Comparison of local influenza vaccine effectiveness using two methods.

Area of Science:

• Epidemiology
• Public Health
• Vaccinology

Background:

• Traditional research methods for influenza vaccine effectiveness (VE) may be impractical due to cost or time constraints.
• Administrative databases offer a potential alternative for VE assessment using viral testing and vaccination data.

Purpose of the Study:

• To compare VE estimates derived from outpatient research databases versus administrative databases.
• To evaluate the feasibility and accuracy of using administrative data for VE studies.

Main Methods:

• A test-negative, case-control design was employed for the 2017-2018 and 2018-2019 influenza seasons.
• Data were collected via consent and specimen collection (research) and an administrative database with electronic immunization records.
• Multivariable logistic regression calculated odds ratios for influenza infection by vaccination status.

Main Results:

• VE was significant against all influenza and influenza A (37-49%) across both seasons.
• Research participants differed demographically and in enrollment site compared to administrative database participants.
• Significant differences in VE estimates were observed when analyzing emergency department enrollment sites separately.

Conclusions:

• The choice of method for VE estimation depends on factors like sample size and subgroups of interest, with research estimates potentially offering greater generalizability.
• Research databases may reduce selection bias and misclassification of vaccination status.
• Administrative databases provide advantages in speed and cost for VE assessments.