Palliative Care Programs in Children’s Hospitals

This study used organizational data that was public and deidentified; therefore, the University of Nebraska Office of Regulatory Affairs and Institutional Review Board determined the study to not constitute human subject research. During an 8 month period between April and December 2021, the Children’s Hospital Association (CHA) surveyed 231 children’s hospitals in the United States about services and programs from fiscal year 2020.⁸  CHA research staff sent e-mail invitations to CHA membership hospitals confirming correct contact information and requesting participation in the Annual Benchmark Report (ABR). When the ABR was ready for data entry, an e-mail solicitation asked administrative leadership at each pediatric care setting to name a reliable data submission contact person. Only 1 ABR could be completed per facility. Reminders were sent electronically in 3 week intervals for 60 days or until survey completion. Financial incentives were not provided for survey completion.

The ABR questions mirrored the American Hospital Association Survey, with questions reworded for pediatric appropriateness. The ABR items of interest were framed as: “Indicate whether palliative care is available through your hospital” with a binary response option; likewise, “Indicate whether hospice care is available through your hospital” with the same yes or no format. The survey results were aggregated and blinded. A data use agreement was completed by the study team, and CHA provided membership data for research specific to this PPC benchmark analyses. The CHA ABR was selected as the PPC national benchmark data source because of the inclusion of all children’s hospitals in the United States.

Our study team determined a priori to analyze specific questions on the ABR based on relevance to specialty PPC provision. Demographic (state, region, number of staffed beds, free-standing, or embedded), presence of a hospice, and respite coverage data were obtained. Four a priori domains were prioritized: (1) Operational (trauma level, NICU level, association with a skilled nursing facility, long-term care facility, home health, or medical day program); (2) Missional (accreditation status, Magnet designation, care coordination for primary medical home model, family advisory board, social determinant of health screening, or family resource center); (3) Educational (Accreditation Council for Graduate Medical Education[ACGME] residency program affiliated or American Medical Association medical school affiliation); and (4) Financial (participation in Accountable Care Organization [ACO], designation as a federally qualified health center [FQHC]). These categories were predetermined to align with a prior pediatric hospice framework and PPC conceptual model previously published by members of the study team.^9–11 

PC SAS version 9.4 was used to summarize and analyze the data. Categorical variables were summarized using frequencies and percentages. These variables were analyzed using χ² tests or Fisher’s Exact Tests when expected cell counts fell below 5. The staffed bed variable was summarized using the median, quartiles, and minimum and maximum. This variable was analyzed using the nonparametric Mann-Whitney U test. A multivariable logistic regression to predict presence of a PPC program was performed, including the terms ACO and FQHC in the model which used a Firth’s penalized maximum likelihood estimation to reduce bias in the parameter estimates.

A total of 231 CHA membership hospitals received ABR survey requests, with 148 submitted successfully (64% response rate) for the 2020 fiscal year. All 50 states were represented in the survey response. Of the 148 pediatric care settings contained in the registry (Table 1), 119 (80%) reported having a PPC program. Eight states presented 5 or more PPC programs: California n = 11, Florida n = 9, Ohio n = 7, Texas n = 6, Pennsylvania n = 5, New York n = 5, Missouri n = 5, and North Carolina n = 5. The Midwest region housed a higher prevalence of PPC programs (n = 31 of 35, 89%) compared with the Northeast (n = 19 of 24, 79.2%), South (n = 42 of 56, 75%), or West (n = 23 of 29, 79%); however, there was not a statistically significant difference in the likelihood of having a PPC program by geographic region.

Free-standing children’s hospitals (n = 42 of 148, 28%) were more likely to report the presence of a PPC program (n = 40 of 42, 95%, P = .004) compared with pediatric settings embedded within adult facilities (n = 79 of 106, 75%). For those care settings with PPC programs, the median number of staffed beds was 185 (minimum 25, maximum 782, 25^th quartile 119, 75^th quartile 303) compared with 49 median number of staffed beds for those without PPC (minimum 12, maximum 281, 25^th quartile 30, 75^th quartile 81).

Over half of respondent facilities reported not having access to hospice services for children (n = 82 of 148, 55%). Access to hospice services correlated with the presence of PPC (P ≤.0001). Likewise, the majority of respondents reported not having access to respite care services for family caregivers (n = 117 of 148, 79%). Access to respite care correlated with the presence of PPC (P = .0098).

Higher-level trauma center accreditation correlated with the presence of PPC (P ≤.0001). Level I trauma centers, recognized as tertiary care regional resources, were most likely to report the presence of PPC (n = 64 of 67, 96%) compared with facilities without accredited trauma center designation (n = 34 of 57, 60%). Similarly, higher NICU designation correlated with the presence of PPC (P <.0001). Only a quarter of care settings without a NICU had PPC (n = 7 of 27, 26%) compared with 97% (86 of 89) of facilities with a level IV NICU.

Almost half of respondents (n = 67 of 148, 45%) reported access to home health services for pediatric patients, with a higher likelihood of home health existing in settings with PPC (P = .0001). The presence of a medical day program for children with complex needs was reported by 30 facilities, although this was not more likely in settings with a PPC program compared with those without PPC (P = .1001). Presence or absence of PPC programs was not statistically significantly associated with pediatric care settings offering skilled nursing affiliations for children (P = .3665) or long-term care facility options for children (P = .2843).

Of the 132 facilities reporting accreditation by the Joint Commission, 104 (79%) had PPC. There was not a significantly significant difference in the presence of PPC between accredited and nonaccredited settings (P = .1982). A total of 78 of 148 facilities reported having Magnet Nursing Designation. Magnet recognition was correlated with the presence of PPC (P ≤.0001). The presence of care coordination services for pediatric primary medical homes was reported by 130 of 148 facilities (88%) although coordination of a primary medical home model was not associated with the presence of a PPC program (P = .5276).The presence of a family advisory board (n = 123 of 148, 83%) occurred more often in the presence of PPC as did screening for social determinant of health (n = 103 of 148, 70%) and the existence of a family resource center (82%), with P <.0001 for each measure.

Hospitals affiliated with ACGME residency programs were more likely to report presence of PPC (n = 95 of 148, P <.0001) as were those with American Medical Association medical school affiliations (n = 101 of 148, P <.0001).

Among the 40% of pediatric hospitals that self-designated as ACOs, 54 (92%) reported specialty PPC services. Access to PPC was statistically significantly associated with ACO status (P = .0055). Likewise, each of the 23 of 148 care settings that self-designated as a FQHC reported the presence of a PPC program (n = 23 of 23, 100%, P = .0079). Using a logistic regression model, ACO is a statistically significant (P = .0191) predictor of the presence of a PPC program.

This study describes the prevalence of PPC programs through analysis of national ABR data aggregated by the CHA in 2020, representing children’s hospitals across every state. In this national sample, we found that 1 in 5 children’s hospitals in the United States lacked access to specialty PPC services. Since 2006, access to specialty PPC has been included as a key criterion for ranking the quality of US children's hospitals by US News and World Report¹² ; yet more than 15 years later, a sizeable percentage of pediatric hospitals remain without access to PPC services.

During the past decade and a half, several studies have attempted to describe the landscape of PPC across the United States, with notable limitations. In 2008, the Children’s Oncology Group surveyed its member hospitals and found that 108 (60%) of respondents reported having PPC services available specifically for children with cancer¹³ ; however, sampling within this population limited generalization of PPC program prevalence rates beyond the Children’s Oncology Group domain. A study conducted in 2012 surveyed 226 hospitals identified by the National Association of Children’s Hospitals and Related Institutions, revealing that 112 (70%) of respondent children’s hospitals reported having a PPC program.⁴  In 2020, an important cross-sectional study used annual survey data from the American Hospital Association to assess the prevalence of palliative care in the United States.⁷  Although this study included free-standing children’s hospitals as a predictor, the study did not examine factors specifically within children’s hospitals relevant to the presence or absence of PPC.⁷  Consistent characteristics between this PPC prevalence study and the prior cross-sectional national study included palliative association with larger hospitals, teaching programs, and hospice presence.⁷ 

Historically, the presence of specialty PPC services within children’s hospitals in the United States has been described through data reported by representatives from existing PPC programs, likely underreporting the scope and reach of PPC across pediatric care settings nationally. In 2018, the National Palliative Care Registry⁵  solicited self-reported information from PPC program leaders and clinicians at children’s hospitals, publishing a pediatric summation report from 54 PPC programs.⁶  Whereas data gathered from self-reported annualized PPC program entry adds value and insight into PPC process, structures of care, and staffing models, it also risks overestimating PPC implementation and reach because of selection or desirability biases.¹⁴ 

Importantly, equipoise also warrants attentiveness to organizational environments not reporting the presence of PPC programs. This study is the first to compare the presence or absence of PPC and other operational, missional, and financial educational characteristics represented by children’s hospitals. We found several institutional and resource factors associated with the existence of PPC programs at a children’s hospital. Access to PPC services correlated with larger environments with more acute care (operations), comprehensive care (operations), and higher standards of care (mission, educational, and financial). Tertiary or quaternary facilities with higher ratio of human capital and trauma, intensive care, or acuity level also were more likely to have PPC. Similarly, larger, free-standing, academically-affiliated children’s hospitals were more likely to offer PPC. Conversely, services traditionally established for children with complex chronic needs (eg, pediatric skilled nursing facility, medical day program, or a medical home model) were not correlated with the presence of PPC services. This finding underscores the need to delineate roles and relationships between complex care for children and PPC.¹⁵  Future research might examine utilization of these services by PPC patients to better understand the complexity of care and need for care coordination.

With respect to missional and educational characteristics, our findings revealed that children’s hospitals providing PPC maintained high standards, as demonstrated by Magnet designation. Unsurprisingly, Magnet nursing designation represents a commitment to quality patient care, nursing excellence, and care innovations that represent values congruent with PPC priorities.^16–18  Access to PPC services correlated with family-centered practices, such as family advisory boards, family resource centers, and social determinant of health screening practices. PPC programs, by their nature, may encourage development and sustainment of these core initiatives; conversely, children’s hospitals prioritizing family-centric initiatives may be more likely to support PPC program development.

Hospital-level accreditation by the Joint Commission, however, was not associated with a higher likelihood of the presence of PPC. Although the Joint Commission currently prioritizes certain performance metrics relevant to PPC, such as pain and dyspnea screening,¹⁹  consideration should be placed on PPC quality and practice standards as a Joint Commission accreditation expectation for pediatric care settings. One-third of PPC programs reported interest in maintaining or seeking Joint Commission palliative care advanced certification.⁴  Integration of PPC in national metrics matters because when PPC becomes a contributing criterion, institutions are incentivized to establish and maintain PPC programs. Additional consideration should be given to ways to measure the quality and outcomes of PPC in children’s hospitals.²⁰ 

Academic pediatric centers were more likely to have access to PPC services, and this phenomenon should be leveraged to strengthen formal PPC exposure and training for medical students and residents. Exposure to PPC can bolster provision of primary palliative care across pediatric patient populations.²¹  Additionally, exposure may translate into career interest to further expand the PPC workforce. Shortage of PPC clinicians remains a pressing concern, particularly in rural settings. Out of 54 inpatient PPC programs included in the National Palliative Care Registry, the minority (37%) met minimum standards of practice for staffing.⁶  The majority of National Palliative Care Registry respondent PPC programs (60%) shared an inability to meet clinical demand with available staffing.⁶ 

With respect to financial characteristics, although PPC programs demonstrated potential cost-reduction by moving from inpatient care to more coordinated outpatient or even home-based care,^22–24  economic environment of PPC for hospitals remains variable across settings. PPC was available in all settings that qualify as FQHC and was correlated with ACO status, perhaps representing recognition of the financial and nonfiscal benefits of PPC for patients, families, hospitals, and communities. As FQHCs are safety net providers that qualify for federal funding to provide comprehensive primary care and serve medically underserved areas, increases in funding for FQHCs may boost the availability of PPC for underserved children and help accommodate increases in demand for PPC with improved access. Moreover, PPC programs operated in organizations that predominately provided primary care services to underserved areas. Children’s hospitals may have begun to integrate PPC into primary care through FQHC mechanisms, as recent federal regulation has changed the payment structure between FQHC and hospices.^25,26  Presently, children’s hospitals are poised to leverage this relationship as federal regulations expand in future years.²⁷  Specific to the pediatric patient, specialized PPC programs consistently reveal a positive impact on patient quality of life and communication, which represents immeasurable value for the recipient child and family.^28–30 

The gold standard of PPC, hospice, and respite access as a universal standard of care^31,32  remains unrealized for too many children and families in the United States.³³  As PPC positively impacts the care quality, communication, cost, and service coordination,^21,30  lack of specialty PPC in children’s hospitals may perpetuate inequities in care.³¹  Sustainable PPC programmatic expansion requires creation of a definition and standards for PPC, specifying eligibility criteria, clarifying service expectations, implementing quality measures, and leveraging policy mechanisms to pay for care. PPC policy should prioritize service coverage, payment reform, and workforce expansion through education.

Presently, few federal and state policies focus on funding or facilitating PPC. Existing federal legislation has concentrated on Medicaid end-of-life and home-based hospice services, evidenced by the 2019 Advance Care for Exceptional Kids Act and the Concurrent Care for Children (2010 Patient Protection and Affordable Care Act, Section 2302). The Advance Care for Exceptional Kids Act recognized in theory that coverage for children should include care coordination between palliative and hospice transitions, as earlier initiation of PPC results in greater impact.³⁴  However, this Act has yet to be implemented. The 2010 Patient Protection and Affordable Care Act allowed for concurrent provision of curative and palliative care with hospice care for children enrolled in Medicaid or the Children's Health Insurance Program.²⁷  The proposed federal Children's Program of All-Inclusive Coordinated Care Act aims to provide access to PPC for children whose medical care is reimbursed by any payer source, although has not yet been enacted.³⁵  State policies aimed at funding PPC through Medicaid programs also are limited; Florida, Massachusetts, and Illinois offer rare examples of state PPC funding policies,^34,36,37  yet these efforts necessitate years of relationship development between key stakeholders and Medicaid offices,²⁷  and as a consequence, only a few states currently fund PPC.¹  Thus, no consistent or uniform federal or state approach to PPC payment exists, which significantly limits provision and access to PPC.

Access to PPC and hospice requires strengthening education mechanisms to improve staffing models.³⁸  The federal Palliative Care and Hospice Education and Training Act would invest in the palliative care workforce so there are more trained professionals to care for patients with serious illness at all ages. Recognizing the geographic and rural-urban gaps in PPC coverage, clinicians should be incentivized to work in low-resource settings through education loan repayment and public service funding mechanisms.

As this study was focused on the national prevalence of PPC and associated organizational factors, limitations include lack of PPC program descriptions, such as staffing, funding, coverage hours, scope, setting, and patient services. Prior studies have shown extensive variability in PPC service reach and models.⁶  Important factors known to correlate with the prevalence of palliative care, such as hospital ownership and rurality,⁷  were not available in the dataset. Although ABR is noted for knowledgeable responders, some degree of unmeasured error may persist because of incomplete knowledge or inaccurate data entry. The ABR did not solicit information related to bereavement or ethics services, which would have been interesting considerations. Patient demographic variables including ethnic and racial diversity, rural and urban, economic background would further expand insight into potential inequities in PPC service reach. Despite these limitations, the findings from this study have important clinical and policy implications for hospital administrators, pediatric clinicians, and PPC policy advocates across the United States.

Development of new PPC programs is an urgent priority to provide optimal care to children with serious illness and their families at the 1 in 5 children’s hospitals that currently lack PPC services. Effective and sustainable PPC care delivery models, inclusive of hospice (which 50% currently lack) and respite services (which 80% currently lack) are integral to achieving equity and justice for vulnerable patient populations. Findings from this national PPC benchmark study should stimulate not only conversations, but also actionable PPC programmatic coverage for children and families.^39–42  The field of PPC must focus on to how to create and sustain quality PPC programs across care environments to include smaller settings without intensive care or trauma services and without academic affiliations. Further research is needed to broaden knowledge around PPC clinical demand and implementation of efficient care models to support patients, families, healthcare professionals, and communities.

We thank Ms. Stacie Sinclair (Center to Advance Palliative Care) for her contribution as well as the Children’s Hospital Association for their original data collection.

ABR

Annual Benchmark Report

CHA

Children’s Hospital Association

PPC

Pediatric Palliative Care