Improving the representativeness of UK’s national COVID-19 Infection Survey through spatio-temporal regression and post-stratification

Add to your list(s) Download to your calendar using vCal

• Koen Pouwels, Oxford
• Friday 17 February 2023, 12:00-13:00
• Zoom.

If you have a question about this talk, please contact Paula Smith.

To inform decisions around SARS CoV2 mitigation and vaccination measures, policy makers ideally would have population-representative estimates of how many people are infected with SARS CoV2 and have antibody levels above meaningful thresholds in small areas at different time points. Here, we apply spatiotemporal regression and post-stratification models to UK’s national COVID19 Infection Survey (CIS) to obtain representative temporal estimates of PCR positivity and antibody prevalence a the (sub)national level, and for different ages and ethnicities. Between Monday 7 December 2020 and Wednesday 4 May 2022, 6,496,052 PCR test and 1,941,333 anti-spike IgG antibody tests results were available for analyses. Not accounting for vaccination status – as done in virtually all outputs from the ONS - through post-stratification led to small underestimation of PCR positivity, but more substantial overestimations of antibody levels in the population. There was marked variation in the contribution of different areas and age-groups to each wave. Future analyses of infectious disease surveys should take into major drivers of outcomes of interest, with vaccination being an important factor to consider. A large part of the variation in the ranking of small-areas in terms of their SARS CoV2 prevalence could by explained by the degree of urbanicity and deprivation, highlighting the inequality in risk of SARS CoV2 infections.

This talk is part of the Worms and Bugs series.

This talk is included in these lists:

• All CMS events
• All Talks (aka the CURE list)
• Cambridge Immunology
• Cambridge Infectious Diseases
• Worms and Bugs
• Zoom
• bld31

Note that ex-directory lists are not shown.