It is useful to study vaccine effectiveness using patient medical records data (e.g., using Clinical Practice Research Datalink) to assess the effect of the vaccinations against new virus mutations and to assess the continued success of these vaccines.
One issue with using patient medical records to study vaccination success is that individuals who regularly access healthcare services for routine health check-ups or vaccinations are likely to have better health outcomes compared to those who never or irregularly visit their healthcare system. This is therefore not a fair comparison when you want to assess the effect of a vaccination by comparing people who are vaccinated to people who are not vaccinated, unless differences in health-seeking behaviour and healthcare access can be accounted for.
Therefore, this study will aim to assess the association between markers or proxies of health-seeking behaviour and healthcare access with vaccinations and COVID-19 and influenza infections. Then the study will assess the impact of accounting for these markers in a COVID-19 and influenza vaccine effectiveness study. This study will be conducted amongst people aged 66 years or older in England.
This work is part of the Health Protection Research Unit in Vaccines and Immunisation, a partnership between the London School of Hygiene & Tropical Medicine and the UK Health Security Agency.
Observational studies of vaccine effectiveness (VE) are important for assessing VE particularly of new strains of influenza and COVID-19 and assessing duration of protection. However, cohort studies of VE are susceptible to confounding, including by health-seeking behaviour/healthcare access, which can result in over or underestimated VE. This study will explore whether identified markers of health-seeking behaviour/healthcare access (e.g., uptake of screening in nationwide programmes) are associated with influenza/COVID-19 vaccination and infections, and the effect of adjusting for these markers in cohort studies of influenza/COVID-19 VE.
The primary exposures of interest will be COVID-19 and influenza vaccinations, whilst the primary outcomes of interest will be COVID-19 infections, COVID-19-related hospitalisation, COVID-19-related death, acute respiratory infection or influenza/influenza-like-illness (ARI/ILI) infections, ARI or ILI-related hospitalisation and ARI or ILI-related death. This will be conducted in a population aged ≥66 years identified in CPRD Aurum. The CPRD Aurum dataset will be linked to Hospital Episode Statistics (HES) to identify COVID-19 and ARI/ILI related infections, as well as markers of health-seeking behaviour/healthcare access. The data will also be linked to Office for National Statistics (ONS) data to identify ARI/ILI or COVID-19 related deaths and to the Index of Multiple Deprivation (IMD) to adjust for differences in socioeconomic status. The study analyses will include multivariable logistic regression to assess associations between markers of health-seeking behaviour/healthcare access and each of the exposures and outcomes and Poisson regression to assess COVID-19 and influenza VE with and without adjusting for a combination of markers of health-seeking behaviour/healthcare access.
This study will ensure a better understanding of confounding in VE studies, which will allow for better interpretation of current and future estimates of VE, allowing for better policy-decisions on vaccination strategy. This work is part of the HPRU in Vaccines and Immunisation, a partnership with the UKHSA.
Primary outcomes:
• COVID-19 infections, which will include the following:
o COVID-19-related primary care record; or
o COVID-19-related hospitalisations; or
o COVID-19-related death.
• COVID-19 related hospitalisations or COVID-19-related death.
• COVID-19 related death.
• ARI or ILI infection, which will include the following:
o ARI or ILI-related primary care record; or
o ARI or ILI-related hospitalisations; or
o ARI or ILI-related death.
• ARI or ILI-related hospitalisations or ARI/ILI-related death.
• ARI or ILI-related death.
Secondary outcome:
• COVID-19 PCR test, regardless of the result (positive, negative or void).
Helen McDonald - Chief Investigator - London School of Hygiene & Tropical Medicine ( LSHTM )
Sophie Graham - Corresponding Applicant - Evidera, Inc
Dorothea Nitsch - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )
Jemma Walker - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )
Nick Andrews - Collaborator - UK Health Security Agency (UKHSA)
Sophie Graham - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation