Is seasonal influenza vaccine effectiveness reduced in people receiving repeat vaccinations? A retrospective cohort study using the Clinical Practice Research Database

Date of ISAC Approval: 
15/05/2018
Lay Summary: 
Influenza (or flu) is a viral infection that generally affects the upper airways (e.g. the nose and throat) and can rarely affect the lungs. It manifests with fever, sore throat, widespread muscle and joint pains. In the last few years, studies suggest that influenza vaccine effectiveness is waning, especially in elderly subjects. Some studies suggest people who have received the seasonal flu vaccine over successive years (e.g. elderly) may be less protected from influenza in a given season compared to first time recipients of the seasonal flu vaccines, although the protection provided is still worthwhile compared to no vaccination. Understanding whether this phenomenon exists is important to consider when developing vaccination policy. This study will look at the occurrence of influenza in those vaccinated in the 2015/2016 influenza season. We will look at whether the proportion of cases of influenza that were prevented is different among people who had a history of receiving the influenza vaccine in one or more of the previous five seasons, and compare this to people who were vaccinated for the first time in the 2015/2016 winter season. We will repeat this analysis for seasons prior to 2015/2016, but looking at a two-year history of influenza vaccines.
Technical Summary: 
A debate has re-emerged on the possible effect of previous influenza vaccinations on the vaccine effectiveness (VE) of seasonal influenza vaccine (SIV). The aim of this study is to examine whether VE was lower in people with repeat vaccination histories relative to first time vaccinees in the 2015/2016 winter season. We will conduct a retrospective cohort study of people vaccinated in the 2015/2016 winter season aged >65 years old and those aged 12 to 65 in clinical risk groups. The analysis will examine the occurrence of influenza infection, influenza-like-illness and acute respiratory infection among those who have a five-year history of SIV and those who are first time SIV recipients. Hazard ratios will be estimated using Cox regression, relative VE of the 2015/2016 influenza vaccine will be calculated as 1 - Hazard ratio and by previous vaccination history type (in one to five previous seasons). Relative VE will be adjusted for sex, age, comorbidities, previous influenza infection in the past five years, smoking status, index of multiple deprivation (IMD), body mass index (BMI), use of statins and immunosuppressive drugs. We will repeat the analyses for previous winter seasons back to 2011-12, but looking at two-year history of SIV.
Health Outcomes to be Measured: 
#NAME?
Collaborators: 

Professor Jonathan Van-Tam - Chief Investigator - University Of Nottingham
Dr Abhishek Abhishek - Collaborator - University Of Nottingham
Dr Georgina Nakafero - Collaborator - University Of Nottingham
Hikaru Bolt - Corresponding Applicant - Public Health England
Puja Myles - Collaborator - CPRD
Sudhir Venkatesan - Collaborator - University Of Nottingham

Linkages: 
Practice IMD (Standard);Pregnancy Register