DOI: 10.12688/f1000research.183931.1 ISSN: 2046-1402

Ethnic reporting and representation in UK COVID-19 consenting observational studies; a systematic review

Naomi V Bradbury, Sarah Booth, Laura J Gray, Angus Jennings, Sangyu Lee, Daniel S March, Urvi Modha, Elnaz Saeedi, Rahma Said, Aiden Smith, Rachael Stannard, Lucy Teece
Show PDF Cite
Background Evidence suggests ethnic minority populations are disproportionately impacted by COVID-19, however ethnic minority groups are often under-represented in research with ethnicity associations not being investigated. Observational studies are an important tool for understanding the health inequalities of ethnic minority populations. We aimed to assess the reporting of ethnic composition and representation of ethnic minority groups in COVID-19 consented observational studies in the UK. Methods A systematic review of COVID-19 studies was conducted in OVID MedLine electronic database (1 st January 2020-22 nd November 2022). Observational, opt-in consenting studies of COVID-19 exposures and outcomes with a minimum of 96 participants in the UK only were eligible. Studies that were non-medical, COVID-19 vaccination studies and studies investigating the impact of COVID-19 policies were excluded as were studies that registered only children (<18 years). Data analysis was conducted in R v4.3.3. Results This systematic review comprised of 138 studies of which 100 (72.5%) reported participant ethnicity, most commonly through self-reporting by participants. Only 15 of the studies reporting ethnicity (15%) used the recognised ONS ethnicity categories for reporting participant ethnicity. Studies acquired their data either through self-conducted studies or from large, observational study databases. UK-Biobank data was the most common database being used for 69 (50%) of the studies. Conclusion Reporting of participant ethnicity in consenting, observational COVID-19 studies in the UK was found to be inconsistent or missing despite early evidence of ethnic minority groups being disproportionately affected by the disease and most studies having access to, at a minimum, ONS five high-level ethnicity data through their use of large databases such as UK-Biobank. This made quantitative assessment of representation challenging. However, many studies under-represented ethnic minority participants compared to the 2021 England and Wales Census data.

More from our Archive