Abstract
Background: The projected 5y survival rates for pediatric hematologic malignancies exceed 85% (SEER statistics, 2017), in large part due to risk-stratified intensive multi-agent therapeutic approaches. However, these regimens result in chemotoxicity, often requiring hospitalization. However, the burden of chemotoxicity-related hospitalizations in children with hematologic malignancies remains understudied.
Methods: Using an administrative claims database (Truven Marketscan ®), we describe chemotoxicity-related hospitalizations within the first 12 mo from first claim of chemotherapy in children with hematologic malignancies who were <21y at diagnosis. Eligibility included (i) incident acute lymphoblastic leukemia (ALL; ICD-10-CM code: C91), acute myeloid leukemia (AML; C92), Hodgkin lymphoma (HL; C81) or non-Hodgkin lymphoma (NHL; C83.0, C83.3, C83.5, C83.7, C84.4, C85) diagnosed between 2011 and 2018; and (ii) continued insurance coverage 30d prior to and 365d after cancer diagnosis. Chemotoxicities (identified using ICD-9 and -10 codes) were grouped into organ systems (hematologic, infectious, gastrointestinal, renal, allergic, pulmonary, central nervous system, cardiovascular and miscellaneous). A hospitalization was considered chemotoxicity-related if a toxicity diagnosis was the primary reason for admission (i.e., first billing code) or occurred in the second billing position only if the cancer diagnosis was in the primary billing position. Hospitalizations for chemotherapy administration for primary or relapsed cancer were excluded. Logistic regression was used to examine the following factors for their association with chemotoxicity-related hospitalization: age at cancer diagnosis, sex, year of diagnosis (2011-2014; 2015-2018), insurance (commercial, Medicaid) and primary cancer diagnosis.
Results: We identified 897 eligible patients (ALL: n=461, AML: n=79, HL: n=202, NHL: n=155). Median age at diagnosis was 15y (range, 0-21), 52.7% were male, 54.2% were diagnosed between 2011 and 2014, and 70.2% had commercial insurance. Medicaid patients were younger than those with commercial insurance (median age at diagnosis: 12y [range, 0.9-21] vs. 17y [0-21y]), and were less likely to carry a diagnosis of ALL (46% vs. 54%, P<0.001) but more likely to have AML (15% vs. 6%, P<0.001). Over the 4,736 person-months of follow-up, 360 patients (40.1%) had 636 chemotoxicity-related hospitalizations (ALL: 56.9%, AML: 8.9%, HL: 20.9%, NHL: 13.2%); 164 (18.3%) patients had ≥1 chemotoxicity-related hospitalization. Median time to first chemotoxicity-related hospitalization from start of therapy was 37d (interquartile range, 12-78). Chemotoxicity claims included hematologic toxicities (63.6%), infections (22.7%) and gastrointestinal toxicities (6.1%); the figure shows the distribution of claims during these hospitalizations by hematologic malignancy. The average length of stay (LOS) for chemotoxicity-related hospitalization was 6.8±8.7d [5.8±7.8d (HL) to 8.6±12.6d (AML)]. Multivariable logistic regression analysis identified Medicaid insurance (vs. commercial) to be associated with lower odds of chemotoxicity-related hospitalization (odds ratio=0.68, 95% confidence interval 0.56-0.84, P<0.001).
Conclusions: Over 40% of children with hematologic malignancies require chemotoxicity-related hospitalizations during the first year of treatment. These data could be used to provide guidance to patients and their families and inform healthcare policy decisions.
No relevant conflicts of interest to declare.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal