The challenge

When Covid-19 started to take hold in the UK, it was difficult to know how widespread the epidemic had become. With little to no testing in the community, the earliest indications came from data on hospital admissions and deaths. But hospital data missed the huge proportion of people who had Covid symptoms but were not ill enough to be admitted. Data on hospital admissions and deaths also presented a delay that made it impossible to react quickly to changes in the pandemic - there are usually several weeks between a person contracting Covid and becoming seriously ill.

What we did

We counted suspected diagnoses of Covid-19 by searching key terms in the electronic health records held by GPs, using data from 1.2 million patients across east London. Our informaticians presented the data as interactive dashboards showing near real-time suspected infections by borough, Primary Care Network and individual GP practice. This knowledge gave local authorities an early indication of the rising epidemic.

Age, gender and ethnicity are already recorded in most GP records, so we could compare the rates of infection in different demographics, and between different localities. CEG produced one of the earliest research papers that demonstrated the population was not being affected equally by Covid-19. People from South Asian and Black ethnic backgrounds were catching the virus at a much higher rate than the white population. During the second wave, in Tower Hamlets, people of South Asian descent were dying from Covid-19 at three times the rate of the white population.

As the vaccination programme began, the primary care data also revealed South Asian and Black populations had far more vaccine hesitancy. In the over-80s in Tower Hamlets, for example, vaccine uptake was 71% in the white population but only 37% in the South Asian population and 26% in the Black population. Worryingly, the same groups at a higher risk of catching Covid-19 and dying from it were also the least likely to be vaccinated.

Reducing inequalities

We alerted our network of GPs, commissioners and local authorities across north east London and supported them to act on the evidence. Across north east London, our dashboards informed public health initiatives to build trust and understanding among communities where vaccine uptake was lowest. These included videos for social media, interactive webinars, a new vaccination centre at the East London Mosque, and working with imams to allay fears and address misinformation.

Six weeks later, the gap in vaccination uptake by ethnicity had already started to narrow. Vaccine uptake in Tower Hamlets stood at 86% in the white population, 71% in the South Asian population and 58% in the Black population.

National impact

Our figures were highlighted on national television and in Parliament, contributing to the wider picture and nationwide effort for vaccine equity.

Team: Sally Hull, Crystal Williams, Chris Carvalho, Kambiz Boomla and John Robson in collaboration with Mark Ashworth, King’s College London. Special thanks to Luke Readman, Regional Director of Digital Transformation at NHS London, and Somen Banerjee, Director of Public Health at Tower Hamlets Council, for their efforts in disseminating our analyses and enabling teams on the ground to act on the evidence.

Published research: Prevalence of suspected COVID-19 infection in patients from ethnic minority populations: a cross-sectional study in primary care.