UK LLC - COVID-19: Methodological enhancement and documentary analysis of the UKLLC

Study code

Lead researcher
Andy Boyd

Study type
Data only

Institution or company
University of Bristol

Researcher type

Speciality area
Infection, Public Health and Prevention, COVID


[The NIHR BioResource contributes to a major national study of the long-term impacts of COVID-19, through the UK Longitudinal Linkage Collaboration.]

The UKRI funded Longitudinal Health and Wellbeing National Core Study aims to understand the health, social and economic impacts of the COVID-19 pandemic by uniting longitudinal population study data from established studies with participants national de-personalised electronic health records. It has built the UK Longitudinal Linkage Collaboration a sa new infrastructure to host de-personalised data from many UK longitudinal studies and to link these to participants routine records.This will form a new national Trusted Research Environment for secure research using linked longitudinal data across studies. Approved researchers will use the UK LLC for COVID policy relevant research.This is a globally unique initiative and new for the UK. It is therefore important that the data managers building and maintaining the resource conduct a series of internal data management tasks. In summary, these tasks will help ensure the UK LLC is a useful resource which works well alongside other UK COVID Research resources, to understand the characteristics of participants whose data are in the UK LLC and to allow us to manage the data and help ensure it is of high quality. We will:1) recreate the definitions of diseases and social status used in other research resources (e.g. code lists which define conditions such as COVID-19 or Asthma) within the UK LLC so approved researchers can make use of these tools.2) to conduct quality assessments which compares data from across different sources to understand if the data linkage and processing is accurate and efficient.3) to conduct descriptive statistics which describe the characteristics of all the participants in the UK LLC so researchers understand how similar the participants whose data are in the UK LLC are to the UK population and whether certain groups (such as those of different ages or ethnicities or health status) are either over or under-represented.4) to create derived data – using study data and linked routine health and administrative records - which more efficiently describes participants health and social circumstances and patterns of study activity. 5) to structure, store and process the data so it is held in a secure way where the data will be accurate for current and future users and where researchers (within the UK LLC) can efficiently bring together and use the data for the same participants from different sources.6) to understand if participants take part in multiple studies and to enable researchers to understand this and to account for it in their analysis so this does not introduce error and can help to understand data quality issues. The results of this work will improve the organisation and management of data in the UK LLC. The results will be used to write research user guides, to promote the resource to research users, to describe how the resource works to participants and the public and to write academic publications and reports which document the nature of the resource and how it works.1. Gilbert R, Lafferty R, Hagger-Johnson G, et al. GUILD: GUidance for Information about Linking Data sets†. J Public Health. 2018;40(1):191-198. doi:10.1093/pubmed/fdx037