Economic and clinical burden associated with respiratory viral infections after allogeneic hematopoietic cell transplant in the United States

Abstract Background Allogeneic hematopoietic cell transplant (allo‐HCT) recipients are at increased risk for respiratory viral infections (RVIs), which invoke substantial morbidity and mortality. Limited effective antiviral options and drug resistance often hamper successful RVI treatment, creating additional burden for patients and the health care system. Methods Using an open‐source health care claims database, we examined differences in clinical outcomes, health resource utilization, and total reimbursements during the 1‐year period following allo‐HCT in patients with and without any RVI infection (respiratory syncytial virus, influenza, parainfluenza virus, and human metapneumovirus). RVIs were diagnosed at any time ≤1 year after allo‐HCT and identified by International Classification of Disease codes. Analyses were stratified by the presence or absence of acute or chronic graft‐versus‐host disease (GVHD). Results The study included 13 363 allo‐HCT patients, 1368 (10.2%) of whom had a diagnostic code for any RVI. A higher proportion of patients with any RVI had pneumonia ≤1 year after allo‐HCT compared to patients without any RVI, with or without GVHD. Patients with any RVI had higher all‐cause mortality risk, longer length of post‐allo‐HCT hospital stay, higher readmission rate, and higher number of hospital days after allo‐HCT compared to patients without the infection (all p < .05). Total unadjusted median reimbursements were higher for those with any RVI and each specific RVI assessed than those without the specific infection, with or without GVHD. Conclusion Allo‐HCT patients with RVIs had significantly worse clinical outcomes and increased health resource utilization and reimbursements during the year following allo‐HCT, with or without GVHD.


INTRODUCTION
Allogeneic hematopoietic cell transplant (allo-HCT) recipients are at an increased risk for respiratory viral infections (RVIs), 1 with respiratory syncytial virus (RSV) being the most common cause of RVI in this population (incidence of up to 50%), followed by influenza (up to 40%), parainfluenza virus (PIV; up to 27%), and human metapneumovirus (hMPV; up to 11%). 2 RVIs are associated with substantial morbidity and mortality in HCT recipients, often related to the progression from an upper respiratory tract infection (URTI) to a lower RTI (LRTI). [2][3][4][5][6][7][8][9][10] As with other types of infections, RVIs occur more frequently after allo-HCT than after autologous HCT, 11,12 owing to delayed immune reconstitution 13 and lack of humoral and T-cellmediated immunity. 1,14,15 In addition to increased frequency, mortality associated with RVIs is higher after allo-HCT than after autologous HCT. 16 A limited number of effective antivirals and the emergence of drug resistance, likely due to prolonged viral shedding in the setting of antiviral monotherapy, often hamper successful treatment of RVIs. 1,17,18 In order to better understand the clinical and economic impact of RVIs in allo-HCT recipients, this retrospective observational study utilized real-world claims database to compare clinical outcomes, health resource utilization, and health care reimbursement in allo-HCT recipients with and without RVIs, including RSV, influenza, PIV, and hMPV.

METHODS
Patients who underwent allo-HCT between January 1, 2012 and

Statistical analysis
For all study outcomes, patients were stratified by presence/absence of any RVI, as well as each type of RVI (RSV, influenza, PIV, and hMPV).
Separate analyses were reported for those with acute or chronic graft-

Study population
A total of 17 520 patients with first allo-HCT procedure codes between January 1, 2012 and December 31, 2017, and no prior allo-HCT procedures in the 1-year prior, were identified within the database. Those without a corresponding allo-HCT hospitalization (n = 2933) and who underwent index allo-HCT in an outpatient setting (n = 1224) were excluded, leaving a total of 13 363 eligible for inclusion in the final study cohort ( Figure S1). Demographic and transplantation characteristics of the patient population can be seen in Table S1. Mean ages of patients were 44 and 47 years in the group with and without infection, respectively; the majority were male and had malignant disease, peripheral blood as the stem cell source, and ≥1 comorbidity.

Clinical outcomes
A significantly higher proportion of patients with any RVI experienced pneumonia within the first year following allo-HCT compared to those with no RVI, irrespective of presence of GVHD (p < .0001; Figure 1). Similar significant findings were seen for each RVI evaluated (p < .0001 for all comparisons).

Health resource utilization
Patients with any RVI had significantly higher health resource utilization (overall LOS, hospital readmission rate) compared to those with no RVI (p < .0001 for all comparisons; Figure 3). Significant findings were also seen for each RVI evaluated (all p ≤ .0002 for overall LOS and p < .0001 for readmission rate).

Health care reimbursement
Patients with any RVI had significantly higher total health care reim-  Figure 4).

DISCUSSION
This retrospective observational database study confirmed that patients with RVIs demonstrated significantly higher morbidity (i.e., pneumonia) and mortality, irrespective of GVHD presence. Health resource utilization, specifically overall hospital LOS and readmission rate, was also significantly higher in those with RVIs. Consistent with these findings, total health care reimbursement was significantly higher for patients undergoing allo-HCT with an RVI compared to those without, with significantly higher total reimbursement found for all RVIs evaluated (RSV, influenza, PIV, and hMPV). In conclusion, our study found that allo-HCT patients with an RVI had worse clinical outcomes and a significantly higher burden on the health care system in terms of health resource utilization and total reimbursement. The adjusted analyses showed that mortality and total reimbursement were higher for allo-HCT patients with an RVI, regardless of the presence of GVHD. Antivirals and other strategies may improve patient outcomes and reduce health resource utilization and costs after allo-HCT.

ACKNOWLEDGMENTS
The authors would like to thank Dr Francisco Marty for his contributions to the design and data collection for this study prior to his passing. This project represents a contracted effort by Certara with AlloVir. Medical writing assistance and revision of the manuscript under the direction of the authors was provided by PRECISIONscientia and compensated by AlloVir.