Despite the widespread availability of several injectable inactivated influenza vaccines (IIV), including the trivalent standard-dose (IIV3-SD) and high-dose (IIV3-HD), and the quadrivalent (IIV4), the US Advisory Committee on Immunization Practices does not currently recommend one over another. The objective of this study was to assess the relative vaccine effectiveness (rVE) of IIV3-HD and IIV4 versus IIV3-SD.
rVE was estimated from a retrospective cohort study of Veterans aged 65 years and older who received an IIV during the 2014-15 influenza season. Veterans Health Administration (VHA) electronic medical records were linked with Centers for Medicare and Medicate Services administrative claims to capture the study outcomes of hospitalizations and baseline characteristics. The inverse probability of treatment weight (IPTW) method was used to adjust for potential confounding due to unmeasured factors associated with IIV3-SD, IIV3-HD, or IIV4 vaccination. The probability was estimated based on patient socio-demographic characteristics, comorbidities, pre-influenza season hospitalizations, prior season influenza vaccination, and use of immunosuppressive medication.
Our study population included 782,346 VHA patients vaccinated during the 2014-15 season. Of these, 10,543 (1%) received IIV4, while 59,536 (8%) received IIV3-HD and 712,267 (91%) received IIV3-SD. 11,626 (1.5%) were female and 588,324 (76%) were non-Hispanic white. Compared to those that received IIV3-SD vaccine, the IPTW-adjusted rVE for IIV3-HD was 7% (95% CI, 9%-21%) against all-cause, 15% (95% CI, 10%-17%) against cardiorespiratory-associated, and 13% (95% CI, 8%-17%) against influenza/pneumonia-associated hospitalization. For those that received IIV4, the IPTW-adjusted rVE was 4% (95% CI, 1%-4%), 1% (95% CI, -2%-5%), and 0% (95% CI, -9%-8%), respectively.
IIV3-HD is more effective than, and IIV4 is as effective as, IIV3-SD vaccination in preventing influenza/pneumonia-associated, cardiorespiratory, and all-cause hospitalizations. Additional studies that employ methods to control for unmeasured confounding are warranted as the use of IIV4 expands.
N. Neupane, None
M. Lewis, None
Y. Halchenko, None