The test-negative design for estimating influenza vaccine effectiveness

Area of Science:

• Epidemiology
• Vaccinology
• Biostatistics

Background:

• The test-negative design is increasingly used for observational studies of influenza vaccine effectiveness (VE).
• Formal methodological development of this design has been lacking.
• Understanding its assumptions is crucial for accurate VE estimation.

Purpose of the Study:

• To formally develop the rationale and underlying assumptions of the test-negative study design for influenza VE.
• To identify conditions under which the test-negative design yields unbiased VE estimates.
• To compare the test-negative design with traditional methods like case-control and cohort studies.

Main Methods:

• The study defines the test-negative design where individuals seeking care for acute respiratory illness (ARI) are tested for influenza.
• VE is calculated as the ratio of vaccination odds in influenza-positive versus influenza-negative individuals.
• The derivation involves analyzing assumptions related to non-influenza ARI causes and healthcare-seeking behavior.

Main Results:

• The test-negative design can generalize VE estimates to the source population if non-influenza ARI distribution is independent of vaccination status and VE is constant across healthcare-seeking behaviors.
• Biased VE estimates can occur if studies include ARI cases during non-circulating periods or fail to adjust for calendar time.
• The design is less prone to misclassification and confounding by healthcare-seeking behavior than traditional methods.

Conclusions:

• The test-negative design offers advantages in reducing bias compared to traditional epidemiological study designs.
• Key assumptions include similar incidence of non-influenza respiratory infections and consistent VE across care-seeking strata.
• Careful consideration of these assumptions is necessary for reliable influenza VE estimation.