Influenza (flu) is a seasonal respiratory viral infection which can cause mild to severe illness. Occasionally, life-threatening complications develop requiring hospitalisation and sometimes causing death. With weakening of the immune system with age, older people are at increased risk for influenza-related complications. Each year everyone aged 65 and over is offered a free flu vaccine to protect them from influenza. A range of new vaccine types designed to be more effective at preventing infection in the elderly are now becoming available. For the 2019-2020 influenza season, we expect one of two influenza vaccines to be offered to people 65 and over:
a) aTIV, containing a chemical (adjuvant) to boost the bodys immune response to the vaccine;
or
b) QIVc, containing influenza virus components grown in specially cultured cells instead of in eggs traditionally used for vaccine production.
The aim of this study is to look at the occurrence of influenza-like illness and influenza complications in older people given one of these new vaccines. Over three influenza seasons we will look at whether the proportion of people with influenza-like illness differs between people receiving the different vaccine types. As further vaccine types are introduced during the course of the study, we will include them in our analyses to see how the protection they offer compares to the established vaccines. The results of this study will help to inform the government when it makes decisions about which vaccines to recommend for older people in future years.
Objective
To estimate the relative vaccine effectiveness (RVE) of influenza vaccines licensed in adults aged 65 and above during a given influenza season.
Study design
Primary-care based cohort study using data from CPRD Aurum/GOLD and HES.
Methods
Specific codes in the CPRD GOLD/Aurum medical dictionary will be used to collect information on types of influenza vaccine administered. Adults 65 and above vaccinated during the influenza seasons of interest with one of the currently recommended influenza vaccines (e.g. adjuvanted trivalent, cell-grown quadrivalent) will be included. Additionally, we will explore the effect on patient numbers of including people aged 60-64 who may have received one of the vaccines of interest. CPRD Aurum will be used to determine GP consultations for influenza-like illness and clinically-diagnosed influenza in people receiving the vaccines of interest each season. HES data will be used to determine influenza-related hospitalisations and intensive-care admissions. Individuals will be considered vaccinated 14 days or more after receipt of influenza vaccine, and will be followed up until one of the following: outcome of interest, end of study period, death, subsequent influenza vaccination within the same influenza season.
We anticipate requiring data for three influenza seasons to adequately power the study (2019-2020, 2020-2021, 2021-2022). We believe adjuvanted vaccine will be predominant during the 2019-2020 season and that vaccine batch numbers will unavailable in Aurum during that season, so we will use CPRD GOLD at the end of the 2019-2020 season to assess data quality and describe the study population, informing CPRD Aurum analyses in subsequent seasons.
Data analysis
If there is no apparent bias in the study population in the type of vaccination given, we will conduct multi-variable Poisson regression analyses. Adjusted RVE will be defined as (1-rate ratio) x100%. Subgroup analysis of people 75 and above and 85 and above will be performed.
Preliminary:
For the first influenza season we will conduct a preliminary exploration of the data in CPRD Gold and describe the study population receiving the different vaccines types in terms of age, gender, co-morbidities, socio-economic status, geographical location and timing of vaccination according to these factors.
Primary:
GP medically-attended influenza-like illness or influenza (clinically diagnosed)
Secondary:
Influenza-related hospital admissions, listing an International Classification of Diseases Tenth Revision, Clinical Modification, code for influenza (Codes J09.xx, J10.xx, J11.xx, and J129) (see appendix for codes);
Influenza-related Intensive Care Unit admissions
Timing of outcomes: from 14 days after receipt of influenza vaccination to the end of the influenza season each year. The influenza season will be defined by Public Health Englands virological surveillance data. Using the Moving Epidemic Method (MEM)(1) to define baseline thresholds, the start of the influenza seasons will be defined when the virological surveillance indicates that the weekly positivity rate of samples reaches the baseline threshold. The end of the influenza season will be defined by when the data indicate the weekly positivity rates of samples returns below the baseline threshold, plus 14 days.
Jonathan Van-Tam - Chief Investigator - University of Nottingham
Louise Lansbury - Corresponding Applicant - University of Nottingham
Heather Whitaker - Collaborator - Public Health England
Nick Andrews - Collaborator - Public Health England
Richard Pebody - Collaborator - Public Health England
Tricia McKeever - Collaborator - University of Nottingham
Wei Shen Lim - Collaborator - Nottingham University Hospitals
HES Admitted Patient Care;Practice Level Index of Multiple Deprivation