Estimating COVID-19 incidence and prevalence using lateral flow tests in England and Scotland, 2023-2024

TL;DR

This study developed a statistical method to estimate SARS-CoV-2 infection rates in England and Scotland using lateral flow tests during 2023-2024.

Contribution

The study introduces a novel statistical method to adjust for false negatives in LFD tests for accurate incidence and prevalence estimation.

Findings

The WCIS survey enabled fortnightly estimates of community SARS-CoV-2 infection levels.

LFD tests were found useful for large-scale surveillance despite their high false negative rate.

The method allowed stratified analysis by age, location, and sex.

Abstract

SARS-CoV-2 continues to cause substantial morbidity and mortality, particularly in winter. During the SARS-CoV-2 pandemic, community prevalence surveys provided detailed monitoring of infection levels. The Winter Coronavirus (COVID-19) Infection Survey (WCIS), conducted in England and Scotland from the 14th November 2023 to the 7th March 2024, enabled the UK Health Security Agency to publish fortnightly estimates of community infection levels in England and Scotland. Unlike previous community prevalence surveys, WCIS conducted testing using Lateral Flow Device (LFD) tests, and featured a repeat testing design that enabled estimation of key epidemiological parameters. LFD tests have a substantially lower cost per unit than Polymerase Chain Reaction (PCR) tests which were used in previous SARS-CoV-2 prevalence surveys; however, they have a high false negative rate that must be accounted for to produce reliable estimates. In this manuscript, statistical methods to robustly estimate incidence and prevalence while adjusting for time-varying false negative rates are developed. This enabled timely and robust inference of the incidence and prevalence of SARS-CoV-2, stratified by age group, location and sex. Overall, the study design of WCIS overcame key limitations of earlier large-scale community prevalence studies and demonstrated the utility of LFD tests in infectious disease surveillance.

Estimating respiratory infection rates in the community is challenging as testing is usually limited to people with more severe infections. Here, the authors develop a statistical method to estimate infection rates using data from a community survey that performed lateral flow testing in England and Scotland in 2023-24.

Figures