Comparison between Rural and Urban Appalachian Children in Hospice Care

Abstract

Objective:

The goal of this study was to compare rural and urban pediatric hospice patients in Appalachia.

Methods:

Using a retrospective, nonexperimental design, we sought to compare characteristics of Appalachian rural and urban children younger than 21 years enrolled in the Medicaid hospice benefit. Descriptive statistics were calculated on the demographic, hospice, and clinical characteristics of children from Appalachia. Comparisons were calculated using Pearson χ² for proportions and the Student t test for means.

Results:

Less than half of the 1788 Appalachian children admitted to hospice care resided in rural areas (40%). Compared with children in urban areas of Appalachia, rural children were significantly younger (8 years vs 9.5 years) and more often had a complex chronic condition (56.0% vs 35.1%) and comorbidities (38.5% vs 17.0%) with technology dependence (32.6% vs 17.0%). Children in rural Appalachian were commonly from communities in the southern region of Appalachia (27.9% vs <10.0%), with median household incomes <$50,000/year (96.7% vs 22.4%). Significant differences were present in clinical care between rural and urban Appalachian children. Rural children had longer lengths of stays in hospice care (38 days vs 11 days) and were less likely to use the emergency department during hospice admission (19.0% vs 43.0%). These children more often visited their primary care provider (49.9% vs 31.3%) and sought care for symptoms from nonhospice providers (18.1% vs 10.0% ) while admitted to hospice.

Conclusions:

Our results suggest that children admitted to hospice care in rural versus urban Appalachia have distinct characteristics. Rural children are admitted to hospice care with significant medical complexities and reside in areas of poverty. Hospice care for rural children suggests a continuity of care with longer hospice stays and fewer transitions to the emergency department; however, the potential for care fragmentation is present, with frequent visits to primary care and nonhospice providers for symptom management. Understanding the unique characteristics of children in Appalachia may be essential for advancing knowledge and care for these children at the end of life. Future research examining geographic variation in hospice care in Appalachia is warranted.

In the United States, hospice care often is underutilized among children. Although ~30,000 children die each year of health-related causes, fewer than one-third (27%) of patients use pediatric hospice care.^1-3 Hospice patients have a prognosis of ≤6 months to live and are cared for by a multidisciplinary team of hospice nurses, physicians, pharmacists, social workers, and child life specialists. Pediatric hospice care offers pain and symptom management, along with other psychosocial services such as respite and bereavement care that extends across a variety of care settings. Evidence suggests that children receive quality symptom management (eg, dyspnea, end-stage restlessness, pain) and parents are satisfied with the comprehensive care provided for their children^4,5; however, access to pediatric hospice care often is limited for rural children. It can be difficult to access hospice services with clinicians who are comfortable working with children, particularly when children have complex conditions and technology dependence. Technology dependence is the need for a medical device to compensate for the loss of a vital bodily function to avert death or further disability. For children at the end of life, technology may include mechanical ventilators, oxygen supplementation, and/or tracheostomy tubes.⁶ In a recent study, fewer than 4% of children in hospice care received services from a hospice provider that served a rural community, and many rural areas lack providers who will enroll pediatric patients.^4,7,8 As such, access to hospice care may differ between rural and urban residence even within Appalachia.

Appalachia is a rural area that historically has been considered underserved by hospice care. Approximately 8% of the US population resides in the Appalachian region, which extends >1000 mi and across 13 states, from Mississippi to New York.⁹ Almost half (42%) of the population of the region is rural, with 87% of the 420 counties having >1.5 times the US poverty rate.⁹ Appalachia has struggled with health and health outcomes for more than a century, especially rural Appalachia. The health of children and adolescents in this area is consistently worse than the national average. Infant mortality rates are 16% higher than in the general population, with some rural counties ranging from 17.1 to 500.0 deaths per 1000 live births.⁹ Lack of access begins at birth with some counties, such as those in rural West Virginia, not having any prenatal care resources, and continues to the end of life, with no counties in rural eastern Tennessee accepting pediatric hospice patients.^7,10 As one of the most medically underserved areas in the nation, Appalachia, especially rural Appalachia, has a shortage of healthcare providers and limited healthcare infrastructure.¹¹ Rural Appalachian children and adolescents are in poor health and often do not receive needed high-quality care.

In comparison to children residing in urban settings, rural children face numerous barriers in accessing hospice care. One of them is the lack of access to hospice services. Despite unchanged hospice needs in the pediatric population of Appalachia, the supply of pediatric hospice agencies continues to be sparse in rural areas and does not meet these needs.⁸ According to Campbell and colleagues,¹² only 36% of rural counties have at least one certified hospice facility, and between 2007 and 2017, the total number of hospices located in rural areas declined from 965 to 878.¹³ Research reveals that hospice care for children does not exist in the southern rural region of Appalachia, meaning that providing pediatric hospice care may be a challenge socially and geographically. Another problem is the lack of professional training for hospice workers, especially in the field of psychosocial support of patients.¹⁴ High poverty rates in rural areas lead to the closure of rural healthcare providers,¹⁵ and many rural hospices face challenges in recruiting and retaining hospice personnel because of the complicated working environment, which includes long commuting times, constantly changing hospice census, the need to provide continuous hospice care to patients residing in geographically isolated communities, and the lack of hospice specialists.¹⁶ Finally, many rural residents lack trust in hospice providers.¹⁶ From the few adult studies, we know that Appalachian families have distinct views about death and dying. These families often want to care for their loved ones at home to facilitate a sense of normalcy even when dying, which may include the use of folk medicine care (ie, herbal medicine and remedies that have been handed down through the years in the family).¹⁷ Hospice care is frequently considered “giving up” and is the last resort for a family.^7,17 Appalachian families also have distinct ideas about hospice care, integrating hospice as a last resort only when the family has become exhausted from caring for the severely ill person.

We still lack information about children in rural Appalachia and how they may differ from urban children. There is a gap in our knowledge about their demographic profile, community characteristics, hospice providers, and ultimately the hospice care they receive in Appalachia. Improving our understanding of rural-urban differences among children in hospice care is clinically important and relevant to policy making. Symptom management and family-centered care at home is important for alleviating distress and facilitating good end-of-life care for children with severe illness.¹⁸ As such, the purpose of this study was to compare demographic/community, hospice, and clinical characteristics of rural and urban pediatric hospice patients in Appalachia.

Methods

We conducted a retrospective, comparative study of demographic, community, hospice, and clinical characteristics of children admitted to hospice care in Appalachia. Inclusion criteria for children were being younger than 21 years and living in a ZIPcode in a designated Appalachian county as defined by the Appalachian Regional Commission.¹⁹ Exclusion criteria were missing date of birth or death or no service-level Medicaid claims. The final sample was 1788 children in Appalachia who were admitted to hospice care. Because our sample included children from the entire Appalachian region and beneficiaries of the largest insurer of children at end of life (Medicaid), it is representative of the target population. This study was reviewed and approved by the University of Tennessee, Knoxville institutional review board.

Data Sources

This study used four data sources. The primary data source was the Centers for Medicare & Medicaid Services (CMS) 2011–2013 Medicaid files. The Medicaid Analytical eXtract files are person-level, administrative claims files from all 50 states and the District of Columbia.²⁰ We used four Medicaid Analytical eXtract files, which included personal summary (enrollment and demographic information), other therapy (diagnosis and procedure information), inpatient data (hospitalization data on service dates and procedure codes), and prescription drug information (prescription fill dates and national drug codes). Medicaid claims data were used because they are one of the only data sources that includes nationwide hospice information on people younger than 21 years. The years 2011–2013 were used because these were the most current data available. Another data source was the publicly available 2010 US Census files, which provided information on community characteristics. Hospice provider information was extracted from the CMS Hospice Provider of Services and CMS Hospice Utilization and Payment files.

Measures

Demographic and Community Characteristics

“Rural/urban” was defined using the Health Resources and Services Administration definition.^21,22 “Age” was the age of the child when admitted to hospice care. “Sex” was female or male. “Race” was operationalized as White or non-White. Whether a child had private insurance in addition to Medicaid insurance was the measure of “co-insurance.” “Complex chronic condition” was whether a child had a complex chronic condition as defined by Feudtner and colleagues.²³ Individual condition measures were created for neurological/neuromuscular, cardiovascular, cancer, respiratory, and gastrointestinal. We defined “comorbidities” as two or more complex chronic conditions.²⁴ “Mental/behavioral disorder” was whether a young adult had a mental or behavioral diagnosis.²⁵ Requiring medical technology or devices was the definition of “technology dependence.”²⁶ Southern Appalachian region was defined as “residence in either Alabama, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, Tennessee, Virginia, or West Virginia.” “Household income” was the percentage of county residents with a median household income less than or more than $50,000/annually.

Hospice Characteristics

The “size of a hospice” was dichotomized as ≤50 employees or >50 employees. “Hospice ownership” was for-profit versus nonprofit and government. The number of years the hospice was a licensed hospice provider was the measure of “organizational age.” “Pediatric program” was whether the hospice had a dedicated program for children and adolescents.

Clinical Characteristics

Whether children used concurrent hospice care (medical care during hospice enrollment) was the measure of “concurrent hospice care.” The “length of stay in hospice care” was calculated as the total number of days a child was admitted to hospice care. “Emergency department transition” was defined as whether the emergency department was used while the child was admitted in hospice care. A single-day admission to hospice care was the measure of “1-day hospice admission.” Whether a child visited the pediatrician or primary care provider was “primary care visit.” “Nonhospice symptom management” was operationalized as receiving care for constipation symptoms from a nonhospice provider during hospice admission.

Data Analysis

The primary aim of the study was to compare rural and urban Appalachian children in hospice care. We conducted a secondary analysis of the data for this study and descriptive statistics were calculated on the demographic, hospice, and clinical characteristics of children from Appalachia. The children were stratified by rural versus urban for the comparison analysis.²¹ The Pearson χ² test for differences in proportions and Student t test for differences in means were used to provide comparisons between those who resided in rural versus urban areas. The results are presented in the form of univariate distributions and means ± standard deviations. Per our data use agreement, small sample size in some categories—results with <10% or 10 observations—were reported as <10.00%. All of the analyses were conducted using STATA version 15 (StataCorp, College Station, TX).

Results

Among the 1788 Appalachian children admitted to hospice care, 40% resided in rural areas and 60% in urban areas. The demographic characteristics of the children from Appalachia are presented in Table 1. In the overall sample, the average age was 9 years and the sex was evenly divided between females and males. A majority of the children were White (87.53%) and slightly more than one-third had private insurance along with Medicaid insurance. Fewer than half of the children had a complex chronic condition. The most common condition was cardiovascular disease (17.06%), followed by neurological/neuromuscular conditions (15.88%). One-fourth of children had comorbidities, fewer than half (43.18%) had a mental/behavioral condition, and 23% were technology dependent. More than 15% of children resided in the southern Appalachian region, whereas half lived in communities with a median household income of <$50,000/year.

Table 1.

Demographic and community characteristics (N = 1788)

Variables	Appalachian children	Rural/urban
		Rural n = 691	Urban n = 1097	P
Age	9.02 (6.51)	8.21 (6.75)	9.50 (6.33)	<0.001
Sex, %				0.387
Female	50.11	51.37	49.27
Male	49.89	48.63	50.73
Race, %				0.295
White	87.53	88.57	86.89
Non-White	12.47	11.43	13.11
Co-insurance, %	36.63	43.85	32.06	<0.001
Complex chronic condition, %	43.12	56.01	35.06	<0.001
Neurological/neuromuscular	15.88	24.60	10.47	<0.001
Cardiovascular	17.06	10.47	12.20	0.265
Cancer	<10.00	13.02	<10.00	<0.001
Respiratory	12.08	18.81	<10.00	<0.001
Gastrointestinal	13.03	21.85	<10.00	<0.001
Comorbidities, %	25.28	38.49	17.03	<0.001
Mental/behavioral conditions, %	43.18	45.59	41.62	0.099
Technology dependence, %	22.99	32.56	17.03	<0.001
Southern Appalachia region, %	15.44	27.93	<10.00	<0.001
Household income <$50,000/y, %	51.12	96.67	22.40	<0.001

Boldface type indicates statistical significance.

Demographic differences were noted between rural and urban children (Table 1). Rural Appalachian children were younger than urban Appalachian children (8 vs 9.5 years) and often had co-insurance (43.85% vs 32.06%). These children also experienced higher frequencies of complex conditions (56.01% vs 35.06%), specifically in neurological/neuromuscular (24.60% vs 10.47%), cancer (13.02% vs <10.00%), respiratory (18.81% vs <10.00%), and gastrointestinal (21.85% vs <10.00%) conditions. Compared with urban children, those who resided in rural Appalachia had higher comorbidities (38.49% vs 17.03%). Rural Appalachian children were more likely to be technology dependent (32.56% vs 17.03%). These children were commonly from the southern Appalachian region (27.93% vs <10.00%) and resided in communities with household income <50,000/year (96.67% vs 22.40%). There were no other significant differences between rural and urban Appalachian children.

The hospice characteristics of hospice providers for children in Appalachia are listed in Table 2. A majority of hospices had >50 employees (64.88%) and were for-profit (86.52%). Hospices had been in operation on average for >22 years and almost half had a pediatric program (43.62%). Rural children in Appalachia were more likely to seek care from hospices with >50 employees (40.81% vs 31.97%) and a pediatric program (46.60% vs 41.71%) compared to urban children. There were no differences in ownership or organizational age.

Table 2.

Hospice characteristics (N = 1788)

Variables	Appalachian children	Rural/urban
		Rural n = 691	Urban n = 1097	P
Size, %				<0.001
<50 employees	35.35	59.48	68.21
>50 employees	64.88	40.81	31.97
Ownership, %				0.056
For profit	86.52	88.42	85.25
Non-profit/government	13.53	11.58	14.85
Organizational age, y	22.87 (5.96)	22.74 (6.01)	22.96 (5.93)	0.449
Pediatric program, %	43.62	46.60	41.71	0.042

Boldface type indicates statistical significance.

The clinical characteristics of the hospice care received by children from Appalachia are displayed in Table 3. More than one-third of the children used concurrent hospice care. On average, their length of stay in hospice was 46 days. One-third of children used the emergency department during hospice admission and almost half (47.82%) experienced a 1-day hospice admission. More than one-third of children continued to see their pediatrician/primary care physician (38.53%) and 13% sought treatment for symptoms from a nonhospice provider while admitted to hospice care.

Table 3.

Clinical characteristics (N = 1788)

Variables	Appalachian Children	Rural/urban
		Rural n = 691	Urban n = 1097	P
Concurrent hospice care, %	38.53	40.81	37.16	0.123
Hospice length of stay, %	45.53 (77.70)	38.36 (103.63)	11.42 (51.83)	<0.001
Emergency department transition, %	33.72	18.96	42.99	<0.001
1-Day hospice admission, %	47.82	36.03	55.19	<0.001
Primary care visit, %	38.53	49.93	31.33	<0.001
Nonhospice symptom management	12.98	18.09	10.00	<0.001

Boldface type indicates statistical significance.

Significant differences were present in clinical care between rural and urban Appalachian children. Rural children had longer length of stays in hospice (38 vs 11 days), more often visited their primary care provider (49.93% vs 31.33%), and sought care for symptoms from nonhospice providers (18.09% vs 10.00%). Conversely, urban children were more likely to use the emergency department (42.99% vs 18.96%) and experience a 1-day hospice admission (55.19% vs 36.03%). There was no difference in the use of concurrent hospice care.

Discussion

As this is one of the first studies to examine Appalachian children in hospice care, the purpose of this retrospective, descriptive study was to compare demographic, community, hospice, and clinical characteristics and differences in Appalachian rural and urban children. We found significant differences between these two groups, with rural Appalachian children being more financially vulnerable, having more fragile health than their urban peers, and having higher rates of technology dependence. The revealed profile of rural and urban children provides baseline data on an underserved and understudied population at the end of life.

One major finding from the study revealed that rural children in Appalachia were more commonly technology dependent on equipment and devices than their urban peers. One-third of Appalachian children in hospice versus 17% of urban children were technology dependent. Although these findings are in the range of prior literature,^27,28 this finding was unexpected. We anticipated that urban children in hospice care with proximity to urban children’s hospitals and specialty care (eg, respiratory care specialist) would more often be sent home to hospice care with technology as compared with rural children. From recent coronavirus disease 2019 data, we know that resources to manage technology (eg, ventilator-dependent patients) in rural hospitals often are scarce, and these patients have no resources to outreach within their community; thus, one possible explanation for our finding may relate to the advances in home technology. For example, home ventilation has progressed during the past 20 years, and the skills needed to operate a home ventilator are not as complex today as when first introduced.²⁹ This may enable the parents of rural pediatric patients to more quickly learn how to operate the equipment. In addition, these parents may be motivated to learn this technology so they can bring their children home in accordance with their beliefs of preserving a sense of normalcy at home during the dying phase. It also may reduce the amount of travel to the children’s hospital. Future research on parental motivation and skills in managing their child’s technology is warranted.

Among the Appalachian children in the study, almost half of them resided in communities with household income <$50,000/year, and in rural Appalachia almost 100% of children resided in communities of poverty. The Appalachian Regional Commission collects education, employment, economic status, and industrial data on the region. The Appalachian Regional Commission reports that rural Appalachians are increasingly receiving college educations and experiencing increases in overall household income; however, many rural Appalachian counties still have per capita incomes <75% of the US average. Rural Appalachia has been hit hard with continued large income disparities, relatively high unemployment, outmigration of families to work, and relative isolation because of limited communications and lack of transportation.³⁰ As such, children in hospice care living in rural Appalachian communities may be receiving their care in an environment of extreme poverty. Hospice clinicians caring for this population may conduct a thorough assessment of the home, family, and community (social determinants) upon admission to ensure high-quality comfort care.

There were other interesting results from the analysis regarding clinical characteristics. Children in rural Appalachia more often visited their primary care physicians during hospice enrollment than did urban Appalachian children. Although many rural Appalachian communities are deemed medically underserved by healthcare providers,³¹ it is common in rural communities that the local primary care doctor has been caring for these children since they were born. Many have a long-term relationship with their patients, especially among those with terminal illness. These physicians are typically trusted members of the community, who in many cases may be seen as a trusted friend.^32,33 These relationships may explain why rural Appalachian children more frequently visited their physician versus urban children. Additional research exploring the relationship between the children, their family, and the primary care physician is needed.

This study has several clinical implications. These findings have important clinical implications for hospice nurses and clinicians. Given the significant differences between Appalachian children residing in urban and rural areas, it is important for hospice professionals to be aware of the significance of place where children come from in the delivery of end-of-life care. In comparison to their urban peers, rural children have more fragile health with more complex chronic conditions and a higher likelihood of having technology dependence. At the same time, families of rural children have significantly fewer economic and healthcare resources for end-of-life care for their children. Considering that hospice services are covered by Medicaid, hospice is the ultimate option for providing high-quality end-of-life care to rural children.

There are several limitations to the study. First, the findings have limited generalizability. The study included Medicaid beneficiaries, so the results are not generalizable to children covered by private insurance. However, Medicaid is the predominate insurer of children in the United States and these study findings have relevance to children covered by other insurers. Second, selection bias may be present in the study. Infants in the sample may be underrepresented, because life-threatening diseases develop very quickly in this age and parents may not have enough time to enroll them in the US Medicaid benefit, as the enrollment process can take several months. The data used for the study also were a limitation. The Medicaid data available at the time of our study was from 2011 to 2013. Although these data are less than a decade old, the findings should be interpreted with caution; however, these files represent the only comprehensive data on pediatric hospice patients available to researchers. Finally, because this study used a Medicaid claims dataset, it had a limited number of contextual variables, such as family history and family characteristics.

In summary, this study sought to compare demographic, hospice, and clinical characteristics of children admitted to hospice care in rural and urban Appalachia. Findings provided important baseline information on rural and urban Appalachian children. They revealed that in comparison to children in urban areas, rural children have more fragile health and higher dependence on medical technology while also having limited access to high-quality hospice services. Because Appalachian is a geographically large and diverse area that covers 13 US states, future research should focus on regional differences in the utilization of pediatric hospice services. This work would help to improve access to high-quality hospice care for children with complex medical conditions.

Key Points.

• The finding from the 1788 Appalachian children studied showed that pediatric decedents were generally young and received care from established hospices with pediatric programs, in which they had a unique clinical hospice experience.

• Significant differences emerged in demographic and clinical characteristics between rural and urban Appalachian children.

• Rural Appalachian children were more likely to be dependent on technology. The same children were commonly from the southern Appalachian region and resided in communities with a household income of <$50,000/year.

• Rural children more often visited their primary care provider and sought care for end-of-life symptoms from nonhospice providers.