Introduction
The ways in which children directly involved with life-limiting conditions understand dying and death remain poorly understood. Most studies explaining children’s understanding of dying and death report the perspective of surviving, bereaved parents (Kennedy et al. Reference Kennedy, Chen and Valdimarsdóttir2018; Sheehan et al. Reference Sheehan, Burke Draucker and Christ2014, Reference Sheehan, Hansen and Stephenson2019, Reference Sheehan, Mayo and Christ2016; Vázquez-Sánchez et al. Reference Vázquez-Sánchez, Fernández-Alcántara and García-Caro2019). Pediatric palliative care (PPC) research has primarily explored parental or provider perspectives (Singh et al. Reference Singh, Raj and Paschal2015; Spruit et al. Reference Spruit, Bell and Toly2018). The paucity of literature exploring dying and death from seriously ill children’s self-perspective presents a significant gap, given that nearly 8 million children worldwide would potentially benefit from PPC (Connor et al. Reference Connor, Downing and Marston2017).
In 2015, the International Children’s Palliative Care Network conducted a Delphi study with 758 international, multidisciplinary members from 52 countries identifying PPC research priorities. Of the 70 research areas presented to respondents, the need to “understand death and dying” emerged as the top priority, illustrating both the urgency with which PPC providers require better evidence to guide care for seriously ill children and the continued silence of the affected child’s voice (Downing et al. Reference Downing, Knapp and Muckaden2015).
Researchers have sought to understand the childhood cancer experience from the child’s perspective but have not explicitly focused on how children process dying and death while struggling with, or watching a sibling struggle with, other life-limiting conditions (Bjork et al. Reference Bjork, Norstrom and Hallstrom2006; Vindrola‐Padros Reference Vindrola‐Padros2012). Children’s understanding of dying and death differs from that of adults in numerous ways, including age, developmental stage, prior experiences with death, and illness variables (diagnosis and time since onset), as well as cultural, religious, and familial beliefs (Panagiotaki et al. Reference Panagiotaki, Hopkins and Nobes2018). How living with the illness in themselves or their sibling affects their understanding of dying and death remains unclear. This study’s purpose was to understand how children process dying and death in 3 cultural contexts to ensure that PPC services can meet the needs of children and families in culturally competent ways.
Methods
Participants
Approvals were obtained from the institutional review board at a hospital in the Midwest USA where it was carried out, the Hospice Africa Uganda Research Committee, the Uganda National Council of Science and Technology, and the National Review Board of Bioethics from the Health Department of Haiti. These countries were chosen to provide diverse locations because the investigators had relationships with local collaborators, facilitating the identification of potential participants and fulfilling local ethical review processes. Participants were children who received care through one of 3 sites in the USA, Haiti, and Uganda or were a sibling of an enrolled US child with a serious illness. Children were eligible if they were 5–18 years old with life-limiting conditions (Uganda, 8–18 years), were able to assent, had a caregiver providing consent, had a Karnofsky/Lansky score of >50, were developmentally able to express themselves, and had the stamina to be interviewed. Siblings were included as the conceptualization of the project was to assess and compare how children with exposure to the medical community, either in themselves or in their family, understand dying and death. Also, in many cases, the patients themselves did not have the ability to participate due to an underlying diagnosis of severe neurologic or cognitive impairment. While the intent was to have siblings in all countries participate, for safety reasons in Haiti, this was not possible.
Procedure
Families were approached by a member of the research team who informed them of their eligibility and offered to further describe the study to them. Haitian participants over 10 years old were aware of their diagnosis. Ugandan participants knew their diagnoses in some, but not all, cases. US participants were normally aware of their diagnoses. After obtaining informed consent, 28 individual semi-structured interviews (which included a collection of demographic data) were conducted between March 2016 and December 2018 in the hospice/hospital where the child received care or in the child’s home. Interviews were conducted in the USA and Haiti by P.M. and overseen in Uganda by J.D. Sample questions included, “Have you ever known anyone or been around someone who has died?”; “What happens to a person after s/he dies?”; and “Do you believe people/animals have souls?” (see Supplementary material for the complete interview guides). The core analytic team included 1 doctorally prepared PPC scientist with pediatric chaplaincy experience and 2 doctorally prepared nurse scientists. A medical interpreter was used for Haitian children when English was not their primary language to ensure accuracy and understanding. Children in Haiti were interviewed by a medical interpreter in Creole (required by the national bioethics committee); Ugandan children were interviewed by nurse researchers who spoke Luganda and Madi. Semi-structured interview guides with specific probes and focused questions directed conversations with the children to ensure inclusion of key areas for discussion. Interviews and field notes were digitally recorded, translated into English when necessary, and transcribed verbatim. The interviewer reviewed each transcript for accuracy. NVivo 12.0 was used to manage the data.
Analysis
The investigators used grounded theory methodology to understand how ill children and their siblings process dying and death from their own perspectives. Grounded theory is a qualitative method of iterative and inductive strategies that guide researchers in understanding the social processes of a specific phenomenon. Systematic analytic procedures lead investigators to produce an explanatory or theoretical model of the phenomenon that is “grounded” in the data itself, giving voice to the participants immersed in the experience (Charmaz Reference Charmaz2014). Data from interview narratives and field notes were analyzed using grounded theory methods to focus on how children processed dying and death using constant comparison analysis to compare emerging codes across transcripts. Four levels of coding (i.e., line-by-line, focused, axial, and theoretical) were used to transform the data into an explanatory model (Charmaz Reference Charmaz2014; Glaser Reference Glaser1978; Strauss and Corbin Reference Strauss and Corbin1990). During first-level coding, any phrases, sentences, or paragraphs revealing information related to the coping strategies of adolescents were labeled with code words that captured the essence of participants’ remarks. The participants’ words were used as labels when possible. A list of codes was developed, and display grids were used to organize codes attributable to the 2 participant groups. Several procedures were used to enhance the trustworthiness of our findings (Lincoln and Guba Reference Lincoln and Guba1985). The perspectives of the children and their siblings were used to construct the explanatory model. The analysis was conducted by an interdisciplinary research team. Each member analyzed the data individually followed by peer debriefing and discussion of emerging findings during data analysis until a consensus was reached. All methodological decisions were documented in an audit trail.
Results
Demographic data for the participants are presented in Table 1. A theoretical model explaining how this sample of children from 3 geographical and culturally diverse regions processed serious illness, dying, and death emerged from the data (Figure 1). Loss emerged as a central theme described by both ill children and siblings. Resilience, altruism, and spirituality were used by ill children as strategies to manage their losses and anticipated death. Major themes and exemplars are presented in Table 2.
Participant demographic characteristics
a One participant did not know his age and was not included in the calculation of the mean age of Ugandan participants.
Children’s understanding of dying and death.
Major processes and exemplar quotations
Loss
Loss was a central theme, with loss of others and loss of normalcy emerging as 2 subcategories of losses most often described.
Loss of others
Culturally distinct differences were identified. Haitian children were more likely to witness another child’s death because they were hospitalized in larger wards and therefore nearby when a child died. Although witnessing another ill child’s death was not distressful for all participants, most described the experience as “sad,” “uncomfortable,” or “scary.”
Ugandan children were more likely to experience the serious illness or death of a parent. An 8-year-old Ugandan girl recounted her father’s death from 3 years earlier: “My father died when I was five years old, many people come and he was buried in the soil. People were crying and I also started crying but my Aunt took me away at that time. He died because of sickness – of HIV.”
American children typically described the loss of family elders (e.g., grandparents) and saw their deaths as expected with statements like, “because he was old” (9-year-old US male) or “she was sick and she was old … she looked peaceful so it wasn’t as hard” (14-year-old US female). Because they had prior relationships with the deceased relative, several participants found comfort by focusing on positive traits. For example, a 17-year-old American male remembered his great-grandmother as “independent – she didn’t really rely on anyone to do things for her … I admired that.”
Loss of normalcy
Losses of normalcy included the loss of health, school, time with friends, and activities. Perceived loss of health often led to other types of loss, such as for a 15-year-old Haitian male who could not play soccer “… when I was sick.” The loss of health obstructed the routines of the ill child and siblings. Similarly, a 14-year-old American female said her brother’s illness kept her from “… hang[ing] out with my friends as much as I would have like [sic] to.” The loss of health obstructed the routines of the ill child and siblings.
The seriously ill children from all 3 cultures missed attending school and spending time with friends. A 17-year-old US male said, “I remember crying about not seeing my friends and then I thought I might be able to go back later to school but I ended up not being able to.” Several Ugandan children reported being bullied at school by their peers or teachers when peers or teachers feared the child’s diagnosis. A 9-year-old female described being “chased from school” which made her “feel bad”: “When I go to school, children begin abusing me and telling me that [I] am infectious and even teachers don’t like me. They normally chase me away from class and even when [I] go for prayers I sit outside.”
Several American and Haitian children said that they missed participating in sports. A 15-year-old American male described how pain kept him from running like he used to. He said, “… I don’t like to run now because it hurts” and wished he could “trade my body for the weekend.” Siblings were affected by the illness when they assumed more home responsibilities because parents were at the hospital. This interfered with the healthy sibling’s activities, such as a 12-year-old American female who missed cheerleading practices because of her brother’s illness.
Resilience
Resilience was used by several older children as a process to manage their loss and anticipate death. Resilience, which may differ across cultures, is an adaptive response to stress in which the individual is able to recover from the stressor (Rutter Reference Rutter2012).
Resilience and loss
Several processes in this model were deemed bidirectional when data showed that each variable influenced the other. For example, data showed that the process between loss and resilience was bidirectional (i.e., resilience influenced loss and loss influenced resilience). Resilience appeared to affect how participants experienced loss if they could see their situation positively or as a new normal. A 17-year-old American male referred to his illness as, “It’s just a fact of life at this point. It’s become such a subconscious thing like breathing to me….” However, there were still times when he was less certain: “[there are] certain things I still don’t want to accept like not being able to walk. I accept it but I don’t accept it.” A 15-year-old American male described his resilience as “there’s a big journey in life. I’ve been getting more aware of things and how I’m going to react to them. Sometimes I’m bad at reacting to them but over the years I’m getting better and better.” When asked to explain what types of things he has learned, he said to “… be aware of the surprises in your life.”
Resilience is about adaptation, and several children discussed how they adapted from their losses. A 17-year-old American male stated, “I kinda had to grow up faster – I think it made me a stronger person. I’ve become more determined to get stuff done. The positives definitely outweigh the negatives.”
Resilience and anticipating death
Data showed that the process between resilience and anticipating death was bidirectional. Anticipated death refers to the uncertainty that participants had about whether the serious illness would ultimately take the patient’s life, and if so, when. Resilience influenced how participants anticipated death. A 15-year-old Haitian male explained, “… to live a good life I have to pass this situation. I have to pass this moment.” His belief that “this moment” could be overcome shows his resilience adapting to new situations. Conversely, a 13-year-old Ugandan female showed that anticipating her death had a negative influence on resilience, because as she said, “I fear death because I still have things to do.” A 15-year-old American male kept a positive attitude despite knowing that he could die. “It’s not fun and every day you think you’re not gonna wake up, but you just thank the Lord every day that you are woke up and you try to have a good time.”
Altruism
Altruism is motivated by compassion and empathy and occurs when someone commits to improving someone else’s welfare. Several children used altruism to process their loss or anticipate death.
Altruism and loss
Data showed that the process between loss and altruism was bidirectional. Altruism influenced how loss was experienced by children wanting to do good deeds for others despite their own loss. A 15-year-old American male wanted to help others although not feeling well himself: “Almost every day I feel tired because of my medicine but I like to make people happy … I like to joke around a lot.” In another example, a 7-year-old American female did not discuss her great-grandmother’s death because “my grandma cries when I talk about her.” For her, not discussing death was something nice she could do for her grandmother. Loss also influenced altruism when children did good deeds for deceased family members, such as a 9-year-old American female who “… prays for [her deceased grandfather] at night.” This girl continues to be committed and motivated to improve her grandfather’s well-being, even after death.
Altruism and anticipated death
Altruism and anticipated death were evident in a 15-year-old Haitian male who witnessed another child’s death. Rather than focusing on himself during the stressful time, he opted to say a prayer of protection for the other children as well: “Pray for those that are still living not to pass the way the kid that died passed.” This was a profoundly altruistic act when it would have been understandable to think solely of his own health and well-being. A 16-year-old Ugandan male provided a good example of how his anticipation of death led to altruistic acts for others. He taught other youth “about HIV and how we can prevent ourselves because when I teach the youth, it gives me hope that … we shall reduce the number of children who get HIV.”
Spirituality
Spirituality was used by several older children to manage loss and anticipate death. Spirituality is defined according to the dimensions of meaning, beliefs, connections, self-transcendence, and value that interrelate to form the essence of spirituality (Stephenson and Berry Reference Stephenson and Berry2015).
Spirituality and loss
Data showed that the process between loss and spirituality was bidirectional. Spirituality influenced how loss was experienced when spiritual beliefs provided comfort for loss. A 12-year-old American female explained this best when she described the loss of her grandfather: “I cried just a little bit because I knew he would be safe wherever he’s going – like heaven.”
Loss also reinforced spiritual beliefs for a 13-year-old American male who attributed the unusual events in his home to a continued bond with his deceased grandmother. A story about his mother’s wedding ring showed the extent to which he believed his grandmother was still around: “My mom had her wedding ring … it’s gone. I think my grandma has something to do with it and she hid it somewhere … that’s what I think.” In fact, whenever something unusual occurred in his home, the participant was quick to believe that his grandmother was somehow involved: “The funny thing is [at] the balloon launch, mom’s balloon goes like [flop, flop sound] and all over the place, it’s like, “grandma stop! I know it’s her messin’!” Even if there was another potential reason for the occurrence, he appeared to be comforted by the possibility that it was his grandmother.
Spirituality and anticipating death
Data showed that the process between spirituality and anticipating death was bidirectional. Spirituality influenced how a 9-year-old Haitian boy anticipated death when asked to explain what it was like to see another child die. He admitted that seeing the deceased child was “… hard and frightening,” but he used his belief in prayer to help: “So when I see that kid die in front of me, I pray to God to get me out of the hospital so I can get healed.” Several Haitian participants feared negative manifestations of death such as “zombies when it’s dark” (5-year-old Haitian male) and “demons at the hospital” (5-year-old Haitian female). Responding to a question about having seen demons at the hospital, the girl said she “tells them to go away and they listen.” Consideration about spirituality’s meaning within the context of death was described by a 17-year-old American male who said he was “scared to die because you don’t really know what is going to happen.” He expanded on his uncertainty when he said, “I mean there are descriptions of it but no one’s ever been able … you can’t go and come back to tell people.”
Process of loss and anticipated death
Data showed that previous loss affected how the children anticipated death. Concern that their illness could result in death was evident for many children who described a “fear” of dying but was compounded for several Ugandan children whose parents struggled with or died from the same illness they also had. An 18-year-old Ugandan female described her mother’s epilepsy: “… it normally throws her down and at times throws them in fire, biting of tongue and all her body has scars.” She explained that “[grandmother] also has the same condition” making it multigenerational, and, for this participant, “this sickness attacks me every day.”
The stories from the Ugandan children indicated that they had more exposure to death than did the American and Haitian children. For example, a 16-year-old Ugandan male said, “my dad died … many people have died…,” while another informed the interviewer that, “yes, even just today my relative has gone for burial – someone has died.” While some of the Ugandan children were not necessarily bothered by witnessing the death of someone else, others were, similar to the Haitian children reported previously. A 13-year-old Ugandan male said that when he “was next to my dead auntie, it didn’t feel nice. I felt like it could be me dead.” In another case, a 10-year-old Ugandan female described seeing another child who had hung herself and speculated on a possible reason why. “Maybe the father had annoyed her and she went up, when going to school, climbed up on the children’s bed and onto the stool.” However, after considering the child’s death, the participant said they “felt bad” and decided “let me take medication because I also don’t take medicine, I can also die.” Participants reflected on their own behaviors after seeing the consequences of actions that could lead to death.
Discussion
We present results of a multinational grounded theory study of how children understand dying and death, an identified global palliative care research priority (Downing et al. Reference Downing, Knapp and Muckaden2015). Our results deepen the understanding of how children think about dying and death in the context of serious illness, whether their own, or a family member’s. The children in the present study were very aware of death, had experience with dying and death, and were able to conceptualize dying, death, and, in some cases, an afterlife, within the vocabulary they had to express themselves. While some providers are hesitant to address dying and death topics directly with children, our results suggest that children are already aware of these issues at some level (Cotton et al. Reference Cotton, Grossoehme and Bignall2014; Friebert et al. Reference Friebert, Grossoehme and Baker2020). The question then shifts to how rather than whether to talk with children about dying and death. Curricula and frameworks are available to guide providers with these conversations (American Academy of Pediatircs n.d.; Dalton et al. Reference Dalton, Rapa and Ziebland2019; Stein et al. Reference Stein, Dalton and Rapa2019).
Our findings offer guidance about children’s concerns related to dying and death. The topic of loss, both of normalcy and of relationships, was a significant theme, deserving clinical attention. Our results suggest a proactive approach because it may bolster children’s resiliency and thus their ability to cope and ultimately grieve well. We refer to bidirectional relationships when the categories influence each other. The bidirectional relationships between loss and resiliency, altruism, and spirituality provide 3 potential topics from which to move toward a discussion of loss and its meaning for a child. The data show that some children express their understanding of dying and death through altruism, such as with spontaneous peer-to-peer support, thus showing the value of creating and/or expanding peer programs.
Our findings also underscore the presence and importance of children’s rich spiritual lives (Coles Reference Coles1991). While spirituality is frequently seen as a positive resource for coping, children also experience spiritual struggles which include troubling beliefs or questions and dissonance between beliefs and experience (Cotton et al. Reference Cotton, Pargament, Grossoehme, Piedmont and Village2013; Grossoehme et al. Reference Grossoehme, Teeters and Jelinek2016). Children may not spontaneously disclose these thoughts to providers but may be able to respond openly and honestly in their own ways to adults who inquire (Cotton et al. Reference Cotton, Grossoehme and McGrady2012; Pendleton et al. Reference Pendleton, Cavalli and Pargament2002; Schreiner et al. Reference Schreiner, Grossoehme and Friebert2020). Thus, clinicians should be encouraged to explore these crucial dimensions with seriously ill children to bring their full selves into resilience and healing.
In addition to clinical implications for proactively addressing loss and grief, as well as attending to children’s rich spiritual lives, the results inform engaging children with serious illness and siblings so that they can collaborate in decision-making about both their treatment and their end of life. Palliative care and hospice programs should be designed – and care delivered – with cultural humility, embracing differences in understanding between individuals with diverse backgrounds. The results also inform anticipatory grief support for children undergoing treatment for serious illness and their siblings. Awareness of the concepts that they understand and how they construct meaning and grow (resilience, altruism, and spirituality) can enable children to be supported with appropriate strategies highlighting their strengths while honoring the grief and loss that they experience.
National guidelines call for the competent integration of cultural and spiritual issues for quality hospice and palliative care (National Consensus Project for Quality Palliative Care 2018). Our data present the voices of children, and while their expressions may differ from adult descriptions of the experience, we see that issues of loss (normalcy and relationships), resiliency, altruism, and spirituality are not that different from how other age groups anticipate death.
This study has the following limitations. One aim of this study was to give children a preliminary opportunity to contribute their voice to the problem of serious illness in their native languages and with the aid of interpreters. The investigators’ goal was to articulate a baseline understanding of the major issues the children were affected by or concerned about. Unfortunately, the limited sample size did not allow in-depth consideration of cultural differences. Further research with a larger and culturally stratified sample would enable a deeper understanding of cultural diversity and further refine this conceptual model. There was a significant delay between the data collection and completion of the analysis due to the logistics of a multinational partnership, data translation, and the COVID-19 pandemic. The sample size in each setting was relatively low, and siblings of hospitalized children in Haiti or Uganda were not available. Despite the interpreters’ best efforts, differences in linguistic structure, vocabulary, and culture may have affected analysis. A deeper exploration of ways in which resilience is experienced or expressed across cultures was beyond the scope of this study; however, those differences may be present. Finally, the majority of participants were Christians. Further research with greater religious and spiritual diversity would deepen understanding of how children understand dying and death. Nevertheless, important conclusions may be drawn. We suspect that some children may exhibit early signs of post-traumatic growth, but because this sample was actively involved in treatment, it may be too soon to assess. Additional research with survivorship may be warranted to fully explore post-traumatic growth for seriously ill children. Extending this exploratory study will increase understanding of the extent to which these themes are culturally bound or more universal. While this study does not permit specific practice recommendations regarding how to talk with children from diverse ages and backgrounds, it does support that, even at very young ages, children understand dying and death and highlights the importance of creating space for their questions. It is important to assess the child’s cultural context and to reinforce the need for cultural humility when training providers to maximize equitable care. Further research about altruism, resilience, and spirituality across cultures may enhance culturally competent PPC.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1478951523000287.
Acknowledgments
The authors gratefully acknowledge the participation of the children and their parents who agreed for them to be interviewed; Kristine Allmendinger-Goertz, Collins Ankunda, Jennifer Deuble, Judith Ellis, Zachary Fryda, Liz Grant, Gabriel Gue, Rachel Jenkins, Ruth Kemigisha, Priscilla Lebu, Mhoira Leng, Liz Nabirye, Ruth Nakayiwa, Liz Namukwaya, Juliet Namyeso, and Valerie Shaner; and the team at the African Palliative Care Research Network meeting at the 2013 African Palliative Care Association conference.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflicts of interest
None of the authors has any conflict of interest to disclose.
