Flu vaccination is a key component of national health policy, targeted at those most at risk including people aged over 64 and younger adults with chronic medical conditions such as diabetes and respiratory disease. Currently, more than 70% of adults aged over 64 are vaccinated each year against flu, helping to avert between 5,500 and 9,000 premature deaths per year.
Although the flu vaccine is generally effective, recent research indicates that patients who take statins may have a reduced immune response to the vaccine compared with patients who do not take statins. Statin medications are a very common preventative treatment in older adults for lowering cholesterol and reducing the risk of heart disease with about 1.8 million people in the over 75y age group alone taking statins.
Previous studies have suggested statins and other treatments at the time of flu vaccination might reduce the effectiveness of the vaccine, although unadjusted differences between statin-users and non-users may have led to bias in those results. We aim to strengthen the evidence by making efficient use of patients’ records stored electronically at Doctors’ surgeries and hospitals. In particular, we will investigate potential biases through modern statistical methods that uses patient outcomes in the period before vaccination to account for unmeasured differences between statins users and non-users. It is important to emphasise that this work does not challenge the well-established clinical benefits of flu vaccination or statins, but seeks better strategies for improving the benefit from both.
Background
Influenza vaccination is routinely advised for many at-risk groups including adults aged≥65 years and, and younger patients with diabetes and cardiovascular disease. Many of these patients are also likely to be prescribed statins for control of cardiovascular risk. However, the mechanism by which statins protect against heart disease may inhibit the immune response to vaccination, potentially lowering its efficacy. This is further complicated by statins themselves potentially offering some protection against influenza. Robust evidence is needed to guide clinical practice and ensure vaccine response is maximised.
Objectives
To evaluate the potential modifying effect of statins on the effectiveness of the influenza vaccine and as a prophylaxis against influenza.
Methods
Annual cohorts of patients eligible for the influenza vaccine will be selected from 2010-20. Our primary analysis will test for an association between statin use and influenza outcomes, amongst patients receiving the influenza vaccine. Secondary analyses will estimate influenza vaccine effectiveness separately among statin users and nonusers, and the extent of any prophylactic effect of statins. A composite influenza outcome will be primary-care consultations for influenza-like illness and hospital admissions for influenza. Cox regression will be used to model the time to influenza with weighting by propensity scores to adjust for confounders observed in the data. We anticipate residual bias from unmeasured confounders so will apply a before-and-after quasi-experimental study design, the prior event rate ratio method, to adjust for this. We shall evaluate any remaining residual bias, applying the analysis protocol to a negative control outcome, such as shingles, for which we expect no association with influenza vaccination. Since polypharmacy is likely to be prevalent in frail and older populations, further sensitivity analysis will screen the five other most common medications to assess the extent to which any observed interaction effect may be attributable to other, concomitantly prescribed medications.
Composite outcome: hospital admissions for influenza; consultations in primary care for symptoms consistent with influenza-like illness
Negative control outcome: shingles
Adam Streeter - Chief Investigator - University of Exeter
Adam Streeter - Corresponding Applicant - University of Exeter
Andrea Shelly - Collaborator - University of Exeter
Fergus Hamilton - Collaborator - University of Bristol
Lauren rodgers - Collaborator - University of Exeter
Sarah Walker - Collaborator - University of Exeter
William Hamilton - Collaborator - University of Exeter
William Henley - Collaborator - University of Exeter
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation