Referrals to a New Pediatric Palliative Care Team: Details of the First 12 Months of Operation

Introduction

In North America palliative care has been rapidly growing over the last three decades, whereas pediatric palliative care (PPC) has been in existence for approximately a decade in the United States and just under two decades in Canada.^1,2 Growth of PPC in the United States is related to calls by the American Academy of Pediatrics and Institute of Medicine to increase access to palliative care for children with life-threatening or life-limiting conditions.^3,4 The oldest established North American programs have been in existence for just less than 20 years, but most programs are still in their infancy. Programs that have reported on their first year of operation describe slow starts with variable growth across a wide range of patient diagnoses.^5–9

In 2010, Vadeboncoeur and colleagues ⁷ described the first nine years of operation of their PPC program in Ottawa, Ontario, Canada, which started in 1999. Their team demonstrated a steady increase in number of referrals over a nine-year period, a change in patients' underlying disease process, and a shift in location of death. In 2011, Feudtner and colleagues ¹⁰ published a multicenter prospective cohort analysis of data collected in 2008. Feudtner's paper was the first to describe the patient population followed by PPC teams and is the only prospective study of the PPC population in North America. The study included data collected across six well-established, hospital-based PPC teams. Reports showed predominant clinical conditions were genetic/congenital and neuromuscular disease (40.8% and 39.2%), 19.8% of patients suffered from malignancies, and a majority of patients were male (54%). During the 12-month follow-up period, 30.5% of the cohort died.

Most recently, Edlynn and colleagues ¹¹ reported on their process of creating a new PPC service at a large, urban children's hospital. They discuss program start-up, including the establishment of a palliative care advisory board and a pilot phase of service delivery that led to their program's success. Their team saw 117 new patients during year one and 236 during year two but did not comment further on disease processes of referrals or patient demographics. There are currently no published reports on the time it takes for a new team to achieve patient populations and referral patterns similar to those described by Feudtner and colleagues.

Nemours/Alfred I. duPont Hospital for Children (N/AIDHC) in Wilmington, DE is a 150-bed, tertiary care, freestanding children's hospital with 80 to 90 inpatient deaths per year. The majority of our patient population comes from 27 counties across four states, including Pennsylvania, Delaware, New Jersey, and Maryland. Our program began in August 2011 and was guided by the Center to Advance Palliative Care Palliative Care Leadership Centers (PCLC) program (Akron Children's Hospital site) with mentorship from the Children's Hospital of Pennsylvania, Philadelphia, PA.

The PPC team at our institution consists of a full-time medical director, nurse practitioner, social worker, and additional part-time physician staffing. The team provides inhospital coverage on weekdays and telephone coverage on nights and weekends. The team conducts home visits for home hospice patients, and comprehensive hospice services are provided in partnership with community hospice organizations. Prenatal consultation was not done by our PPC team in the first year of operation.

A great deal of administrative preparation and educational training began well in advance of our PPC program launch, including invited grand rounds speakers, a daylong palliative medicine conference for local clinicians held annually for three years prior to the PPC team starting, as well as an informal team of interdisciplinary palliative-minded practitioners who worked to self-educate through journal clubs, weekly clinical conferences, and case reviews. Based on the number of inpatient deaths our hospital sees annually as well as the number of new oncologic diagnoses per year, the administration projected that our team would receive 100 new patient referrals per year once fully operational. Based on previously published reports, a period of quiescence was expected on start-up.^5–9 Therefore a goal was set by the administration of receiving one quarter of the target yearly consultation rate (25 referrals) in the first year with an increase in the number of referrals in subsequent years. Referral numbers were expected to increase as educational efforts continued and awareness of the PPC team grew. This is a description of our first 12 months of operation as compared with previously published reports.^5–9

Methods

Patient referral and demographic data were recorded from the time of the first referral to the palliative and supportive care team. Data from this internal database were de-identified and analyzed in aggregate. We performed descriptive statistics, including counts and percentages. Diseases were classified according to Feudtner's Complex Chronic Conditions (CCC) categories, ¹² and patients were categorized based on the underlying disease process that led to the PPC referral. The project was approved by the Nemours/Alfred I. duPont Hospital for Children institutional review board.

Results

From August 2011 to July 2012, 97 patients were referred for consultation (see Table 1), of whom 94 were seen in consultation by our team (two patients died before being seen, and one consultation request was canceled by the primary team). Thirty-nine patients were ages 1 to 9 years (42.9%), and 28 were ages 10 to 18 years (30.5%) (see Table 2). Most of our patients were male (n=57, 59.3%) and Caucasian (n=49, 50.5%), although Hispanic patients represented a significant percentage of our cohort (n=20, 20.6%).

The two most common underlying diagnoses in our cohort were neuromuscular disorders (n=42, 32.3%) and genetic/congenital disorders (n=24, 26.1%). Twenty-six children (20%) had malignant diagnoses (see Table 3).

There were a significant number of patients referred while in the intensive care unit (ICU) (n=36, 37.5%). Of the 36 patients referred while in the ICU—includes pediatric, neonatal, and cardiac ICUs—six of those referrals came from the patient's primary subspecialty (oncology, pulmonology, and metabolism) with agreement from the attending ICU physician. We saw a notable number of children who were outpatient at the time of referral (n=19, 19.8%) (see Table 4). Lastly, 39 patients (39%) died during the observation period, and of those, 29 (74%) died in an inpatient setting, nine (23%) died at home with hospice services, and one (3%) died in a local emergency department (see Table 5).

Discussion

When we launched our program, we expected to have trends similar to reports published by other newly developed PPC programs—an initial period of quiescence followed by growth in later years.^5–9 We were surprised to be immediately operating beyond our projected year-one patient volume. We completed twice the number of consultations during our first year compared to published reports from most other newly developed PPC teams (see Table 1), and the total number of new patient consultations requested and performed was nearly four times the volume of patients we projected for our first 12 months. We speculate the rapid growth experienced during year one was due to multiple factors. First, our program was well designed by senior hospital leadership based on PCLC mentorship. From the program's inception we had tremendous clinical, administrative, and financial support from our institution. Additionally, a large educational effort was launched in advance of the program as previously described. However, we also believe our program's rapid growth can be attributed to a broader cultural shift. Palliative care is more universally accepted in pediatric hospitals in comparison to reports of programs that started in prior decades.

Our demographics and the age distribution of the majority of our patients did not differ significantly from those described by Feudtner and colleagues. ¹⁰ There was a difference, however, in our percentage of Hispanic patients. Our institution sees a population that is 8% to 10% Hispanic patients, a percentage that is still lower than what our PPC team experienced (20%). This is likely because our institution has a foundation that funds surgical and oncologic procedures for children from developing nations, largely from South America. The procedures are often high risk, which increases the percentage of Hispanic patients seen in consultation by our PPC team out of proportion to our hospital's general population.

The most common underlying diagnoses seen by our team were neuromuscular and congenital/genetic diseases, representing half of all patients referred. Malignancies represented only one-fifth of our total patient population. Additionally, nearly two-thirds of those patients referred to us were alive at the end of the observation period, reflecting trends seen by the most established PPC programs. ¹⁰ We attribute our wide array of referring subspecialties and underlying diagnoses during the observed time period to successful, proactive hospital-wide education as well as an overall shift in medical culture and understanding regarding the variety of services a PPC team may provide.

The only data that closely reflected the pattern of a start-up PPC team were the location of patients at the time of referral and the location of patients at the time of their death. A majority of the deaths occurred in the acute care hospital setting. This is similar to the data presented by Vadeboncoeur ⁷ during their first year of development over a decade ago. We suspect this high rate of inpatient rather than in-home or outside hospice facility death is related to accessibility. Although we provide home visits for children in hospice, inpatient hospice is not an option for children in our area, as no facilities exist for children under the age of 18 years. This is also likely due to the large number of patients referred while receiving care in the ICU setting—a percentage significantly greater than described in the multicenter cohort trial (37.5% versus 17.5%). ¹⁰

Conclusions

Gaining widespread acceptance of a new PPC program can be challenging for a developing team. Many hospital administrators with developing palliative care programs are concerned about a period of quiescence at the team's inception. This concern is based on the time it may take a hospital to understand, accept, and incorporate PPC into the care of patients with life-threatening and life-limiting conditions. In our experience, because of educational efforts that preceded the start of our team and also a larger, cultural shift in the understanding of the role of PPC in the care of pediatric patients, there was no lag time between the start of the team and the request for consults. Health care professionals were ready, willing, and eager for palliative care involvement with their patients. We saw this reflected in our referral pattern, which mirrored the distribution of referrals as reported by North America's most established PPC programs.