“Some things are even worse than telling a child he is going to die”: Pediatric oncology healthcare professionals perspectives on communicating with children about cancer and end of life

Abstract

Introduction:

This study explored pediatric oncology healthcare professionals’ (HCPs) perspectives on direct communication with children with advanced disease about their disease, palliative care, and end-of-life (EOL) communication.

Methods:

Forty-six pediatric oncologists, nurses, psychosocial team members, and other HCPs from six hospital centers in Israel participated in semi-structured interviews. The Grounded Theory method was used. Data were analyzed line-by-line with codes and categories developed inductively from participants’ narratives.

Results:

HCPs viewed communication about disease progression and EOL as vital because children were often aware of their prognosis, because lack of communication could lead to emotional distress, and because communication is a prerequisite for shared decision-making. HCPs identified several barriers for communication including HCP barriers (such as emotional strains, lack of training), parental barriers, guardianship law, and language and culture. HCPs also described strategies to promote EOL communication. Direct strategies include tailoring communication, allowing for silence, echoing children’s questions, giving information gradually, and answering direct questions honestly. Indirect strategies included encouraging parents to talk to their children and teamwork with colleagues.

Conclusions:

Open communication with children who have cancer is essential. Nevertheless, multiple barriers persist. The rising accessibility of online information calls for urgent training of HCPs in communication so that children will not turn to unmediated and potentially misleading information online in the absence of HCP communication. Evidence-based effective communication training modules and emotional support should be offered to HCPs. Knowledge about children’s development, age-appropriate communication, and cultural sensitivity should be included in this training.

1 |. INTRODUCTION

The United Nations Convention on the rights of the child states that children have a right to be heard and involved in matters that affect their lives, including being informed of their disease, prognosis, and decisions regarding their healthcare.^1,2 However, this right can become complicated by factors such as the patient’s age, their cognitive and emotional development,³ parents’ rights and attitudes,^4,5 healthcare professionals’ (HCPs) attitudes and communication skills,^6,7 and cultural norms.⁸ Consequently, patient/child-centered communication and the ethical concept of pediatric consent and/or assent are not always fulfilled.⁹

Communicating with children in pediatric oncology facing life-threatening conditions is even more complicated due to the emotional nature of treating children, 20% of whom will not survive the disease.^5,10,11 As such, communicating about disease recurrence or stopping curative treatment is a concern of many pediatric oncologists.⁶ The International Society of Pediatric Oncology states that if children are willing, they should be informed about the seriousness of their situation, and be involved in decision-making processes about their care.^4,12 Evidence-based guidelines have been formulated to achieve this goal,^4,13 as thoughtful discussions of prognosis have proven to do no harm,¹⁴ whereas uncertainty can augment the child’s fears about death, pain, or suffering.⁷ Additionally, open communication can reduce distress, loneliness, depression, and anxiety, and can promote treatment adherence and enable participation in decision-making.^3,15–18 Studies with seriously ill children and adolescents have found that they would like to be informed and involved in managing their disease throughout the illness trajectory,^16,19–21 and during end-of-life (EOL) situations.^19,22

To add to this complexity, while clinicians should not be barrier to clear communication about prognosis and EOL, not all children wish to be informed in the same way. Some children may prefer to know as little as possible, and that all communication should be filtered through their parents.^16,22 The extent and means by which information is provided is multifactorial, including the child’s age and developmental stage.^19,21

Unfortunately, communication barriers remain prominent due to professionals’ lack of training,²³ emotional strains,^6,24 and the perception that avoiding communication will preserve a child’s hope.^16,24 Parents also tend to limit the information shared with their children,²¹ and as a result, children are often informed at a later stage of their disease, and at times by their peers, instead of by their parents or HCPs.¹⁶ To date, qualitative studies have emphasized that children feel disempowered due to the difficulty in obtaining information regarding their illness, noting that they turn to online resources to acquire it.¹⁸ Indeed, technological advancements have made information more accessible to children of all ages. Therefore, concealing information about their disease is not only undesirable, it is often unachievable.

A recent systematic review on communication in pediatric oncology reviewed 109 studies over the last 20 years and identified several gaps in the literature on this topic. This review found that the majority of studies focused on individual-level barriers and that there was a notable dearth of literature exploring non-individual-level barriers.²⁵ The authors identified only five studies relating to policy-level factors and among them none described the impact of legal aspects of guardianship on HCPs’ communication practices. Laws regarding minors’ informed consent and assent varies across countries, cultures, children ages, and illness severity^26–29 and may influence HCPs’ communication. Another knowledge gap relates to the lack of geographical and professional diversity in this research, as majority of publications originate from the United States and Europe^25,30 and less than 20% of publications included HCPs who were not physicians in their samples.³⁰

Our current study aimed to fill some of these knowledge gaps. Israel is located in the Middle East and has a highly modernized public health system including advanced pediatric oncology treatments and state-of-the-art care. Yet, as with other places in the world, pediatric palliative care services and training are underdeveloped.^11,31 Nevertheless, it offers psychosocial and other treatments aiming to improve the well-being of the sick children, enabling us to incorporate multiprofessional HCP perspectives in the study.^31,32 The purpose of this research was to explore multiprofessional pediatric oncology healthcare communication practices and attitudes to talking to children with advanced disease about their disease, palliative care, and EOL.

2 |. METHODS

2.1 |. Study design and participants

This study is part of a larger project that explored pediatric palliative care in Israel,^31,33 using the Grounded Theory method. The study included 46 HCPs from six major pediatric cancer centers in Israel including 21 oncologists, nine nurses, 10 mental health professionals (social workers, psychologists, and psychiatrists), two physicians (palliative care and radiation oncology experts), and four therapists (art therapists, bibliotherapists, and acupuncturists). We used theoretical sampling, which enabled us to determine subsequent recruitment needs based on gaps in developing theory.

2.2 |. Procedure

After approvals were obtained from the research ethics board (No. 1532–1, Date July 28, 2017; No. 4424–17 Date October 3, 2017), co-investigators at the centers emailed information about the study to potential participants. Recruitment began in November 2017 and continued through January 2019. The inclusion criterion included working within the field of pediatric oncology. Forty-six eligible HCPs agreed to participate after learning about the goals of the research. Interviews were conducted by the first author (Anat Laronne), a doctoral student and a health psychologist. Participants signed a consent form, and a semi-structured interview guide was used. The initial interview guide was developed based on the literature and in consultation with the study team that includes a pediatric oncologist, a health psychologist with extensive experience doing research on pediatric oncology, a mental health clinician working with oncology patients, and an expert in pediatric palliative care. The interview guide was then revised as needed during the data collection and analysis to allow for exploring new emerging themes and categories. Questions focused on how, if at all, HCPs discuss healthcare and palliative care issues with patients and parents, as well as barriers and facilitators to communication with pediatric oncology patients (Supporting Material S1). Interviews lasted between 35–100 minutes and were recorded and transcribed verbatim.

2.3 |. Data analysis

Data were collected and analyzed concurrently. Line-by-line coding allowed the authors to develop codes and categories inductively from participants’ narratives. Constant comparison helped us understand the relationships between the codes. Coding was done by the first author, with weekly consultation and discussion of each code and category until agreement was reached with the study principal investigator (Leeat Granek). During this process, codes were reworded, collapsed, or moved around as needed based on these discussions until a final coding scheme was developed. During data collection and analysis, the developed themes were discussed by the study team. Data collection stopped once data saturation was reached and no new codes were identified. NVivo 12 software was used to code, organize, and store the data.

3 |. RESULTS

Forty-six pediatric oncology professionals were interviewed (Table 1). Three major categories developed: (a) Importance of communicating with the child about their poor prognosis with the themes of child’s awareness of impending death, child’s emotional distress due to lack of prognostic disclosure, and information as a prerequisite for shared decision-making; (b) Barriers to open communication about EOL prognosis with the themes of: HCP and parental barriers; guardianship law; and language and culture barriers; and (c) HCP strategies to promote EOL communication with the themes of direct and indirect communication strategies (Table 2).

TABLE 1.

Participant demographics

Characteristic	No.	%
Sex
Male	17	37
Female	29	63
Age
Mean	48
Range	24–66
Marital status
Married	38	83
Common-law	2	4
Divorced	3	7
Single	2	4
Widowed	1	2
Children
No children	5	11
≥1 Child	41	89
Profession
Pediatric oncologist	21	46
Nurse	9	19
Psychosocial team member	10	22
Other physiciansa	2	4
Other therapistsb	4	9
Years in oncological practice
<5	9	20
5–10	14	30
11–15	8	17
>15	15	33
No. of oncological patients seen per week
<5	3	7
5–15	7	15
16–25	5	11
26–35	10	22
36–40	3	7
>40	18	39

^aPalliative care expert and radiation oncologist.

^bArt therapy, bibliotherapy, and acupuncture therapy.

TABLE 2.

EOL communication with the child: Important barriers and provision strategies

Category	Theme	Subtheme
Importance of communication of EOL prognosis	Healthcare professionals’ perspective of child’s awareness of EOL prognosis	■ Physical/body awareness ■ Nonverbal implicit awareness ■ Awareness due to information accessibility ■ Verbal awareness ■ Awareness and request for inclusion in decision-making processes
	Child’s emotional distress due to lack of prognostic disclosure
	Prerequisite for shared decision-making
Barriers to open communication about the child’s EOL prognosis	Healthcare professionals barriers	■ Emotional strains ■ Lack of training in communicating with children
	Parental barriers
	Guardianship Law
	Language and culture
Healthcare professionals’ strategies to promote EOL communication	Direct communication strategies	■ Tailoring communication ■ Silence ■ Echoing children’s questions ■ Giving information gradually ■ Answering direct questions
	Indirect communication strategies	■ Encouraging parents–children communication ■ Multidisciplinary teamwork

Abbreviation: EOL, end of life.

3.1 |. Importance of communicating with the child of EOL prognosis

3.1.1 |. HCPs’ perspective of child’s awareness of EOL prognosis

Many participants described what they perceived as children’s awareness as a main reason for open communication. They discussed five aspects of perceived child’s awareness.

1). Physical/body awareness:

It referred to children being able to sense their disease state and physical well-being. On this, an oncologist remarked: “It’s their body, they understand that the disease returned, they understand that the tumor grows, that we already stopped chemotherapy … This is why few children ask because they already know the answer” (PO37).

2). Nonverbal implicit awareness:

It referred to the child’s awareness of EOL in a nonverbal way. One example was given by an art therapist who described a child that expressed this awareness through his art. As the disease progressed, he erased shadows and plants. “When the child first chose the painting, he spoke of a solid rock by the sea … At the end, he spoke of the sea disintegrating the rock” (ART24).

3). Awareness due to information accessibility:

HCPs spoke of the difficulty of hiding diagnostic information from children as they had access to the internet. One oncologist reported:

A 15-year-old’s parents didn’t tell him he has cancer … We inserted a central line, he gets chemotherapy, he knows he goes to an oncological institute every day…He reads about it online. (PO15)

4). Verbal awareness:

It was noted as situations when children spoke explicitly of expected death. An oncologist noted, “He was approximately 13… He said to me ‘I’m going to die and all of you will stay here and forget about me’” (PO12).

5). Awareness and request for inclusion in decision-making processes:

Occasionally children explicitly demanded to make treatment choices. An oncologist stated, “It’s rare, but there are teenagers who say ‘It doesn’t help. It is harmful. I don’t want treatment’. And we respect their wishes” (PO37).

3.1.2 |. Child’s emotional distress due to lack of prognostic disclosure

Lack of communication about the child’s prognosis was mentioned as a cause of emotional distress for the children that can be seen in their behaviors. An oncologist illustrated this point:

As teenagers they make a lot of trouble. They scream, they break things, they give their parents hell. It is all a cry of ‘I don’t know what is going on with me. I’m frightened and you are not talking to me’. Or ‘I know, but I have no one to talk to’. (PO38)

Others noted behaviors like avoidance and withdrawal from HCPs as related to distress and mistrust. “Often, near the end, children shut down because they feel that we are lying” (PO37).

3.1.3 |. Prerequisite for shared decision-making

Participants reported that sometimes children are not included in conversations about treatment options. Unaware of their state, they cannot participate in decision-making. A social worker noted, “Sometimes children are not even asked what they want. Sometimes it ends with the parents’ wishes. It isn’t always compatible with the child’s wishes” (SW7).

3.2 |. Barriers to open communication about the child’s EOL prognosis

3.2.1 |. HCP barriers

HCPs reported that barriers to communication included the emotional strains in discussing these difficult issues and lack of training in communicating with children.

Emotional strain:

HCPs spoke of their fears around prognostic disclosure. One oncologist noted, “With children, it is more terrifying to talk of death because you’re constantly hoping for a miracle” (PO9). A social worker remarked: “Even if you dare and the parents gave permission, it is not simple. It is scary … You need training, experience, skill, guidance, all to take it where it should go. And to not leave the patient alone” (SW25).

Participants also mentioned the fear of harming the children by potentially taking away their hope for a cure. On this an oncologist stated that children are better off not knowing their prognosis, considering it needlessly harmful. “The brutality of saying, ‘look no matter what we will do, you will die’, how is it beneficial when you are talking to a child?” (PO36).

Lack of communication training:

It was commonly reported by HCPs. Participants from all professions reported that communication with children about EOL was not addressed. An oncologist remarked, “I don’t have sufficient tools. Nobody taught me” (PO36).

3.2.2 |. Parental barriers

HCPs reported that parents often explicitly wished to conceal information from their sick children and are the key in determining information sharing with the child. An oncologist remarked, “Some parents totally refuse to let us talk to the children about those [EOL] things. Others want us to. It is extremely dependent on the parents, what they want us to convey to the child” (PO22). Participants noted that occasionally, parents were even reluctant to leave otherwise trusted HCPs alone with the child to prevent discussions of this kind.

3.2.3 |. Guardianship Law

The provision of medical treatment to minors in Israel is conditional upon the consent of their guardians. Participants referred to this law as restricting their ability to share information with children under the age of 18. A nurse stressed its complexity.

When the child is young, you turn to the parents and you address them almost exclusively, which is a very big barrier.With teenagers, you truly want to approach the adolescent, but parents are still a barrier because they are the legal guardians (NU39).

Another nurse explained that sharing information is extremely sensitive in EOL situations, due to the need to navigate between the needs of the parents and the child, while also respecting the law.

Let’s say a nurse was asked by a child and she explained that we don’t have any other treatments. Gently, she didn’t say ‘you will die’ … The mother can go to the department head and say, ‘who authorized you to tell my child he is going to die. I want to sue’. I think she will win. (NU20)

3.2.4 |. Language and culture

Communication difficulties caused by language barriers were noted. Israel has a diverse population with 20% of Arab descent. This difficulty was mentioned frequently, despite the native Arabic speakers on staff and availability of medical translators. To address this challenge, participants spoke of designated hospital courses, learning the language on their own personal time, and using translation. One nurse remarked:

Language! Language to speak Arabic. I use Google translate. … I feel I give bad treatment because I don’t have a common language with them. The fact that they don’t ask, doesn’t mean they don’t want to ask. (NU17).

HCPs reported that immigrants and patients from abroad who come to Israel for treatment also face language barriers.

Culture was also mentioned. Some HCPs noted that communication practices are influenced by culture of origin, while others noted that there is as much variation between families within the same culture, as between families from different cultures. A social worker noted, “There is a cultural issue. For instance, Bedouin families are sometimes more protective, they avoid talking more often, reveal less” (SW10). An oncologist described the complexity as follows:

The family background also influences communication. People from Ethiopia, for instance, perceive death differently. It’s harder to talk to them. It’s extremely dependent on the family. There can be one ethnic origin and two entirely different families. (PO12)

3.3 |. HCPs’ direct and indirect communication strategies to promote EOL communication

3.3.1 |. Direct communication strategies

Tailoring communication:

HCPs spoke of the need to tailor and adjust the conversation to the child’s age, developmental state, and personality, as the pediatric population is diverse in all these aspects. One oncologist shared, “Often, I try to check where the child stands. Where they are and where they aren’t [Silence]. I speak to them at a level that enables them to understand what they have and what they want to hear” (PO4). The potential inconsistency in chronological age and the child’s actual maturity/developmental stage, was reflected in the words of another oncologist. “We have 19-year-olds that are like little kids and you are helpless with them. You tell them to sign forms because they are over 18, but it is the parents you talk to” (PO38).

Silence:

Silence was utilized to give children leverage to express their thoughts and concerns. A nurse explained “It takes time. It is not immediate. Very carefully. When I conduct those conversations, I’m usually silent. They talk” (NU1).

Echoing children’s questions:

It is to clarify what they already know and how willing they are to receive the information they request was another strategy used by HCPs. One nurse remarked: “The family doesn’t want [to reveal information] and the child asks. We have been trained to answer with a question. ‘What do you think you have?’ and continue from there” (NU18).

Giving information gradually, as a dynamic process, was another strategy. A nurse noted how this strategy is often used in palliative care.

It is a journey. Palliative care is like giving birth. I need to cross the bridge with them hand in hand … I need the ability to be what they need from me right now. The father needs to hear something, and the mother needs not to hear something, and the child needs me to look in her eyes and say, ‘I’m here’. (NU29)

Answering direct questions honestly:

This category was noted as important in situations where the child was requesting information. An oncologist remarked, “If a child initiates, if he wants to talk about anxieties I talk. What scares him, will it hurt. Usually, they are afraid of pain” (PO9).

3.3.2 |. Indirect communication strategies

Encouraging parents to talk to their children and disclose information about the disease trajectory was noted as a way to share information. HCPs described navigating these difficult conversations and advocating for the child’s inclusion. An oncologist described this as follows:

Often I find myself telling parents ‘talk to your child’ … Some things are even worse than telling him that he is going to die. He might think that he will choke to death, that it will be very painful … We have means to ensure that will not happen. That it will happen peacefully, calmly. If you can say that to the child, I think it is extremely helpful. (PO15)

While a palliative care physician noted

I always say [to the parent] ‘look I’m not going to lie to the child, but come in, let’s talk, just talk to the child’. And then suddenly the child brings it up … then the parents understand that they can’t avoid talking to the child … I won’t say a thing to the child if the child doesn’t want to know. (P21)

Multidisciplinary teamwork:

Requesting and relying on the support of coworkers, including psychosocial teams and palliative care experts, when discussing EOL with a child was another indirect way to ensure that children will get the information they need. On this, one oncologist remarked:

In general, I will not directly tell a child under the age of 8 that he is going to die. The palliative care physician can be in situations in which he has these conversations. It is a process that is often done by the psychosocial team. (PO22)

4 |. DISCUSSION

This study provides detailed insight on HCPs’ perspectives and practices of communication about disease, palliative care, and EOL with children suffering from advanced cancer. Our findings revealed that oncology HCPs from multiple professions viewed communication about the child’s disease and EOL as vital but faced individual-level, policy-level, and cultural barriers to having these conversations. They also described a range of approaches to promote EOL communication using direct and indirect strategies.

Our results, based on a significant sample size from six medical centers, show that policies based on the law, influences HCPs’ communication with children with advanced disease. Study participants referred to the guardianship law as restricting their ability to communicate with children with advanced disease. In Israel, the provision of medical treatment to minors who are under the age of 18 is conditioned upon the consent of their parents, which means that minors are not asked to assent or consent for cancer treatment. This policy curbs opportunities for patient–provider conversations as was the case in our study findings. It is interesting to note that across countries, the age of consent to health care varies. Some have an age limit for medical majority (i.e., in Denmark age 15; Spain, age 16; Greece, age 18), and others do an individual evaluation according to age and maturity (Estonia).²⁹

In Israel, the Dying Patient Law,²⁶ which relates to patients with a life expectancy of less than 6 months, lowers the age of consent to 17 years. Therefore, it allows teenagers at EOL more autonomy at a younger age. This law provides a framework for information sharing and decision-making as well as participation of minors in their health care, providing that they are aware of their situation, are cognitively and mentally capable to understand the meaning of the information given, and ask to participate in medical decisions.²⁶ This framework may provide a promising avenue to pursue new policies and laws on communicating with children about their disease and prognosis. While we found that laws influence HCPs–children communication, it remains understudied within pediatric oncology.²⁵

Some of our findings are supported by other studies in pediatric oncology. HCPs’ acknowledgment of children’s awareness of their death was noted to contribute to an open communication approach.²⁴ Our study adds HCPs’ perspectives on this topic and a more nuanced understanding of child awareness that includes five aspects: physical/body awareness, nonverbal implicit awareness, awareness due to information accessibility, verbal awareness, and awareness and request for inclusion in decision-making. Our findings correspond with Morrissey’s description (1963) of illness anxiety levels in children with life-threatening diseases (verbal, behavioral, physiological, and symbolic).^34,35 However, we emphasize that EOL awareness does not merely provoke anxiety, it is also an avenue for shared decision-making.

Our research also speaks of the challenges of communicating with children in a culturally diverse setting. This issue was also noted in a study about the conflict between Dutch HCP norms of revealing information to the child, and the Muslim minority parents’ requests to conceal information.³⁶ In the United States, cultural barriers were noted when families came from other countries with different norms.³⁷

Finally, while supporting disclosing prognostic information to children, HCPs acknowledged the complexity for children, parents, and themselves. A systematic review on communication during childhood cancer reported this complexity from patient perspectives, stating that: “a delicate balance exists between empowering children or adolescents with information and burdening them with responsibility and decisions.”^18(p712) HCPs’ emotional strains associated with discussing EOL with children was documented in other countries.^6,8,38 In Sweden, the main concern of pediatric oncologists was related to their role as bad news bearers.⁶ In a Canadian study, more than 80% of pediatric oncologists expressed concerns regarding the reaction of the child or parent to bad news.³⁸ In the United States, identified barriers were lack of communication training and skills,^3,7,14 HCPs’ lack of knowledge about child development, death conceptualization, and palliative care.^3,35 Not knowing the right thing to say triggered discomfort, which relates to HCPs’ preference to refer patients to others for advance care planning.^14,22 In sum, the need for emotional support and training in communication with children about EOL is widespread among pediatric oncology HCPs.

Limitations of this study include not incorporating patient’s experiences as we focused on HCPs’ perspectives. Furthermore, while the qualitative approach used for this study is suitable for the research questions we explored, future studies may wish to use other approaches to quantify the prevalence of communication barriers and strategies. Finally, the unique cultural and legal structures within Israel could affect the generalizability of the study findings to other contexts.

Our research has several clinical implications. First, as our participants noted, medical information is increasingly available to young people with cancer¹⁸ being “digital natives.”³⁹ Yet, we have a limited understanding of health-related internet use by pediatric patients.^39–42 Other research has identified negative emotions as driving online searches including feeling isolated and anxious.⁴² Unmediated online information may also add to children’s anxiety and may be misleading. Therefore, asking what children learned online is important as it pertains to their well-being. It may be especially important to learn more of online information gathering in countries and cultures where barriers to direct HCP–child communication persist.

Second, our results suggest that pediatric oncology HCPs need and want training in communication with children about their disease, palliative care, and EOL. HCPs in pediatric oncology treat a wide range of patients from birth to young adulthood. Participants noted this challenge and the need to adapt communication to the child’s age, developmental stage, and maturity. Developmental knowledge such as children’s abilities to comprehend the meaning of death when faced with life-threatening situations^35,43 is also lacking. Third, as talking to children with advanced disease about their disease, palliative care, and EOL results in tremendous distress for HCPs, emotional support should be offered to these teams.

Communication skills training interventions in pediatric oncology are limited. Kaye et al. identified six published interventions, all from the United States.⁴⁴ Interventions included didactic workshops, case-based role play, and bereaved parent teachings. Benefits involved increased comfort with communication about child’s disease, increased palliative care knowledge and consultations. Our participants’ clinical experience illuminated the complexity of child’s awareness of their illness and prognosis. HCPs’ sensitive multiaspect perceptive of children’s understanding of their disease and bodies can be incorporated in didactic HCP communication training modules. Training programs may also benefit from incorporating discussing disclosure with parents in a culturally sensitive way, as parents are the key in determining what information the child will receive and HCPs often aspire to advocate for the children’s autonomy.

Communicating with children with advanced disease about their health, palliative care, and EOL is complicated with no simple straightforward recommendations.²⁴ It involves many aspects including being sensitively attuned to children in order to be able to meet them where they are and addressing parental needs and desires regarding communication with the child. Given the importance of communication with the child with advanced disease about disease, palliative care, and EOL, the persisting communication barriers, and the lack of training reported, evidence-based effective interventions should be offered to pediatric oncology HCPs. These may be established on theory, pedagogy,⁴⁴ and inductively, utilizing the vast clinical experience of HCPs.

Abbreviations:

EOL

end of life

HCP

healthcare professional

DATA AVAILABILITY STATEMENT

Research data are not shared as it involves transcripts with individuals that would be identified. As such, the data are not available due to privacy or ethical restrictions.