Case-control study of vaccine effectiveness in preventing influenza hospitalizations in older adults, United States, 2010-11
Background: Older adults are at increased risk of complications from influenza, including hospitalization, but few observational studies have used laboratory-confirmed illness as an endpoint or adjusted for confounders such as baseline functional status. We conducted a large case-control study to estimate influenza vaccine effectiveness (VE) to prevent laboratory-confirmed influenza hospitalizations among older adults in eleven U.S. Emerging Infections Program sites.
Methods: Cases were adults ≥50 years old hospitalized with RT-PCR-confirmed influenza infections during the 2010-11 influenza season. Community controls were matched by age, county and month of case hospitalization. Medical records were reviewed and a standardized questionnaire administered to all study subjects, including questions on baseline functional status using a physical health composite score of 0 to 100. Subjects were considered vaccinated if they had received an influenza vaccine at least 14 days prior to the hospital admission date of the case, either according to medical records or self-report (if vaccination location was also provided). We estimated VE by comparing vaccination status between cases and matched controls using conditional logistic regression. Adjusted VE (aVE) was estimated as 100 x (1 – adjusted odds ratio).
Results: Overall, 364/368 (99%) cases and 767/773 (99%) controls enrolled had vaccination information available; among these, 72% of cases and 76% of controls were vaccinated. Cases were more likely to be of non-white race (28% of cases vs. 15% of controls), of Hispanic ethnicity (7% vs. 2%), have ≥1 chronic health conditions (94% vs. 66%), have been hospitalized in the previous year (12% vs. 1%), have lower income (59% of cases with annual income <$35,000 vs. 41% of controls), and worse self-reported functional status (mean score of 35 among cases vs. 41 among controls) (P-values <0.01 for all). After adjusting for potential confounders, the aVE for the season was 54.5% (95% confidence interval: 29.5% to 70.5%) and was similar in all age groups (Figure).
Conclusion: During 2010-11, influenza vaccination was associated with a significant reduction of the risk of laboratory-confirmed influenza hospitalization among a moderately well-vaccinated population of older adults regardless of age.
M. M. Farley, None
M. Monroe, None
J. I. Meek, CDC Emerging Infections Program: Investigator, Research grant and Salary
P. Daily Kirley, None
N. Bennett, None
C. Morin, None
D. Aragon, None
A. Thomas, None
M. L. Lindegren, None
S. M. Zansky, None
J. Baumbach, None
J. Ferdinands, None
A. M. Fry, None