Institution: | 1. School of Medicine and Dentistry and Menzies Health Institute Queensland, Griffith University, Parklands Drive, Gold Coast, QLD 4222, Australia;2. Departments of Paediatrics and Infectious Diseases, Gold Coast Health, Gold Coast, QLD 4215, Australia;3. Child Health Research Centre, The University of Queensland, South Brisbane, QLD 4101, Australia;4. Curtin School of Population Health, Curtin University, Bentley, WA 6102, Australia;5. Wesfarmers Centre for Vaccines and Infectious Diseases, Telethon Kids Institute, Nedlands, WA 6009, Australia |
Abstract: | AimTo test the internal validity of the test-negative design (TND) by investigating associations between maternal influenza vaccination, and new virus detection episodes (VDEs), acute respiratory illness, and healthcare visits in their children.MethodsEighty-five children from a birth cohort provided daily symptoms, weekly nasal swabs, and healthcare use data until age 2-years. Effect estimates are summarised as incidence rate ratios (IRR).ResultsThere was no association between maternal vaccination and VDEs in children (IRR = 1.1; 95 %CI = 0.9–1.2). Influenza-vaccinated mothers were more likely than unvaccinated mothers to both report, and seek healthcare for, acute lower respiratory illness in their children, IRR = 2.4; 95 %CI = 1.2–4.8 and IRR = 2.2; 95 %CI = 1.1–4.3, respectively.ConclusionA key assumption of the TND, that healthcare seeking behaviour for conditions of the same severity is not associated with vaccine receipt, did not hold. Further studies of the performance of the TND in different populations are required to confirm its validity. |