Health Care Reimbursement and Service Utilization Among Medicare Beneficiaries with Multiple Myeloma Receiving Autologous Hematopoietic Cell Transplantation in Inpatient and Outpatient Settings

Introduction: Inpatient services are the leading drivers of cost for autologous hematopoietic cell transplantation (HCT), and the number of Medicare beneficiaries who receive autologous HCT is increasing. Using a merged dataset of Center for International Blood and Marrow Transplant Research (CIBMTR) transplant and outcomes data and Centers for Medicare and Medicaid Services (CMS) Medicare administrative claims data, we examined reimbursement and service utilization among Medicare beneficiaries with multiple myeloma (MM) who received IP and OP autologous HCT.

Methods: This was a multicenter retrospective cohort study. A total of 11,358 HCT recipients from 2010-2012 were identified in the CMS Medicare database; 9,055 (80%) were linked with CIBMTR data. Selection criteria included first HCT for MM, diagnosis-to-HCT time between 0 and 18 months, and continuous enrollment for 30 days prior to index and 100 days post-HCT or until death. For IP-HCT, the index period for reimbursement and service utilization was day of admission for HCT through discharge date. For OP-HCT, the index period was day -2 through HCT date to capture the conditioning regimen. Total IP and OP service days from 30 days prior to index and 100 days post-HCT, and subsequent admissions post the HCT index period were calculated. Total reimbursement consisted of all payments made to providers (Medicare payments for Part A & B services, secondary payer, and patient responsibility of deductibles, coinsurance, and copayments), which was adjusted by a weighted generalized linear model (GLM). Patient responsibility was assessed separately and adjusted by the same GLM. Kaplan-Meier method was used for overall survival (OS) analysis; potential factors associated with OS were adjusted by Cox regression modeling.

Results: The final cohort comprised 1,640 patients; 1,445 (88%) received IP-HCT (126 centers) and 195 (12%) OP-HCT (24 centers). Patient characteristics, functional status, disease status, and HCT year were similar between groups except a higher percentage of IP-HCT recipients were 70 years and older (IP-HCT: 31%, OP-HCT: 19%; P=0.0003), and a lower percentage of IP-HCT recipients received full dose melphalan 200 mg/m² (IP-HCT: 68%, OP-HCT: 90%; P=0.0036). There was a significant difference between the cohorts in the utilization of IP services (IP-HCT group: median 19 days, OP-HCT group: 4 days; P < 0.0001) and OP services (IP-HCT group: median 16 days, OP-HCT group: 33 days; P < 0.0001) at day 100. Adjusted total mean reimbursement for the IP-HCT group ($83,380 [95% CI: $78,958-$88,051]) was higher than the OP-HCT group ($55,721 [95% CI: $38,595-$80,446]) (P= 0.0301) (Figure). Factors associated with total reimbursement in the GLM were transplant setting, age, sex, comorbidity index, diagnosis-to-HCT time, and melphalan dose. Adjusted total patient responsibility for the IP-HCT group was $4,567 (95% CI: $4,210- $4,955) and $7,372 ($4,218- $12,884) (P=0.0902) for the OP-HCT group. Within 100 days post-HCT, 107/195 (55%) OP-HCT recipients had at least one subsequent admission, compared to 348/1,445 (24%) IP-HCT recipients (P < 0.0001). OS at 100 days was high for both HCT settings and adjusted OS was not significantly different by transplant setting (IP-HCT 98% [95% CI: 97%-99%]; OP-HCT 99% [95% CI: 98%-100%; P=0.1903)

Conclusions: Reimbursement and service utilization varied by HCT setting for Medicare beneficiaries with MM. Total reimbursement for 100 days post-HCT was $27,659 higher for IP-HCT than OP-HCT, after adjusting for patient and HCT-related characteristics. After the HCT index period, approximately 1 in 4 IP-HCT recipients required re-hospitalization within 100 days, whereas 1 in 2 OP-HCT recipients required subsequent hospitalization. Many factors influence the decision between IP or OP autologous HCT, including: center experience, severity of disease, patient co-morbidities, access to caregivers, proximity of lodging, cost to the patient, and reimbursement for services to the hospital system. The CIBMTR-CMS merged database is a new resource to support ongoing efforts to inform transplant centers and healthcare systems about provision of care options in the Medicare population.

Figure.

View largeDownload slide

Figure.

View largeDownload slide

Close modal