Seasonal influenza vaccination is associated with reduced risk of death among Medicare beneficiaries☆

Timothy G Buchman,Steven Q Simpson,Kimberly L Sciarretta,Kristen P Finne,Nicole Sowers,Michael Collier,Saurabh Chavan,Rose Do,Cheng Lin,Ibijoke Oke,Kiersten E Rhodes,Aathira Santhosh,Alexander T Sandhu,Steve Chu,Sandeep A Patel,Gary L Disbrow,Rick A Bright,Thomas E MaCurdy,Jeffrey A Kelman,Timothy G. Buchman,Steven Q. Simpson,Kimberly L. Sciarretta,Kristen P. Finne,Kiersten E. Rhodes,Alexander T. Sandhu,Sandeep A. Patel,Gary L. Disbrow,Rick A. Bright,Thomas E. MaCurdy,Jeffrey A. Kelman

DOI: https://doi.org/10.1016/j.vaccine.2021.11.016

IF: 4.169

2021-12-01

Vaccine

Abstract:BACKGROUND: Influenza causes substantial mortality, especially among older persons. Influenza vaccines are rarely more than 50% effective and rarely reach more than half of the US Medicare population, which is primarily an aged population. We wished to estimate the association between vaccination and mortality reduction.METHOD: We used the US Center for Medicare and Medicaid Services (CMS) DataLink Project to determine vaccination status and timing during the 2017-2018 influenza season for more than 26 million Medicare enrollees. Patient-level demographic, health, co-morbidity, hospitalization, vaccination, and healthcare utilization claims data were supplied as covariates to general linear models in order to isolate and estimate the association between participation in the vaccination program and relative risk of death.FINDINGS: The 2017-2018 seasonal influenza vaccine reduced (Relative Risk Ratio [RRR] 0.936 [95% CI = 0.918-0.954]) the risk of all-cause death among beneficiaries following a hospitalization for sepsis and moreover the risk of death without a prior hospitalization during the 2.5-month outcome window (RRR 0.870 [95% CI = 0.853-0.887]). We estimate the number needed to vaccinate (NNV) to prevent a death in the ten-week outcome window is between 1,515 beneficiaries (95% CI = 1,351-1,754; derived from the average treatment effect of augmented inverse probability weighting) and 1,960 beneficiaries (95% CI = 1,695-2,381; derived from the average marginal effect of logistic regression). Among beneficiaries requiring hospitalization, the greatest death risk reduction accrued to those 85 + years of age who were hospitalized with sepsis, RRR 0.92 [95% CI = 0.89-0.95]. No apparent benefit was realized by beneficiaries who required custodial (nursing home) care.INTERPRETATION: Seasonal influenza immunization is associated with relative reduction of death risk among non-institutionalized Medicare beneficiaries.FUNDING: All authors are full-time or contractual employees of the United States Federal Government, Department of Health and Human Services, the funding agency.

immunology,medicine, research & experimental