Abstract
Abstract
Background:
Despite recent strides in pediatric palliative care (PPC), there are few published data on community-based care of dying children.
Objective:
Our aim was to describe end-of-life care during the last 6 months of life for children on a community-based PPC program.
Methods:
We conducted a retrospective review of children <21 years old who died while enrolled in a community-based PPC program (CompassionNet) from December 2004 through May 2008. Data were extracted on the last 6 months of life from hospital records (University of Rochester) and records of CompassionNet and its home-nursing team.
Results:
Thirty-six children died while enrolled in the CompassionNet program; diagnoses included cancer 20 (56%), genetic/neurologic disorders 7 (19%), cardiac anomalies 6 (17%), and other 3 (8%). In the last 6 months of life, there were a median of 2 emergency room visits (range 0–12) and 2 unplanned hospitalizations (range 0–6). Home visits by CompassionNet (case manager, physician, nurse) ranged from 1 to 121 (median 24). A “goals of care” discussion was documented for 29 children (81%), occurring a median of 16 days before death (range 0–118). Sixteen children died at home (44%); 20 died in the hospital (56%). Of the 16 families with a documented preference for location of death, 14 children died in this location (88%). Thirty-two families (89%) had bereavement support through CompassionNet.
Conclusions:
Children who die of complex chronic conditions spend the majority of their last 6 months of life at home. Community-based PPC can contribute substantially to their care and comfort.
Introduction
W
Methods
Setting
CompassionNet is a community-based pediatric palliative care program that offers comprehensive psychosocial and medical care to children with life-limiting illnesses and their families. The interdisciplinary team of CompassionNet consists of pediatric nurses, pediatric nurse practitioners, a pediatrician board-certified in hospice and palliative medicine, child life specialists, social workers, chaplain, and expressive therapists.
Study design
The study is a retrospective review of the last 6 months of life of children who died from December 2004 through May 2008 while enrolled in the CompassionNet program. Data on the last 6 months of life (demographic data, services received [hospital care, emergency room visits, home visits by CompassionNet personnel], end-of-life discussion, description of death, and bereavement support) were extracted from the medical records at Golisano Children's Hospital (where the children received medical care) and CompassionNet. Medians and ranges were used as indices of spread of data, and Fisher's exact test for comparisons of categorical data. Medians were compared using the Wilcoxon rank sum test. The study was exempted by the Research Review Board of the University of Rochester Medical Center.
Results
Patient characteristics
Thirty-eight children died while enrolled in the CompassionNet program. Two children were excluded from analysis due to insufficient data describing their deaths. The median age of children at death was 11 years (range 0.06–20 years); their diagnoses included cancer (56%), cardiac anomaly (17%), progressive genetic or neurologic disorder (19%), cystic fibrosis (5%), and immunodeficiency (3%). Children were enrolled in CompassionNet a median of 1.1 years (range 0.03–4.75 years) before death.
Hospital and community care
The 36 children had a median of 3 hospitalizations (range 0–8), and 25.5 hospital days (range 0–125), and 2 emergency department visits (range 0–12) in their last 6 months of life. When elective hospitalizations were eliminated from the analysis (e.g., planned admissions for chemotherapy) they had a median of 2 hospitalizations (range 0–6), and 18.5 hospital days (range 0–125).
Community-based care included a median of 24 home visits (range 1–121) by CompassionNet team members (MD, nurse practitioner, nurse, case managers, child life specialists, social worker, expressive therapists). Details of the number and type of visits are displayed in Table 1. Thirty-two families (89%) received financial support through CompassionNet for co-payments, medical equipment, travel, hospital food and parking, weekly meals for the family, summer camps, gym memberships, respite care, child life services, and wigs.
IDT, interdisciplinary team.
End-of-life care
Pain was the most common symptom these children experienced during the last month of life, documented in 26 of 33 (79%) children. Other common symptoms included dyspnea, fatigue, nausea, vomiting, constipation, diarrhea, and anxiety. In the last 48 hours of life, the most common symptoms were dyspnea, agitation/delirium, vomiting, and decreased responsiveness. During that time frame 25 of 30 children received opioids, and 18 of 30 received a benzodiazepine (data not available on 6 children).
A “goals of care” discussion was documented for 29 children (81%), occurring a median of 16 days prior to death (range 0–116 days). Twenty-five families (69%) had a home do not resuscitate (DNR) document. Fifteen of the children were aware of their impending death, whereas 8 were not (data not recorded for 13 children). The children who were aware of their terminal illness were significantly older (median 16 years versus 1.4 years, p=0.0007, Mann-Whitney U test).
Sixteen children (44%) died at home, and 20 (56%) died in the hospital. Children who died at home were significantly older (median 15 years versus 5.5 years; p=0.04, Mann-Whitney-U test). There was no significant difference in diagnoses between the home and hospital groups. The family's desired location for death was documented for 16 children: 13 (45%) preferred home and 3 (10%) preferred the hospital. Fourteen of the children (88%) died in the desired location. The deaths of 24 children (67%) were described as “peaceful” and/or “in parent's arms.” (There was no documented description of the death for the remaining 12 children.) Thirty-two families (89%) received bereavement support from CompassionNet, and CompassionNet staff attended the funeral or calling hours for 34 children (94%).
Discussion
Terminally ill children with complex chronic conditions spend a great deal of time at home, even if they do not ultimately die at home. Despite this, there is a paucity of data regarding the services they receive at home, the quality of their last months of life, and their actual end-of-life experience. We undertook this review in an effort to get some sense of the type and nature of care that terminally ill children receive at home. In our retrospective review of 36 children who died while enrolled in a community-based PPC program, we found that the care of these children was complex, the services they received in the hospital and at home were many and varied, almost half of the children died at home, and most children (for whom data were available) died in their desired location.
The majority of children in our study had cancer, cardiac anomalies, or neurodegenerative disorders. Although more than half had cancer, this is actually a smaller proportion than in other studies of home care of terminally ill children, many of which focused exclusively on children with cancer.4–9
The care of these children was complex and they spent considerable time in the hospital. However, they spent the majority of their time at home, and received a wide array of home-based services. The median of 24 palliative care team home visits over their last 6 months of life included home visits from physicians, nurse practitioners, nurses, social workers, child life specialists, and expressive therapists. Published data on care at home for terminally ill children are scant. In a remarkable study published more than 30 years ago, Martinson and colleagues reported on 32 children dying of cancer who received home care. 6 Children were on the program a median of 20 days, suggesting that many of these children were near death at the time of enrollment. Twenty-seven children (84%) were able to remain at home. Most of the home care was provided by visiting nurses who made a median of 11 visits per child. Lauer and Cammita reported a similar experience. 9 A handful of other studies have examined home care for terminally ill children, but none with the detail of Lauer and Cammita's and Martinson's reports.4,5,7,10
Sixteen (44%) of the 36 children in our study died at home. This is roughly comparable to the proportion of home deaths for children with complex chronic conditions reported in a number of recent studies.3,7,11 Importantly, of the 16 families for whom the family's desired location of death was documented, 14 children (88%) died in that desired location. Dussel and coworkers reported on this phenomenon in 2008 in their study of 140 parents who lost a child to cancer. 11 They found that the process of planning the location of death (as opposed to the actual location) was associated with more home deaths, fewer hospital admissions, being prepared for the child's death, and being comfortable with the location of death.
There are a number of limitations of our study. These children were followed from 2004 to 2008, and the fields of PPC and home care are changing rapidly; these changes are not reflected in these data. In addition, by its retrospective nature, we are limited to what is recorded in the medical record, which may not tell the whole story. Finally, although we did make some attempt at documenting symptoms, these data are difficult to verify based on what is recorded in the medical record, and in part due to the variability in what is recorded by different types of practitioners in varying settings (home and hospital).
In sum, terminally ill children with complex chronic conditions spend a great deal of time in the hospital, but they spend even more time at home. Comprehensive, interdisciplinary home care services are vital for these children whether they ultimately die at home or in the hospital. Our study provides important insights into the types of services that can be offered in the home. It is imperative, however, to conduct larger, prospective studies of children that document the effectiveness of a home-based palliative care team in end-of-life care planning, coordination of care with hospital-based teams, and the effectiveness of home-based treatment of these children's many symptoms.
Author Disclosure Statement
No competing financial interests exist.