Post by Nadica (She/Her) on Oct 29, 2024 5:48:19 GMT
The Incidence and Prevalence of SARS-CoV-2 in the UK Population from the UKHSA Winter COVID Infection Study - Preprint Posted Oct 23, 2024
Abstract
The SARS-CoV-2 virus continues to cause substantial morbidity and mortality, particularly during the winter period. The Winter Covid Infection Study (WCIS) ran from the 14th of November 2023 to the 7th of March 2024, and enabled the UK Health Security Agency to publish fortnightly estimates of the incidence and prevalence of SARS-CoV-2. Testing was performed using Lateral Flow Devices (LFD), and a repeat testing design was used to estimate key epidemiological parameters of SARS-CoV-2. This facilitated the estimation of time-varying prevalence, incidence, and test sensitivity. A Bayesian multilevel regression and poststratification model was developed to produce representative and unbiased estimates. In England and Scotland, prevalence peaked at 4.54% (95% CI: 3.90 to 5.24), and incidence peaked at 498 (95% CrI: 429 to 585) new infections per 100,000 individuals per day. The average LFD test sensitivity in England and Scotland during the study was estimated to be 72.1% (95% CrI: 70.3, 74.0), though due to epidemic phase bias this varied from a minimum value of 68.6% (95% CrI: 66.4 to 70.7) to a maximum value of 77.2% (95% CrI: 75.3 to 79.2). The novel study design of WCIS addressed key survey design challenges faced by previous large-scale SARS-CoV-2 population prevalence studies. The study demonstrated the utility and cost-benefit of LFD tests in large community surveys of prevalence.
Abstract
The SARS-CoV-2 virus continues to cause substantial morbidity and mortality, particularly during the winter period. The Winter Covid Infection Study (WCIS) ran from the 14th of November 2023 to the 7th of March 2024, and enabled the UK Health Security Agency to publish fortnightly estimates of the incidence and prevalence of SARS-CoV-2. Testing was performed using Lateral Flow Devices (LFD), and a repeat testing design was used to estimate key epidemiological parameters of SARS-CoV-2. This facilitated the estimation of time-varying prevalence, incidence, and test sensitivity. A Bayesian multilevel regression and poststratification model was developed to produce representative and unbiased estimates. In England and Scotland, prevalence peaked at 4.54% (95% CI: 3.90 to 5.24), and incidence peaked at 498 (95% CrI: 429 to 585) new infections per 100,000 individuals per day. The average LFD test sensitivity in England and Scotland during the study was estimated to be 72.1% (95% CrI: 70.3, 74.0), though due to epidemic phase bias this varied from a minimum value of 68.6% (95% CrI: 66.4 to 70.7) to a maximum value of 77.2% (95% CrI: 75.3 to 79.2). The novel study design of WCIS addressed key survey design challenges faced by previous large-scale SARS-CoV-2 population prevalence studies. The study demonstrated the utility and cost-benefit of LFD tests in large community surveys of prevalence.