The UK LLC Trusted Research Environment exposes the long-Covid evidence gap

First published by UK LLC on 12th June 2024

Find out more about the UK LLC Trusted Research Environment (TRE)

Researchers used data collected by eight UK Longitudinal Population Studies (LPS) that were linked with data from electronic healthcare records from NHS England. Data were brought together by UK Longitudinal Linkage Collaboration (UK LLC) and collectively analysed in their Trusted Research Environment, powered by SeRP – the Secure eResearch Platform.

Findings showed that most people with a history of self-reporting long COVID did not have evidence of diagnoses or referrals for the syndrome in their healthcare records. This might reflect unmet clinical needs and is in keeping with wider reports that those with long COVID face difficulties accessing care and having their symptoms recognised by GPs and other practitioners. It also indicates that wider research and planning based on using coded health records for long COVID may severely under-estimate the scale of patient reported illness.

Data from 6405 participants were used to investigate whether individuals with self-reported long COVID from Longitudinal Population Studies (LPS) between July 2020, and October 2021, had received a clinical code for long COVID diagnosis or referral in the English health-care system after 20-32 months of follow-up.

All eight LPS data were combined and accessed in the UK Longitudinal Linkage Collaboration (UK LLC) Trusted Research Environment:

  • 1970 British Cohort Study
  • Child of the New Century
  • Children of the 90s (ALSPAC)
  • National Survey of Health and Development
  • Next Steps
  • National Child Development Study
  • Twins UK
  • Understanding Society

This allowed researchers to consider enough participants reporting long COVID to make their findings accurate. UK LLC hosts these de-identified data from longitudinal population studies and links these to study participant’s healthcare records. De-identified data have personal details (those which can directly identify a person such as their name and address) removed. This is done by replacing or removing these direct identifiers, such as name and full date of birth. Findings Findings showed that out of 896 people who reported long COVID to their longitudinal population study, only 5.4% had long COVID diagnoses in their health records.

These findings therefore illustrates that 94.6% selfreported long COVID but had no evidence of diagnosis.

Read the full study, lay summary and first author blog here.