You are here
Patterns of End-of-Life Care in Children With Advanced Solid Tumor Malignancies Enrolled on a Palliative Care Service
Journal of Pain and Symptom Management, Volume 50, Issue 3, September 2015, Pages 305 - 312
Pediatric patients with solid tumors can have a significant symptom burden that impacts quality of life (QoL) and end-of-life care needs.
We evaluated outcomes and symptoms in children with solid tumors and compared patterns of end-of-life care after implementation of a dedicated institutional pediatric palliative care (PC) service.
We performed a retrospective cohort study of children with solid tumors treated at St. Jude Children's Research Hospital, before and after implementation of the institutional QoL/PC service in January 2007. Patients who died between July 2001 and February 2005 (historical cohort; n = 134) were compared with those who died between January 2007 and January 2012 (QoL/PC cohort; n = 57).
Median time to first QoL/PC consultation was 17.2 months (range 9–33). At consultation, 60% of children were not receiving or discontinued cancer-directed therapy. Within the QoL/PC cohort, 54 patients had documented symptoms, 94% required intervention for ≥3 symptoms, and 76% received intervention for ≥5 symptoms. Eighty-three percent achieved their preferred place of death. Compared with the historical cohort, the QoL/PC cohort had more end-of-life discussions per patient (median 12 vs. 3; P < 0.001), earlier end-of-life discussions, with longer times before do-not-resuscitate orders (median 195 vs. 2 days; P < 0.001), and greater hospice enrollment (71% vs. 46%, P = 0.002).
Although children with solid tumor malignancies may have significant symptom burden toward the end of life, positive changes were documented in communication and in places of care and death after implementation of a pediatric PC service.
Key Words: Palliative care, pediatric oncology, pediatric, end of life.
Cancer is the leading cause of childhood nonaccidental death in the U.S., with approximately 17% dying of their disease. 1 Children with malignancies often face unique challenges, with some experiencing longer disease trajectories, frequent recurrences, and high symptom burdens.2 and 3 Anticipating transition goals toward comfort care becomes critical when prognosis worsens. Suboptimal communication about prognosis and maintaining goals for cure near the end of life can result in substantial suffering and increased use of aggressive life-sustaining measures.3 and 4 Children diagnosed with advanced high-risk solid tumors are especially prone to poor outcomes and death. Integration of palliative care (PC) for these children improves end-of-life care, enhances advance care planning, and substantially decreases suffering by improving communication, providing optimal symptom control, and emphasizing quality of life (QoL).3, 4, 5, 6, and 7
In 2007, St. Jude Children's Research Hospital (SJCRH) began its integration of the QoL/PC service into the multidisciplinary approach to children with cancer to improve end-of-life care outcomes. The QoL/PC service began as a pilot program in 2007, focusing on the development and provision of quality care for children with solid tumors. Through ongoing assessments and identifying patient and family preferences, goals, and needs throughout a child's illness during consultations, regular inpatient/outpatient follow-up, team meetings, and home visits, the QoL/PC Service quickly expanded into an institution-wide PC program available to all SJCRH patients beginning in the first quarter of 2008. This interdisciplinary team offered symptom management expertise; addressed psychosocial and spiritual needs throughout the continuum of care, coordinated care, and ensured communication between providers at times of transition, facilitated end-of-life/do-not-resuscitate (DNR) discussions, and provided decision-making and bereavement support. The QoL/PC Service provided PC and hospice support services for children and their families once transferred to their home institutions to facilitate continuity of care.
We hypothesized that children with advanced solid tumors will have a high symptom burden, need for psychosocial support, and be likely to achieve desired location of death; furthermore, that integration of the QoL/PC Service in the institution would result in earlier end-of-life/DNR discussions and advance care planning compared with a historical cohort of patients treated before the team's establishment. We also wanted to evaluate end-of-life conversations before and after PC involvement.
This was a retrospective chart review of patients younger than 21 years with solid tumors (QoL/PC cohort, n = 57), enrolled on the PC service through death (January 2007–January 2012). During this period, 10 additional patients also were eligible for enrollment but were not referred for QoL/PC consultation because of various treating physician or parental personal preferences. The historical cohort (n = 134) with solid malignancies and death (July 2001–February 2005) was a published database 8 before January 2007. The SJCRH Institutional Review Board approved this study with parental informed consent waiver because patients were deceased.
The following data for the QoL/PC cohort were obtained: sex, race, religion, birth date, diagnosis, date/cause of death (disease progression vs. treatment-related complication), cardiopulmonary resuscitation or intubation, hospice enrollment, DNR orders, and interventions during the last month of life (antibiotics, antifungals, antivirals, blood products, chemotherapy, haloperidol, fluids, parenteral nutrition/tube feeds, intrusive procedures, laboratory draws, mechanical ventilation, oxygen, radiation, surgery, vasopressors, hemodialysis/continuous veno-venous hemofiltration (CVVH), pain medication, steroids, benzodiazepines, anticonvulsants, and antidepressants). After DNR implementation, the QoL/PC and historical cohorts were compared to identify further care preferences and symptom management strategies.
Additional information from QoL/PC and oncology initial consultations, inpatient and outpatient follow-ups, home visits, and phone service notes included relapses, protocol enrollment, transplant status, symptoms, care goals, hospitalizations, cancer-directed therapy during QoL/PC consultation and the last month of life, sibling and bereavement support, content of end-of-life discussion (e.g., high likelihood of death, transitioning care home), and death location. Data entry accuracy was verified for every third case, with records maintained by additional team members at ≥95% accuracy. If discrepancies were identified, two abstractors reviewed the chart, and consensus determined final data entry. For comparison, historical cohort demographics, death location, and end-of-life discussions were used. 8
An end-of-life discussion was identified by patient record documentation and/or family conference with clinician(s) of any discipline or a member of the interdisciplinary team (e.g., QoL/PC, oncology, social work). The discussion had to specifically address poor prognosis, disease progression and implications, and patient and/or family preferences for care when cure was no longer realistic. The first documented end-of-life discussion with the patient and/or family was defined as the first statement reflecting no realistic chance for cure, Phase I protocol enrollment, home health for terminal care, referral to hospice, or DNR status discussion or DNR order.
Demographics, clinical parameters, and study outcomes were summarized using descriptive statistics. The Wilcoxon signed rank test determined whether end-of-life discussion number differed before/after PC involvement. Fisher's exact tests examined differences for categorical variables between patient subgroups and cohorts. Wilcoxon rank sum and Kruskal-Wallis tests determined differences for continuous variables between patient subgroups and cohorts. Patient subgroup comparisons were completed for the QoL/PC cohort only and included comparisons of sex, age at diagnosis (12 years and younger, older than 12 years), post-transplant status, Phase I protocol enrollment, race, and religious preference. Statistical analyses used SAS, version 9.3 (SAS Institute, Inc., Cary, NC). All statistical tests were two sided, and P < 0.05 was considered statistically significant.
Demographically, race, religion, and permanent residence at the time of diagnosis were significantly different between the two cohorts ( Table 1 ). Within the QoL/PC cohort alone, 91% of patients died of progressive disease, 4% died of treatment-related complications, and 5% died of undetermined causes. Most patients were white, and 74% (40 of 54) had metastatic disease at the time of initial diagnosis.
|Male||37 (65)||70 (52)|
|Female||20 (35)||64 (48)|
|Age at diagnosis||0.105|
|12 years old or younger||30 (53)||88 (66)|
|Older than 12 years||27 (47)||46 (34)|
|White||30 (53)||88 (66)|
|Black||18 (32)||37 (28)|
|Other||9 (16)||5 (4)|
|Catholic||10 (18)||26 (19)|
|Protestant||6 (11)||38 (28)|
|Other||38 (67)||61 (46)|
|None||1 (2)||2 (1)|
|Unknown||2 (4)||7 (5)|
|Permanent residence at diagnosis||0.002|
|St. Jude Affiliate b||6 (11)||31 (23)|
|Local or in state||25 (44)||23 (17)|
|Out of state||22 (39)||68 (51)|
|International||4 (7)||12 (9)|
|Stem cell transplant recipient||0.125|
|No||44 (77)||86 (64)|
|Yes||13 (23)||47 (35)|
|Location of death||0.060|
|Home||32 (56)||73 (54)|
|Home hospital||5 (9)||24 (18)|
|St. Jude inpatient||16 (28)||17 (13)|
|Other||0 (0)||2 (2)|
|Unknown||4 (7)||18 (13)|
a Comparisons exclude unknown category.
b St. Jude Affiliate Program is associated with six institutions: St. Jude Baton Rouge Affiliate Clinic attached to Our Lady of the Lake Regional Medical Center (Baton Rouge, Louisiana); St. Jude Clinic at Huntsville Hospital for Women and Children (Huntsville, Alabama); St. Jude Tri-Cities Affiliate Clinic in the Johnson Medical Center (Johnson City, Tennessee); St. Jude Jim and Trudy Maloof St. Jude Midwest Affiliate Clinic in OSF-Saint Francis Medical Center (Peoria, Illinois); St. Jude Shreveport Affiliate Clinic in LSU Health Sciences Center (Shreveport, Louisiana); and St. Jude Affiliate Clinic at Mercy Children's Hospital-Springfield (Springfield, Missouri).
QoL/PC = quality of life/palliative care.
The median time to the first QoL/PC service consultation was 17.2 months (interquartile range [IQR] 9.2–33) from the initial diagnosis. The median time from initial consultation to death was 1.6 months (IQR 0.6–3). Primary referral to the QoL/PC service reasons are summarized in Table 2 . At initial consultation, many children were not receiving or were no longer pursuing cancer-directed therapy (60%). The remainder received nonprotocol treatment protocol chemotherapy (26%), enrolled in a Phase I protocol (5%), received alternative/complementary therapies (5%), radiotherapy and nonprotocol treatment protocol chemotherapy (2%), or had unknown treatment status (2%). Initial child/family care goals at initial QoL/PC consultation were cure (7%), prolong life with hope of cure (21%), prolong life with comfort focus (40%), comfort only (30%), and unknown (2%). No differences were identified in time from diagnosis to initial QoL/PC consultation based on sex, age, race, or religious preference. The time between diagnosis and the first PC consultation was significantly longer for stem cell transplant recipients than for nonrecipients, with a median (IQR) time to referral of 37.9 (20.1–61.8) vs. 15.8 (7.4–24) months, respectively (P < 0.006). Patients involved in Phase I protocols before or at initial consultation were also referred later for PC consultation than those not on a protocol, with a median (IQR) time of 23.6 (17.2–39.3) vs. 11.9 (6.3–17.3) months, respectively (P < 0.001).
|N = 56 a||N (%)|
|Care coordination/continuity and symptom management||25 (44)|
|Care coordination/continuity and transition to home||9 (16)|
|End-of-life care/DNR discussion and care coordination/continuity||8 (15)|
|End-of-life care/DNR discussion and transition to home||2 (4)|
|End-of-life care/DNR discussion and transition of care||3 (6)|
|Transition to home and symptom management||2 (4)|
|Decision-making support and communication||5 (9)|
|Communication and symptom management||1 (2)|
|Education/support for the patient or family||1 (2)|
a Information about the primary reason was missing for one patient.
QoL/PC = quality of life/palliative care; DNR = do-not-resuscitate.
Symptoms and Suffering
The QoL/PC team managed 24 different types of symptoms during the last month of life. In the QoL/PC cohort, 54 of 57 patients had documented symptom management available for review. Fifty-one patients (94%) required intervention from the QoL/PC team for three or more symptoms, and 41 patients (76%) received intervention for five or more. All patients received at least one intervention from the QoL/PC team. The most common symptoms treated were pain (100%), nausea/vomiting (63%), constipation (57%), and anxiety (56%) ( Fig. 1 ).
The median time from the first end-of-life discussion to initial PC consultation was 178 days (IQR 11.5–416.5) among patients who had end-of-life discussions before the initial consultation (n = 56). The median time from the first end-of-life discussion to death was 268 days (IQR 57–474), and the median time from the first DNR discussion to death was 45 days (IQR 13–79). No differences in time from first DNR discussion to death were identified between groups based on sex, age, race, religious preference, post-transplant status, or Phase I enrollment. No differences in time from the first end-of-life discussion to death were found based on the above cohort characteristics with the exception of patients enrolled on Phase I studies; in this latter group, time from discussion to death was significantly longer. PC team involvement significantly increased the number of end-of-life discussions addressing end-of-life/DNR and transitioning to home/hospice; conversations about the high likelihood of death decreased significantly after PC team involvement ( Table 3 ).
|End-of-Life Discussions/Patient-Specified Content||N a||Before PC Consultation||After PC Consultation||P|
|Median (IQR)||Median (IQR)|
|End-of-life care/DNR||56||0 (0–1)||1 (0–2)||0.026|
|High likelihood of death||56||2.5 (1.5–4)||1 (1–3)||0.003|
|Transition of care/withdrawal of life support||56||0 (0–1)||0 (0–1)||0.208|
|Transition to home/hospice||56||1 (0–1)||2 (1–4)||0.001|
|Total of all discussions||56||5 (2–8)||5.5 (3–10)||0.386|
a Information about no. of end-of-life discussions was missing for one patient.
QoL/PC = quality of life/palliative care; IQR = interquartile range; DNR = do-not-resuscitate.
Hospice Enrollment and Preferred and Actual Location of Death
Forty patients (71%) in the QoL/PC cohort were enrolled in hospice. Median hospice care time was 44 days (IQR 19–83). No significant differences were detected based on sex, age, race, religious preference, post-transplant status, or Phase I protocol involvement. Preferred and actual place of death was documented for 47 children: 39 (83%) achieved their preferred location; 68% occurred at home ( Table 4 ). Most patients (65%) had at least one inpatient hospitalization during the last month of life; 17% required two or more hospitalizations; three (6%) had ventilator support; and two (4%) were intubated at the time of death.
|Preferred Place of Death||Actual Place of Death|
|SJ ICU, n (%)||SJ IPD, n (%)||SJ Domiciliary (Houses), n (%)||Home, n (%)||Other Hospital, n (%)|
|SJ ICU||0 (0)||1 (2)||0 (0)||0 (0)||0 (0)|
|SJ IPD||0 (0)||6 (13)||0 (0)||0 (0)||1 (2)|
|SJ domiciliary (houses)||0 (0)||1 (2)||0 (0)||0 (0)||0 (0)|
|Home||1 (2)||3 (6)||0 (0)||32 (68)||1 (2)|
|Other hospital||0 (0)||0 (0)||0 (0)||0 (0)||1 (2)|
QoL/PC = quality of life/palliative care; SJ ICU = St. Jude Intensive Care Unit; SJ IPD = St. Jude Inpatient Department.
Preferred or actual place of death is missing for 10 patients.
Comparison of the QoL/PC and Historical Cohorts
Significantly more end-of-life discussions occurred in the QoL/PC cohort compared with the historical cohort (median 12 vs. 3, P < 0.001). These discussions were introduced significantly earlier for the QoL/PC cohort than the historical cohort, in relation to timing of hospice enrollment, DNR documentation, and death ( Table 5 ). The time from the first end-of-life discussion to enrollment in hospice care was significantly longer in the QoL/PC compared with the historical cohort (median 225 vs. 18 days, P < 0.001). The QoL/PC time from the first end-of-life discussion to DNR order was significantly increased compared with the historical cohort (median 195 vs. 2 days, P < 0.001). The first end-of-life discussion time until death was also longer in the QoL/PC cohort than in the historical cohort (median 268 vs. 64 days, P < 0.001).
|End-of-Life Quality Variable||QoL/PC Cohort||Historical Cohort||P|
|Time from first end-of-life discussion to initial PC consultation a|
|Median (IQR)||178 (11.5–416.5)||N/A|
|Total number of end-of-life discussions per patient|
|Median (IQR)||12 (7.5–15.5)||3 (2–4)|
|Time from first end-of-life discussion to enrollment in hospice care (days)|
|Median (IQR)||225 (36–442)||18 (2–98)|
|Time from first end-of-life discussion to DNR order (days)|
|Median (IQR)||195 (20–421)||2 (0–35)|
|Time from first end-of-life discussion to death (days)|
|Median (IQR)||268 (57–474)||64 (11–164)|
|Sibling counseling (N = 102) (patients 21 years or younger) b , c|
|Given, N (%)||23 (50)||16 (16)||<0.0001|
|Not given/not documented, N (%)||23 (50)||86 (84)|
|Bereavement support b (N = 145)|
|Given, N (%)||55 (96)||72 (50)||<0.0001|
|Not given/not documented, N (%)||2 (4)||73 (50)|
a One patient had first end-of-life conversation after QoL/PC consultation.
b QoL/PC cohort compared with the historical cohort of previously published cohort of 145 pediatric patients with cancer who received care at St. Jude and died between January 1, 2000 and June 30, 2001. 28
c Not applicable for 11 patients in the QOL/PC cohort and 43 patients in the historical cohort who did not have siblings or had siblings of 21 years or older.
QoL/PC = quality of life/palliative care; N = number of patients for whom data were available; N/A = not applicable; IQR = interquartile range (25%–75%).
There were no differences from the time of initial DNR order until death, with a median of 21 and nine days in the QoL/PC and historical groups, respectively (P = 0.175). DNR orders at the time of death were significantly greater in the QoL/PC cohort (84% vs. 47%, P ≤ 0.0001). From the time of DNR implementation, the QoL/PC cohort compared with the historical cohort was more likely to receive care interventions. Of the interventions evaluated, only antibiotics (69% vs. 43%, P = 0.008), fluids (75% vs. 40%, P = 0.001), steroids (63% vs. 22%, P = 0.001), and anticonvulsants (40% vs. 6%, P = 0.001) were statistically significant.
Hospice Enrollment, Location of Death, and Bereavement Support
Hospice enrollment was more frequent in the QoL/PC cohort (71% vs. 46%, P = 0.002). No differences were identified between hospice enrollment and time to death between the QoL/PC and historical cohorts, with a median of 44 and 31 days, respectively (P = 0.277). Death at home occurred for 56% QoL/PC and 54% historical patients, respectively. Sibling and bereavement support was provided (50% vs. 16%, P < 0.0001) and documented (96% vs. 50%, P < 0.0001) more frequently in the QoL/PC cohort ( Table 5 ).
A QoL/PC team at SJCRH has proven to be a vital component in comprehensive patient management, demonstrating improvement in areas prioritized by the National Consensus Project for Quality Palliative Care. 9
More end-of-life discussions in the QoL/PC cohort compared with the historical cohort may reflect a change in culture at SJCRH stemming from the implementation of a PC service and from a national initiative to improve end-of-life care for children. Both encourage open and honest communication as part of standard care. The QoL/PC cohort held more conversations addressing thoughts and concerns regarding death compared with the historical cohort. Furthermore, the QoL/PC cohort also had more time between the first end-of-life discussion and DNR order and between the first end-of-life discussion and death than did the historical cohort. These findings should not be undervalued as physicians believe that advance care discussions often occur too late in patients' clinical course as a result of communication barriers that exist between parents and health care providers. 10 When a mutual physician-parent understanding of the incurable cancer exists before death, results can include greater satisfaction with care, more days with a DNR order, better use of hospice, symptom control, and comfort. 11
Although some families and clinicians may be hesitant to suggest a PC referral because of presumed implications of giving up, PC services make provisions for goal-based care. This was evident in the current series by the significantly increased supportive interventions provided to those in the QoL/PC cohort. We demonstrated a substantial increase in hospice enrollment, which has shown to be a proxy for advanced end-of-life care by improving symptom management and addressing the various psychosocial concerns faced by patients and their families.12 and 13 In addition, significantly more sibling and bereavement support services were provided to help facilitate the grieving process.14 and 15 Furthermore, as earlier interventions for end-of-life decision making for patients with advanced malignancies have resulted in earlier DNR orders and fewer hospital deaths, this underscores the importance of early and frequent goal-directed communication to identify patient/family needs as a child's disease and treatment course evolves. 16
Recognizing patient/family preferences is crucial when coordinating care, including death location, which often depends on variations in family perception, preparedness, cultural differences, and available resources.17, 18, and 19 The premise of a home death is that families are better prepared to ensure a comfortable, loving, familiar setting, and are less likely to seek hospitalization.17, 18, and 19 Having a clear understanding of prognosis/treatment options and selecting the end-of-life care location can be emotionally significant.17, 19, and 20 This study indicated that most QoL/PC patients with documented preferences died at their desired location, possibly a result of the increased end-of-life discussions and advance care planning.
There was no significant decrease in inpatient deaths after PC implementation. Approximately 37% of children in the QoL/PC cohort died as inpatients at SJCRH or at outside institutions. Possible explanations include a referral bias whereby QoL/PC consultation was requested for only the most critical patients with the highest symptom burden, whose care required an inpatient setting. The QoL/PC team cooperated with families who preferred the supportive surroundings of the hospital setting they had adopted. For some families, death at home might introduce unimaginable challenges. 21 Further studies are warranted to better understand the preferences when planning location of death.
A common misconception is that the role of the PC team is for patients transitioning in the last months to weeks of life, when in fact, greater benefits are achieved with earlier involvement. 22 Various institutions have implemented policies for early PC in high-risk patients with greater medical complexity and prognostic uncertainty. 23 Despite the ability to provide a variety of patient/family support and services early, initial introduction of PC is restricted because it is commonly mistaken as an exclusive end-of-life service. 24 Although children at SJCRH were referred for palliative consultation for a variety of indications, the lingering misunderstandings and lack of awareness of services offered by the QoL/PC team prevented earlier referral and care for other patients. Ullrich et al. 25 reported that children undergoing stem cell transplantation had a significantly shorter interval between parental realization that there was no realistic chance for cure and the child's death, with decreased end-of-life preparation, and impaired ability to plan death location, discuss resuscitation status, or reduce suffering. Similar to our findings, transplant recipients and those involved in Phase I protocols were referred much later for PC consultation.
The parental role in decision making is challenging throughout an evolving disease process, especially in the context of incurable disease. It is not uncommon for parents to regret the decision to pursue their child's last chemotherapy course. 26 A series from SJCRH 2 identified good parent rationales and definitions from parental interviews of those whose children recently enrolled in a Phase I protocol or decided DNR or terminal care. Parents who chose protocol treatment reported a need to pursue any remaining cancer-directed therapies and staff support in making this decision. These families would likely benefit from early PC assistance in decision-making support, symptom management, and additional provisions to achieve care goals. 18
Research has focused on childhood end-of-life physical suffering and unmet needs, emphasizing inadequate pain prevention, assessment, and management in advanced disease.3 and 27 Optimizing symptom relief is critical for children throughout the course of illness as they face the consequences of their therapy and disease. At the end of life, 89% of children suffered from at least one symptom, commonly pain, fatigue, and dyspnea, and less than 30% achieved adequate control. 3 In our series, the QoL/PC cohort demonstrated similar symptom burden, with pain as the most common symptom requiring intervention. Notably, rates of documented fatigue and depression were lower than generally expressed by caregivers. This may have been a consequence to underreporting and symptom overlap (e.g., treatment and documentation of anxiety in a patient with an underlying and unrecognized depression). No standardized assessment tool was used to evaluate and record responses to interventions applied for symptom management, limiting the scope of this analysis. Future prospective studies are warranted to evaluate the incidence of cancer-related symptoms among various childhood malignancies and evaluation of response from PC intervention.
Our study has several limitations. The retrospective single-institution data did not allow for variable control, introducing potential selection and referral bias. Furthermore, significant demographic differences between the QoL/PC and historical cohorts may introduce uncontrollable variables that may ultimately impact outcomes. Documentation gaps could have resulted in underestimation of clinical events (e.g., actual first end-of-life conversation, DNR order, patient preferences about death location). Failure to document end-of-life discussions in the earlier cohort secondary to barriers in documentation (e.g., lack of accessibility on electronic medical records) may account for some of the differences. Although improvements in end-of-life care with PC team implementation have been demonstrated, causality would require differentiating the impact of the QoL/PC team from that of an overall institutional culture evolution; carefully designed and mixed methodology prospective studies would be necessary. The discordant time frames in which the two cohorts were treated are also an issue: a comparison of the deceased solid tumor patients who either received or did not receive specific interventions delivered by the PC team after January 2007 would have been ideal. Attempts were made to compare the QoL/PC cohort with patients who did not receive PC intervention after January 2007; however, there were not enough patients to draw meaningful comparisons. Finally, qualitative evaluations and experiences from children/families were not included in this review and might have presented different perspectives.
This study describes the positive impact of implementing a pediatric PC service in pediatric patients with solid tumor malignancies. Integration of PC promotes better end-of-life care with more advance care planning and achievement of family preferences. With the continued growth and earlier integration of PC services at SJCRH and across the nation, the QoL for patients with life-threatening illnesses will continue to improve. Further prospective studies are necessary to better define outcomes, identify ways to enhance communication, promote symptom management, and improve overall care for this population.
Disclosures and Acknowledgments
This work was supported in part by NIH Cancer Center Support Core Grant CA-21765 and the American Lebanese Syrian Associated Charities. The authors of this article have no actual or potential conflicts of interest in the article preparation or publication.
- 1 Howlander N, Noone AM, Krapcho M, et al. SEER cancer statistics review, 1975-2010. (Based on November 2012 SEER data submission, posted to the SEER Web site, April 2013). Bethesda, MD: National Cancer Institute. Available at: http://seer.cancer.gov/csr/1975_2010/ . Accessed January 7, 2014.
- 2 S.H. Maurer, P.S. Hinds, S.L. Spunt, W.L. Furman, J.R. Kane, J.N. Baker. Decision-making by parents of children with incurable cancer who opt for enrollment on a phase I trial compared with choosing a do-not-resuscitate/terminal care option. J Clin Oncol. 2010;28:3292-3298 Crossref
- 3 J. Wolfe, H.E. Grier, N. Klar, et al. Symptoms and suffering at the end of life in children with cancer. N Engl J Med. 2000;342:326-333 Crossref
- 4 N. Contro, J. Larson, S. Scofield, B. Sourkes, H. Cohen. Family perspective on the quality of pediatric palliative care. Arch Pediatr Adolesc Med. 2002;156:14-19 Crossref
- 5 J. Wolfe, J.F. Hammel, K.E. Edwards, et al. Easing of suffering in children with cancer at the end of life: is care changing?. J Clin Oncol. 2008;26:1717-1723 Crossref
- 6 J.A. Heath, N.E. Clarke, M. McCarthy, S.M. Donath, V.A. Anderson, J. Wolfe. Quality of care at the end of life in children with cancer. J Paediatr Child Health. 2009;45:656-659 Crossref
- 7 C. Laine, F. Davidoff, C.E. Lewis, et al. Important elements of outpatient care: a comparison of patients' and physicians' opinions. Ann Intern Med. 1996;125:640-645 Crossref
- 8 J.N. Baker, S. Rai, W. Liu, et al. Race does not influence do-not-resuscitate status or the number or timing of end-of-life care discussions at a pediatric oncology referral center. J Palliat Med. 2009;12:71-76 Crossref
- 9 National Consensus Project for Quality Palliative Care. Clinical practice guidelines for quality palliative care. 3rd ed. (National Consensus Project for Quality Palliative Care, Pittsburgh, PA, 2013)
- 10 A. Durall, D. Zurakowski, J. Wolfe. Barriers to conducting advance care discussions for children with life-threatening conditions. Pediatrics. 2012;129:e975-e982 Crossref
- 11 J. Wolfe, N. Klar, H.E. Grier, et al. Understanding of prognosis among parents of children who died of cancer: impact on treatment goals and integration of palliative care. JAMA. 2000;284:2469-2475 Crossref
- 12 K. Fowler, K. Poehling, D. Billheimer, et al. Hospice referral for children with cancer: a survey of pediatric oncologists. J Clin Oncol. 2006;24:1099-1104 Crossref
- 13 J. Wolfe, S. Freibert, J.M. Hilden. Caring for children with advanced cancer: integrating palliative care. Pediatr Clin North Am. 2002;49:1043-1062 Crossref
- 14 U.C. Kreicbergs, P. Lannen, E. Onelov, J. Wolfe. Parental grief after losing a child to cancer: impact of professional and social support on long-term outcomes. J Clin Oncol. 2007;25:3307-3312 Crossref
- 15 J.J. Spinetta, M. Jankovic, T. Eden, et al. Guidelines for assistance to siblings of children with cancer: report of the SIOP working committee on psychosocial issues in pediatric oncology. Med Pediatr Oncol. 1999;33:395-398 Crossref
- 16 R.A. Stein, L. Sharpe, M.L. Bell, F.M. Boyle, S.M. Dunn, S.J. Clarke. Randomized controlled trial of a structured intervention to facilitate end-of-life decision-making in patients with advanced cancer. J Clin Oncol. 2013;31:3403-3410 Crossref
- 17 V. Dussel, U. Kreicbergs, J.M. Hilden, et al. Looking beyond where children die: determinants and effects of planning a child's location of death. J Pain Symptom Manage. 2009;37:33-43 Crossref
- 18 J. Vickers, A. Thompson, G.S. Collins, M. Childs, R. Hain. Place and provision of palliative care for children with progressive cancer: a study by the Paediatric Oncology Nurses' Forum/United Kingdom Children's Cancer Study Group Palliative Care Working Group. J Clin Oncol. 2007;25:4472-4476 Crossref
- 19 A. Kassam, J. Skiadaresis, S. Alexander, J. Wolfe. Parent and clinician preferences for location of end-of-life care: home, hospital or freestanding hospice?. Pediatr Blood Cancer. 2014;61:859-864 Crossref
- 20 P.S. Hinds, L. Oakes, W. Furman, et al. End-of-life decision making by adolescents, parents, and healthcare providers in pediatric oncology: research to evidence-based practice guidelines. Cancer Nurs. 2001;24:122-134 Crossref
- 21 K.I. Stajduhar, B. Davies. Death at home: challenges for families and directions for the future. J Palliat Care. 1998;14:8-14
- 22 J.W. Mack, J. Wolfe. Early integration of pediatric palliative care: for some children, palliative care starts at diagnosis. Curr Opin Pediatr. 2006;18:10-14 Crossref
- 23 A.Z. Doorenbos, H. Starks, E. Bourget, et al. Examining palliative care team involvement in automatic consultations for children on extracorporeal life support in the pediatric intensive care unit. J Palliat Med. 2013;16:492-495 Crossref
- 24 M.K. Song, A. De Vito Danns, S.M. Studer, R.M. Arnold. Palliative care referrals after lung transplantation in major transplant centers in the United States. Crit Care Med. 2009;37:1288-1292 Crossref
- 25 C.K. Ullrich, V. Dussel, J.M. Hilden, J.W. Sheaffer, L. Lehmann, J. Wolfe. End-of-life experience of children undergoing stem cell transplantation for malignancy: parent and provider perspectives and patterns of care. Blood. 2010;115:3879-3885 Crossref
- 26 J.W. Mack, S. Joffe, J.M. Hilden, et al. Parents' views of cancer-directed therapy for children with no realistic chance for cure. J Clin Oncol. 2008;26:4759-4764 Crossref
- 27 J.J. Collins, H.E. Grier, H.C. Kinney, C.B. Berde. Control of severe pain in children with terminal malignancy. J Pediatr. 1995;126:653-657 Crossref
- 28 G. Bradshaw, P.S. Hinds, S. Lensing, J.S. Gattuso, B.I. Razzouk. Cancer-related deaths in children and adolescents. J Palliat Med. 2005;8:86-95 Crossref
a Department of Radiation Oncology, University of Florida Proton Therapy Institute, Jacksonville, Florida, USA
b Division of Solid Tumors, Department of Oncology, St. Jude Children's Research Hospital, Memphis, Tennessee, USA
c Division of Quality of Life and Palliative Care, Department of Oncology, St. Jude Children's Research Hospital, Memphis, Tennessee, USA
d Department of Biostatistics, St. Jude Children's Research Hospital, Memphis, Tennessee, USA
e Division of Anesthesiology, St. Jude Children's Research Hospital, Memphis, Tennessee, USA
∗ Address correspondence to: Justin N. Baker, MD, Division of Quality of Life and Palliative Care, Department of Oncology, St. Jude Children's Research Hospital, MS 260, Room C6034, 262 Danny Thomas Place, Memphis, TN 38105-3678, USA.
© 2015 American Academy of Hospice and Palliative Medicine, Published by Elsevier B.V.