Introduction: We previously demonstrated that receiving all cancer treatment at specialized cancer centers (SCCs; versus community hospitals) through end of therapy is associated with better leukemia-specific survival in children (0-18 years) and young adults (YA; 19-39 years) diagnosed with ALL in California. While both age groups benefit from care at a SCC, we found only a minority of YAs (36%), compared with children (84%), received care at a SCC. Our prior findings highlight the need for these patients to be referred to and treated at SCCs, but there are potential cost differences associated with this recommendation that have not been previously examined.

Methods: Using the California Cancer Registry linked to the Office of Statewide Health and Planning and Development (OSHPD) statewide hospitalization database, we identified children and YAs (19-39 years) with first primary ALL who received inpatient treatment from 1995-2014 and had at least 3 years of follow-up. Patients were classified as receiving all or part/none of their treatment at a SCC (Children's Oncology Group or National Cancer Institute-designated cancer centers for children and National Cancer Institute-designated cancer centers for YAs) within 3 years of diagnosis to capture the full time of potential primary treatment. Total charges for each admission and hospital level financial information were used to calculate costs (adjusted for inflation to 2016 US dollars) for each admission, excluding peri-partum admissions and those associated with traumatic accidents. One large health maintenance organization system in California did not report charges to OSHPD and was excluded from the analysis (13% of the patient population). In addition, patients needed to have at least 80% of charge data to be included in the primary analysis, resulting in a study population of 5,167 ALL patients. We determined the number of inpatient days and cumulative inpatient costs within 3 years of diagnosis. Mean and median costs overall and per day by age group and location of care were compared using t-tests and Kruskal Wallace tests. We conducted sensitivity analyses 1) limiting our analyses to only patients with all charge data available (n=5,118) and 2) excluding patients with stem cell transplant (n=693) to determine the impact of these factors on study findings.

Results: The mean cost for children receiving all care at SCCs vs non-SCCs was $216,439 (median=$121,039) vs $191,082 (median=$84,529) (p mean = 0.008; p median = <0.001). The cost per day was higher at SCCs (mean=$2,840; median=$2,529) than non-SCCs (mean=$2,283; median=$1,865) (p mean < 0.001; p median < 0.001). In children, the mean number of inpatient days within 3 years of diagnosis was similar for those who did (n=70 days) and did not (72 days) receive all cancer care in SCCs (p=0.70). Among YAs, the mean cost for patients receiving all treatment at SCCs was $380,556 (median=$308,864) vs $346,706 (median=$241,847) at non-SCCs (p mean=0.02; p median < 0.001). The cost per day was higher at SCCs (mean=$3,730; median=$3,537) than non-SCCs (mean=$3,224; median=$2,917) (p mean < 0.001; p median < 0.001). YAs receiving all cancer care at SCCs (99 days) had a similar mean number of inpatient days to those receiving care at non-SCCs (101 days) (p=0.97). In the sensitivity analyses excluding patients receiving a transplant, the mean cost was lower at both SCCs and non-SCCs, but the differences in costs in children and YAs by location of care remained. In addition, results were similar when analyses were limited to patients with complete charge data.

Conclusion: In this large, population-based cohort of pediatric and YA patients with ALL, we found that inpatient costs and number of inpatient days were higher among YAs than children with ALL. In addition, in each age group, the costs of inpatient care during the full course of therapy for primary ALL was higher in patients receiving all of their care at SCCs compared patients receiving part or none of their care at an SCC. As inpatient costs do not reflect the total burden associated with cancer care, future studies should consider how location of care impacts outpatient, emergency department and out-of-pocket costs. Given findings of better outcomes among children and YAs receiving all care at SCCs, we believe the marginal increased cost should be considered in view of the better outcomes at SCCs.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution