Economic Burden of Infections Following Allogeneic Hematopoietic Cell Transplant for Hematologic Malignancies

Background: Allogeneic hematopoietic cell transplantation (HCT) is a potentially curative option for many hematologic malignancies, but is associated with numerous complications, including infection.  While the clinical consequences of infection post-HCT have been well characterized, details of their economic burden have not.

Methods: Using a large US healthcare claims database, we identified patients aged ≥2 years who underwent allogeneic HCT between October 2009 and March 2013.  The hospital admission during which HCT was performed was deemed the “Index Admission”.  We excluded patients without continuous health plan enrollment and evidence of a hematologic malignancy in the 6-month period prior to the Index Admission.  Patients were followed from discharge of the Index Admission until death, plan disenrollment, or one year, whichever occurred first (“Follow-Up”).  We ascertained the incidence of infections during the Index Admission through 100 days after discharge (“Early Infections”), and examined by etiology (bacterial, viral, fungal, other/unknown) using relevant ICD-9-CM diagnosis codes.  We then ascertained total healthcare costs (inpatient care, outpatient care, outpatient pharmacy) and hospital length of stay (LOS) during the Index Admission and Follow-Up.  Reimbursed amounts (plan payment plus patient liability) were used as a proxy for healthcare costs. We compared total healthcare costs and LOS during the Index Admission, Follow-Up, and Overall (Index Admission + Follow-Up) between patients who did versus did not experience an Early Infection using Student’s t-tests (unadjusted analyses), and analyses of covariance (ANCOVA) to adjust for differences in age, sex, plan type, geography, year of Index Admission, type of malignancy, and comorbidities.

Results: A total of 1,635 patients were identified who underwent allogeneic HCT and met all selection criteria; mean age was 48.4 years, 55.6% were men, and acute myeloid leukemia was the most common malignancy (41.8%). A total of 77.7% of patients (n=1,270) experienced ≥1 Early Infections: bacterial (45.5%), viral (33.2%), fungal (19.5%), and other/unknown (54.1%).  Univariate findings are presented in the table.  Compared to patients without Early Infections, multivariate-adjusted mean total healthcare costs among patients with Early Infection were $69,044 greater during Index Admission, $74,307 greater during Follow-Up, and $143,351 greater Overall; corresponding increases in LOS were 8 days, 10 days, and 18 days, respectively (all p<.001).

Conclusions:  Nearly four-in-five patients who undergo allogeneic HCT experience Early Infections.  These infections are associated with significant increases in healthcare costs and hospital LOS. Our findings suggest new therapies capable of preventing infection – especially across etiologies– could be impactful from an economic perspective.

Table. Mean (SD) Total LOS and Healthcare Costs, by Time Since Allogeneic HCT and Early Infection Status

Type of Infection	Overall			Index Admission			Follow-Up Period
	WITH Infection	WITHOUT Infection	p-value	WITH Infection	WITHOUT Infection	p-value	WITH Infection	WITHOUT Infection	p-value
Total LOS, days
Any Infection	56.9 (41.8)	38.0 (25.2)	<.001	35.2 (22.7)	26.7 (14.6)	<.001	21.7 (34.5)	11.3 (22.0)	<.001
Bacterial	63.9 (46.4)	43.4 (29.6)	<.001	37.4 (23.9)	29.9 (18.5)	<.001	26.4 (38.6)	13.5 (24.7)	<.001
Viral	64.5 (48.8)	46.8 (32.4)	<.001	37.7 (24.3)	31.1 (19.6)	<.001	26.8 (40.3)	15.7 (27.0)	<.001
Fungal	68.1 (52.0)	49.0 (34.9)	<.001	40.4 (27.2)	31.6 (19.4)	<.001	27.7 (41.7)	17.4 (29.4)	<.001
Other/ Unknown	59.3 (43.5)	45.0 (32.6)	<.001	36.4 (23.9)	29.7 (17.4)	<.001	22.8 (35.6)	15.3 (27.7)	<.001
Total Healthcare Costs, $
Any Infection	479,162 (387,142)	310,911 (203,337)	<.001	262,290 (231,969)	178,560 (105,550)	<.001	216,871 (294,604)	132,350 (171,085)	<.001
Bacterial	517,802 (413,054)	377,972 (297,137)	<.001	275,178 (238,242)	217,228 (186,010)	<.001	242,624 (317,973)	160,743 (224,804)	<.001
Viral	561,509 (460,937)	381,977 (281,459)	<.001	296,565 (268,145)	217,260 (174,118)	<.001	264,944 (350,792)	164,716 (219,322)	<.001
Fungal	552,653 (464,501)	414,682 (326,023)	<.001	293,649 (263,916)	231,466 (197,292)	<.001	259,003 (354,145)	183,216 (248,892)	<.001
Other/ Unknown	495,675 (396,898)	377,794 (302,142)	<.001	270,943 (236,232)	211,331 (177,328)	<.001	224,731 (307,157)	166,462 (225,358)	<.001