Bereaved Parents’ Narratives of their Child who Died from Cancer: A Qualitative Study

Abstract

Background and Objectives:

Humans use storytelling to create meaning from suffering, including after the death of a child. The elicitation of and response to stories remains underutilized in medicine as a whole, and in particular, within parental bereavement. Thus, we sought to explore how bereaved parents choose to share the story of their child.

Methods:

This is a secondary analysis of a dual site, survey-based study of parents’ experiences following their child’s death from cancer 6–24 months prior. Our qualitative inquiry focuses on inductive, iterative analysis of free-text responses to the question: “If you would be willing to share, please tell us about your child. What would you like us to know about him/her?”

Results:

One-hundred twenty-eight parents completed the survey; 101 parents (79%) representing 81 children shared about their child. Responses took the form of narratives, varying in terms of structure (e.g., verb tense, subject, narrative arc) and content (e.g., character development, narration/voice). Parents highlighted their child’s appearance, traits, and passions. Most did not detail cancer history or death. Parent narratives progressed through themes of medicalization, humanization, and supernaturalization in the “character development” of their child, who was frequently characterized as “otherworldly.”

Conclusions:

Bereaved parents whose child died from cancer desire to tell their child’s story. When provided the opportunity, nearly 80% of parents in this study chose to share. Thematic progression may be a framework for narrative repair. Parental narratives of their deceased child provide new insight into their bereavement experience and may help to develop novel supportive interventions.

INTRODUCTION

Parents of children who die from cancer experience lifelong grief and are at increased risk of significant long-term psychosocial and physical morbidities.^1–4 The bereavement literature most often focuses on parents many years following the death of their child, leaving a gap in understanding of early bereavement processes (i.e., within the first two years after a death), where parents are known to be highly vulnerable.^5–8 Bereavement practices and guidelines vary across both providers and institutions,⁹ leading to disjointed bereavement support that often lacks evidence-based interventions.¹⁰ Many bereaved parents desire to maintain connection with the medical team, but providers often do not know when or how to begin in their support.^11–13 There is also a lack of training for pediatric clinicians in the process and experience of grief and bereavement, leaving many well-intended providers unsure of how, when, and what to say to bereaved families. Therefore, afraid to say the “wrong” thing, clinicians and community members alike often choose the perceived safety of silence.^12,14,15

Humans are innate storytellers, making sense of the world through narratives - written or spoken.¹⁶ Bereaved parents of children are no exception. They desire to tell the story of their child and hope their child will not be forgotten by others.^11,12,17 Narratives are used to “transform what feels like a shapeless, endless, and meaningless experience into something with structure, finitude, and implication,¹⁸ and help “build connections between past events and the future.”¹⁹ A narrative approach to parental bereavement is used informally in the form of journaling in support groups but remains underutilized in clinical and research settings.

Narratives may provide critical insight into parental conceptualization of meaning, identity, stigma, and social functioning—dimensions poorly captured by traditional quantitative measures.⁷ In this secondary analysis of a larger study⁸ exploring the early grief experiences of cancer-bereaved parents⁸, we sought to explore the specific narrative strategies parents employ as they share the story of their child. By identifying narrative approaches used in early parental bereavement, we aimed to inform the development of targeted narrative-based bereavement interventions.

METHODS

Study Design, Setting, and Participants

This is a secondary analysis of a dual site, survey-based study of parents whose child died from cancer 6 to 24 months prior. Details on the survey development, procedures, and primary findings have been previously published.^7,8 We enrolled a total of 128 parents representing 88 children, with a household response rate of 37%. Study methods and findings are reported in accordance with the Consolidated Criteria for Reporting Qualitative Research (COREQ)²⁰ (Supplemental Table S1). This study was IRB-approved at both institutions.

Data Collection

Parents from two large academic centers were mailed a survey including quantitative scales and open-ended questions.⁸ Demographic data, including gender, caregiver role, ethnicity, the child’s oncologic diagnosis, and the child’s age at death were parent-reported. Our qualitative inquiry and analysis focus on free-text responses to the final question: “If you would be willing to share, please tell us about your child. What would you like us to know about him/her?”

Quantitative Data Analysis

The study team selected quantitative measures previously evaluated and published in the primary analysis⁸ for (1) overlap with the code book of this secondary analysis and (2) potential to provide insight into psychosocial wellbeing and/or values of the parent writer. These measures included meeting DSM-5 criteria for post-traumatic stress disorder (PTSD) and/or prolonged grief, the child’s age at death, and reported degrees of spirituality and religiosity on a 4-point Likert scale. These demographic characteristics of parents and their decreased child were summarized using frequencies and proportions.

Qualitative Data Analysis

The two authors serving as primary coders (C.M.L. and B.H.) examined a subset of the free-text responses to develop a preliminary codebook. which was reviewed and finalized by a third coding team member (J.S.), before being applied to all transcripts. The coding team engaged in regular discussions, enabling them to challenge and refine interpretation of the free-text responses and application of codes. Coding was performed using qualitative software (Dedoose, version 10.0.35, 2025). Through an inductive, iterative review process,²¹ the two primary coders discussed discrepancies until consensus was reached and patterns were identified in the data to develop overarching themes. The third team member verified consensus, resolved any disputes, and finalized the thematic analysis. Thematic analysis was informed by prior research and literature on the concept of narrative repair²² and established components of stories²³ (i.e., stage setting/context, plot progression, tension, structure, character, and perspective/voice), in addition to expert input from a PhD-trained narrative ethicist (M.M.). After this inductive process of qualitative inquiry was complete, we compared our findings to existing theories and situated them into an existing conceptual framework.

The authors who conducted coding and the thematic analysis are experienced in qualitative methods and maintain awareness of personal and professional experience they bring. Of the primary coders, C.M.L. is a female post-doctoral pediatric hematology/oncology fellow, and B.H. is a male post-graduate research assistant. The third senior coding team member who refined and finalized the code book and thematic analysis, J.S., is a female pediatric hematology/oncology and palliative care physician-researcher.

The frequency of code occurrence was calculated by dividing the total number of parents using the code of interest (i.e., code presence rather than code count) in their narrative divided by the total number of parents completing question of interest (N=101) for all codes except for “no-answer,” wherein the total number of parents completing the survey (N=128) was used.

RESULTS

A total of 128 parents completed the survey in its entirety; 101 (79%) parents representing 81 children shared about their child. The median time from child’s death was 12 months (range 8–26 months). Twenty-seven (21%) parents representing 24 children either did not answer or shared an inability to answer the question. Most participants were white (N=84, 83%), non-Hispanic (N=92, 91%), and mothers (N=86, 85%) who identified as the primary caregiver (N=72, 71%). Nearly two-thirds of parents held an associate or bachelor’s degree or higher (N=66, 65%); all were native English speakers. The child’s age at time of death ranged from 0.5 to 29 years, with a mean of 13.6 years, and represented a variety of cancer diagnoses (Table 1). There was no difference in parent role for non-responders, with 14 mothers and 13 fathers forgoing the question of interest (Supplemental Table S2).

TABLE 1.

Participant demographics

Parent Information (N = 101)
Variable	Response	Number (%)
Gender	Female	86 (85)
	Male	15 (15)
Age (years)	18–24	0
	25–34	15 (15)
	35–44	28 (28)
	45–54	33 (33)
	55–64	22 (22)
	64+	2 (2)
	Declined to answer	1 (1)
Racial and Ethnic Background	Asian	1 (1)
	American Indian / Alaskan Native	1 (1)
	American Indian / Alaskan Native and White	4 (4)
	Black or African American	5 (5)
	Native Hawaiian or other Pacific Islander	1 (1)
	White, non-Hispanic	79 (78)
	White, Hispanic	5 (5)
	Other (e.g., Hispanic, Middle Eastern, Brazilian)	4 (4)
	Incomplete	1 (1)
Highest Level of Education	High school diploma or equivalent	35 (35)
	Associate degree	11 (11)
	Bachelor’s degree	28 (28)
	Master’s degree	20 (20)
	Doctorate (PhD, etc.)	1 (1)
	Professional (MD, JD, DDS, etc.)	3 (3)
	Other (e.g., LPN, some college)	3 (3)
Caregiver Role	Primary caregiver	72 (71)
	Secondary caregiver	28 (28)
	Other	1 (1)
Reported Degree of Spirituality	Very	36 (36)
	Somewhat	45 (45)
	Not very	12 (12)
	Not at all	6 (6)
	Declined to answer	2 (2)
Reported Degree of Religiosity	Very	23 (23)
	Somewhat	43 (43)
	Not very	18 (18)
	Not at all	15 (15)
	Declined to answer	2 (2)
Post-Traumatic Stress Disorder (PTSD)	Met DSM-5 criteria	41 (41)
	Did not meet DSM-5 criteria	60 (59)
Prolonged Grief	Met DSM-5 criteria	27 (27)
	Did not meet DSM-5 criteria	73 (72)
	Incomplete	1 (1)
Deceased Child’s Information (N = 81)
Average Age at Death	13.6 ± 7.7 years (range 0.5–29 years)-
Median Time from Diagnosis to Death	29 months (range 1–288 months)
Oncologic History	CNS tumor	28 (35)
	Solid Tumor, non-CNS	26 (32)
	Hematologic Malignancy	25 (31)
	Multiple	2 (2)

General Structure and Plot of Parental Narratives

Narratives were on average 61 words in length, ranging from 2 to 732 words, with a median of 39 words. Parental narratives of their children varied in terms of structure (e.g., verb tense, narrative arc) and content (Table 2). Parents tended to use a singular tense, i.e., either present (N=12, 12%) or past (N=92, 91%) tense verbs; nine parents (9%) used a combination of verb tenses. Most parents (N=91, 70%) made their child the direct subject of the narrative, whereas others implied that their child was the subject (N=10, 10%); less than half of parents (N=49) mentioned their child by name. Few parents (N=13, 13%) detailed their child’s cancer diagnosis or treatment, and only 12 parents (12%) explicitly mentioned their child’s death.

TABLE 2.

Qualitative analysis of a parent’s use of structure and plot in the narrative of their deceased child

Theme	Code	Definition	Example Quotes	Frequency‡
Structure and Plot	Explicit subject	Parent begins with a sentence using a name, pronoun, or “my child.” The child is the subject.	“His name was [Name].” (288A) “She was a gifted athlete and musician.” (570B)	91/101 (90%)
	Implicit Subject	Parent begins with a sentence that does not directly name or reference the child.	“Loved to sing Loved everyone. Loved Jesus.” (344B) “I could write a book.” (312A)	10/101 (10%)
	My child is	Parent refers to their child in the present tense.	“He is amazing and loves his family and friends. He loves soccer and dancing and being funny. He is my best friend.” (329B)	92/101 (72%)
	My child was	Parent refers to their child in the past tense.	“She maintained her sass and wit right up to the end.” (570A)	12/101 (12%)
	Explicit Mention of Death	Parent explicitly mentioned that the child died or is dead.	“They day before he died we went to his favorite park to take family pictures and he asked to go to Wal-mart.” (201A) “He died on August 17, 2018.” (506A)	12/101 (12%)
	Incorporation of Humor	Parent details moments of laughter and levity in the narrative of their child.	“After his first surgery to install his port his sister and I went in to sit with him in recovery and as he was waking up he was cracking jokes and he looked us straight in the eyes and seriously said that we weren’t much fun since he quit drinking. Needless to say everyone in the recovery room that was awake was laughing.” (258B)	12/101 (12%)

^‡Frequency is defined as the total number of parents using the code of interest (i.e., code presence rather than code count) divided by the total number of parents writing narratives.

Character Development of Parental Narratives

Narratives also varied by strategies employed for character development of the child (Tables 3). Parents highlighted their child’s appearance, traits, passions, humorous anecdotes, and faith: “She had the biggest brown eyes I’ve ever seen and I called her my ‘brown eyed girl’” (286A) and “Brave. Strong. Funny. Kind. Friend. Smart. Feisty. Honest. Sincere. Original. Awesome” (312A). Parents served as narrators of the story to emphasize the cancer battle, child’s legacy, and their hope to see their child again. One parent shared, “Can’t wait until my time on this earth is through and I can get a big [child’s name] hug again” (201A). Another noted, “We are going to start a scholarship in his name” (563B). Other parents emphasized the loss of a future and stolen time (N=24, 24%), with one parent writing: “Unfortunately, he was only 6 months old and was just beginning life” (333B). While a few parents (N=7, 7%) expressed helplessness in their child’s suffering, others (N=12, 12%) incorporated humorous anecdotes, ranging from passions for “knock-knock jokes and talking about poop” (523A) and pranks, to disinhibited post-operative conversations in anesthesia recovery. One parent recalled how her daughter “told everyone [doctor’s name] told her she needed to drink beer for medical reasons” (291A).

TABLE 3.

Qualitative analysis of a parent’s character development strategies of their deceased child

Theme	Code	Definition	Example Quotes	Frequency‡
Character Development: Humanization	Physical / Sensory Characteristics	Parent emphasizes the physical characteristics of their child, for example, their hair, smile, giggle, or eyes to evoke a visual and/or auditory representation of the child in the reader’s imagination.	“He had an infectious giggle and smile. Brightened everyone’s day. Red-headed and Elmo-loving!” (231B) “She had electric blue eyes, that nearly everyone commented on and she liked to dress in crazy outfits.” (573A)	21/101 (21%)
	Passions and Personality	Parent focuses on how the child lived and loved, including what they liked and/or did not like, and elements of their personality.	“He was quiet, but very smart & logical. He loved science, history, and all things related to Star Wars.” (262A) “She was smart and talented. She played the guitar and the french horn. She was creative and imaginative and expressed herself through her creativity… Everyone who met her fell in love with her.” (307A)	89/101 (88%)
	Funny	Parent describes the child as funny, silly, goofy, quirky, or a prankster. This may or may not be in the context of a humorous anecdote.	“He was a prankster. He loved scaring people.” (288A)	17/101 (17%)
	Believer	Parent emphasizes the child’s faith and may mention words such as “God,” “Lord,” “Jesus,” and church. Applies to the faith of the child.	“He loved the Lord Jesus Christ with all his heart.” (582A) “She was up there in front of everyone, on her oxygen, with her Harry Potter wand, leading the hymns.” (274A)	12/101 (12%)
Character Development: Medicalization	Diagnosis / Treatment	Parent details on the cancer diagnosis (e.g., name of cancer, extent of cancer), medical complications, and/or manner of death rather than personal characteristics.	“He was a high school senior while in the hospital ([date range]). He lived on 6 floor North for 8 months. Ask any doctor, NPA, RN or anyone that knew [name] on 6 North. They will tell you about him. ” (506B)	13/101 (13%)
Character Development: Narration	Otherworldly	“Children with cancer are just built different, not of this world.” Parents describe their child as an “angel,” “old soul,” “beyond their years,” “living life to the fullest,” and/or “lights” for others, setting them apart from their siblings, peers, and/or fellow human beings.	“He came into this world as an ‘old soul,’ he saw things with a unique perspective as soon as he could articulate things. He had a huge spirit, bright light ascending spirit.” (612A) “Endlessly optimistic cheerful, steady. We all (friends, family) wear a wrist band in his honor that lists his “super powers”: Resilience - humility - kindness – laughter. He was and is a gift to all who knew him.” (550A)	54/101 (53%)
	Battle Language	Parent uses “battle language” that describes their child as a hero, superhero, fighter, and/or protector.	“He was already in pain but he was so strong. He’s my Golden Warrior.” (320A) “He quickly started going by ‘[Child’s name] the Brave!’ He truly lived up to the name! Most days I do not even want to think about getting out of bed, but then I force myself up because if my infant son could fight cancer majority of his life with a HUGE smile on his face and love in his heart, than (sic) the LEAST I can do is get up and TRY to be brave with the rest of my days, days good or bad.” (231A)	19/101 (19%)
	Stolen Future	Parent emphasizes or includes the age of the child at their death. They may describe that their child died too young or offer insight into what the child would have been if they grew-up.	“too soon.” (572A) “She deserved to live a full life.” (590B) “The world would have been a better place with her.” (584A)	24/101 (24%)
	Parental Helplessness	Parent details regret that their child suffered and a sense of helplessness and/or loss of control.	“A fighter, he wanted to live so bad I hope I could feel better about this someday, I wish I could of helped him, my poor [Child’s name].”(544B) “My child didn’t deserve this horrible situation.” (237A)	7/101 (7%)
	Missed	Parent explicitly notes that they and/or siblings miss the child.	“We will never forget her sweet, gentle soul and miss her every hour of every day.” (574A) “I miss him so much with tears in my eyes as I write this.” (289A) “He made us better. We are worse without his joy and magic. I miss him so much.” (262A)	18/101 (18%)
	“See you again”	Parent describes faith that they will be with the child again one day. Applies to the faith of the parent.	“He loved the Lord Jesus Christ with all his heart and we know we will see him in heaven someday.” (582A)	2/101 (2%)
	Legacy	Parent details the legacy their child leaves behind. The parent may seek to be their child’s legacy moving forward. Parents may offer tangible legacy items, including social media posts, quotes, obituaries, prayer cards, etc.	“Her love lives on.” (570A) “I could write a book. [Child’s name] is the greatest miracle of my life. He taught me more in his 9 years than I could ever imagine.” (312A)	34/101 (34%)
	No answer	Parent chose to not answer the question or expresses the inability to answer.	Blank “This question is hard for me. It seems a little trite. I can talk easily of [child’s name] and never tire of describing him. But not this way.” (205A)	27/128 (21%)

^‡Frequency is defined as the total number of parents using the code of interest (i.e., code presence rather than code count) divided by the total number of parents writing narratives. For “no answer,” N = 128, the total number of parents who completed the survey and viewed the question, choosing to provide an answer or not.

Over half of parents (N=54, 53%) characterized their child as “otherworldly,” noted to be an “angel,” “old soul,” “beyond [their] years,” and/or “lights.” A parent wrote that her daughter:

“Loved to sing. Loved everyone. Loved Jesus. Was not afraid to die. Very giving. The day after she was told there was nothing else that could be done, she was at [hospital] clinic…supporting the children. She donated her tumors to [hospital], hoping they would be the tumors that researchers found the cure with”

(344B).

Another parent reflected, “I don’t think you can teach other people how to be the kind of person she inherently was. She had always figured out many of life’s secrets, long before she became sick” (272B).

Emergence of Themes within a Framework of Narrative Repair

Codes within the “character development” theme fell into the following three thematic categories: medicalization, humanization, and supernaturalization, which we situated in a conceptual model (Figure 1). Medicalization included narrative elements focused on the child’s medical diagnosis, history, and treatment. Codes related to humanization centered on a child’s personhood and humanity, including their personality, quirks, passions, and the physical appearance or experience of knowing them. Finally, supernaturalization included codes reflecting “otherworldliness,” indicating the child’s existential impact and legacy, beyond their physical existence alone.

Figure 1. Proposed progression through “narrative repair” in early parental bereavement following the death of a child from a chronic illness, including cancer.

Parents who employed medicalization rarely included elements of humanization and supernaturalization. A second subset of parents utilized humanization alone, and a third utilized humanization and supernaturalization. We posit that parents oscillate through disequilibrium (i.e., dissonance) and equilibrium (i.e., consonance) among these three narrative strategies, suggestive of dynamic, progressive narrative repair.

Of the 101 parent narratives analyzed, about half (N=48, 48%) utilized humanization only, and more than a third (N=38, 37%) employed both humanization and early supernaturalization. Only one parent used only medicalization and 13 (13%) narratives featured both medicalization and humanization. Only one parent narrative utilized all three strategies, with the narrative progressing in sequence from medicalization to humanization to supernaturalization (Table 4).

TABLE 4.

Representative quotes characteristic of the proposed stages of narrative repair outlined in Figure 1

Stage	Stage Characteristics	Representative Quote	Additional Timpoint Information for Parent Exemplar
I	Medicalization predominates with minimal (if any) humanization. If humanization is employed, it is a flicker, and the crux of the narrative is medicalization. Supernatrualization is not employed.	“[Name] was a [hospital] baby at 18 months old in [year] (#13975). Diagnosis was Neuroblastoma and Adrenal Carcinoma growing on top of each other on her adrenal gland and the Adrenal had spread to her lung. she was never sympathetic (sic)…caught by chance during a heart cath for her heart murmur. 2 surgery (sic) later it was over…no chemo. No radiation… [Name] had Turner’s Syndrome and Li-Fraumeni Syndrome…I had her for 26 years that she was healthy and happy and saw more in her short life than most see in a lifetime. We had said our piece and in the end there was nothing left to say other than See Ya Soon.” (311A)	12 months from child’s death to parent completion of question 288 months elapsed between child’s diagnosis and death Child was 25 years old at death
II	Medicalization remains present but is employed as a minor strategy in construction of the narrative, with the focus being on humanization. Supernatrualization is not employed.	“He greeted everyone with a smile. He smiled even when he experienced pain. He was a fighter from the time he was born at 26 weeks. Despite his disability and cancer diagnosis, he worked hard to his milestones. He loved school and making friends. His best friend was his grandpa who watched him after school and days when school was out. He loved for grandpa to tell him stories. He enjoyed watching game shows and TV westerns with his Dad and me. He also loved listening to music. His taste in music ranges from Barney to Snoop Dog to Mariah Carey. I will miss his smile, laugh, and kisses (which he gave often).”(259A)	17 months from child’s death to parent completion of question 70 months elapsed between child’s diagnosis and death Child was 22 years old at death
III	Medicalization, humanization, and supernatrualization co-exist.	“[Name] was the joy of my (our) life. He was such a happy go lucky little boy. With all he had been through since he was 12 months old, it (cancer) never took the excitement or happiness from his life. Since he was diagnosed at such a young age, it was all he ever knew hospitals, tests, procedures. As long as we went through this, once the procedure was done, we would say ‘Lets hug it out!’ and comforted him, he was ok and would go back to playing and smiling. He was so blessed that in spite of all his chemo treatments, radiation, 4 facial surgeries, he never complained, or had any bad side effects or post-op issues and never succumbed to any problems. We were able to still let him be a baby, toddler, and little boy and experience life and the fun and beauty of this world because he was so healthy through all of his 3.5 of treatments. Even last year, when we had to make the decision to remove his left eye, we turned it into a positive - he loved his superheros and spider man, Hulk, and flash, and Cyborg (he had only one eye) were his favorites. [Name] was like a Super Hero - fearless, kind, patriotic, strong, fearless, Happy, a friend to everyone! One of the Greatest Blessings this world has known!!” (552A)	9 months from child’s death to parent completion of question 39 months elapsed between child’s diagnosis and death Child was 4 years old at death
IV	Humanization is thre predominant narrative strategy. There are references to supernatrualization, but the focus of the narrative is humanuzation.	“[Name] was for sure a mama’s boy. He and I had a strong bond even before his diagnosis. Will was very smart. He was very introverted but if you were one of the lucky ones he let in you saw a very talented and very comical guy. [Name]’s strength during the entire 6.5 years was amazing. He never complained until the last few months. Even then I don’t think I would say he complained. He was so very tired of feeling bad and hurting all of the time. He fought so very hard. He was an inspiration to so many people. Will was and always will be my hero! I try my best to live each day and be more like [name].” (278A)	17 months from child’s death to parent completion of question 76 months elapsed between child’s diagnosis and death Child was 19 years old at death
V	Humanization and supernatrualization are both prominent strategies used in the narrative. The parent has a greater focus on supernaturalization as compared to the prior stage.	“[Name] was a saint. Catholic priests in Boston invoke her often in their prayers. She started a foundation “BuildtheFaith.org” to bring the faith to impoverished locations around the world. The foundation has already built 3 churches, established multiple spiritual retreats, and continue to bring many people closer to the faith (including members of her care team). Her prayer groups in Facebook ([page name]) has > 50 thousand members. People around the world pray to [name]. She was the happiest person I have ever known.” (539B)	21 months from child’s death to parent completion of question 61 months elapsed between child’s diagnosis and death Child was 11 years old at death

DISCUSSION

In our secondary analysis of parents’ responses to the prompt “Tell us about your child,” nearly 80% of participants chose to share the story of their deceased child, suggesting an important role for narrative in early parental bereavement. When parents in our study were asked to talk about their child, the responses took the form of narratives. Although these parental narratives varied in structure, they consistently centered on their child as a person, humanizing and narrating the child’s life, battle, and continued legacy, with only rare mentions of the cancer diagnosis, medical treatment, or even death.

We hypothesize that the life-threatening illness or death of a child “ruptures” the story of the child’s parents and family,²⁴ leading to application of narrative strategies. Our results corroborate prior research using storytelling to explore and create meaning in bereavement. Although the impact of storytelling may be challenging to quantify, a narrative-based qualitative study of caregivers experiencing chronic stress found that the opportunity for reflection and integration afforded by storytelling was associated with lower psychometric and physiologic measures of stress.²⁵ Case in point, guided writing in bereaved undergraduates, aimed at helping participants express their thoughts and emotions, was associated with a reduced incidence of prolonged grief disorder and symptoms of depression and PTSD.²⁶ Therefore, offering opportunities for reflection and storytelling may be perceived as both meaningful therapeutic to bereaved parents.

Two noteworthy themes included (1) the lack of parents’ focus on the death of their child in their conceptualization of “plot” and (2) varied use of verb tense. Parents served as narrators who emphasized their child’s otherworldliness, battle, legacy, and being missed, yet few parents conveyed helplessness or signs of a stolen future. Only 12 (12%) parents even mentioned their child’s death--instead focusing on life after the death of a child—suggesting parents may not view their child’s death as essential to the plot of their child’s story. Additionally, parents used a spectrum of verb tenses, with nearly a third of parents using either present tense or a combination of past and present tense. Practically speaking, clinicians often worry about saying the wrong thing when speaking to bereaved parents, including verb tense, but out data demonstrates that there may not be one “right” verb tense. Narrative analysis of parental narratives of their child at different timepoints during bereavement, focusing on the inclusion (or omission) of plot points and patterns in verb tense, may provide greater insight into parents’ bereavement experiences and narrative reconstructions.

The parent narratives in our analysis revealed a continuum of medicalization, humanization, and supernaturalization (Figure 1), reflective of “narrative repair,”^27–29 a process originally conceptualized within the social sciences literature surrounding the therapeutic role for narrative in psychopathology. Narrative repair was most notably depicted by American psychiatrist R.J. Lifton in Vietnam War veterans, characterized as a “psychic rebuilding” towards order, hope, and meaning after a transformative event.^24,30 Such transformative events, including war (or the death of a child), required building a bridge between the self and the rest of the world in order to reintegrate, with narrative repair serving as the constructive mechanism to achieve resolution.^24,30 Though Lifton’s psychohistories first illuminated the concept of narrative repair, sociologist Arthur Frank introduced the term^31,32, and narrative ethicist, Martha Montello, identified and described Lifton’s three stages of narrative repair as relevant to clinical medicine.^24,30 The first stage is connection, meaning an individual begins the process of leaving isolation after a trauma, seeking connection with other human beings. The second stage is cohesion, wherein an individual begins to create meaning within their narrative, binding it together again after it has ruptured. The final stage is movement, best understood as “living past an ending,” where an individual is able to move forward (though never fully on), working towards a state of resolution rather than a pure solution.^24,30 Our findings bring the work of Lifton, Frank, and Montello into pediatrics, a previously unexplored landscape of “narrative wreckage,”³² and anchor narrative repair in data rather than observations or anecdotes. In our proposed conceptual model (Figure 1), we posit that parents employ medicalization in the early stages of narrative repair—framing their child’s story in the clinical details of their diagnosis—as a way to connect with other parents who have faced similar experiences with their child. This medicalization stage of narrative reconstruction may serve as a bridge out of the isolation that often accompanies serious illness and child death. Humanization reflects a parent’s shift towards cohesion and integration, with less emphasis on the medical facts (i.e., connection with others) and more on their child’s personhood and humanity (i.e., connection with the child). Supernaturalization marks a further transition: the ability to “live past the ending” of the story and move beyond medical and human realms. Although Montello’s clinical framework identified from Lifton’s psychohistories consists of three discrete stages (i.e., connection, cohesion, movement), we believe that bereaved parents oscillate between medicalization, humanization, and supernaturalization. Parents exist in states of both disequilibrium and equilibrium to achieve resolution as they repair their narratives.

Our proposed model of narrative repair (Figure 1) is relevant for any parent whose child died from a chronic illness, cancer or otherwise, as well as the pediatric providers who care for them. Chronic, life-threatening pediatric illnesses, cause medicalization of the child to become the “new normal,” irreparably altering how a parent conceptualizes their child, role as a parent, and confines of their social circle.³³ Parents are even given a new vocabulary, most often focused on the diagnosis and treatment of illness, rather than the maintenance of humanity. These physical, psychosocial, existential, and linguistic alterations do not end with the life of the child and contribute to the existence of the medicalization narrative. Thus, our framework of narrative repair not only has psychological relevance in understanding a bereaved parent’s unique experience, but also clinical implications for acting on this new approach to framing and conceptualizing bereavement. Medical and psychosocial providers may be able to help parents in the medicalization phase progress through to humanization by asking them to share in reflections and memories of their child as a narrative act. Most parents in our study existed in the humanization phase, suggesting that bereaved parents desire to tell their story, particularly a story centered on their child’s personhood rather than their diagnosis or death. Our data also suggest that humanization and supernaturalization may co-exist in early parental bereavement. It is unclear if this narrative restructuring is linear or if supernaturalization is, or should be, the ultimate ‘goal.’ Continued engagement of parents in narrative reflection is needed to determine if humanization and supernaturalization co-exist or diverge throughout bereavement and how this narrative structure relates to other bereavement outcomes.

Our findings must be considered in the setting of the following limitations to this secondary analysis, which include recall bias and limitations of existing data. As with most prior bereavement research, respondents were primarily white mothers. In addition, the overall participation rate for the initial survey⁸ which included our question of interest was 37%. Thus, it is possible that the results of this study underrepresent the needs of families who did not participate, and as such, should be interpreted with a lens aimed at identifying priorities for further work rather than generalizability. Demographic data were self-reported by respondents; therefore, we do not have demographic data for the survey non-respondents to compare. Our study did not evaluate the psychologic or physiologic impact of employing a narrative approach in parental bereavement, an area that reflects an opportunity for further exploration. Additionally, while the framework of narrative repair may not approximate or be generalizable to all parental bereavement, particularly if a child did not have significant interaction with the medical system leading up to their death, it functions as a starting point for exploring grief in other bereaved parents whose children had non-oncologic chronic health conditions. Our study is not statistically powered to analyze patterns between stages of narrative repair (Figure 1; Table 4) and time since the child’s death, and this represents an opportunity for future investigation in a larger, more diverse sample size of parents. Within these limitations, qualitative methods enable rich ideas and concepts to emerge from bereaved parents’ experiences through hypothesis generation rather than hypothesis testing.

CONCLUSION:

Integration of storytelling in parental bereavement, informally or formally, may provide a unique benefit for caregivers, clinicians, researchers, and communities.³⁴ Narrative yields a nuanced understanding of grief, enabling the writer to regain some control after a transformative event and helping the reader to observe the life of another, explore things that he or she may have been blind to, and find solidarity with others.²² Many available parental bereavement supports, such as groups and workshops, have financial and temporal barriers to entry that prevent access for all families.³⁵ There are no physical barriers to engaging in storytelling; all that is required to begin is a prompt or question, big or small, enabling a richer participant pool. Employing a narrative approach in early parental bereavement may be beneficial in a parent’s search for meaning after the death of a child, serve as an opportunity for engagement with parents who may otherwise be excluded from traditional community bereavement supports, and function as a means for providers to identify who, when, and how to provide additional support, perhaps even in the form of storytelling itself.

Supplementary Material

Supplement 1

Supplemental Table S1. Consolidated criteria for reporting qualitative studies (COREQ): 32-item checklist. The COREQ checklist is summarized in this table, developed from: Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care. 2007. Volume 19, Number 6: pp. 349 – 357.

Supplemental Table S2

Supplemental Table S2. Demographics for parent non-responders to the question of interest

Article Summary.

We asked bereaved parents whose child died from cancer, “Tell us about your child,” yielding rich narratives of repair, progressing through medicalization, humanization, and supernaturalization.

What’s Known on This Subject

Narrative medicine and narrative competence are emerging fields within clinical medicine, emphasizing the power of subjective experiences through storytelling. Humans use narratives to create meaning, and these stories serve as points of resonance with and insight into the human experience.

What This Study Adds

Our results situate within a framework of narrative repair; bereaved parents use their narrating voices to progress through a staged reentry into the world after their child’s death. This framework may help to better understand and support parents in bereavement.

Abbreviations:

DSM

Diagnostic and Statistical Manual of Mental Disorders

IRB

Institutional Review Board

PTSD

Post-Traumatic Stress Disorder

Data Sharing Statement:

Deidentified individual participant data will made available upon reasonable request.