J Infect Dis
. 2023 Jan 23;jiad015.
doi: 10.1093/infdis/jiad015. Online ahead of print.
Vaccine effectiveness against influenza-associated urgent care, emergency department, and hospital encounters during the 2021-2022 season, VISION Network
Mark W Tenforde 1 , Zachary A Weber 2 , Malini B DeSilva 3 , Edward Stenehjem 4 , Duck-Hye Yang 2 , Bruce Fireman 5 , Manjusha Gaglani 6 7 , Noah Kojima 1 , Stephanie A Irving 8 , Suchitra Rao 9 , Shaun J Grannis 10 11 , Allison L Naleway 8 , Lindsey Kirshner 2 , Anupam B Kharbanda 12 , Kristin Dascomb 4 , Ned Lewis 5 , Alexandra F Dalton 1 , Sarah W Ball 2 , Karthik Natarajan 13 14 , Toan C Ong 9 , Emily Hartmann 15 , Peter J Embi 10 16 , Charlene E McEvoy 3 , Nancy Grisel 4 , Ousseny Zerbo 5 , Margaret M Dunne 2 , Julie Arndorfer 4 , Kristin Goddard 5 , Monica Dickerson 1 , Palak Patel 1 , Julius Timbol 5 , Eric P Griggs 1 , John Hansen 5 , Mark G Thompson 1 , Brendan Flannery 1 , Nicola P Klein 5
Affiliations
- PMID: 36683410
- DOI: 10.1093/infdis/jiad015
Abstract
Background: Following historically low influenza activity during the 2020-2021 season, the United States saw an increase in influenza circulating during the 2021-2022 season. Most viruses belonged to the influenza A(H3N2) 3C.2a1b 2a.2 subclade.
Methods: We conducted a test-negative case-control analysis among adults ≥18 years of age at three sites within the VISION Network. Encounters included emergency department/urgent care (ED/UC) visits or hospitalizations with ≥1 acute respiratory illness (ARI) discharge diagnosis codes and molecular testing for influenza. Vaccine effectiveness (VE) was calculated by comparing the odds of influenza vaccination ≥14 days before the encounter date between influenza-positive cases (type A) and influenza-negative and SARS-CoV-2-negative controls, applying inverse probability-to-be-vaccinated weights, and adjusting for confounders.
Results: 86,732 ED/UC ARI-associated encounters (7,696 [9%] cases) and 16,805 hospitalized ARI-associated encounters (649 [4%] cases) were included. VE against influenza-associated ED/UC encounters was 25% (95% confidence interval (CI): 20-29%) and 25% (95%CI: 11-37%) against influenza-associated hospitalizations. VE against ED/UC encounters was lower in adults ≥65 years of age (7%; CI: -5-17%) or with immunocompromising conditions (4%, CI:-45-36%).
Conclusions: During an influenza A(H3N2)-predominant influenza season, modest VE was observed. These findings highlight the need for improved vaccines, particularly for A(H3N2) viruses that are historically associated with lower VE.
Keywords: COVID-19; bias; influenza; test-negative design; vaccine effectiveness.