Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Oct 1.
Published in final edited form as: Am J Hosp Palliat Care. 2021 Dec 4;39(10):1129–1136. doi: 10.1177/10499091211056039

Effectiveness of pediatric concurrent hospice care to improve continuity of care

Lisa C Lindley 1, Melanie J Cozad 2, Jennifer W Mack 3, Jessica Keim-Malpass 4, Radion Svynarenko 5, Pamela S Hinds 6
PMCID: PMC9166145  NIHMSID: NIHMS1768340  PMID: 34866426

Abstract

Background:

The 2010 Patient Protection and Affordable Care Act (ACA) mandated landmark hospice care legislation for children at end of life. Little is known about the impact of pediatric concurrent hospice care.

Objective:

The purpose of this study was to examine the effect of pediatric concurrent versus standard hospice care on end-of-life care continuity among Medicaid beneficiaries.

Methods:

Using national Medicaid data, we conducted a quasi-experimental designed study to estimate the effect of concurrent versus standard hospice care to improve end-of-life care continuity for children. Care continuity (i.e., hospice length of stay, hospice disenrollment, emergency room transition, inpatient transition) were measured via claims data. Exposures were concurrent hospice versus standard hospice care. Using instrumental variable analysis, the effectiveness of exposures on care continuity was compared.

Results:

Concurrent hospice care affected care continuity. It resulted in longer lengths of stays in hospice (β = 2.76, p < 0.001) and reduced hospice live discharges (β = −2.80, p < 0.05), compared to standard hospice care. Concurrent care was not effective at reducing emergency room (β = 2.09, p < 0.001) or inpatient care (β = 0.007, p < 0.05) transitions during hospice enrollment.

Conclusion:

Our study provides critical insight into the quality of care delivered for children at end of life. These findings have policy implications.

Keywords: Pediatric hospice care, concurrent hospice care, care continuity, hospice live discharge, hospice length of stay, emergency department use, inpatient admission, Medicaid

Among the 30,000 pediatric deaths each year in the United States (U.S.), approximately 8,000 or 27% of children and their families utilize hospice care at end of life.1–3 It is common for these children to have multiple chronic, complex, life-limiting conditions such as chromosomal anomalies, congenital malformations, complications related to prematurity, neurological conditions, and/or cancer.4 They often have functional limitations and require technology for activities of daily living such as oxygen, gastronomy tubes, and assistive devices.5 Children at end of life have complicated care plans that extend across numerous clinical specialties,6 and a consequence of their care is often substantial caregiving burden and family strain.7

The 2010 Patient Protection and Affordable Care Act (ACA) mandated landmark hospice care legislation for children at end of life. ACA, Section 2302, also known as Concurrent Care for Children, required that Medicaid and Children’s Health Insurance Program (CHIP) change hospice eligibility requirements for children under 21 years. Under concurrent hospice care, families having a terminally ill child with a 6-month to live prognosis may choose to continue medical treatments for the illness while enrolled in hospice care.8 For those patients and families who do not wish to continue medical treatments, standard hospice care with its focus on comfort is still available for children.

What little is known about pediatric concurrent hospice care has focused on describing the approach to care.9–13 However, there is new evidence about the connection between concurrent hospice care and care continuity for children at end of life. End-of-life care continuity is defined as a series of healthcare events that are experienced as coherent and sequentially related to patients’ needs and personal context.14,15 In prior work, researchers have found that standard hospice care for children is frequently fragmented and discontinuous.16 Almost a third of the children in standard hospice care disenrolled from hospice to seek more aggressive care, while 9% transitioned to other providers during hospice enrollment to obtain health care services. The average hospice length of stay was as brief as 14 days. Specific to concurrent care, the assumption reflected in a policy brief was that concurrent hospice care would improve care continuity because it offered continued communication and ongoing connections for the child and family with their original medical team while receiving hospice care.17 However, no pediatric studies were identified that investigated the impact of concurrent care on any patient outcomes, including care continuity.

This study was guided by the Donabedian Model of Health Care Quality (Figure 1). The Donabedian Model posits that quality of health care can be drawn from the structures, processes, and outcomes of health services.18–22 The current study sought to test the effect of concurrent versus standard hospice care on end-of-life care continuity for children. More specifically, we tested the hypotheses that children who used concurrent hospice care would have longer length of stay in hospice, reduced disenrollment from hospice, and fewer transitions to the emergency room or inpatient setting, compared to those using standard hospice care.

Figure 1.

Figure 1.

Open in a new tab

Conceptual Model of Pediatric Hospice Care Quality

Methods

Design and Data Sources

We conducted a retrospective, quasi-experimental study using observational data. The primary data were the 2011 to 2013 Medicaid Analytic Extract (MAX) claims files. The Medicaid Analytical Extract (MAX) files are collected and created by the Centers for Medicare and Medicaid Services (CMS). They are person-level, administrative claims files from data submitted by all 50 states and the District of Columbia.23 The study used four Medicaid files: Personal Summary (enrollment and demographic information), Other Therapy (diagnosis and procedure information), Inpatient (hospitalization data on service dates and procedure codes), and Prescription Drug (information on prescription fill dates and national drug codes). Medicaid claims data were used because they were one of the only nation-wide data sources to include hospice information on children. Other data sources were the publicly-available 2010 U.S. Census files, which provided information on community characteristics, and the CMS Hospice Provider of Services files and CMS Hospice Utilization and Payment files, which provided information on hospice provider characteristics. This study was reviewed and approved by the University of Tennessee, Knoxville Institutional Review Board.

Sample

The sample was limited to pediatric decedents who were younger than 21 years with a hospice service claim from January 1, 2011 to December 31, 2013. Children who had no date of birth, date of death, or service level Medicaid claims were excluded. The year 2011 was chosen because it was the first full year of concurrent hospice care implementation within state Medicaid programs and the year 2013 was chosen because it was the most current year that CMS MAX data were available. The sample was limited to participants who were 20 years old or younger because concurrent hospice care under Medicaid regulations is not available beyond age 20.

Measures

Structures.

The structures in the model included demographic, hospice, and community characteristics using the Medicaid, U.S. Census, CMS Hospice Provider of Services and CMS Hospice Utilization and Payment files. From the Medicaid data, demographics for children included age, gender, race (Caucasian, Black, Other), ethnicity (Hispanic, non-Hispanic), complex chronic condition, comorbidities (i.e., having 2 or more complex chronic conditions), mental/behavioral conditions, and technology dependence. Hospice characteristics were organizational size (i.e., ≤ 50 employees, > 50 employees), ownership (i.e., for-profit, non-profit, & government), years of operation, and pediatric program, using the CMS Hospice Provider of Services and CMS Hospice Utilization and Payment files. U.S. Census data was used for Education level (no high school, high school, college), median household income (i.e., ≤ $50,000/yr, > $50,000/yr), U.S. census regions (i.e., Northeast, Midwest, South, West), and rural/urban were characteristics of the community. The study years, 2011 to 2013, were also included.

Processes.

Our processes were concurrent hospice care or standard hospice care use. Concurrent hospice care as mandated by the ACA, section 2302 states that Medicaid/CHIP children may receive medical services for their terminal illness during hospice care. Using data from the Medicaid MAX Other Therapy and Prescription Drug files, we created a hospice episode, which was the consecutive days of hospice admission based on per diem hospice claims.20,24 The episode dates were used to evaluate whether a child received medical services during a hospice episode. Because all hospice services are included in the Medicaid per-diem, any medical services received on the same day as a hospice claim would constitute concurrent hospice care use. We coded hospice episodes with medical services as concurrent care, and episodes without medical services as standard hospice care.

Outcomes.

Using the Medicaid files, we created 4 measures of end-of-life care continuity based on the hospice and end-life-literature: hospice length of stay, hospice disenrollment, emergency room transition, and inpatient transition.9,16,19,21,22,25,26 Hospice length of stay, which was defined as the total number of days a child was enrolled in hospice care during the study timeframe. This variable was transformed to logarithmic form. Hospice disenrollment was the number of times a child disenrolled and re-enrolled in hospice care. In other words, they left hospice care alive and reenrolled into hospice at a later date. This variable was transformed to logarithmic form. Emergency room (ER) transition was defined as whether the child used the emergency room during hospice care (coded yes/no). Inpatient transition was defined as disenrollment from hospice care in order to be admitted to inpatient care (coded yes/no).

Analyses

For this study, we relied on retrospective, observational data to assess the effectiveness of pediatric concurrent hospice care versus standard hospice care on care continuity using instrumental variable analysis.27–29 The challenge in using the instrumental variable approach is identifying instruments that are significantly correlated to the exposure (concurrent versus standard hospice care), but unrelated to the outcomes of end-of-life care continuity.30 We identified the size of the medical team as an instrumental variable.31,32 We hypothesized children and their families who have a large medical team would decide to maintain that relationship and use concurrent hospice care. Medical team size was operationalized as the number of medical providers upon hospice admission.33 Prior to conducting the instrumental variable analysis, we tested the assumptions for identification of a suitable instrumental variable.34–37

To compare the effectiveness of the concurrent versus standard hospice care on end-of-life continuity of care, a simultaneous two-equation bivariate probit for binary dependent variables was used. Generalized Method of Moment (GMM) models were applied for those models with continuous dependent variables.34,38 The analysis assessed the relationship between concurrent hospice versus standard hospice care on hospice length of stay, hospice disenrollment, and transitions, while controlling for demographic, hospice, and community characteristics. Our analysis was adjusted to account for children with multiple hospice episodes. Separate analyses were conducted for each continuity of care outcome. All analyses were performed using Stata 15.0.39,40

Results

The characteristics of the 18,152 pediatric decedents in the sample are displayed in Table 1. The average age of the children was 7.5 years and boys and girls were evenly distributed. Half the sample was Caucasian, 27% Black, and 22% Hispanic. Less than half of children had a complex chronic condition (48.8%); while 28% suffered comorbidities and 34% had mental/ behavioral conditions. A quarter of the children were technology dependent. Approximately a third of the children in the sample received care at well-established hospices (18.3 years) with more than 50 employees (35.6%). A third of hospices were non-profit (38.7%) and had a pediatric program (33.7%). The children were primarily from the Northeast (57.6%). Most children lived in communities where residents had a college education (61.8%). More than 30% of children resided in a community with an average median annual household income below $50,000/yr. A third of children lived in rural areas. Standard differences greater than 10 were slightly less than half of the variables suggesting that our exposure groups were modestly balanced.

Table 1.

Baseline Characteristics of Study Cohort (N=18,152)

Variables Total % or mean(SD) Concurrent Hospice Care Use % or mean(SD) Standard Hospice Care Use % or mean(SD)
n=6,243 n=11,909 Std Dif
Demographic Characteristics
 Age 7.4 (6.3) 7.8 (6.3) 7.4 (6.3) 6.1
 Female (%) 48.8 48.5 49.0 1.0
 Caucasian (%) 53.8 51.1 54.9 7.6
 Black (%) 27.2 23.9 28.1 9.6
 Other Race (%) 19.0 25.0 17.0 19.7
 Hispanic (%) 22.0 23.2 22.3 2.2
 Complex Chronic Condition (%) 48.4 56.4 46.6 19.7
 Comorbities (%) 28.0 38.6 25.1 29.2
 Mental/Behavioral Conditions (%) 34.1 41.7 32.1 20.0
 Technology Dependence (%) 22.9 31.9 20.5 26.2
Hospice Characteristics
 >50 employees (%) 35.6 38.7 34.7 8.4
 Non-profit/Government (%) 38.7 38.7 39.9 0.4
 Years of Operation 18.3 (9.8) 18.6 (9.6) 18.2 (10.0) 4.1
 Pediatric Program (%) 33.7 37.2 31.6 12.0
Community Characteristics
 No High School Education (%) 14.0 14.5 13.9 1.7
 High School Education (%) 24.2 24.9 24.0 2.3
 College Education (%) 61.8 60.6 62.1 3.1
 ≤ $50,000/yr Household Income (%) 35.6 40.4 33.0 15.5
 Northeast 57.6 49.4 61.8 25.3
 Midwest (%) 24.3 22.9 23.8 2.2
 South (%) 11.8 17.8 <10.0 26.9
 West (%) <10.0 10.0 <10.0 16.5
 Rural (%) 33.4 38.7 31.7 14.7
Open in a new tab

Note: Std Dif, standard difference; <10 = Unable to report less than 10% per Medicaid Data Use Agreement

The frequencies of the care continuity outcomes are listed in Table 2. In the overall sample, the average length of stay in hospice was 57 days. Children disenrolled from hospice on average 20 times. Transitions to the ER were 16.5% and inpatient care were 10% for all children. The average hospice length of stay varied from 88.5 days for children in concurrent care versus 48.8 days for those in standard hospice. The mean hospice disenrollment was similar for children in concurrent (19.5) and standard hospice care (20.7). Nineteen percent of concurrent care children transitioned to the ER versus 13.4% for standard hospice, and 11% of children in concurrent care transitioned to inpatient care versus 10% of standard hospice children.

Table 2.

Frequency of Care Continuity Outcomes (N=18,152)

Variables Total % or mean(SD) Concurrent Hospice Care Use % or mean(SD) Standard Hospice Care Use % or mean(SD)
 Hospice Length of Stay (days) 57.0 (72.5) 88.5 (109.8) 48.8 (58.7)
 Hospice Disenrollment (episodes) 20.4 (6.70) 19.5 (10.7) 20.7 (7.5)
 Emergency Room Transition (%) 16.5 19.5 13.4
 Inpatient Transition (%) 10.0 10.9 10.0
Open in a new tab

Note: SD, standard deviation

The correlation results between the instrumental variable and the exposure are displayed in Table 3. The first-stage equation found that the instrumental variable was highly correlated with the exposure. Medical team size was significantly related to concurrent hospice care (β = 0.07, p< 0.001). In addition, our test of the instrumental variable strength, which was run simultaneously with the first-stage equation, suggested that medical team size was a strong instrument with a statistically significant F-statistic > 10 and partial R2 < 0.05. These results suggest that medical team size was a valid instrumental variable.

Table 3.

Assessment of Instrumental Variable – First Stage Equation (N=18,152)

Variables Concurrent Hospice Care Use
β (95%CI)
Instrument
 Medical Team Size 0.07 (0.064,0.076) ***
Demographic Characteristics
 Age 0.01 (0.001,0.003) ***
 Female −0.01 (−0.024,0.006)
 Caucasian 0.01 (−0.019,0.023)
 Black −0.01 (−0.029,0.011)
 Hispanic 0.01 (−0.007,0.034)
 Complex Chronic Condition −0.05 (−0.065, −0.034) ***
 Comorbities 0.02 (−0.016,0.051)
 Mental/Behavioral Conditions 0.01 (−0.018,0.020)
 Technology Dependence 0.01 (−0.033, 0.044)
Hospice Characteristics
 >50 employees −0.01 (−0.037,0.008)
 Non-profit/Government 0.05 (0.026,0.069) ***
 Years of Operation 0.01 (−0.001,0.002)
 Pediatric Program 0.01 (−0.011,0.034)
Community Characteristics
 No High School Education −0.01 (−0.004,0.001)
 High School Education −0.01 (−0.002,0.001)
 ≤ $50,000/yr Household Income 0.06 (0.039,0.083) ***
 Midwest 0.07 (0.035,0.115) ***
 South 0.12 (0.080,0.160) ***
 West 0.08 (0.033,0.133 ***
 Rural 0.01 (−0.012,0.038)
F-statistic >10
Partial R2 0.0427
Open in a new tab
*

p< 0.05,

**

p <0.01,

***

p< 0.001

Note: CI, Confidence Intervals

Note: All analyses controlled for study years

The results of the analysis estimating the effect concurrent care use had on the end-of-life continuity of care outcomes for children at end of life are listed in Table 4. In the second-stage equation, concurrent hospice care had an effect on care continuity. Children who used concurrent care had significantly longer stays in hospice care (β = 2.76, p < 0.001), compared to their peers in standard hospice care. Concurrent care was negatively associated with hospice disenrollment. Compared to standard hospice care, the probability of a disenrollment was reduced by a factor of three for children in concurrent hospice care (β = −2.80, p < 0.05). Transitions were also affected by concurrent care use. Children in concurrent care had twice the odds of using the emergency room during hospice care (β = 2.09, p < 0.001) or transition out of hospice for inpatient care (β = 1.34, p < 0.001).

Table 4.

Instrumental Variable Analysis– Second Stage Equation (N=18,152)

Variables Hospice Length of Stay Hospice Disenrollment Emergency Room Transition Inpatient Transition
β (95%CI) β (95%CI) β (95%CI) β (95%CI)
Exposure
 Concurrent Hospice Care Use 2.76 (2.480,3.050)*** −2.80 (−3.091,−2.518)*** 2.09 (2.023,2.163)*** 1.34 (1.221,1.464)***
Demographic Characteristics
 Age −0.01 (−0.017,−0.001)* 0.01 (−0.002,0.013) −0.01 (−0.007,−0.003)* −0.01 (−0.010,−0.001)*
 Female 0.01 (−0.097,0.126) −0.03 (−0.146,0.081) 0.04 (−0.004,0.085) 0.02 (−0.036, 0.074)
 Caucasian −0.27 (−0.437,−0.108)*** −0.24 (−0.411,−0.72)** 0.03 (−0.028,0.092) −0.01 (−0.095,0.066)
 Black −0.14 (−0.317,0.027) −0.14 (−0.304,0.015) 0.15 (0.086,0.211)*** 0.11 (0.021,0.196)*
 Hispanic −0.04 (−0.185,0.112) 0.09 (−0.069,0.257) 0.03 (−0.030,0.093) 0.02 (−0.063,0.095)
 Complex Chronic Condition 0.073 (0.585,0.881)*** 0.21 (0.088,0.338)*** −0.04 (−0.102,0.019) 0.42 (0.349,0.492)***
 Comorbities 0.60 (0.410,0.798)*** 0.49 (0.228, 0.752)*** −0.22 (−0.315,−0.123)*** 0.23 (0.138,0.329)***
 Mental/Behavioral Conditions 0.43 (0.300,0.567)*** 0.45 (0.302,0.592)*** −0.06 (−0.109,−0.003)* 0.10 (0.037,0.154)***
 Technology Dependence 0.16 (−0.033,0.343) 0.19 (−0.097,0.483) 0.10 (−0.002,0.200) −0.01 (−0.099,0.092)
Hospice Characteristics
 >50 employees 0.11 (−0.045,0.261) −0.17 (−0.343,0.003) −0.16 (−0.229,−0.101)*** 0.08 (0.003,0.149)*
 Non-profit/Government −0.03 (−0.161,0.102) 0.26 (0.117,0.402)*** −0.26 (−0.327,−0.188)*** −0.13 (−0.212,−0.046)**
 Years of Operation 0.03 (0.018,0.034)*** 0.03 (0.017,0.037)*** 0.01 (−0.001,0.008) −0.01 (−0.011,−0.001)*
 Pediatric Program −0.01 (−0.213,0.212) 0.29 (0.098,0.476)** −0.03 (−0.088,0.033) 0.06 (−0.013,0.139)
Community Characteristics
 No High School Education 0.02 (0.002,0.038)* −0.01 (−0.027,0.007) −0.01 (−0.011,0.002) 0.01 (−0.003,0.011)
 High School Education 0.02 (0.006,0.028)** 0.01 (−0.004,0.014) 0.03 (0.024, 0.031)*** 0.01 (0.001,0.010)*
 ≤ $50,000/yr Household Income −0.57 (−0.728,−0.410)*** −0.19 (−0.349,−0.025) −0.23 (−0.294,−0.173)*** −0.06 (−0.134,0.011)
 Midwest −0.60 (−0.978,−0.222)** −0.74 (−1.100,0.391)*** 0.30 (0.190, 0.411)*** 0.14 (0.013,0.278)*
 South 0.34 (−0.010,0.696) −0.35 (−0.723,−0.029) 0.10 (−0.018,0.214) −0.02 (−0.158,0.117)
 West 1.13 (0.758,1.502)*** 1.39 (0.974,1.810)*** 0.17 (0.033,0.306)** −0.04 (−0.186,0.100)
 Rural −0.21 (−0.394,−0.017) −0.05 (−0.235,0.137) −0.33 (−0.402,−0.267)*** −0.12 (−0.198,−0.041)**
Open in a new tab
*

p< 0.05,

**

p <0.01,

***

p< 0.001

Note: CI, Confidence Interval

Note: All analyses controlled for study years

Discussion

In this novel analysis, we used rigorous statistical methods and a conceptual framework to compare the effectiveness of concurrent hospice care versus standard hospice care use on care continuity outcomes for children. Our findings demonstrated that concurrent care was more effective at improving multiple outcomes. Compared to standard hospice care, concurrent care use was significantly related to increased hospice length of stay and reduced hospice disenrollment. However, it was not effective at reducing transitions to the emergency room or inpatient care during hospice enrollment.

The study results revealed that concurrent hospice care use was more effective than standard hospice care in increasing hospice length of stay. The average number of days in hospice doubled for children who used concurrent hospice care, compared to standard hospice care (89 days vs. 49 days). We expected that concurrent hospice care would lengthen the time a child was in hospice care. It is possible that the medical treatments children receive during extended stays in concurrent hospice care for their terminal illness may extend their life expectancy. In fact, concurrent care is often referred to as life-prolonging care or disease-directed care. For the children with cancer, who are able to continue their chemotherapy and/or radiation under concurrent care, these treatments might slow down metastasis or reduce the physical burden of the tumor on the body. Children who use concurrent care might have longer hospice stays because their care is life-prolonging and their prognosis is extended. Hospice length of stay represents the time available to provide appropriate and needed services near the end-of-life, and a longer length of stay offers children and their families a greater period of comfort and closure at the end of life. Clearly, additional research is needed to understand the influence of concurrent-hospice-care lengthened stays on quality of care at end of life.

Concurrent hospice care utilization was more effective at reducing hospice disenrollment for children. Overall, we found children experienced on average 20 occurrences where they disenrolled from hospice and reenrolled again at a later date, which was consistent with pediatric hospice literature.16 Among the children in the study, however, those who used concurrent hospice care had statistically significant lower frequency of disenrollment (19.5 versus 20.7). Concurrent care might provide more effective continuity of care through continued communication and bonds with the medical team.17 Communication between family and the health care provider about the child’s health condition is a critical determinant of parent satisfaction and their perception “of being a good parent”.41,42 These communication channels include regular, clear, and accurate information about the child’s health condition.43 Maintaining open communication with the treating physician and medical team matters to families. A recent study reported greater parental satisfaction with hospice care when the primary oncologist provides appropriate and understandable information about the child’s medical condition.32 Thus, concurrent hospice care might reduce the fragmented care of cycling in and out of hospice care for children and their families.

The combination of medical treatment and hospice care under concurrent care was not more effective at reducing transitions to the emergency room or inpatient care. One possible reason for this finding might be the difference in care coordination between concurrent and standard hospice care. It is common for standard hospice care to include care coordination of any non-hospice care, usually by a hospice nurse. If a hospice patient needs to seek care in the emergency room or be admitted to the hospital for acute care, the hospice will typically be notified before care is sought. The hospice team can respond by authorizing care, sending a nurse to the home, or changing the patient to hospice inpatient hospice care. However, children who use concurrent care may be directed to the emergency room or inpatient care by their medical team. If a child is experiencing uncontrollable epistaxis (nosebleed) because of their chemotherapy treatments, for example, the oncologist or otolaryngologist might admit the child to the hospital.12 Thus, emergency room and inpatient use may represent the blended care of comfort and curative among children receiving concurrent hospice care. Additional research is needed to understand the episodes of care under which children are transitioning during hospice care and who is authorizing the care. This might include exploring the number of emergency room and inpatient procedures, along with the costs of these transitions. More information on the care coordination process during concurrent hospice care is also warranted.

This study had several notable limitations. First, our results may not generalize to pediatric non-Medicaid populations. The findings may differ among children with private, commercial insurance. However, Medicaid is the largest insurer of children in the US, especially among children with terminal illnesses.44 Second, this study was based on data between 2011 and 2013. Family and provider views, along with cultural norms about concurrent hospice care, may have shifted since the timeframe of this study. Our study provided critical information on the initial years of ACA, Section 2302, and yet, future research is clearly warranted that examines children in current health care environments. Finally, the Medicaid data set contained no information on family perspective of concurrent hospice care. Although this information is important because the family makes the decision to use pediatric concurrent care or not, the results of this study might provide important insight to assist families in making those decisions and guide future studies that integrate family data.

Despite the study limitations, there are key policy implications from our findings. ACA, Section 2302 eliminated the hospice eligibility criteria that pediatric patients stop medical treatment for their terminal illness upon enrollment into hospice. However, Section 2302 did not eliminate the requirement that patients entering hospice care have a 6-month or less to live prognosis. A portion of our sample was enrolled in hospice care for 6 months or longer. Hospice patients who live longer than 6 months are required to recertify with a physician that they are still eligible for hospice. Consequently, these recertifications are frequent targets for Medicaid audits because the child has gone over the 6 months. Thus, if children who use concurrent hospice care have the potential of staying in hospice care longer than 6 months, policy-makers and key stakeholders might examine modifications at the state or federal level to the ACA, Section 2302 language that removes the 6 month to live prognosis requirement for hospice.

In summary, we used multiple years of administrative claims data to conduct an analysis comparing the effectiveness of concurrent hospice care versus standard hospice care use on care continuity outcomes for children. We found that concurrent hospice care was more effective at improving multiple outcomes. Compared to standard hospice care, concurrent care use was significantly related to increased hospice length of stay and reduced hospice disenrollment. It was not effective at reducing transitions to the emergency room or inpatient care during hospice enrollment. These findings have important policy implications.

Acknowledgements:

Special thanks to Ms. Jamie Butler for her assistance in preparing the manuscript.

Funding Source:

This publication was made possible by Grant Number R01NR017848 from the National Institute of Nursing Research (PI: Lindley). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institute of Nursing Research or National Institutes of Health.

Contributor Information

Lisa C. Lindley, College of Nursing, University of Tennessee, Knoxville, Knoxville, Tennessee 37996.

Melanie J. Cozad, Department of Health Services Policy and Management, University of South Carolina, Columbia, SC 29201.

Jennifer W. Mack, Department of Pediatric Oncology and Division of Population Sciences, Dana-Farber Cancer Institute, Boston Children’s Hospital, Boston, MA 02214.

Jessica Keim-Malpass, School of Nursing, University of Virginia, Charlottesville, Virginia 22908.

Radion Svynarenko, College of Nursing, University of Tennessee, Knoxville, Knoxville, Tennessee 37996.

Pamela S. Hinds, The William and Joanne Conway Chair in Nursing Research, Executive Director, Department of Nursing Science, Professional Practice, and Quality Outcomes, Research Integrity Officer, Children’s National Hospital, Washington, D.C. 20010, Professor, Department of Pediatrics, The George Washington University.

References

  • 1.Keim-Malpass J, Cozad MJ, Svynarenko R, Mack JW, Lindley LC. The medical complexity of children receiving pediatric concurrent hospice care. J Spec Pediatr Nurs. (2021). Epub. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lindley LC, Cohrs AC, Keim-Malpass J, Leslie DL. Children enrolled in hospice care under commercial insurance: A comparison of different age groups. Am J Hosp Palliat Care. 2019;36(2):123–129. [DOI] [PubMed] [Google Scholar]
  • 3.Murphy SL, Mathews TJ, Martin JA, Minkovitz CS, Strobino DM. Annual summary of vital statistics: 2013–2014. Pediatrics. 2017;139(6). doi: 10.1542/peds.2016-3239. [DOI] [PubMed] [Google Scholar]
  • 4.Lindley LC, Lyon ME. A profile of children with complex chronic conditions at end of life among Medicaid-beneficiaries: Implications for healthcare reform. J Palliat Med. 2013;16(11):1388–1393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Lindley LC, Mack JW, Bruce DJ. Clusters of multiple complex chronic conditions: A latent class analysis of children at end of life. J Pain Symptom Manage. 2016;51(5):868–874. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Boyden JY, Curley MAQ, Deatrick JA, Ersek M. Factors associated with the use of U.S. community–based palliative care for children with life-limiting or life-threatening illnesses and their families: An integrative review. J Pain Symptom Manage. 2018;55(1): 117–31. [DOI] [PubMed] [Google Scholar]
  • 7.Mooney-Doyle K, Lindley LC. Family and child characteristics associated with caregiver challenges for medically complex children. Fam Community Health. 2019;43(1):74–81. [DOI] [PubMed] [Google Scholar]
  • 8.U.S. Department of Health & Human Services. Letter to state Medicaid directors on hospice care for children in Medicaid and CHIP. 2010. http://www.scribd.com/doc/37423817/HHS-Letter-to-State-Medicaid-Directors-on-Hospice-Care-for-Children-in-Medicaid-and-CHIP. Accessed September 20, 2010.
  • 9.Keim-Malpass J, Lindley LC. Repeal of the Affordable Care Act will negatively impact children at end of life. Pediatrics. 2017;140(3):e20171134. [DOI] [PubMed] [Google Scholar]
  • 10.Laird JM, Keim-Malpass J, Mack JW, Cozad MJ, Lindley LC. Examining variation in state Medicaid implementation of ACA: The case of concurrent care for children. Health Aff. 2020;39(10):1770–1775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Miller EG, Laragione G, Kang TI, Feudtner C. Concurrent care for the medically complex child: Lessons of implementation. J Palliat Med. 2012;15:1281–3. [DOI] [PubMed] [Google Scholar]
  • 12.Lindley LC, Morvant A. How to hospitalize a child receiving concurrent hospice care. NHPCO Pediatric e-Journal. 2020;60:61–64. [Google Scholar]
  • 13.Lindley LC, Richar CS, Hoit T, Steinhorn DM. The cost of pediatric concurrent hospice care: An economic analysis of relevant cost components, review of the literature, and case illustration. J Palliat Med. 2021;epub. [DOI] [PubMed] [Google Scholar]
  • 14.D’Angelo D, Mastroianni C, Hammer JM, et al. Continuity of care during end of life: An evolutionary concept analysis. Intl J Nurs Knowled. 2014;26(2):80–89. [DOI] [PubMed] [Google Scholar]
  • 15.Lorenz K, Ettner S, Rosenfeld K, Carlisle D, Leake B, Asch S. Cash and compassion: Profit status and the delivery of hospice services. J Palliat Med. 2002;5(4),507–514. [DOI] [PubMed] [Google Scholar]
  • 16.Lindley LC, Keim-Malpass J. Quality of pediatric hospice care for children with and without multiple complex chronic conditions. Int J Palliat Nurs. 2017;23(5):112–119. [DOI] [PubMed] [Google Scholar]
  • 17.Lindley LC. Health care reform and concurrent curative care for terminally ill children: A policy analysis. J Hosp Palliat Nurs. 2011;13(2):81–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Donabedian A The quality of care: How can it be assessed? JAMA. 1988;260(12): 1743–8. [DOI] [PubMed] [Google Scholar]
  • 19.Wang SY, Aldridge MD, Gross CP, et al. Transitions between healthcare settings of hospice enrollees at end of life. JAGS. 2016;64:314–322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Mor V, Wagner TH, Levy C, et al. Association of expanded VA hospice care with aggressive care and cost for veterans with advanced lung cancer. JAMA Oncol. 2019;5(6):810–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Wang SY, Dang W, Aldridge MD, Canavan M, Cherlin E, Bradley E. Associations of hospice disenrollment and hospitalization with continuous home care provision. Med Care. 2017;55:848–855. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Wu S, Volker DL. Live discharge from hospice: A systematic review. J Hosp Palliat Nurs. 2019;21(6):482–488. [DOI] [PubMed] [Google Scholar]
  • 23.Ruttner L, Borck R, Nysenbaum J, Williams S. Guide to MAX data. Medicaid Policy Brief #21. Mathematica Policy Research; 2015. [Google Scholar]
  • 24.Mor V, Joyce NR, Cote DL, et al. The rise of concurrent care for veterans with advanced cancer at end of life. Cancer. 2016;22(5):782–90. [DOI] [PubMed] [Google Scholar]
  • 25.Teno JM, Gozalo P, Trivedi AN, , et al. Site of death, place of care, and health care transitions among US Medicare beneficiaries, 2000–2015. JAMA. 2018;320(3):264–271. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Nageswaren S, Radulovic A, Anania A. Transitions to and from the acute inpatient care setting for children with life-threatening illness. Pediatr Clin N Am. 2014;61:761–783. [DOI] [PubMed] [Google Scholar]
  • 27.Penrod JD, Goldstein NE, Deb P. When and how to use instrumental variables in palliative care research. J Palliat Med. 2009;12(5):471–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Teno JM. Measuring end-of-life care outcomes retrospectively. J Palliat Med. 2005;8(suppl.1):s-42–s-49. [DOI] [PubMed] [Google Scholar]
  • 29.Bagiella E, Karamlou T, Chang H, Spivack J. Instrumental variable methods in clinical research. J Thorac Cardiovasc Surg. 2015;150(4):779–782. [DOI] [PubMed] [Google Scholar]
  • 30.Lousdal ML. An introduction to instrumental variable assumptions, validation and estimation. Emerg Themes Epidemiol. 2018;15:1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Ertefaie A, Small DS, Flory JH, Hennessy S. A tutorial on the use of instrumental variables in pharmacoepidemiology. Pharmacoepidemiol Drug Saf. 2017;26:357–67. [DOI] [PubMed] [Google Scholar]
  • 32.Zimmerman K, Bergstraesser E, Engberg S, et al. When parents face the death of their child: A nationwide cross-sectional survey of parental perspectives on their child’s end-of-life care. BMC Palliat Care. 2016;epub. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kennedy EH, Lorch S, Small DS. Robust causal inference with continuous instruments using the local instrumental variable curve. J R Stat Soc Series B Stat Methodol. 2019;81:121–143. [Google Scholar]
  • 34.Brookhart MA, Rassen JA, Schneeweiss S. Instrumental variable methods in comparative safety and effectiveness research. Pharmacoepidemiol Drug Saf. 2010;19(6):537–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Xian Y, Holloway RG, Chan PS, et al. Association between stroke center hospitalization for acute ischemic stroke and mortality. JAMA. 2011;305(4):373–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Baum CF, Schaffer ME, Stillman S. Enhanced routines for instrumental variables/generalized method of moments estimation and testing. Stata J. 2007;7(4):465–506. [Google Scholar]
  • 37.Staiger D, Stock JH. Instrumental variables regression with weak instruments. Econometrica. 1997;65(2):557–586. [Google Scholar]
  • 38.Uddin MJ, Groenwold RH, Ton de Boer, Belitser SV, Roes KC, Klungel OH. Instrumental variable analysis in epidemiologic studies: An overview of the estimation methods. Pharm Anal Acta. 2015;6:353. [Google Scholar]
  • 39.StataCorp LLC. Stata statistical software: Release 15. College Station, TX: StataCorp LP. 2017. [Google Scholar]
  • 40.Lindley LC, Svynarenko R, Profant TL. Data infrastructure for sensitive data: Nursing’s role in the development of a secure research enclave. Comput Inform Nurs. 2020;38(9),427–430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Feudtner C Collaborative communication in pediatric palliative care: A foundation for problem-solving and decision-making. Pediatr Clin N Am. 2007;54:583–607. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.October TW, Fisher KR, Feudtner C, Hinds PS. The parental perspective: “being a good parent” when making decisions in the PICU. Pediatr Crit Care Med. 2014;15(4):291–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Mack JW, Wolfe J. Early integration of pediatric palliative care: For some children palliative care starts at diagnosis. Curr Opin Pediatr. 2006;18(1):10–4. [DOI] [PubMed] [Google Scholar]
  • 44.Kaiser Family Foundation. Distribution of Medicaid enrollees by enrollment group, FY2014. https://www.kff.org/medicaid/state-indicator/distribution-of-medicaid-enrollees-by-enrollment-group/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D. Accessed January 3, 2021.

RESOURCES