Speaker
Description
Household-level vaccine uptake patterns are known to have serious impacts on infectious disease dynamics, with clusters of susceptibility in low-uptake households driving global transmission and potentially leading to inequitable burdens of disease\cite{geard_effects_2015}. Anticipating these uptake patterns could improve the design of vaccine rollout campaigns, but they remain poorly understood. The ONS COVID-19 Infection Survey (ONS-CIS) recorded regular COVID-19 test results from 2020-2023 for over 200,000 participating UK households along with vaccination dates and sociodemographic factors, offering an unparalleled source of data on individual- and household-level uptake behaviour. In this initial modelling analysis of vaccination data from ONS-CIS, we aim to quantify household-level clustering of uptake within the survey cohort. We compare the observed uptake patterns with two theoretical extremes: uniform random vaccination and a worst-case scenario in which entire households are either vaccinated or unvaccinated. To project the epidemiological impact of these patterns we calculate weekly estimates of the household reproductive ratio, a key invasion parameter for household epidemic models, over the course of the rollout period. We find that while the UK’s vaccine rollout was intermediate between the two extreme scenarios in terms of uptake, the likely effect on overall spread of infection appears to have been closer to the suboptimal by-household uptake scenario.
Bibliography
@article{geard_effects_2015,
title = {The effects of demographic change on disease transmission and vaccine impact in a household structured population},
volume = {13},
issn = {17554365},
url = {https://linkinghub.elsevier.com/retrieve/pii/S175543651500081X},
doi = {10.1016/j.epidem.2015.08.002},
language = {en},
urldate = {2026-03-14},
journal = {Epidemics},
author = {Geard, Nicholas and Glass, Kathryn and McCaw, James M. and McBryde, Emma S. and Korb, Kevin B. and Keeling, Matt J. and McVernon, Jodie},
month = dec,
year = {2015},
pages = {56--64},
}
