End-of-Life Decisions in Pediatric Cancer Patients

Abstract

Background:

End-of-life decisions (ELDs) have been investigated in several care settings, but rarely in pediatric oncology.

Objective:

The aims of this study were to characterize the practice of end-of-life decision making in a Dutch academic medical center and to explore pediatric oncologists' perspectives on decision making.

Methods:

Between 2001 and 2010, in a specified period of 2 years, 57 children died of cancer. The attending pediatric oncologists of 48 deceased children were eligible for this study. They were requested to complete a retrospective questionnaire on characteristics of ELDs that may have preceded a child's death. ELDs were defined as decisions concerning administering or forgoing treatment that may unintentionally or intentionally hasten death.

Results:

In 31 of 48 cases (65%) one or more ELDs were made. In 20 of 31 cases potentially life-prolonging treatments were discontinued or withheld, and in 22 of 31 cases drugs were administered to alleviate pain or other symptoms in potentially life-shortening dosages. Frequently mentioned considerations for making ELDs were no prospects of improvement (n=21;68%) and unbearable suffering without a curative perspective (n=13;42%). ELDs were discussed with parents in all cases, and with the child in 9 of 31 cases. After the child's death, the pediatric oncologist met the parents in all ELD cases and in 11 of 17 non-ELD cases. Pediatric oncologists were satisfied with care around the child's death in 90% of the ELD cases versus 59% of the non-ELD cases.

Conclusions:

In two-thirds of cases, ELDs preceded the death of a child with cancer. This is the first study providing insights into the characteristics of ELDs from a pediatric oncologist's point of view.

Introduction

Although survival rates of childhood cancer have increased during the past decades, approximately 25% of children diagnosed with cancer eventually die.¹ Palliative care for children with cancer is guided by a multidisciplinary pediatric palliative care team, consisting of a pediatric oncology team (a pediatric oncologist, pediatric oncology nurses, a pediatric psychologist, a spiritual worker, a social worker, a child life specialist, and a pain team) as well as community health care professionals, that is, a general practitioner (GP) and community nurses.

According to the World Health Organization (WHO), one of the main objectives of pediatric palliative care is to prevent and relieve suffering of both the child and the family.² To improve the quality of life during the terminal phase, end-of-life decisions (ELDs) can be made. ELDs are defined as decisions concerning administering or forgoing treatment that may unintentionally or intentionally hasten death.^3,4 In clinical practice, these ELDs range from decisions to abstain from potentially life-prolonging treatments, such as mechanical ventilation or drip feeding, to decisions to alleviate pain or other symptoms by using drugs in dosages that may involve a life-shortening effect.^3,4 In the Netherlands, ELDs may also include decisions for euthanasia or physician-assisted dying. According to the Euthanasia Act, euthanasia and physician-assisted dying can be legal for minors aged 12 to 16 years if parents agree with the request of the child, and for minors aged 16 or 17 years if parents are informed.⁵ For children below the age of one year old, the use of drugs in neonates with the intention to hasten death can be acceptable if five requirements according to the Groningen Protocol for Life-termination in Newborns are fulfilled.⁶

Several studies have been performed investigating ELDs in children in a variety of settings and causes of death.^3,4,7–16 Many of these studies examined in particular ELDs in neonatal care or ELDs in pediatric intensive care units.^8,11–16 Among the studies that focussed on different pediatric settings, a Dutch study performed in 2005 showed an ELD rate of 36% in all dying children suffering from a variety of diseases with an age range from 1 to 17 years old.³ This finding was quite similar to a Belgian study from 2010 that found an ELD rate of 36.4%.⁷ Another study, also Dutch, reported that treatment was withheld in 44% of patients and withdrawn in 56% of patients.⁴

Decision making occurs in the majority of cases in conjunction with communication with the child's parents or caregivers.^17–19 Involvement of children, however, does not always take place, although even young children might understand what is happening and can participate in this communication.^20–22 Besides communication with patient and parents, health care professionals frequently speak to other physicians about ELDs.³ In pediatrics, ELDs are recognized as team decisions in consultation and agreement with the families.⁴

No attention has been given in the literature to ELDs solely from a pediatric oncologist's point of view. Children dying of a malignancy represent a distinct population as death is usually anticipated.²³ Therefore, characteristics such as frequency, considerations, and communication around ELDs might be different from those for other severely ill children. Moreover, to the best of our knowledge, satisfaction of pediatric oncologists with pediatric palliative care has never been established, and it is not yet clear whether pediatric oncologists perceive palliative care for children differently when ELDs are made.

The aims of this study are to investigate the frequency and considerations for ELDs in a pediatric oncology setting, to explore communication around ELDs, and to investigate whether pediatric oncologists were satisfied with care around the child's death.

Methods

Sample

In a specified period of two years between 2001 and 2010, 57 children died after treatment for cancer at the Department of Pediatric Hematology/Oncology Erasmus MC-Sophia Children's Hospital in Rotterdam, the Netherlands. To maintain both patients' and pediatric oncologists' anonymity the specific 2 years cannot be revealed. Relative to the child's death, the attending pediatric oncologist completed the questionnaire within 2 years afterwards.

At the start of the study, all 10 attending pediatric oncologists were guided with instructions on how to complete the questionnaire on the basis of one of their patients who died in the specified period of 2 years. Afterwards, the pediatric oncologists were able to complete the questionnaires for the remaining patients themselves.

Questionnaire

The questionnaire used in the present study was based on a previous study by Vrakking and colleagues and included questions regarding patient and disease characteristics and potential ELDs.³ They adapted their pediatric questionnaire from a validated questionnaire used in an adult patient setting.³ For the current study, we used the previously established definition of ELDs: “decisions that may unintentionally or intentionally hasten death. These include decisions to withhold or withdraw medical treatment and decisions to alleviate pain or other symptoms by using drugs with a possible life-shortening effect.”⁴ Specific examples of ELDs are, for instance, withholding or withdrawing certain medications such as antibiotics, chemotherapy, or other treatments such as nutrition, intubation, or resuscitation.

When such a decision had been made, pediatric oncologists were additionally asked for their consideration(s) for the ELDs, the estimated amount of time that life was shortened, communication about the ELDs with the patients, the parents, and health care professionals, and what they perceived as the most impressive ELD. For all cases it was assessed whether the pediatric oncologist met the parents after the child's death, and whether they were satisfied with care around the child's death. Satisfaction was rated on a 4-point Likert scale (1=very satisfied; 4=not satisfied at all).

The following patient characteristics were collected in the questionnaire: gender, age at death, place of death, and diagnosis. Diagnosis was categorized into three groups: hematological malignancies (including acute lymphoblastic leukemia, acute myeloid leukemia, non-Hodgkin lymphoma, hemophagocytic lymphohistiocytosis, myelodysplastic syndrome, and aplastic anemia), solid tumors (neuroblastoma, clear cell sarcoma, rhabdomyosarcoma, Ewing's sarcoma, osteosarcoma, and Wilms' tumor), and brain tumors (medulloblastoma, ependymoma, primitive neuroectodermal tumor, astrocytoma, and pontine glioma).

Data analysis

Data were analysed using the Statistical Package for the Social Sciences, version 21.0 (SPSS Inc., Chicago, IL). Patient characteristics are presented either as median and range or number and percentages. Information about characteristics of ELDs, such as frequency and considerations as well as information about communication and evaluation after the child's death, are presented as number and frequency.

Results

Patient characteristics

For 9 of the 57 deceased children, no complete information was available about the ELDs, for instance because they were transferred to another hospital or ward. Therefore, these 9 cases were excluded, leaving 48 cases eligible for the current study. All 10 attending pediatric oncologists completed the questionnaire for one or more of their deceased patients. The distribution of questionnaires among the pediatric oncologists is presented in Table 1.

Table 1.

Distribution of Questionnaires among Pediatric Oncologists

Cases	No. questionnaires
#1	5
#2	1
#3	1
#4	2
#5	4
#6	10
#7	6
#8	6
#9	9
#10	4

Of the 48 deceased children, 32 were male (67%) and 16 were female (33%). Median age at death for the 48 patients was 7 years (range 0–19 years). Twenty patients (42%) were younger than 6 years, 17 (35%) were between 6 and 11 years, and 11 (23%) were older than 11 years. Twenty-three patients (48%) suffered from a hematological malignancy, 13 (27%) from a solid tumor, and 12 (25%) from a brain tumor. Twenty-nine children (60%) died at home, 10 children (21%) died at the intensive care unit of the hospital, six children (13%) died on the pediatric oncology ward/pediatric emergency department of the hospital, and three children (6%) died in another hospital.

Frequencies of ELDs

In 31 cases (65%) one or more ELDs were made (Table 2). In 20 cases it was decided to withhold or withdraw possible life-prolonging treatment (forgoing therapy group). In 13 of these cases a treatment was withheld and in 12 cases a treatment was withdrawn, while taking into account hastening death. In the oncologists reporting, in one case a treatment was withheld and in six cases a treatment was withdrawn with the understanding that death could be hastened. Withholding treatment included hospital admission (n=5), ventilation (n=5), chemotherapy (n=4), drip feeding (n=3), other nonspecified medication (n=2), radiotherapy (n=1), parental feeding (n=1), surgery (n=1), antibiotics (n=1), blood transfusion (n=1), and others (n=1). Withdrawing treatment included ventilation (n=7), chemotherapy (n=4), other nonspecified medication (n=1), and others (n=1).

Table 2.

Frequency and Characteristics of End-of-Life Decisions (ELDs)

	No.	Most impressive decision
No ELD	17
One or more ELD(s)	31
Type of ELD^a
1) Forgoing therapy group	20
Withholding treatment, taking into account hastening death	13	8
Withholding treatment with the explicit intention of hastening death	1	-
Withdrawing treatment, taking into account hastening death	12	5
Withdrawing treatment with the explicit intention of hastening death	6	5
2) Administering drugs	22
To alleviate pain or other symptoms, taking into account the possibility of hastening death	19	11
To alleviate pain or other symptoms, partly with the intention of hastening death	3	1
To alleviate pain or other symptoms with the explicit primary intention of hastening death	0	-
Considerations for ELDs^a
Without prospects of improvement	21
Unbearable suffering without a curative perspective	13
Request of parents	9
Expected to die soon	8
Treatment too intensive	6
Complications of the treatment	5
Expected poor quality of life	3
Shortening of the dying process	3
(Supposed) request of the patient	3
Parents could not handle the situation anymore	1
Other	5
Estimated shortening of life
<24 hours or none	11
1–7 days	9
>1 week	8
Not known	3
Communication about ELD^a
Child	9
Parents	30
Colleague pediatrician	27
Multidisciplinary meeting	24
Nursing staff	21
Social worker	4
Spiritual assistant	1
Legal expert	1
Others	5

More answers possible. Right column represents most impressive decision according to pediatric oncologists.

In 22 cases drugs were administered to alleviate pain or other symptoms, in three cases partly with the intention to hasten death. In none of the patients, a drug was administered with the primary intention to hasten death. The following pain medication were prescribed: morphine (n=17), midazolam (n=6), fentanyl transdermal (n=2), pentobarbital (n=2), tramadol (n=1), and ketamine (n=1).

In 17 cases (35%), no ELD was made. The majority of these children (n=12; 71%) died at home.

Considerations for ELDs

Frequently mentioned considerations for ELDs according to the pediatric oncologist were (more answers possible): no prospects of improvement (n=21; 68%), unbearable suffering without a curative perspective (n=13; 42%), request of parents (n=9; 29%), and/or expected to die soon (n=8; 26%) (Table 2). In most cases pediatric oncologists estimated that ELDs had shortened life by less than one week (n=20; 65%) (Table 2).

Communication about ELDs

ELDs were discussed with 9 of the 31 children, who had a median age of 10 years (range 7–17 years). Pediatric oncologists mentioned as reasons for not discussing ELDs with the child (more answers possible): child was too young (n=14), child was unconscious (n=8), and no time for discussion (n=1). The emotional state of the child was never mentioned as a reason for not discussing ELDs. All documented ELDs were discussed with the parents (in one ELD case this information is lacking).

ELDs were in 27 cases discussed with a colleague pediatrician (n=1 missing). In 25 of the 27 cases, colleague(s) agreed on the decision and in two cases one colleague agreed whereas another disagreed. In 24 of 31 cases ELDs were discussed within a multidisciplinary meeting, and in two cases there was no multidisciplinary meeting (n=5 missing). In addition, ELDs were also discussed with the nursing staff (n=21), a social worker (n=4), a spiritual assistant (n=1), a legal expert (n=1), and/or others (n=5).

Evaluation

Several weeks after the child died, the pediatric oncologist met the parents in 11 of the 17 cases (65%) in which no ELDs were made, and in all cases (n=31) in which ELDs were made. The children of parents who did not meet the pediatric oncologist afterwards all died at home.

In general, pediatric oncologists were (very) satisfied with care around the child's death. In cases in which ELDs were made they were more often satisfied than when no ELDs were made (n=28; 90% [data of one parent missing] versus n=10; 59% [data of one parent missing]). Most pediatric oncologists recognized that the decision of administering drugs taking into account the possibility of hastening death was the most impressive decision for them (n=11). Five of the six pediatric oncologists who had withdrawn treatment with the explicit intention of hastening death reported this as the most impressive decision (Table 2).

Discussion

The present study reports on pediatric oncologists' perspectives on ELDs and characterizes the practice of end-of-life decision making in a Dutch academic medical center. To our knowledge, this is the first descriptive evaluation providing insights into the characteristics of ELDs solely from a pediatric oncologist's point of view. We showed that in two-thirds of cases, ELDs were made preceding the death of a child with cancer. Often a possible life-prolonging treatment was forgone and/or pain or other symptoms were alleviated with drugs that potentially hastened death.

The percentage forgoing treatment as well as alleviation of pain and symptoms by using drugs in dosages that may involve a life-shortening effect in children with incurable cancer was higher compared with the Dutch death certificate study (65% versus 36%).³ This difference may be explained by our selected group of patients, that is, children dying of cancer. In this particular patient category, unexpected deaths occur less often, and pain and symptom management are considered extremely important.^24–27 Still, in one-third of deaths in the present study no ELDs (defined as decisions concerning administering or forgoing treatment that may unintentionally or intentionally hasten death) were made. It is plausible that in these cases deaths were unexpected, treatment was continued until the child died, or no life-prolonging treatment options were left and alleviation of pain or other symptoms with drugs that might hasten death were not necessary. In addition, the majority of these children in which no ELD was made died at home. Home death implicates that the burden of care lies with community health care professionals, which can be the GPs and community nurses. In some cases, the GP could have made one or more ELDs that were not discussed with the pediatric oncologists. This is, however, less likely as most patients and families benefit from a close collaboration between members of the multidisciplinary pediatric palliative care team.

Frequently mentioned considerations for ELDs were no prospects of improvement and unbearable suffering without a curative perspective. This is in agreement with previous studies that showed that for oncologists, biomedical aspects such as the child's prognosis and comorbid conditions, were essential considerations in ELDs.^28–31 In approximately one-third of the cases in which ELDs were made, parents requested the ELDs. A previous Dutch study reported that 51% (n=32 of 63) of the pediatricians received at some point a request from parents to end the life of their severely ill child.³ Forty-four percent (n=14) had ever granted such request from the parents.³ In Canada, the United States, and the United Kingdom oncologists were less often requested by parents to end the life of a child, and 9% ever granted such a request.³¹ This difference is significant, and perhaps related to how palliative care is administered in different countries, as well as religious affiliation and/or beliefs on both the part of the physicians and the parents.

In most cases in our study pediatric oncologists estimated shortening of life due to the ELDs in children to be less than a week. In comparison, in a previous study including all causes of pediatric death, shortening of life was estimated to be less than a week in 3 of the 76 cases.³¹ In terminally ill children with advanced cancer, it is difficult to assess whether or not drugs such as opioids or benzodiazepines shorten life. In fact, it is sometimes stated that they might also prolong life by making a child more comfortable, having to work less hard to breathe.

In the current study, all ELDs were discussed with the parents (data of one parent missing). This is in line with former studies in parents and caregivers of children with incurable cancer or other life-limiting diseases.^17–19 In the present study, children were included in communication whenever feasible. This is a meaningful finding as the involvement of children is of great importance.³² Reasons for not communicating decision making were that the child was unconscious or perceived too young, congruent with previous studies.^20,33,34 This communication in pediatrics is different in comparison with adult palliative care.³⁵ The more pronounced involvement with parents is obviously a reflection of the fact that parents are responsible, being guardians of their child, more than relatives are responsible for adult patients.³⁶

Several weeks after the child died, the majority of parents responded to the invitation to meet the pediatric oncologist. In the cases in which parents did not meet the pediatric oncologist afterwards, no ELD was made, and all six children died at home. These parents might have had less intensive contact with the pediatric oncologists and therefore felt less need to meet afterwards than other parents did.

The finding that the pediatric oncologists felt more often satisfied with care when ELDs were made, is notable. In cases in which ELDs were performed, a comprehensive decision-making process takes place with discussions with the parents (and child when feasible). The positive perspective of pediatric oncologists could partially be due to being allowed to be part of the family's most emotional, intimate period. In addition, it might feel gratifying to be able to provide care after not being able to cure the disease. However, it could also be due to the feeling of the pediatric oncologist to have more control over the situation, or, once the ELD is accepted, due to calmer guidance of the dying process of the child and guidance for the parents. On the other hand, still nearly 60% of pediatric oncologists were satisfied with care in the cases where no ELD was made. It is possible that in the majority of these cases, the child died peacefully.

Future studies should explore the emotional, spiritual, and psychological states of physicians in terms of characterizing their increased satisfaction with palliative care in cases where one or more ELDs were made. Moreover, the effects that a close collaboration between GPs and oncologists have on families from the standpoint of providing continuity of care are important to explore in future studies.

A limitation of this single-center study is the retrospective study design, which could lead to recall bias. We tried to minimize recall bias by asking the pediatric oncologists to get their information from the patient charts. However, certain questions, such as satisfaction with care, could be obtained from memory only. Moreover, we did not separate the sudden, unexpected deaths from the nonsudden deaths. In situations where death was sudden, ELDs were not possible. Therefore, the amount of ELDs as presented in the present study could be an underestimation. Future studies should analyse these groups independently to get more insights into the characteristics of ELDs. Lastly, the fact that euthanasia is legalized in the Netherlands for children from the age of 12 years old might also have influenced clinical practices and hampers the generalizability of these results to other countries.

Conclusion

In summary, in two-thirds of cases, ELDs preceded the death of a child with cancer. This was almost twice as often as in children who die from all causes, underscoring that children who die from cancer represent a distinct subgroup. In most terminally ill children with cancer, potentially life-prolonging treatments were discontinued or withheld, and/or drugs were administered to alleviate pain or other symptoms in potentially life-shortening dosages. Frequently mentioned considerations for ELDs were no prospects of improvement and unbearable suffering without a curative perspective. In all documented cases ELDs were discussed with parents, and children were involved whenever feasible. Pediatric oncologists were more satisfied with care around the child's death in cases in which one or more ELDs were made.

Footnotes

Acknowledgments

We would like to acknowledge our gratitude toward all participating pediatric oncologists. We are grateful to Suzanne van de Vathorst for valuable suggestions and her input on ethics in pediatric medicine.

Author Disclosure Statement

No conflicting financial interests exist.

References

Gatta

, Botta

, Rossi

, et al.: Childhood cancer survival in Europe 1999–2007: Results of EUROCARE-5—a population-based study. Lancet Oncol, 2014; 15:35–47.

World Health Organization: Cancer pain relief and palliative care in children. Geneva, Switserland: World Health Organization, 1998.

Vrakking

, van der Heide

, Arts

, et al.: Medical end-of-life decisions for children in the Netherlands. Arch Pediatr Adolesc Med, 2005; 159:802–809.

de Vos

, van der Heide

, Maurice-Stam

, et al.: The process of end-of-life decision-making in pediatrics: A national survey in the Netherlands. Pediatrics, 2011; 127:e1004–e1012.

Act of April 12, 2001: Review procedures of termination of life on request and assisted suicide and amendment to the penal code and the Burial and Cremation Act (The Termination of Life on Request and Assistance With Suicide [Review Procedures] Act). Staatsblad, 2001; 1:1–8.

Verhagen

, Sauer

: The Groningen protocol—euthanasia in severely ill newborns. N Engl J Med, 2005; 352:959–962.

Pousset

, Bilsen

, Cohen

, et al.: Medical end-of-life decisions in children in Flanders, Belgium: A population-based postmortem survey. Arch Pediatr Adolesc Med, 2010; 164:547–553.

Burns

, Mitchell

, Outwater

, et al.: End-of-life care in the pediatric intensive care unit after the forgoing of life-sustaining treatment. Crit Care Med, 2000; 28:3060–3066.

Provoost

, Cools

, Mortier

, et al.: Medical end-of-life decisions in neonates and infants in Flanders. Lancet, 2005; 365:1315–1320.

10.

Vrakking

, van der Heide

, Provoost

, et al.: End-of-life decision making in neonates and infants: Comparison of The Netherlands and Belgium (Flanders). Acta Paediatr, 2007; 96:820–824.

11.

Lago

, Piva

, Garcia

, et al.: End-of-life practices in seven Brazilian pediatric intensive care units. Pediatr Crit Care Med, 2008; 9:26–31.

12.

Ryan

, Byrne

, Kuhn

, Tyebkhan

: No resuscitation and withdrawal of therapy in a neonatal and a pediatric intensive care unit in Canada. J Pediatr, 1993; 123:534–538.

13.

Devictor

, Nguyen

, Groupe Francophone de Reanimation et d'Urgences P: Forgoing life-sustaining treatments: How the decision is made in French pediatric intensive care units. Crit Care Med, 2001; 29:1356–1359.

14.

Garros

, Rosychuk

, Cox

: Circumstances surrounding end of life in a pediatric intensive care unit. Pediatrics, 2003; 112:e371.

15.

de Leeuw

, de Beaufort

, de Kleine

, et al.: Foregoing intensive care treatment in newborn infants with extremely poor prognoses. A study in four neonatal intensive care units in The Netherlands. J Pediatr, 1996; 129:661–666.

16.

Wall

, Partridge

: Death in the intensive care nursery: Physician practice of withdrawing and withholding life support. Pediatrics, 1997; 99:64–70.

17.

Hechler

, Blankenburg

, Friedrichsdorf

, et al.: Parents' perspective on symptoms, quality of life, characteristics of death and end-of-life decisions for children dying from cancer. Klin Padiatr, 2008; 220:166–174.

18.

Meyer

, Burns

, Griffith

, Truog

: Parental perspectives on end-of-life care in the pediatric intensive care unit. Crit Care Med, 2002; 30:226–231.

19.

van der Heide

, van der Maas

, van der Wal

, et al.: The role of parents in end-of-life decisions in neonatology: Physicians' views and practices. Pediatrics, 1998; 101:413–488.

20.

Evans

: Are children competent to make decisions about their own deaths?. Behav Sci Law, 1995; 13:27–41.

21.

Kenyon

: Current research in children's conceptions of death: A critical review. Omega, 2001; 43:63–91.

22.

Wiener

, Ballard

, Brennan

, et al.: How I wish to be remembered: The use of an advance care planning document in adolescent and young adult populations. J Palliat Med, 2008; 11:1309–1313.

23.

Nuss

, Hinds

, LaFond

: Collaborative clinical research on end-of-life care in pediatric oncology. Semin Oncol Nurs, 2005; 21:125–134; discussion 134–144.

24.

Van den Block

, Bilsen

, Deschepper

, et al.: End-of-life decisions among cancer patients compared with noncancer patients in Flanders, Belgium. J Clin Oncol, 2006; 24:2842–2848.

25.

Murray

, Boyd

, Kendall

, et al.: Dying of lung cancer or cardiac failure: Prospective qualitative interview study of patients and their carers in the community. BMJ, 2002; 325:929.

26.

Murray

, Kendall

, Boyd

, Sheikh

: Illness trajectories and palliative care. BMJ, 2005; 330:1007–1011.

27.

Murtagh

, Preston

, Higginson

: Patterns of dying: Palliative care for non-malignant disease. Clin Med, 2004; 4:39–44.

28.

Tomlinson

, Bartels

, Hendershot

, et al.: Factors affecting treatment choices in paediatric palliative care: Comparing parents and health professionals. Eur J Cancer, 2011; 47:2182–2187.

29.

Hinds

, Drew

, Oakes

, et al.: End-of-life care preferences of pediatric patients with cancer. J Clin Oncol, 2005; 23:9146–9154.

30.

Hinds

, Oakes

, Furman

, et al.: Decision making by parents and healthcare professionals when considering continued care for pediatric patients with cancer. Oncol Nurs Forum, 1997; 24:1523–1528.

31.

Hilden

, Emanuel

, Fairclough

, et al.: Attitudes and practices among pediatric oncologists regarding end-of-life care: Results of the 1998 American Society of Clinical Oncology survey. J Clin Oncol, 2001; 19:205–212.

32.

Himelstein

, Hilden

, Boldt

, Weissman

: Pediatric palliative care. N Engl J Med, 2004; 350:1752–1762.

33.

Mack

, Hilden

, Watterson

, et al.: Parent and physician perspectives on quality of care at the end of life in children with cancer. J Clin Oncol, 2005; 23:9155–9161.

34.

Stewart

, Pyke-Grimm

, Kelly

: Making the right decision for my child with cancer: The parental imperative. Cancer Nurs, 2012; 35:419–428.

35.

van der Heide

, Deliens

, Faisst

, et al.: End-of-life decision-making in six European countries: Descriptive study. Lancet, 2003; 362:345–350.

36.

Tates

, Meeuwesen

: Doctor-parent-child communication. A (re)view of the literature. Soc Sci Med, 2001; 52:839–851.