Skip to main content

Community transmission and viral load kinetics of the SARS-CoV-2 delta (B.1.617.2) variant in vaccinated and unvaccinated individuals in the UK: a prospective, longitudinal, cohort study

Our take —

Most SARS-CoV-2 transmission occurs in households. There has been scant high-quality evidence to date about how vaccination status (of both the index case and household contacts) affects household transmission risk since Delta has become the dominant variant. In this community-based study of close contacts of people with COVID-19 in the UK, the authors used daily sampling up to 20 days after initial exposure to estimate differences in the secondary attack rate (the proportion of contacts who become infected) by vaccination status and strain, and to test whether viral loads change differently over time in infected individuals depending on vaccination status and strain. For those contacts exposed to the Delta variant within a household, 25% of fully vaccinated individuals became infected compared to 38% of unvaccinated individuals. Vaccine effectiveness within a household with a delta-infected case was estimated to be 34%. Secondary attack rates were similar regardless of whether the initial case was vaccinated or not. There were no meaningful differences in peak viral load by vaccination status or strain. This study shows that the Delta strain of SARS-CoV-2 can transmit efficiently even within households in which cases and contacts are fully vaccinated; however, vaccinated contacts had a lower secondary attack rate, and this study was not designed to estimate differences in severe disease by vaccination status.

Study design

Prospective Cohort

Study population and setting

This community-based study from the UK was designed to estimate the effect of vaccination status on index case viral load trajectories and close contacts’ risk of infection, particularly with respect to the Delta variant of SARS-CoV-2, between September 13, 2020 and September 15, 2021. The primary aims of the study were: 1) to compare the secondary attack rate (SAR) in households by vaccination status of the infected case and the household contact;, and 2) to compare peak viral load and viral growth/decline rates by vaccination status and variant. Through the UK contact tracing system, 602 close contacts (household and non-household, aged 5 years and older) of 471 people with  COVID-19 were identified; 19 of these index cases were also enrolled in this study. Contacts and the 19 index cases provided 8,145 daily upper respiratory tract samples up to 20 days after exposure, beginning a median of 4 days after exposure. Eligible contacts were identified within 5 days of initial exposure, and agreed to self-swabbing of the upper respiratory tract. Unvaccinated/partially vaccinated/fully vaccinated individuals were defined by having 0/1/2 vaccine doses, respectively, at least 7 days before study enrollment. Log peak viral load, viral growth rates, and viral decline rates were estimated for each group using Bayesian hierarchical models and compared with posterior probabilities of a difference in mean values.

Summary of Main Findings

Of the 621 participants, including the 19 index cases, 163 (26%) had detectable SARS-CoV-2 RNA (55% female, median age 36, 82% white). Of these, 71 (44%) had the Delta variant. All infected participants had non-severe ambulatory illness or were asymptomatic. The secondary attack rate for household contacts exposed to the delta variant was 25% (95% CI: 18% to 33%) for fully vaccinated individuals and 38% (24% to 53%) for unvaccinated individuals. Infected contacts had a longer time period between their second dose and study enrollment relative to uninfected contacts (101 days vs. 64 days). Household secondary attack rates were similar regardless of whether the index case was fully vaccinated. Estimated vaccine effectiveness against the Delta variant within a household was estimated at 34% (-15% to 60%). There were no statistically significant differences by vaccination status or variant in the peak viral load; older age was associated with higher peak viral load. Fully vaccinated individuals with the Delta variant had a faster rate of viral decline (posterior probability >0.75) than unvaccinated individuals with the pre-alpha, alpha, or delta variants. There were 12 cases of transmission of the Delta variant between a fully vaccinated index case and a fully vaccinated household contact (out of 43 such pairs, for an attack rate of 28%).

Study Strengths

This study employed daily sequential testing of individuals for up to 20 days after exposure, permitting a high degree of sensitivity in identifying transmission events and allowing viral load kinetics modeling. Whole genome sequencing allowed comparisons by variant.


There were fairly small numbers of individuals in individual strata defined by vaccination status and variant. Only symptomatic index cases were used to identify close contacts; non-symptomatic cases may have different risks of onward transmission. It is possible that in some cases, the defined index case within a household was infected by another household member who was not identified as having been infected. Trajectories of viral load could not be observed for those participants with prevalent infection at study entry. Unvaccinated contacts were younger, on average, than vaccinated contacts (because of the progressive rollout of vaccination in the UK), which may have confounded estimates of peak viral load toward the null, but would have biased SAR estimates only if age is related to susceptibility to infection.

Value added

This study employed well-characterized data on vaccination status, nature of contact, variant type, and longitudinally collected sampling of the upper respiratory tract to provide estimates of household transmission risk by vaccination status for the Delta variant.

This review was posted on: 15 December 2021