Abstract
Background:
Children, adolescents, and young adults with hematologic malignancies tend to receive high-intensity end-of-life care (HI-EOLC), but sociodemographic and hospital-based predictors of HI-EOLC remain unclear.
Methods:
We conducted a population-based retrospective cohort study using the Premier Healthcare Database. We identified individuals with hematologic malignancies, ages 0-39 years at death, who died between 2010-2017. HI-EOLC was defined as experiencing ≥2 of the following: cardiopulmonary resuscitation, intravenous chemotherapy, hemodialysis, mechanical ventilation, tracheostomy placement, or emergency department visit within the last 30 days of life; and death in the intensive care unit. Multivariable logistic regression models were constructed to identify patient sociodemographic and hospital characteristics associated with HI-EOLC.
Results:
Among 1,454 decedents, over half (55%) experienced HI-EOLC. In multivariable models, patients treated in medium (adjusted odds ratio [aOR]: 1.63; 95% confidence interval [CI]: 1.07, 2.50) or large hospitals (aOR: 2.21; 95% CI: 1.45, 3.39), insured by Medicaid (aOR: 1.40; 95% CI: 1.09, 2.06), or receiving cancer-directed treatment in the Northeast (aOR: 1.50; 95% CI: 1.05-2.15) were more likely to receive HI-EOLC.
Conclusion:
A majority of children, adolescents, and young adults with hematologic malignancies experienced HI-EOLC, and the likelihood of HI-EOLC was influenced by hospital size, type of insurance, and geographic region. Further research is needed to determine how to mitigate these risks.
Keywords: pediatric, hematologic malignancies, pediatric palliative care, end-of-life care, pediatric oncology
Precis:
Many children, adolescents, and young adults with hematologic malignancies experience intense end-of-life care across the United States. Predictors for more intense care may include: (1) receiving care in larger hospitals, (2) having Medicaid insurance, and (3) receiving care in the Northeast.
Introduction
Existing literature suggests that children, adolescents, and young adults (CAYAs) with advanced cancer in North America are likely to receive high intensity end-of-life care (HI-EOLC).1–4 Earle and colleagues developed a set of quality indicators to evaluate intensity of end-of-life care for adults with cancer.5 Indicators deemed feasible to assess at the population level included avoiding chemotherapy, multiple hospitalizations, or intensive care unit (ICU) 3admission near end of life.5 These indicators, primarily drawing upon claims data, have been used extensively to benchmark end-of-life care quality and develop quality improvement initiatives for adults.6,7 While relevant indicators have not been formally developed for children, extant indicators have been adapted to explore end-of-life care patterns among CAYAs with cancer.1,3,4,8,9
Studies in California and New York have shown that 59-75% of childhood cancer decedents experienced at least one indicator of HI-EOLC, most often ICU admissions and mechanical ventilation.3,8,10 Furthermore, individuals of all ages with hematologic malignancies may be particularly prone to receipt of HI-EOLC.8,9,11–13 One hypothesized factor that may lead to higher intensity care among individuals with hematologic malignancies is the potential use of interventions, such as transfusions to alleviate symptoms of anemia, that cannot be administered at home.14 Children with hematologic malignancies may be more likely to continue cancer-directed therapy in the last weeks of life compared to children with solid tumors,15 and less likely to receive palliative care.16 These data underscore a need to better characterize predictors of HI-EOLC in CAYAs with hematologic malignancies.
While population-level studies assessing EOLC in childhood cancer decedents have been conducted in Canada (Ontario),4 Taiwan,9 and South Korea,17 studies based in the US are limited in geographic representation.1–3,10 In addition, few studies have focused specifically on CAYAs with hematologic malignancies.18 To fill these knowledge gaps, we assessed end-of-life care patterns and identified sociodemographic and hospital-based predictors of HI-EOLC in a national cohort of CAYAs with hematologic malignancies.
Methods
Study Design
We conducted a population-based retrospective cohort study using the Premier Healthcare Database (PHD). PHD is a hospital-based, service-level, all-payer administrative database that contains comprehensive healthcare utilization data from both inpatient and outpatient encounters across geographically diverse US hospitals.19 PHD represents approximately 25% of admissions among private and academic hospitals. This study was exempt from review by the Yale Human Research Protection Program.
Study Population
We identified patients with a hematologic malignancy who died between January 1, 2010 and March 31, 2017 and were ages 0-39 years at death (Figure 1). Hematologic malignancies were defined using the following codes from the International Classification of Diseases (ICD), Ninth and Tenth Revisions: Hodgkin lymphoma (ICD-9: 201.xx; ICD-10: C81.xx), non-Hodgkin lymphoma (ICD-9: 202.xx; ICD-10: C82.xx – C86.xx), myeloma (ICD-9: 203.xx; ICD-10: C90.xx), and leukemia (ICD-9: 204.xx – 208.xx, V10.60; ICD-10: C91.xx – C95.xx).20,21 Death was identified according to the following PHD codes: 20 (expired), 40 (expired at home for hospice care), 41 (expired in medical facility for hospice), and 42 (expired, place unknown, for hospice). We used the National Cancer Institute’s definition of adolescents and young adults to establish the upper age limit of patients included in the study cohort.22
Figure 1.
Study cohort: children, adolescents, and young adults who died of a hematologic malignancy between 2010-2017 at ages 0-39 years.
Definition of Outcome
We defined our main outcome of interest, HI-EOLC, as receipt of 2 or more of the following indicators: more than one emergency department (ED) visit, cardiopulmonary resuscitation (CPR), intravenous chemotherapy, hemodialysis, mechanical ventilation, or tracheostomy placement in the last 30 days of life; and death in the ICU.1
PHD provides the month and year of admission. Therefore, HI-EOLC indicators were tracked by capturing all encounters that fell within one month of the patient’s final admission recorded in PHD. While there were no precise codes designating ICU admission, we generated a comprehensive list of PHD charge master codes that suggested ICU admission at the time of death.23 ED visits were identified by determining encounters that were charged to the ED. ICD-9 and ICD-10 procedures codes and Current Procedural Terminology codes were used to identify receipt of CPR, hemodialysis, intubation/mechanical ventilation, tracheostomy placement, and chemotherapy (Supplemental Figure 1).24 We computed the composite number of HI-EOLC indicators received by each patient.
Predictors of Interest
We examined patient sociodemographic and hospital characteristics that might influence EOLC. Patient sociodemographic variables included age at death in years (0-18 versus 19-39), sex (male or female), race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, or other/unknown), and insurance type (Medicaid, managed care, or other [indemnity, self-pay, or Medicare]). Hospital variables included the hospital’s urbanity/rurality, whether it was a teaching hospital, number of beds (0-199=small, 200-499=medium, or ≥500=large), and geographic region (Northeast, Midwest, West, or South).1,3,4,8,9
Statistical Analysis
Categorical variables were characterized using frequencies and proportions. Continuous variables were presented as median and interquartile range [IQR]. Students t-test and Pearson’s Chi-square test were used to evaluate associations between continuous and categorical predictor variables, respectively, and receipt of HI-EOLC.
We conducted bivariate analyses to assess potential associations between patient- and hospital-level characteristics and receipt of HI-EOLC. This was followed by a multivariable logistic regression analysis with backward elimination to produce the final adjusted model. We further stratified analyses by age: children and adolescents (≤ 18 years) and young adults (19-39 years). All patient- and hospital-level characteristics listed above were included initially when building models. Significance tests were two-sided with an -level of 0.05. Analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC).
Results
We included 1,454 decedent patients in our study cohort, with median age of 29 (IQR 22-35) years at death. Patients were predominantly male (n=859, 59.1%) and diagnosed with leukemia (n=886, 60.9%) (Table 1). Median length of stay in the final hospital admission prior to death was 9 (IQR 2-24) days. Most patients received care at urban (n=1,386, 95.3%), teaching (n=882, 60.7%), and large (n=741, 50.9%) hospitals.
Table 1.
Patient and Hospital Characteristics.
| Characteristic | Overall n (%) |
Number of patients meeting high intensity end-of-life-care indicators | ||
|---|---|---|---|---|
| < 2 indicators, n (%) | ≥ 2 indicators, n (%) | p-value | ||
| Patient Characteristics | ||||
| Age (at death) | ||||
| 0 – 18 | 233 (16.0) | 106 (15.5) | 127 (16.5) | 0.59 |
| 19 – 39 | 1221 (84.0) | 579 (84.5) | 642 (83.5) | |
| Gender | ||||
| Male | 859 (59.1) | 396 (57.8) | 463 (60.2) | 0.35 |
| Female | 595 (40.9) | 289 (42.2) | 306 (39.8) | |
| Race/Ethnicity | ||||
| Non-Hispanic White | 652 (44.8) | 329 (48.0) | 323 (42.0) | 0.07 |
| Non-Hispanic Black | 274 (18.8) | 116 (16.9) | 158 (20.6) | |
| Hispanic | 220 (15.1) | 106 (15.5) | 114 (14.8) | |
| Other/unknown | 308 (21.2) | 134 (19.6) | 174 (22.6) | |
| Length of stay at last admission (in days) | ||||
| 0 – 2 | 406 (27.9) | 316 (46.1) | 90 (11.7) | <0.01 |
| 3 – 9 | 334 (23.0) | 229 (33.4) | 105 (13.7) | |
| 10 – 24 | 355 (24.4) | 82 (12.0) | 273 (35.5) | |
| ≥ 25 | 359 (24.7) | 58 (8.5) | 301 (39.1) | |
| Diagnosis | ||||
| Leukemia | 886 (60.9) | 420 (61.3) | 466 (60.6) | 0.96 |
| Hodgkin lymphoma | 132 (9.1) | 60 (8.8) | 72 (9.4) | |
| Non-Hodgkin lymphoma | 411 (28.3) | 194 (28.3) | 217 (28.2) | |
| Myeloma | 25 (1.7) | 11 (1.6) | 14 (1.8) | |
| Insurance Type | ||||
| Other (Indemnity, Self-Pay, Medicare) | 443 (30.5) | 232 (33.9) | 211 (27.4) | 0.01 |
| Medicaid | 591 (40.7) | 255 (37.2) | 336 (43.7) | |
| Managed Care | 420 (28.9) | 198 (28.9) | 222 (28.9) | |
| Hospital Characteristics | ||||
| Hospital vicinity to city center | ||||
| Urban | 1386 (95.3) | 648 (94.6) | 738 (96.0) | 0.22 |
| Rural | 68 (4.7) | 37 (5.4) | 31 (4.0) | |
| Teaching status | ||||
| Teaching | 882 (60.7) | 378 (55.2) | 504 (65.6) | <0.01 |
| Non-teaching | 572 (39.3) | 307 (44.8) | 265 (34.5) | |
| Hospital size | ||||
| Small, 0-199 beds | 106 (7.3) | 65 (9.5) | 41 (5.3) | <0.01 |
| Medium, 200 – 499 beds | 607 (41.8) | 310 (45.3) | 297 (38.6) | |
| Large, ≥ 500 beds | 741 (50.9) | 310 (45.3) | 431 (56.1) | |
| Geographic region | ||||
| West | 277 (10.1) | 137 (20.0) | 140 (18.2) | <0.01 |
| Northeast | 270 (18.6) | 97 (14.2) | 173 (22.5) | |
| Midwest | 141 (9.7) | 66 (9.6) | 75 (9.8) | |
| South | 766 (52.7) | 385 (56.2) | 381 (49.5) | |
Patients receiving ≥ 2 indicators of HI-EOLC had a longer length of stay at last admission (p<0.01) and were more likely to be insured by Medicaid (p<0.01), compared with patients receiving lower-intensity EOLC. Furthermore, patients receiving HI-EOLC were more likely to receive care at teaching hospitals (p<0.01), large hospitals (p<0.01), and hospitals in the Northeast (p<0.01) (Table 1).
EOLC Patterns
Overall, 79.6% (n=1,158) of patients experienced at least 1 indicator of HI-EOLC; 55.2% (n=803) experienced ≥ 2 indicators. The most common end-of-life interventions included mechanical ventilation (n=586, 40.3%), CPR (n=268, 18.4%), and intravenous chemotherapy (n=490, 33.7%). Young adults were less likely than children and adolescents to have multiple ED visits in the last 30 days of life (p<0.01) and to die in the ICU (p<0.01). (Table 2)
Table 2.
Prevalence of Indicators of High-Intensity End-of-Life Care.
| Indicator | Overall N (%) |
Age |
||
|---|---|---|---|---|
| 0-18 years N (%) |
19-39 years N (%) |
p-value | ||
| Within 30 days: | ||||
| CPR | 268 (18.4) | 42 (18.0) | 226 (18.5) | 0.86 |
| > 1 ER visit | 247 (17.0) | 52 (22.3) | 195 (16.0) | <0.01 |
| Hemodialysis | 182 (12.5) | 25 (10.7) | 157 (12.9) | 0.37 |
| Intubation/mechanical ventilation | 586 (40.3) | 95 (40.8) | 491 (40.2) | 0.59 |
| Tracheostomy | 47 (3.2) | 6 (2.6) | 41 (3.4) | 0.54 |
| Intravenous chemotherapy | 490 (33.7) | 87 (37.3) | 403 (33.0) | 0.19 |
| Death in ICU | 257 (17.7) | 54 (23.2) | 203 (16.6) | <0.01 |
Abbreviations: cardiopulmonary resuscitation (CPR), emergency room (ER), intensive care unit (ICU)
Predictors of HI-EOLC
In multivariable logistic regression analysis (Table 3), patients insured by Medicaid were more likely to experience HI-EOLC compared with those insured through other mechanisms (adjusted odds ratio [aOR] 1.40, 95% confidence interval [CI] 1.09, 2.06; p<.01). Patients treated at medium (aOR 1.63, 95% CI 1.07, 2.50; p=0.02) or large (aOR 2.21, 95% CI 1.45, 3.39; p<0.01) hospitals had higher odds of HI-EOLC than their counterparts treated in small hospitals. Patients treated in the Northeast also had higher odds of HI-EOLC compared with those treated in the West (aOR 1.50, 95% CI 1.05, 2.15; p=0.03). When stratifying by age at death, findings were similar to what was observed in the overall cohort (Table 3).
Table 3.
Associations Between Patient Characteristics and Receipt of Two or More Indicators of High-Intensity End-of-Life Care, Unadjusted and Adjusted (Stratified by Age Group).
| Characteristic | Unadjusted Analyses | Adjusted Analyses |
||||
|---|---|---|---|---|---|---|
| Overall | Ages 19-39 years* | |||||
|
| ||||||
| OR (95% CI) |
P | aOR (95% CI) |
P | aOR (95% CI) |
P | |
|
| ||||||
| Age (at death) | ||||||
| 0 – 18 | 1.00 | -- | -- | -- | -- | |
| 19 – 39 | 0.93 (0.67, 1.23) | 0.59 | -- | -- | -- | -- |
| Gender | ||||||
| Male | 1.00 | -- | -- | -- | -- | |
| Female | 0.91 (0.73, 1.12) | 0.35 | -- | -- | -- | -- |
| Race/Ethnicity | ||||||
| Non-Hispanic White | 1.00 | -- | -- | -- | -- | |
| Non-Hispanic Black | 1.39 (1.04, 1.84) | 0.02 | -- | -- | -- | -- |
| Hispanic | 1.10 (0.81, 1.49) | 0.45 | -- | -- | -- | -- |
| Other/unknown | 1.32 (1.01, 1.74) | 0.04 | -- | -- | -- | -- |
| Diagnosis | ||||||
| Leukemia | 1.00 | -- | -- | -- | -- | |
| Hodgkin Lymphoma | 1.08 (0.75, 1.56) | 0.68 | -- | -- | -- | -- |
| Non-Hodgkin Lymphoma | 1.01 (0.80, 1.27) | 0.95 | -- | -- | -- | -- |
| Myeloma | 1.15 (0.52, 2.55) | 0.74 | -- | -- | -- | -- |
| Insurance Type | ||||||
| Other (Indemnity, Self-Pay, Medicare) | 1.00 | 1.00 | 1.00 | |||
| Medicaid | 1.45 (1.13, 1.86) | <0.01 | 1.40 (1.09, 2.06) | <0.01 | 1.34 (1.02, 1.76) | 0.03 |
| Managed Care | 1.23 (0.94, 1.61) | 0.13 | 1.12 (0.72, 1.75) | 0.62 | 1.12 (0.83, 1.49) | 0.47 |
| Hospital vicinity to city center | ||||||
| Urban | 1.00 | -- | -- | -- | -- | |
| Rural | 0.74 (0.45, 1.20) | 0.22 | -- | -- | -- | -- |
| Teaching status | ||||||
| Teaching | 1.00 | -- | -- | -- | -- | |
| Non-teaching | 0.65 (0.52, 0.80) | <0.01 | -- | -- | -- | -- |
| Hospital size | ||||||
| Small, 0-199 beds | 1.00 | 1.00 | 1.00 | |||
| Medium, 200 – 499 beds | 1.52 (1.00, 2.32) | 0.05 | 1.63 (1.07, 2.50) | 0.02 | 1.76 (1.08, 2.88) | 0.02 |
| Large, ≥ 500 beds | 2.20 (1.45, 3.35) | <0.01 | 2.21 (1.45, 3.39) | <0.01 | 2.42 (1.48, 3.98) | <0.01 |
| Geographic region | ||||||
| West | 1.00 | 1.00 | 1.00 | |||
| Northeast | 1.75 (1.24, 2.46) | <0.01 | 1.50 (1.05, 2.15) | 0.03 | 1.40 (1.09, 2.06) | <0.01 |
| Midwest | 1.12 (0.74, 1.67) | 0.61 | 1.10 (0.73, 1.67) | 0.64 | 1.12 (0.72, 1.75) | 0.62 |
| South | 0.97 (0.64, 1.28) | 0.82 | 0.88 (0.66, 1.17) | 0.37 | 0.87 (0.64, 1.18) | 0.37 |
When stratifying by age group, there were no significant predictors associated with HI-EOLC among patients aged 0-18 years at death.
Discussion
In this population-based cohort study of CAYAs with hematologic malignancies who died over an 8-year period, we found that more than half experienced HI-EOLC. Over one-third of patients were mechanically ventilated, and nearly one-third received intravenous chemotherapy near the end of life. Young adults were less likely than children and adolescents to have multiple ED visits or to die in the ICU. Notably, patients who had Medicaid insurance, received treatment in the Northeast, and were treated at medium or large hospitals were more likely to experience HI-EOLC.
The proportion of patients receiving intravenous chemotherapy and mechanical ventilation near the end of life in our study is higher than previously reported in the US among CAYAs with cancer.1,3,10 Our study captured a national sample with a more expansive age range, thus reflecting national trends more commonly seen in adults with cancer.25 Furthermore, our cohort only included patients with hematologic malignancies, who may be less likely to receive palliative care compared to patients with solid tumors26 and often receive intermittent hospice services.27 Collectively, these factors may yield delays in advance care planning and ultimately drive higher use of intensive interventions in this population.25,28,29 Notably, we were unable to determine the intent of intravenous chemotherapy in PHD. As palliative chemotherapy may often be administered for individuals with hematologic malignancies,30 it is possible that chemotherapy was used to relieve symptoms, rather than with curative intent. Furthermore, our study also suggests that young adults may be less likely to have multiple ED visits or to die in the ICU compared with children and adolescents. While the exact drivers of differential ED and ICU use are unclear, this finding may reflect greater autonomy among young adults in decision-making regarding care preferences near the end of life.31 Age-related differences in EOLC preferences and resource use require further characterization in a subsequent study.
We additionally found that Medicaid insurance predicts receipt of HI-EOLC in this population. Prior studies similarly suggest that adolescents and young adults with cancer who are insured by Medicaid may be more likely to receive intensive end-of-life interventions and die in the hospital.1–3,8,10 Due to expansions in public insurance eligibility through the Affordable Care Act (ACA), Medicaid coverage among children with cancer increased during our study period.32 The ACA offers a range of cost sharing subsidies, consequently reducing out-of-pocket expenses for beneficiaries.33,34 Reduced cost sharing likely improves access to primary care and preventive health services,35,36 while also potentially increasing hospital resource use.37,38 Future research should further interrogate the relationship between Medicaid expansion in the ACA and intensive end-of-life care services for CAYAs with cancer.
Patients treated at medium or large hospitals were more likely to receive HI-EOLC compared with those treated in hospitals with fewer than 200 beds. In surgical oncology, institutions with higher surgical volumes tend to demonstrate more favorable outcomes.39 In medical oncology, the association between volume and outcome is less clear. Morden et al. previously reported that adult patients with cancer treated in larger volume hospitals were more likely to receive HI-EOLC compared with patients treated in smaller hospitals; in addition, factors such as cancer center designation or for-profit status may influence care intensity.40 Illness severity, and by extension, average severity level of any given hospital’s procedures (i.e., case mix), may also play a role.41 Since we did not adjust for case mix in this study, we are unable to draw conclusions regarding whether higher illness severity in patients treated at larger hospitals yielded HI-EOLC. Exploring the interplay of case mix with hospital characteristics may better elucidate the factors effecting HI-EOLC in larger hospitals.
We identified marked regional differences in receipt of HI-EOLC. A number of factors, including patient preferences, social determinants of health, physician practice patterns, and health system culture, may collectively influence regional variation in end-of-life care service receipt.42 Sullivan et al. identified differential hospice enrollment by region, with lower rates of hospice enrollment among adults in the Northeast.43 Reduced or abbreviated hospice enrollment bears the potential to heighten end-of-life care intensity, though overall, we identified low rates of hospice use in our cohort.44 A dearth of literature in the childhood cancer context serves to explain the observed regional variation, as rates of pediatric cancers are fairly uniform across the US.45 Therefore, this finding of regional variation calls for further assessment of patient, healthcare professional, and hospital factors that may impact region-specific trends.
A strength of this study is its use of contemporary, population-based data, by which we examined a number of sociodemographic and hospital-level characteristics as possible factors associated with intense end-of-life care. Our study also has the advantage of being the first to analyze a national cohort that evaluates an extensive age range of children, adolescent, and young adults with hematologic malignancies.
Among our key limitations is that PHD restricts our analyses to billing and coding data. Data for individual patient encounters did not have exact admission dates; rather, we were given month and year of admission. Therefore, we cannot guarantee that we comprehensively accounted for all encounters and interventions received in the final 30 days of life. We did not have data on the circumstances of death, so we were unable to ascertain whether deaths were cancer-related and/or anticipated. Because claims data are dependent on professional ICD coding, it is also possible that some diagnoses may have been missed, coding patterns may differ across hospitals, and that not all coding is accurate.46 Other important intensity indicators, such as hospice use and ICU admissions,5 were not available in sufficient detail in the PHD and could not be validated through chart review, which may limit our scope of understanding. Our chosen retrospective study design limits the ability to make causal inferences. Prospective studies may be more optimally poised to render causal inferences in this context. Furthermore, we were limited to patients who died in the hospital or outside the hospital while enrolled in hospice care. Although research shows that around 40% of children with cancer die at home,47 our database did not capture any decedents who died at home without hospice care. Indeed, only 1.1% of our cohort (n=16) died at home or at another facility for hospice care. Finally, young adults were heavily represented in our cohort compared with children and adolescents (84.0% versus 16.0%, respectively). While we did not identify significant predictors for receipt of HI-EOLC in children and adolescents aged 0-18 years at death, limited statistical power may have contributed to the null finding.
Conclusion
This study confirms that most CAYAs with hematologic malignancies receive multiple intensive interventions near the end of life. Insurance type, hospital size, and hospital region appear to significantly influence receipt of HI-EOLC. Mitigation strategies may include earlier integration of palliative and/or hospice care, when feasible, as well as policy change that enables interventions such as transfusions to occur in settings outside the hospital.25,48,49 Importantly, metrics used to gauge end-of-life care intensity in this study are largely extrapolated from studies in adults. Future studies should ascertain whether existing definitions of HI-EOLC5 are applicable to CAYAs.
Supplementary Material
Funding Support:
This work was supported by CTSA Grant UL1 TR001863 from the National Center for Advancing Translational Science (NCATS), a component of the National Institutes of Health (NIH), and NIH roadmap for Medical Research. Its contents are solely the responsibility of the authors and do not necessarily represent the official view of NIH.
Footnotes
Conflicts of Interest: XM received research funding (institutional) from Celgene/Bristol Myers Squibb (BMS) and consults for BMS. None of the activities are related to the present study.
References
- 1.Johnston EE, Alvarez E, Saynina O, Sanders L, Bhatia S, Chamberlain LJ. End-of-Life Intensity for Adolescents and Young Adults With Cancer: A Californian Population-Based Study That Shows Disparities. J Oncol Pract. 2017;13(9):e770–e781. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Johnston EE, Alvarez E, Saynina O, Sanders LM, Bhatia S, Chamberlain LJ. Inpatient utilization and disparities: The last year of life of adolescent and young adult oncology patients in California. Cancer. 2018;124(8):1819–1827. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mack JW, Chen K, Boscoe FP, et al. High Intensity of End-of-Life Care Among Adolescent and Young Adult Cancer Patients in the New York State Medicaid Program. Med Care. 2015;53(12):1018–1026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Kassam A, Sutradhar R, Widger K, et al. Predictors of and Trends in High-Intensity End-of-Life Care Among Children With Cancer: A Population-Based Study Using Health Services Data. Journal of Clinical Oncology. 2016;35(2):236–242. [DOI] [PubMed] [Google Scholar]
- 5.Earle CC, Park ER, Lai B, Weeks JC, Ayanian JZ, Block S. Identifying potential indicators of the quality of end-of-life cancer care from administrative data. J Clin Oncol. 2003;21(6):1133–1138. [DOI] [PubMed] [Google Scholar]
- 6.Emanuel EJ, Young-Xu Y, Levinsky NG, Gazelle G, Saynina O, Ash AS. Chemotherapy use among Medicare beneficiaries at the end of life. Ann Intern Med. 2003;138(8):639–643. [DOI] [PubMed] [Google Scholar]
- 7.Wright AA, Hatfield LA, Earle CC, Keating NL. End-of-life care for older patients with ovarian cancer is intensive despite high rates of hospice use. J Clin Oncol. 2014;32(31):3534–3539. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Johnston EE, Alvarez E, Saynina O, Sanders L, Bhatia S, Chamberlain LJ. Disparities in the Intensity of End-of-Life Care for Children With Cancer. Pediatrics. 2017;140(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tang ST, Wu S-C, Hung Y-N, Chen J-S, Huang E-W, Liu T-W. Determinants of Aggressive End-of-Life Care for Taiwanese Cancer Decedents, 2001 to 2006. Journal of Clinical Oncology. 2009;27(27):4613–4618. [DOI] [PubMed] [Google Scholar]
- 10.Mack JW, Chen LH, Cannavale K, Sattayapiwat O, Cooper RM, Chao CR. End-of-Life Care Intensity Among Adolescent and Young Adult Patients With Cancer in Kaiser Permanente Southern California. JAMA Oncol. 2015;1(5):592–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hui D, Didwaniya N, Vidal M, et al. Quality of end-of-life care in patients with hematologic malignancies: a retrospective cohort study. Cancer. 2014;120(10):1572–1578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ho TH, Barbera L, Saskin R, Lu H, Neville BA, Earle CC. Trends in the Aggressiveness of End-of-Life Cancer Care in the Universal Health Care System of Ontario, Canada. Journal of Clinical Oncology. 2011;29(12):1587–1591. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kirtane K, Downey L, Lee SJ, Curtis JR, Engelberg RA. Intensity of End-of-Life Care for Patients with Hematologic Malignancies and the Role of Race/Ethnicity. Journal of Palliative Medicine. 2018;21(10):1466–1471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Odejide OO, Salas Coronado DY, Watts CD, Wright AA, Abel GA. End-of-life care for blood cancers: a series of focus groups with hematologic oncologists. J Oncol Pract. 2014;10(6):e396–403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jalmsell L, Forslund M, Hansson MG, Henter JI, Kreicbergs U, Frost BM. Transition to noncurative end-of-life care in paediatric oncology--a nationwide follow-up in Sweden. Acta Paediatr. 2013;102(7):744–748. [DOI] [PubMed] [Google Scholar]
- 16.Kassam A, Skiadaresis J, Alexander S, Wolfe J. Differences in end-of-life communication for children with advanced cancer who were referred to a palliative care team. Pediatr Blood Cancer. 2015;62(8):1409–1413. [DOI] [PubMed] [Google Scholar]
- 17.Park JD, Kang HJ, Kim YA, et al. Trends in the aggressiveness of end-of-life care for Korean pediatric cancer patients who died in 2007-2010. PLoS One. 2014;9(6):e99888. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Johnston EE, Muffly L, Alvarez E, et al. End-of-Life Care Intensity in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation: A Population-Level Analysis. J Clin Oncol. 2018;36(30):3023–3030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Premier Applied Sciences. Premier Healthcare Database: Data That Informs and Performs. 2019. Available from: https://products.premierinc.com/downloads/PremierHealthcareDatabaseWhitepaper.pdf
- 20.Centers for Disease Control and Prevention. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). Available from: https://www.cdc.gov/nchs/icd/icd9cm.htm.
- 21.World Health Organization. ICD-10 Version: 2019. Available from: https://icd.who.int/browse10/2019/en. [Google Scholar]
- 22.Institute NC. Adolescents and Young Adults with Cancer. [Google Scholar]
- 23.Premier Healthcare Database: Data that Informs and Performs. Premier Applied Sciences;2018 [Google Scholar]
- 24.Barnato AE, Farrell MH, Chang CC, Lave JR, Roberts MS, Angus DC. Development and validation of hospital “end-of-life” treatment intensity measures. Med Care. 2009;47(10):1098–1105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wang R, Zeidan AM, Halene S, et al. Health Care Use by Older Adults With Acute Myeloid Leukemia at the End of Life. J Clin Oncol. 2017;35(30):3417–3424. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Howell DA, Shellens R, Roman E, Garry AC, Patmore R, Howard MR. Haematological malignancy: are patients appropriately referred for specialist palliative and hospice care? A systematic review and meta-analysis of published data. Palliat Med. 2011;25(6):630–641. [DOI] [PubMed] [Google Scholar]
- 27.Sexauer A, Cheng MJ, Knight L, Riley AW, King L, Smith TJ. Patterns of hospice use in patients dying from hematologic malignancies. J Palliat Med. 2014;17(2):195–199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Mannis GN, McNey LM, Gupta NK, Gross DM. The transfusion tether: Bridging the gap between end-stage hematologic malignancies and optimal end-of-life care. Am J Hematol. 2016;91(4):364–365. [DOI] [PubMed] [Google Scholar]
- 29.Ananth P, Melvin P, Feudtner C, Wolfe J, Berry JG. Hospital Use in the Last Year of Life for Children With Life-Threatening Complex Chronic Conditions. Pediatrics. 2015;136(5):938–946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.LeBlanc TW, O’Donnell JD, Crowley-Matoka M, et al. Perceptions of palliative care among hematologic malignancy specialists: a mixed-methods study. J Oncol Pract. 2015;11(2):e230–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Sansom-Daly UM, Wakefield CE, Patterson P, et al. End-of-Life Communication Needs for Adolescents and Young Adults with Cancer: Recommendations for Research and Practice. J Adolesc Young Adult Oncol. 2020;9(2):157–165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Barnes JM, Barker AR, King AA, Johnson KJ. Association of Medicaid Expansion With Insurance Coverage Among Children With Cancer. JAMA Pediatr. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Brooks GA, Hoverman JR, Colla CH. The Affordable Care Act and Cancer Care Delivery. Cancer J. 2017;23(3):163–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Glied SA, Collins SR, Lin S. Did The ACA Lower Americans’ Financial Barriers To Health Care? Health Aff (Millwood). 2020;39(3):379–386. [DOI] [PubMed] [Google Scholar]
- 35.Kirby JB, Davidoff AJ, Basu J. The ACA’s Zero Cost-Sharing Mandate and Trends in Out-of-Pocket Expenditures on Well-Child and Screening Mammography Visits. Med Care. 2016;54(12):1056–1062. [DOI] [PubMed] [Google Scholar]
- 36.Dixon RB, Hertelendy AJ. Interrelation of preventive care benefits and shared costs under the Affordable Care Act (ACA). Int J Health Policy Manag. 2014;3(3):145–148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Keim-Malpass J, Constantoulakis L, Letzkus LC. Variability In States’ Coverage Of Children With Medical Complexity Through Home And Community-Based Services Waivers. Health Affairs. 2019;38(9):1484–1490. [DOI] [PubMed] [Google Scholar]
- 38.Shigeoka H The Effect of Patient Cost Sharing on Utilization, Health, and Risk Protection. The American Economic Review. 2014;104(7):2152–2184. [Google Scholar]
- 39.Medicine Io, Council NR. Interpreting the Volume-Outcome Relationship in the Context of Cancer Care. Washington, DC: The National Academies Press; 2001. [PubMed] [Google Scholar]
- 40.Morden NE, Chang CH, Jacobson JO, et al. End-of-life care for Medicare beneficiaries with cancer is highly intensive overall and varies widely. Health Aff (Millwood). 2012;31(4):786–796. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Barnato AE, Cohen ED, Mistovich KA, Chang CC. Hospital end-of-life treatment intensity among cancer and non-cancer cohorts. J Pain Symptom Manage. 2015;49(3):521-529 e521–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Keating NL, Huskamp HA, Kouri E, et al. Factors Contributing To Geographic Variation In End-Of-Life Expenditures For Cancer Patients. Health Aff (Millwood). 2018;37(7):1136–1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sullivan DR, Ganzini L, Lapidus JA, et al. Improvements in hospice utilization among patients with advanced-stage lung cancer in an integrated health care system. Cancer. 2018;124(2):426–433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Wang SY, Aldridge MD, Gross CP, et al. End-of-Life Care Intensity and Hospice Use: A Regional-level Analysis. Med Care. 2016;54(7):672–678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Siegel D, Li J, Henley S, et al. Geographic Variation in Pediatric Cancer Incidence—United States, 2003–2014. 2018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Tyree PT, Lind BK, Lafferty WE. Challenges of using medical insurance claims data for utilization analysis. Am J Med Qual. 2006;21(4):269–275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Feudtner C, Feinstein JA, Satchell M, Zhao H, Kang TI. Shifting place of death among children with complex chronic conditions in the United States, 1989-2003. JAMA. 2007;297(24):2725–2732. [DOI] [PubMed] [Google Scholar]
- 48.Ranallo L Improving the Quality of End-of-Life Care in Pediatric Oncology Patients Through the Early Implementation of Palliative Care. J Pediatr Oncol Nurs. 2017;34(6):374–380. [DOI] [PubMed] [Google Scholar]
- 49.Snaman JM, Kaye EC, Lu JJ, Sykes A, Baker JN. Palliative Care Involvement Is Associated with Less Intensive End-of-Life Care in Adolescent and Young Adult Oncology Patients. J Palliat Med. 2017;20(5):509–516. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.