostered through good listening skills facilitated the development of familiarity, security, and trust and enabled open and honest discussions around diagnosis and prognosis (Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016; Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017).
“Well there's sort of a bond or connection between you. If you know someone fairly well it's easier to do things with them, work out plans. But if you're more like a stranger, um they really don't know what you might like or what's best for you and you don't really understand them.” (Patient in Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014, p. 4)
Working in partnership with the patient/family by involving them in end-of-life discussions gave them control in decision-making which reflected person-centered care and attenuated paternalism (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Millar et al., Reference Millar, Reid and Porter2013; Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). Involving the family in discussions was important, as they were pivotal in providing physical and emotional support to the patient and were potential future decision-makers (Millar et al., Reference Millar, Reid and Porter2013; Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014).
Where a prior relationship did not exist, the generalist doctors relied on testing the patient's current knowledge of their illness to frame the direction of the conversation and did not disclose any prognostic information or discuss the severity of the illness without a direct request from the patient (Anderson et al., Reference Anderson, Kools and Lyndon2013).
Reading cues
Whether or not to break the bad news to patients often depended on whether patients gave cues about their readiness to know more (Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). Requests for further information or acknowledgment of dying took the form of overt verbal requests (Anderson et al., Reference Anderson, Kools and Lyndon2013; McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Millar et al., Reference Millar, Reid and Porter2013; Henselmans et al., Reference Henselmans, Smets and Han2017), vague comments or euphemisms (Anderson et al., Reference Anderson, Kools and Lyndon2013; Henselmans et al., Reference Henselmans, Smets and Han2017), or non-verbal cues such as touch and a “look” (Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). Other times, patients expressed their emotional distress which if explored further by the doctor enabled the patient to acknowledge the terminal illness, opening the door to an end-of-life conversation (Anderson et al., Reference Anderson, Kools and Lyndon2013; Millar et al., Reference Millar, Reid and Porter2013).
At times health professionals needed to cast out their cues to test patients’ readiness for the diagnostic, prognostic, or deeper end-of-life discussions, sometimes referred to as giving a “warning shot” (Anderson et al., Reference Anderson, Kools and Lyndon2013; McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). Patients felt the doctor had a difficult task looking for and understanding cues for readiness, and suggested the doctor simply put out their cues to test the patient's response. Despite overt requests for further information by patients/families, some doctors practiced avoidance and dismissed or deflected their requests (Anderson et al., Reference Anderson, Kools and Lyndon2013; Henselmans et al., Reference Henselmans, Smets and Han2017). When neither doctor nor patient initiated cues, then the truth was not disclosed and opportunities for end-of-life discussions were missed (Henselmans et al., Reference Henselmans, Smets and Han2017). Breaking bad news before the patient was ready was considered to be potentially emotionally harmful and risked damage to the doctor–patient relationship (Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014) yet had to be balanced against urgency (Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017).
Alternatively, patients gave cues about their unwillingness to enter conversations about end-of-life issues or worsening news and while within the patient's rights caused dilemmas for health professionals at times (Henselmans et al., Reference Henselmans, Smets and Han2017; Bergqvist and Strang, Reference Bergqvist and Strang2019). This is explored further under the heading of Truth avoidance.
Acknowledgment
Without a prior relationship, the generalist doctor addressed the end-of-life issues after patients acknowledged their terminal diagnosis and the possibility of dying, otherwise no doctor moved conversations forward without recognizing the patient's readiness (Anderson et al., Reference Anderson, Kools and Lyndon2013; Henselmans et al., Reference Henselmans, Smets and Han2017). Acknowledging the terminal illness was not the same as the patient acknowledging its full gravity (Millar et al., Reference Millar, Reid and Porter2013) or that they were dying from it (Anderson et al., Reference Anderson, Kools and Lyndon2013). Yet at other times after bad news was delivered, the patient then requested prognostic information around life expectancy (Henselmans et al., Reference Henselmans, Smets and Han2017).
Language/delivery
Requesting the patient to describe in their own words what they knew about their illness allowed the doctor to assess the patient's knowledge, health literacy, and language style (Anderson et al., Reference Anderson, Kools and Lyndon2013). Simple, straightforward language without medical jargon or “sugar coating” was expressed by some patients and health professionals (Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016; Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017) as being important and is illustrated by this remark: “He [the doctor] was very open and very clear about what he was telling me. He didn't sort of hide away from it or pull punches” (male patient in Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014, p. 4). Yet not all patients were in agreeance as illustrated by this patient's comments: “But she said there are metastases here and in the lung and … I thought … if I may say so, that she was brutal … Should you hand it out like that?” (female patient in Bergqvist and Strang, Reference Bergqvist and Strang2019, p. 750).
Contradictory beliefs from different patients reinforce the dilemma health practitioners are faced with and confirm the importance of the therapeutic relationship and reading cues. Tailoring the information presented to the patient with the information being requested also required skill on behalf of the health professional (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Henselmans et al., Reference Henselmans, Smets and Han2017). Incremental disclosure of information allowed better absorption of bad news (Millar et al., Reference Millar, Reid and Porter2013); thus, ongoing conversations were important (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Millar et al., Reference Millar, Reid and Porter2013).
The impartation of suitable and absorbable knowledge by the health professional to a level of understanding by the patient/family was essential. Unfortunately, nurses observed that some patients did not understand what was being said to them concerning treatment, or were aware of the extensiveness of their terminal illness, due to poor health literacy, general education, or dementia (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Millar et al., Reference Millar, Reid and Porter2013).
When doctors used vague terminology, there was the risk that some patients would create a false hope by focusing on certain positively worded phrases (McLennon et al., Reference McLennon, Uhrich and Lasiter2013). When doctors avoided using the words “death or near dying,” patients/families did not develop full awareness and remained unprepared (McLennon et al., Reference McLennon, Uhrich and Lasiter2013). Doctors who conveyed overly optimistic prognoses allowed patients/families to set unrealistic expectations, which made follow-up conversations difficult for nurses (McLennon et al., Reference McLennon, Uhrich and Lasiter2013).
Language can present a barrier to effective end-of-life conversations for non-English speaking patients and families (McLennon et al., Reference McLennon, Uhrich and Lasiter2013). While simple body language and mobile phone translation apps assisted with nursing tasks and care, in-depth prognosis-related conversations could not safely be undertaken without an interpreter present, which diminished impromptu conversations (McLennon et al., Reference McLennon, Uhrich and Lasiter2013).
Appropriate time/place
The timing for the delivery of bad news depended on contexts such as the urgency of the consultation, in the case of stroke (Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017), or whether the doctor would be visiting the patient daily (Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016) and could deliver news in increments (Millar et al., Reference Millar, Reid and Porter2013; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). At other times, the doctor delayed the delivery of bad news, so the family could be present (Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017).
The appropriate location to deliver bad news was only mentioned in one paper and took the form of a complaint (Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014). Not only was discussing end-of-life issues during the doctors’ rounds in the presence of a multitude of interns an example of lack of privacy and sensitivity, but patients also felt pounced on and unprepared (Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014).
Qualities
Certain personal qualities were expected in the health professional such as being compassionate, empathetic, non-judgmental, sensitive, kind, and comfortable in exploring the patient's fears and distress (Anderson et al., Reference Anderson, Kools and Lyndon2013; Millar et al., Reference Millar, Reid and Porter2013; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). Palliative care/oncology nurse practitioners saw themselves as qualified and experienced to deliver bad news and with experience gained over time, clinical nurse specialists believed they gained confidence and were, therefore, more willing to partake in breaking bad news and providing support to patients and families (Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). Yet many nurses felt inadequately trained and believed all health professionals should be well educated in the art of delivering bad news (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Millar et al., Reference Millar, Reid and Porter2013). Nurses tended to use communication guidelines flexibly and with experience developed their own style (Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). In contrast, other nurses claimed being placed in this position was against their workplace norms and practices and it was the role of the doctors to engage in diagnostic, prognostic, and in-depth end-of-life discussions (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Millar et al., Reference Millar, Reid and Porter2013; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016; Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017).
Collaboration among staff was essential for the smooth running of the department, mutual support, and presenting a united front to the patient/family (McLennon et al., Reference McLennon, Uhrich and Lasiter2013). Timely sharing of information and prompt completion of the doctor's notes assisted nurses to know exactly what the doctor had told the patient, so they could assist with translating meanings and treatments and provide emotional support after the news was broken (McLennon et al., Reference McLennon, Uhrich and Lasiter2013).
Truth avoidance
Truth Avoidance was evidenced in six of the eight papers (Anderson et al., Reference Anderson, Kools and Lyndon2013; McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Millar et al., Reference Millar, Reid and Porter2013; Henselmans et al., Reference Henselmans, Smets and Han2017; Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017; Bergqvist and Strang, Reference Bergqvist and Strang2019) and is a broad concept incorporating reasons why the disclosure of the truth was avoided and its consequences.
Some patients and families used denial as a defense mechanism to disassociate themselves from the seriousness of their condition which took the form of only seeking positive news or reducing the information they sought in case bad news was broached (Bergqvist and Strang, Reference Bergqvist and Strang2019). Some patient's perception of a good doctor was one who only presented good news (Henselmans et al., Reference Henselmans, Smets and Han2017; Bergqvist and Strang, Reference Bergqvist and Strang2019) and likewise in their view a bad doctor presented bad news. While internet searching was undertaken, patients chose to look for vague diagnoses or isolated symptoms which enabled them to disassociate themselves from the seriousness of their condition (Bergqvist and Strang, Reference Bergqvist and Strang2019).
A game of “common collusion” was played out between doctors, patients, family, and friends, all avoiding the reality of approaching death (Bergqvist and Strang, Reference Bergqvist and Strang2019). Instead of having open and honest discussions about end-of-life preferences or treatment options, futile treatment continued (Bergqvist and Strang, Reference Bergqvist and Strang2019). Over time, patients relinquished their autonomy and handed over decision-making to others, often expressed in a desire to please the doctor, friends, and family (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Bergqvist and Strang, Reference Bergqvist and Strang2019). Despite some patients already aged in their 90s with weeks to live, treatment continued, causing poorer quality of life than the patient could enjoy had they been told the truth and accepted their prognosis (McLennon et al., Reference McLennon, Uhrich and Lasiter2013).
Some generalist doctors during hospital admission, despite direct information requests from the patient, also deflected end-of-life conversations. Truth avoidance occurred by claiming it was not their role as there were other more qualified specialists such as oncologists or their primary care doctor that the patient should be discussing such topics with (Anderson et al., Reference Anderson, Kools and Lyndon2013) and opportunities were lost. Oncologists were either vague with their time estimates, overly optimistic, spoke of life expectancy in years rather than in months, or avoided prognostication altogether (Henselmans et al., Reference Henselmans, Smets and Han2017). Sometimes this was due to a patient's request, other times it was despite it (Henselmans et al., Reference Henselmans, Smets and Han2017). Nurses who prognosticated tended to be more realistic with their time estimates, although their qualifications and experience were not stated (McLennon et al., Reference McLennon, Uhrich and Lasiter2013).
Idealism versus reality was evident wherein health professionals described truth-telling as virtuous, to be practiced at all costs despite the consequences, yet at other times truth disclosure was handled with caution and was selective, being withheld to protect the family from perceived emotional outbursts or to protect themselves (nurses) from an uncomfortable task (Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017). Doctors also acknowledged lying to the family or acting more optimistic to give the family hope, while the patient was still alive (Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017). Most nurses told “half-lies, white lies or outright lies” to avoid truth disclosure to family while waiting for tests to be confirmed or the right timing to do so (Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017). Alternatively, to avoid lying, nurses justified any delays in truth disclosure by purposefully not checking results (Rejnö et al., Reference Rejnö, Silfverberg and Ternestedt2017).
Nurses and dieticians avoided discussing cachexia in cancer patients due to lack of confidence, education, and experience and wanted to avoid the subsequent distress this might cause to the patient/family (Millar et al., Reference Millar, Reid and Porter2013). A more pressing issue in truth avoidance for these staff members was knowing such conversations would invariably lead toward the patient asking about prognostic questions, the result of which they did not believe was their duty to divulge (Millar et al., Reference Millar, Reid and Porter2013). The consequences were the patient/family remained uninformed of the seriousness of their illness until the doctor had these important conversations with them. Another consequence of this avoidance affected the collegiality of the team. Nurses were reluctant to have end-of-life conversations with the patients when they did not know what the doctor had told them for fear of giving conflicting information or divulging too much (McLennon et al., Reference McLennon, Uhrich and Lasiter2013).
Truth avoidance was practiced by doctors due to many factors, some being a lack of time especially when conversations needed to be delivered incrementally and repeated or not having an established doctor–patient relationship (Millar et al., Reference Millar, Reid and Porter2013). The result is the patient is sent away under a pretense of false hope and the doctor and patient both avoid facing reality and any emotional outburst (Millar et al., Reference Millar, Reid and Porter2013). Another reason for some doctors’ reluctance stemmed from a “fix-it culture” in preference to providing psychological support (Millar et al., Reference Millar, Reid and Porter2013).
Nurses faced ethical dilemmas in truth disclosure when the family wanted information withheld from the patient or overrode the wishes of the patient by insisting on continuing unwanted treatment (McLennon et al., Reference McLennon, Uhrich and Lasiter2013). Patient's cultural and religious preferences also impacted whether nurses could have end-of-life discussions with the patient and created dilemmas in whose wishes the nurse was obliged to adhere to, or accepting the patient's right to wait for a miracle instead of commencing end-of-life preparations (Anderson et al., Reference Anderson, Kools and Lyndon2013; McLennon et al., Reference McLennon, Uhrich and Lasiter2013).
Discussion
The qualitative meta-synthesis explored the views and experiences of health professionals and patients and families with truth disclosure during the delivery or receipt of bad news in palliative care. While family voices were not directly obtained, their opinions and behaviors were discussed in the papers. Each Enabler of breaking bad news was believed to be an important factor in the development of a conducive environment in which to improve the success of the delivery of bad news but were not always achieved. The development of a therapeutic relationship, for example, in the person-centered care model, was recognized as an important mechanism in which to care, build trust, and communicate to patients and families and it follows that bad news should be more readily received and accepted within this environment (Beste, Reference Beste2005; Mishelmovich et al., Reference Mishelmovich, Arber and Odelius2016). Nurses are in a unique position to discuss end-of-life issues with patients, as they spend concentrated time with the patient and family (Pontin and Jordan, Reference Pontin and Jordan2013). Yet not all patients may want to take control of their health or be responsible for decision-making as found in this review (Gaston and Mitchell, Reference Gaston and Mitchell2005; Bergqvist and Strang, Reference Bergqvist and Strang2019).
Other than the patient group mentioned above, most of the patients in the synthesis valued openness and honesty and understood the benefit of having a prior relationship with their doctor (Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014). They imagined the stress the doctor felt while waiting for patients to cue readiness and suggested the doctor just send out testing phrases and gauge patient responses (Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014). Regarding ongoing treatment, if patients are to participate in shared decision-making, they need to understand their medical condition, treatment options, side effects, and how the benefits outweigh the risks, which is part of the consenting process (Sarafis et al., Reference Sarafis, Tsounis and Malliarou2014) — this is easier said than done (Wilson et al., Reference Wilson, Ingleton and Gott2014). Not all patients will have full cognitive ability (McLennon et al., Reference McLennon, Uhrich and Lasiter2013; Sarafis et al., Reference Sarafis, Tsounis and Malliarou2014), highlighting the need for early discussions in the disease trajectory and the completion of advance care directives. Informed patients with “realistic” hope can decide whether tolerating treatment with unpleasant side effects is worth the trade-off against quality time with loved ones. Through open and honest end-of-life conversations, patients may avoid futile treatment resulting in cost savings (Zhang et al., Reference Zhang, Wright and Huskamp2009), improved the quality of life for themselves and their bereaved carers (Selman et al., Reference Selman, Brighton and Hawkins2017), tie up their financial affairs, complete their bucket list, and plan for end-of-life. Therefore, truth-telling is a vitally important aspect for achieving quality outcomes for palliative patients and their families.
Communicating bad news in palliative care is not linear but a circular process (Duke and Bailey, Reference Duke, Bailey, Payne, Seymour and Ingleton2008) and requires returning frequently to the facilitating factors (as depicted in Figure 2.), not only as the disease trajectory unfolds, but in daily follow-up conversations. Just because truth disclosure has taken place once, the patient's receipt of the bad news or comprehension of it may not be fully realized (if at all) and the process will need to start again. When closing a consultation where bad news was given, it was essential to assess the understanding of the patient/family confirming that the message has been received and understood, as patients can take on a completely different meaning than the health professional intended (Fried et al., Reference Fried, Bradley and O’ Leary2003; Xafis and Wilkinson, Reference Xafis and Wilkinson2019).
The findings demonstrate how difficult the task is for health professionals to break bad news while being fully truthful in the process. What is perceived as a straightforward, honest, and open disclosure of bad news by one patient is perceived as blunt, rude, and inappropriate by another, proving the need for careful individual tailoring of communicating information by health professionals (Abdul-Razzak et al., Reference Abdul-Razzak, You and Sherifali2014; Bergqvist and Strang, Reference Bergqvist and Strang2019). Assessing readiness or reading the patient's cues for further information indicates to the health professional there is some degree of receptivity on the patient's part. Yet for patients afraid or not sure how to ask, topics may not be broached, thus robbing the patient of time to put their affairs in order (Henselmans et al., Reference Henselmans, Smets and Han2017) or continue with futile treatment. Emotional numbness caused by the shock of receiving bad news may create difficulties in paying attention and retaining information which may also be a factor for the patient not pursuing end-of-life discussions (Maciejewski and Prigerson, Reference Maciejewski and Prigerson2013), yet this does not negate gentle ongoing conversations to help move the patient forward.
The crossroads in the conceptual model illustrate how the health professional is faced with the need to decide whether or not to disclose the truth which may be influenced by personal issues. These personal issues include stress and anxiety when having to deliver bad news due to inexperience, personal fears with illness and death, concerns about how the news would impact the patient/family, discomfort in handling emotional outbursts, uncertainty about the illness trajectory, or family requesting the withholding of information (Hagerty et al., Reference Hagerty, Butow and Ellis2005; Hancock et al., Reference Hancock, Clayton and Parker2007; Kogan et al., Reference Kogan, Enguidanos and Penido2013; Sarafis et al., Reference Sarafis, Tsounis and Malliarou2014; Johnston and Beckman, Reference Johnston and Beckman2019).
Health professionals may err on the side of caution in truth disclosure for fear of taking away the patient's hope (Beste, Reference Beste2005; Sarafis et al., Reference Sarafis, Tsounis and Malliarou2014). However, hope can be fostered by honesty and support developed during a therapeutic relationship (Beste, Reference Beste2005), assuring the patient that they will not be abandoned by the health professional as they transition in palliative care (van Vliet et al., Reference van Vliet, van der Wall and Plum2013), setting goals other than seeking a cure (Clayton et al., Reference Clayton, Hancock and Parker2008), or having prognostic information framed in “hope for the best and expect the worst” case scenarios (Kiely et al., Reference Kiely, Soon and Tattersall2011). Trying to protect patients by withholding the truth from them will eventually increase anxiety and confusion, as the patient's condition deteriorates and is not in the patient's best interests (Fallowfield et al., Reference Fallowfield, Jenkins and Beveridge2002; Clayton et al., Reference Clayton, Hancock and Parker2008). Enzinger et al. (Reference Enzinger, Zhang and Schrag2015) found that doctor–patient relationship ratings were not negatively affected, nor did the patients show an increase in sadness or anxiety due to truthful prognostication.
Only papers written in the English language and representing Western culture were included in this synthesis which may have produced different results. A strength of the findings of this meta-synthesis stemmed from the examination of the voices of patients; overall, (n = 77) patients were interviewed together with (n = 235) health professionals with broad experience and qualifications. Families were referred to by participants, but they were not directly interviewed in any of the included papers.
Conclusion
Breaking bad news truthfully is a difficult task and requires forethought, skill, and time. Creating a conducive environment using the Enablers described in the conceptual model leads the health professional to decide whether to disclose the truth or avoid it. Personal issues such as fear and anxiety impact decision-making at this point and can become a barrier. It is a circular process and requires the Enablers to be revisited in recurring conversations throughout the illness trajectory. The truth about the patient's illness status needs to be disclosed openly and honestly, and the patient's understanding confirmed, for them to exercise their autonomy and make informed decisions about their treatment and end-of-life goals.
Author contributions
The first author designed the project, searched, and screened the literature, coded, analyzed, synthesized, and wrote the manuscript. The second and third authors assisted with project design, screening, synthesis, and edits of the manuscript and supervised throughout the project. All authors have read and agree to the final manuscript.
Funding
The first author is supported by an Australian Government Research Training Program (RTP) Stipend and RTP Fee-Offset Scholarship through Federation University Australia.
Conflict of interest
There are no conflicts of interest.
