Association between herpes zoster vaccination and autoimmune rheumatic diseases in UK: A regression discontinuity analysis

Study type
Protocol
Date of Approval
Study reference ID
23_003378
Lay Summary

Autoimmune rheumatic diseases (ARD) are serious conditions that cause pain and disability. More and more people are struggling with these diseases over the past decades, posing great medical and socioeconomic burdens. These diseases have become more common and pose big challenges for affected people. There's a thought that certain viruses, like the one causing shingles, might be linked to these diseases. In the UK, folks between 70 and 79 have been advised to get a shingles vaccine since 2013.

This study wants to see if getting the shingles vaccine changes the chances of getting these diseases.

We will investigate the difference in the incidence among people vaccinated against herpes zoster compared with unvaccinated people. We will identify differences in outcomes across populations.

By conducting this research, we hope to better understand if the shingles vaccine affects these diseases. The findings will be valuable for shaping public health policies, guiding vaccination strategies, and planning future interventions.

Technical Summary

To date, no population-level studies have investigated the impact of herpes zoster vaccination on autoimmune rheumatic diseases. The study will seek to explore the causal effect of herpes zoster vaccination on the occurrence of autoimmune rheumatic disease.

Specifically, we aim to
1) investigate if causality exists between zoster vaccination and the risk of developing autoimmune rheumatic diseases
2) assess which autoimmune rheumatic diseases have the strongest association with zoster vaccination
3) identify any variations in the above outcomes across different population groups

The planned study design is a longitudinal retrospective observational cohort study, conducted from 1st Sep 2013 to 31st Aug 2019. To provide causal as opposed to merely correlational evidence on this question, this study will use a regression discontinuity (RD) design which is a quasi-experimental method to establish causality and assigns to groups according to a cutoff value for a continuous variable, taking advantage of the fact that eligibility for the herpes zoster vaccine was determined based on the age of individuals. We will identify the eligibility of participants based on their year of birth, which is collected in CPRD. The main analysis will estimate the effect of being eligible for the zoster vaccine and the secondary analysis will estimate the effect of receiving the zoster vaccine on composite and individual incidence of ARD. Both sharp and fuzzy RD models will be used in local linear regression with mean squared error (MSE) optimal bandwidth. The sensitivity analysis will apply different bandwidth sizes, use polynomials of varying degrees, and adjust for covariates.
The findings of this study will provide evidence of a more effective and cost-effective intervention, the zoster vaccine, for preventing ARD and are helpful for health policy making.

Health Outcomes to be Measured

A composite measure of the incidence of any of the following autoimmune rheumatic diseases during the study follow-up period:
Rheumatoid arthritis, Psoriatic arthritis, Undifferentiated inflammatory arthritis, Ankylosing spondylitis, Axial spondyloarthropathy, Polymyalgia rheumatica, Giant cell arteritis, large vessel vasculitis, ANCA vasculitis, Small vessel vasculitis, Systemic Lupus Erythematosus, Sjogren syndrome.

The individual incidence of the aforementioned autoimmune rheumatic diseases during the study time period.

Collaborators

James Galloway - Chief Investigator - King's College London (KCL)
Zijing Yang - Corresponding Applicant - King's College London (KCL)
Katie Bechman - Collaborator - King's College London (KCL)
Mark Russell - Collaborator - King's College London (KCL)
Sam Norton - Collaborator - King's College London (KCL)
Sinead Langan - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )

Linkages

Patient Level Index of Multiple Deprivation