Cost comparison by treatment arm and center-level variations in cost and inpatient days on the phase III high-risk B acute lymphoblastic leukemia trial AALL0232.

Amanda M Dinofia,Naomi J Winick,Stephen P Hunger,Yuan‐Shung V Huang,Richard Aplenc,Alix E Seif,William L Carroll,Matthew Hall,Viviane Cahen,Eric C Larsen,Meenakshi Devidas,Yimei Li,Brian T Fisher

doi:10.1002/cam4.1206

Abstract

The Children's Oncology Group (COG) develops and implements multi‐institutional clinical trials with the primary goal of assessing the efficacy and safety profile of treatment regimens for various pediatric cancers. However, the monetary costs of treatment regimens are not measured. AALL0232 was a COG randomized phase III trial for children with acute lymphoblastic leukemia that found that dexamethasone (DEX) was a more effective glucocorticoid than prednisone (PRED) in patients younger than 10 years, but PRED was equally effective and less toxic in older patients. In addition, high‐dose methotrexate (HD‐MTX) led to better survival than escalating doses of methotrexate (C‐MTX). Cost data from the Pediatric Health Information System database were merged with clinical data from the COG AALL0232 trial. Total and component costs were compared between treatment arms and across hospitals. Inpatient costs were higher in the HD‐MTX and DEX arms when compared to the C‐MTX and PRED arms at the end of therapy. There was no difference in cost between these arms at last follow‐up. Considerable variation in total costs existed across centers to deliver the same therapy that was driven by differences in inpatient days and pharmacy costs. The more effective regimens were found to be more expensive during therapy but were ultimately cost‐neutral in longer term follow‐up. The variations in cost across centers suggest an opportunity to standardize resource utilization for patients receiving similar therapies, which could translate into reduced healthcare expenditures.

Highlights

The adult healthcare community has embraced health economics as a necessary tool in maximizing the value of medical care, which is the optimal patient outcome at the lowest cost [1]
Of the 3,083 patients Children’s Oncology Group (COG) enrolled on AALL0232, 935 patients matched within the COG and Pediatric Health Information System (PHIS) databases and had inpatient data within the COG treatment period, which is 91% of the patients enrolled by COG and treated at PHIS institutions and 30% of the total patients enrolled by COG (Fig. 1)
An increased number of patients in this region likely receive oncology care in non-PHIS institutions compared to other United States (US) regions

Summary

Introduction

The adult healthcare community has embraced health economics as a necessary tool in maximizing the value of medical care, which is the optimal patient outcome at the lowest cost [1]. While the pediatric literature has been slower to adopt cost research, pediatric healthcare providers have increasingly become interested in the quality and costs of care delivery. Though pediatric research in this area lags significantly behind the adult literature, a growing body of economics research does exist in pediatric oncology. Few studies identifying pediatric cancer costs exist in bone marrow transplant, acute lymphoblastic leukemia (ALL), and rhabdomyosarcoma [2,3,4,5,6]. There are data, not in pediatric cancer, demonstrating variations in costs among hospitals delivering similar care [7,8,9]. No study has compared costs of care for patients randomized to different treatment arms of a large ALL clinical trial or compared costs of care for patients receiving the same on-study therapy but at different centers

Objectives

Methods

Results

Conclusion