Mechanistic models of humoral kinetics following COVID-19 vaccination (preprint)

Introduction: Future COVID-19 vaccine programmes need to take into account the variable responses elicited by different vaccines and their waning protection over time. Existing descriptions of antibody response to COVID-19 vaccination convey limited information about the mechanisms of antibody production and maintenance. Methods: We describe the antibody dynamics elicited by COVID-19 vaccination with two biologically-motivated mathematical models of antibody production by plasma cells and subsequent decay. We fit the models using Markov Chain Monte Carlo to seroprevalence data from 14,602 uninfected individuals collected via the primary care network in England between May 2020 and September 2022. We ensure our models are structurally and practically identifiable when using anti- body data alone. We analyse the effect of age, vaccine type, number of doses, and the interval between doses on antibody production and longevity of response. Results: We find evidence that individuals over 35 years of age who received a second dose of ChAdOx1-S generate a persistent antibody response suggestive of long-lived plasma cell induction, while individuals that receive two doses of BNT162b2, or one dose of either vaccine do not. We also find that plasamblast productive capacity, the likely driver of short-term antibody responses, is greater in younger people than older people (≤ 4.5 fold change in point estimates), people vaccinated with two doses than people vaccinated with one dose (≤ 12 fold change), and people vaccinated with BNT162b2 than people vaccinated with ChAdOx1-S (≤ 440 fold change). The effect of age on antibody dynamics is more pronounced in people vaccinated with BNT162b2 than people vaccinated with ChAdOx1-S. We find the half-life of an antibody to be between 23 - 106 days. Conclusion: Routinely-collected seroprevalence data are a valuable source of information for characterising within-host mechanisms of antibody production and persistence. Extended sampling and linking seroprevalence data to outcomes would allow for powerful conclusions about how humoral kinetics protect against disease.

Syndromic case definitions for Lower Respiratory Tract Infection (LRTI) are less sensitive in older age: an analysis of symptoms among hospitalised adults (preprint)

Background Lower Respiratory Tract Infections (LRTI) pose a serious threat to older adults but may be underdiagnosed due to atypical presentations. Here we assess LRTI symptom profiles and syndromic (symptom-based) case ascertainment in older (≥65y) as compared to younger adults (<65y). Methods We included adults (≥18y) with confirmed LRTI admitted to two acute care Trusts in Bristol, UK from 1st August 2020- 31st July 2022.  Logistic regression was used to assess whether age ≥65y reduced the probability of meeting syndromic LRTI case definitions, using patients’ symptoms at admission. We also calculated relative symptom frequencies (log-odds ratios) and evaluated how symptoms were clustered across different age groups. Results Of 17,620 clinically confirmed LRTI cases, 8,487 (48.1%) had symptoms meeting the case definition. Compared to those not meeting the definition these cases were younger, had less severe illness and were less likely to have received a SARS-CoV-2 vaccination or to have active SARS-CoV-2 infection. Prevalence of dementia/cognitive impairment and levels of comorbidity were lower in this group. After controlling for sex, dementia and comorbidities, age ≥65y significantly reduced the probability of meeting the case definition (aOR=0.67, 95% CI:0.63-0.71). Cases aged ≥65y were less likely to present with fever and LRTI-specific symptoms (e.g., pleurisy, sputum) than younger cases, and those aged ≥85y were characterised by lack of cough but frequent confusion and falls. Conclusions LRTI symptom profiles changed considerably with age in this hospitalised cohort. Standard screening protocols may fail to detect older and frailer cases of LRTI based on their symptoms.

Relative vaccine effectiveness (rVE) of mRNA COVID-19 boosters in the UK vaccination programme, during the Spring-Summer (monovalent vaccine) and Autumn-Winter 2022 (bivalent vaccine) booster campaigns: a prospective test negative case-control study (preprint)

Background Understanding the relative vaccine effectiveness (rVE) of new COVID-19 vaccine formulations against SARS-CoV-2 infection is an urgent public health priority. A precise comparison of the rVE of monovalent and bivalent boosters given during the 2022 Spring-Summer and Autumn-Winter campaigns, respectively, in a defined population has not been reported. We therefore assessed rVE against hospitalisation for the Spring-Summer (fourth vs third monovalent mRNA vaccine doses) and Autumn-Winter (fifth BA.1/ancestral bivalent vs fourth monovalent mRNA vaccine dose) boosters. Methods A prospective single-centre test-negative design case-control study of [≥]75 year-olds hospitalised with COVID-19 or other acute respiratory disease. We conducted regression analyses controlling for age, gender, socioeconomic status, patient comorbidities, community SARS-CoV-2 prevalence, vaccine brand and time between baseline dose and hospitalisation. Results 682 controls and 182 cases were included in the Spring-Summer booster analysis; 572 controls and 152 cases for the Autumn-Winter booster analysis. A monovalent mRNA COVID-19 vaccine as fourth dose showed rVE 46*9% (95% confidence interval [CI] 14*4-67*3) versus those not boosted. A bivalent mRNA COVID-19 vaccine as fifth dose had rVE 46*4% (95%CI 17*5-65), compared to a fourth monovalent mRNA COVID-19 vaccine dose. Interpretation Both fourth monovalent and fifth BA.1/ancestral mRNA bivalent COVID-19 vaccine doses demonstrated benefit as a booster in older adults. Bivalent mRNA boosters offer equivalent protection against hospitalisation with Omicron infection to monovalent mRNA boosters given earlier in the year. These findings support the current UK immunisation programme that advises the use of bivalent booster doses.

Severity of Omicron (B.1.1.529) and Delta (B.1.1.617.2) SARS-CoV-2 infection among hospitalised adults: a prospective cohort study (preprint)

Limited data exist assessing severity of disease in adults hospitalised with Omicron SARS-CoV-2 variant infections, and to what extent patient-factors, including vaccination and pre-existing disease, affect variant-dependent disease severity. This prospective cohort study of all adults ([≥]18 years of age) hospitalised at acute care hospitals in Bristol, UK assessed disease severity using 3 different measures: FiO2 >28%, World Health Organization (WHO) outcome score >5, and hospital length of stay (LOS) >3 days following admission for Omicron or Delta variant infection. Independent of other variables, including vaccination, Omicron variant infection was associated with a statistically lower severity compared to Delta; risk reductions were 58%, 67%, and 16% for FiO2, WHO score, and LOS, respectively. Younger age and vaccination with two or three doses were also independently associated with lower COVID-19 severity. Despite lower severity relative to Delta, Omicron infection still resulted in substantial patient and public health burden following admission.

Severity of Omicron (B.1.1.529) and Delta (B.1.1.617.2) SARS-CoV-2 infection among hospitalised adults: a prospective cohort study (preprint)

Limited data exist assessing severity of disease in adults hospitalised with Omicron SARS-CoV-2 variant infections, and to what extent patient-factors, including vaccination and pre-existing disease, affect variant-dependent disease severity. This prospective cohort study of all adults (≥18 years of age) hospitalised at acute care hospitals in Bristol, UK assessed disease severity using 3 different measures: FiO2 >28%, World Health Organization (WHO) outcome score >5, and hospital length of stay (LOS) >3 days following admission for Omicron or Delta variant infection. Independent of other variables, including vaccination, Omicron variant infection was associated with a statistically lower severity compared to Delta; risk reductions were 58%, 67%, and 16% for FiO2, WHO score, and LOS, respectively. Younger age and vaccination with two or three doses were also independently associated with lower COVID-19 severity. Despite lower severity relative to Delta, Omicron infection still resulted in substantial patient and public health burden following admission.

Impact of voluntary risk-mitigation behaviour on transmission of the Omicron SARS-CoV-2 variant in England (preprint)

Abstract Background The Omicron variant of SARS-CoV-2 infection poses substantial challenges to public health. In England, "plan B" mitigation measures were introduced in December 2021 including increased home working and face coverings in shops, but stopped short of restrictions on social contacts. The impact of voluntary risk mitigation behaviours on future SARS-CoV-2 burden is unknown. Methods We developed a rapid online survey of risk mitigation behaviours during the winter 2021 festive period and deployed in two longitudinal cohort studies in the UK (Avon Longitudinal Study of Parents and Children (ALSPAC) and TwinsUK/Covid Symptom Study (CSS) Biobank) in December 2021. Using an individual-based, probabilistic model of COVID-19 transmission between social contacts with SARS-CoV-2 Omicron variant parameters and realistic vaccine coverage in England, we describe the potential impact of the SARS-CoV-2 Omicron wave in England in terms of the effective reproduction number and cumulative infections, hospital admissions and deaths. Using survey results, we estimated in real-time the impact of voluntary risk mitigation behaviours on the Omicron wave in England, if implemented for the entire epidemic wave. Results Over 95% of survey respondents (N_ALSPAC=2,686 and N_Twins=6,155) reported some risk mitigation behaviours, with being fully vaccinated and using home testing kits the most frequently reported behaviours. Less than half of those respondents reported that their behaviour was due to "plan B". We estimate that without risk mitigation behaviours, the Omicron variant is consistent with an effective reproduction number between 2.5 and 3.5. Due to the reduced vaccine effectiveness against infection with the Omicron variant, our modelled estimates suggest that between 55% and 60% of the English population could be infected during the current wave, translating into between 15,000 and 46,000 cumulative deaths, depending on assumptions about vaccine effectiveness. We estimate that voluntary risk reduction measures could reduce the effective reproduction number to between 1.8 and 2.2 and reduce the cumulative number of deaths by up to 24%. Conclusions We conclude that voluntary measures substantially reduce the projected impact of the SARS-CoV-2 Omicron variant, but that voluntary measures alone would be unlikely to completely control transmission.

Short-term Projections based on Early Omicron Variant Dynamics in England (preprint)

Throughout the ongoing COVID-19 pandemic, the worldwide transmission and replication of SARS-COV-2, the causative agent of COVID-19 disease, has resulted in the opportunity for multiple mutations to occur that may alter the virus transmission characteristics, the effectiveness of vaccines and the severity of disease upon infection. The Omicron variant (B.1.1.529) was first reported to the WHO by South Africa on 24 November 2021 and was declared a variant of concern by the WHO on 26 November 2021. The variant was first detected in the UK on 27 November 2021 and has since been reported in a number of countries globally where it is frequently associated with rapid increase in cases. Here we present analyses of UK data showing the earliest signatures of the Omicron variant and mathematical modelling that uses the UK data to simulate the potential impact of this variant in the UK. In order to account for the uncertainty in transmission advantage, vaccine escape and severity at the time of writing, we carry out a sensitivity analysis to assess the impact of these variant characteristics on future risk.

Impacts of vaccination and asymptomatic testing on SARS-CoV-2 transmission dynamics in a university setting (preprint)

We investigate the impact of vaccination and asymptomatic testing uptake on SARS-CoV-2 transmission in a university student population using a stochastic compartmental model. We find that the magnitude and timing of outbreaks is highly variable under different vaccine uptake levels. With low level interventions (no asymptomatic testing, 30% vaccinated), 53-71% of students become infected during the first term; with high interventions (90% using asymptomatic testing, 90% vaccinated) cumulative incidence is 7-9%, with around 80% of these cases estimated to be asymptomatic. Asymptomatic testing is most useful when vaccine uptake is low: when 30% of students are vaccinated, 90% uptake of asymptomatic testing leads to almost half the case numbers. Under high levels of vaccine uptake (70-90%), case numbers in the student population are largely driven by community importation. Our findings suggest that vaccination is critical for controlling SARS-CoV-2 transmission in university settings with asymptomatic testing being a useful supporting measure.

Estimating SARS-CoV-2 prevalence from large-scale wastewater surveillance: insights from combined analysis of 44 sites in England (preprint)

Accurate surveillance of the COVID-19 pandemic can be weakened by under-reporting of cases, particularly due to asymptomatic or pre-symptomatic infections, resulting in bias. Quantification of SARS-CoV-2 RNA in wastewater (WW) can be used to infer infection prevalence, but uncertainty in sensitivity and considerable variability has meant that accurate measurement remains elusive. Data from 44 sewage sites in England, covering 31% of the population, shows that SARS-CoV-2 prevalence is estimated to within 1.1% of estimates from representative prevalence surveys (with 95% confidence). Using machine learning and phenomenological models, differences between sampled sites, particularly the WW flow rate, influence prevalence estimation and require careful interpretation. SARS-CoV-2 signals in WW appear 4–5 days earlier in comparison to clinical testing data but are coincident with prevalence surveys suggesting that WW surveillance can be a leading indicator for asymptomatic viral infections. Wastewater-based epidemiology complements and strengthens traditional surveillance, with significant implications for public health.

Population disruption: estimating changes in population distribution in the UK during the COVID-19 pandemic (preprint)

Mobility data have demonstrated major changes in human movement patterns in response to COVID-19 and associated interventions in many countries. This can involve sub-national redistribution, short-term relocations as well as international migration. In this paper, we combine detailed location data from Facebook measuring the location of approximately 6 million daily active Facebook users in 5km2 tiles in the UK with census-derived population estimates to measure population mobility and redistribution. We provide time-varying population estimates and assess spatial population changes with respect to population density and four key reference dates in 2020 (First lockdown, End of term, Beginning of term, Christmas). We also show how the timing and magnitude of observed population changes can impact the size of epidemics using a deterministic model of COVID-19 transmission. We estimate that between March 2020 and March 2021, the total population of the UK has declined and we identify important spatial variations in this population change, showing that low-density areas have experienced lower population decreases than urban areas. We estimate that, for the top 10% highest population tiles, the population has decreased by 6.6%. Further, we provide evidence that geographic redistributions of population within the UK coincide with dates of non-pharmaceutical interventions including lockdowns and movement restrictions, as well as seasonal patterns of migration around holiday dates. The methods used in this study reveal significant changes in population distribution at high spatial and temporal resolutions that have not previously been quantified by available demographic surveys in the UK. We found early indicators of potential longer-term changes in the population distribution of the UK although it is not clear how these changes may persist after the COVID-19 pandemic.

Early epidemiological signatures of novel SARS-CoV-2 variants: establishment of B.1.617.2 in England (preprint)

The rapid emergence of SARS-CoV-2 mutants with new phenotypic properties is a critical challenge to the control of the ongoing pandemic. B.1.1.7 was monitored in the UK through routine testing and S-gene target failures (SGTF), comprising over 90% of cases by March 2021. Now, the reverse is occurring: SGTF cases are being replaced by an S-gene positive variant, which we associate with B.1.617.2. Evidence from the characteristics of S-gene positive cases demonstrates that, following importation, B.1.617.2 is transmitted locally, growing at a rate higher than B.1.1.7 and a doubling time between 5-14 days. S-gene positive cases should be prioritised for sequencing and aggressive control in any countries in which this variant is newly detected.

Vaccine escape in a heterogeneous population: insights for SARS-CoV-2 from a simple model (preprint)

As a counter measure to the SARS-CoV-2 pandemic there has been swift development and clinical trial assessment of candidate vaccines, with subsequent deployment as part of mass vaccination campaigns. However, the SARS-CoV-2 virus has demonstrated the ability to mutate and develop variants, which can modify epidemiological properties and potentially also the effectiveness of vaccines. The widespread deployment of highly effective vaccines may rapidly exert selection pressure on the SARS-CoV-2 virus directed towards mutations that escape the vaccine induced immune response. This is particularly concerning whilst infection is widespread. By developing and analysing a mathematical model of two population groupings with differing vulnerability and contact rates, we explore the impact of the deployment of vaccine amongst the population on R, cases, disease abundance and vaccine escape pressure. The results from this model illustrate two insights (i) vaccination aimed at reducing prevalence could be more effective at reducing disease than directly vaccinating the vulnerable; (ii) the highest risk for vaccine escape can occur at intermediate levels of vaccination. This work demonstrates a key principle that the careful targeting of vaccines towards particular population groups could reduce disease as much as possible whilst limiting the risk of vaccine escape.

Limits of lockdown: characterising essential contacts during strict physical distancing (preprint)

COVID-19 has exposed health inequalities within countries and globally. The fundamental determining factor behind an individuals risk of infection is the number of social contacts they make. In many countries, physical distancing measures have been implemented to control transmission of SARS-CoV-2, reducing social contacts to a minimum. Characterising unavoidable social contacts is key for understanding the inequalities behind differential risks and planning vaccination programmes. We utilised an existing English longitudinal birth cohort, which is broadly representative of the wider population (n=6807), to explore social contact patterns and behaviours when strict physical distancing measures were in place during the UKs first lockdown in March-May 2020. Essential workers, specifically those in healthcare, had 4.5 times as many contacts as non-essential workers [incident rate ratio = 4.42 (CI95%: 3.88-5.04)], whilst essential workers in other sectors, mainly teaching and the police force had three times as many contacts [IRR = 2.84 (2.58-3.13)]. The number of individuals in a household, which is conflated by number of children, increases essential social contacts by 40%. Self-isolation effectively reduces numbers of contacts outside of the home, but not entirely. Together, these findings will aid the interpretation of epidemiological data and impact the design of effective SARS-CoV-2 control strategies, such as vaccination, testing and contact tracing.

Assessing the Effectiveness of BNT162b2 and ChAdOx1nCoV-19 COVID-19 Vaccination in Prevention of Hospitalisations in Elderly and Frail Adults: A Single Centre Test Negative Case-Control Study (preprint)

Background: On 8th December 2020, deployment of the first vaccine against SARS-CoV-2 authorised for UK use, the mRNA-based vaccine BNT162b2, began, followed by the adenoviral vector vaccine ChAdOx1nCoV-19 on 4th January 2021. Initially care home-residents and staff, frontline healthcare workers and adults from age 80 were targeted. In phase 3 trials, BNT162b2 and ChAdOx1nCoV-19 demonstrated 95% and 70% efficacy, respectively, after two doses against symptomatic SARS-CoV-2 infection. However, few data exist regarding the effectiveness of these vaccines in elderly frail people. Evaluation following implementation to determine the effectiveness of one dose in reducing hospitalisations due to SARS-CoV-2 infection in elderly adults is urgent.Methods: A prospective single-centre test-negative design case-control study of adults aged ≥80 years hospitalised with COVID-19 disease or other acute respiratory disease. We conducted logistic regression controlling for time (week), gender, index of multiple deprivations (IMD), and care residency status (CRS), and sensitivity analyses matched for time and gender using a conditional logistic model adjusting for IMD and CRS.Findings: First dose vaccine effectiveness of BNT162b2 was 71.4% (95% confidence interval [CI] 46.5-90.6). ChAdOx1nCoV-19 first dose vaccine effectiveness was 80.4% (95% CI 36.4-94.5). When effectiveness analysis for BNT162b2 was restricted to the period covered by ChAdOx1nCoV-19, the estimate was 79.3% (95% CI 47.0-92.5).Interpretation: A single dose of either BNT162b2 or ChAdOx1nCoV-19 vaccine resulted in substantial reductions in the risk of COVID-19-related hospitalisation in elderly, frail patients with extensive co-morbid disease.Funding: The AvonCAP study is an investigator-led project funded under a collaborative agreement by Pfizer.Conflict of Interest: CH is Principal Investigator of the Avon CAP study which is an investigator-led University of Bristol study funded by Pfizer and has previously received support from the NIHR in an Academic Clinical Fellowship. JO is a Co-Investigator on the Avon CAP Study. AF is a member of the Joint Committee on Vaccination and Immunization (JCVI) and chair of the World Health Organization European Technical Advisory Group of Experts on Immunization (ETAGE) committee. In addition to receiving funding from Pfizer as Chief Investigator of this study, he leads another project investigating transmission of respiratory bacteria in families jointly funded by Pfizer and the Gates Foundation. The other authors have no relevant conflicts of interest to declare.Ethical Approval: The study was approved by the Health Research Authority Research Ethics Committee (East of England, Essex), REC 20/EE/0157, including data collection under Section 251 of the 2006 NHS Act authorised by the Confidentiality Advisory Group.

Increased hazard of mortality in cases compatible with SARS-CoV-2 variant of concern 202012/1 - a matched cohort study (preprint)

ObjectivesTo establish whether there is any change in mortality associated with infection of a new variant of SARS-CoV-2 (VOC-202012/1), first detected in UK in December 2020, compared to that associated with infection with circulating SARS-CoV-2 variants. DesignMatched cohort study. Cases are matched by age, gender, ethnicity, index of multiple deprivation, lower tier local authority region, and sample date of positive specimen, and differing only by detectability of the spike protein gene using the TaqPath assay - a proxy measure of VOC-202012/1 infection. SettingUnited Kingdom, Pillar 2 COVID-19 testing centres using the taqPath assay. Participants54,773 pairs of participants testing positive for SARS-CoV-2 in Pillar 2 between 1st October 2020 and 29th January 2021. Main outcome measures - Death within 28 days of first positive SARS-CoV-2 test. ResultsThere is a high probability that the risk of mortality is increased by infection with VOC-202012/01 (p <0.001). The mortality hazard ratio associated with infection with VOC-202012/1 compared to infection with previously circulating variants is 1.7 (95% CI 1.3 - 2.2) in patients who have tested positive for COVID-19 in the community. In this comparatively low risk group, this represents an increase of deaths from 1.8 in 1000 to 3.1 in 1000 detected cases. ConclusionsIf this finding is generalisable to other populations, VOC-202012/1 infections have the potential to cause substantial additional mortality over and previously circulating variants. Healthcare capacity planning, national and international control policies are all impacted by this finding, with increased mortality lending weight to the argument that further coordinated and stringent measures are justified to reduce deaths from SARS-CoV-2.

Contact patterns before and during the UK's Autumn 2020 COVID-19 lockdown among university students and staff (preprint)

Introduction Contact patterns are an important determinant of infection transmission. CONQUEST (COroNavirus QUESTionnaire) is an online survey of contacts for University of Bristol (UoB) staff and students. Results from 23/06/2020-24/11/2020 are used to investigate contact patterns among staff/students throughout various UK COVID-19 guidance periods. Methods Responses captured self-reported contacts on the previous day, with COVID-19 guidance periods split: post-first lockdown (23/06/2020-03/07/2020), relaxed guidance period (04/07/2020-13/09/2020), "rule-of-six" period (14/09/2020-04/11/2020), and the second lockdown (05/11/2020-25/11/2020). Results 722 staff (4199 responses) (mean household size: 2.6) and 738 students (1906 responses) (mean household size: 4.5) were included in the study. For staff, median contacts on the previous day were higher in the relaxed guidance and "rule-of-six" periods (3) than the post-lockdown and second lockdown periods (2). Mean contacts were higher during these two middle periods (4-6 per week) than the post-lockdown and second lockdown periods (~3). Mean contacts dropped between the last two periods (5.4 to 3.3), driven by a mean reduction in contacts in non-home/university locations (2.9 to 1.2). Few students responded until October. After 05/10/2020, the median was 2 and the mean was around 6-7 each week, until the second lockdown when it dropped to ~4. Mean number of contacts dropped between the last two periods (6.3 to 4.0) driven by a reduction in all contact types, including UoB contacts (3.5 to 2.5). Discussion The mean and median number of contacts for UoB staff and students were low compared to pre-COVID-19 studies throughout the survey period and lowest during the second lockdown.

Modelling pooling strategies for SARS-CoV-2 testing in a university setting (preprint)

Pre-symptomatic and asymptomatic transmission of SARS-CoV-2 are important elements in the Covid-19 pandemic, and until vaccines are made widely available there remains a reliance on testing to manage the spread of the disease, alongside non-pharmaceutical interventions such as measures to reduce close social interactions. In the UK, many universities opened for blended learning for the 2020-2021 academic year, with a mixture of face to face and online teaching. In this study we present a simulation framework to evaluate the effectiveness of different asymptomatic testing strategies within a university setting, across a range of transmission scenarios. We show that when positive cases are clustered by known social structures, such as student households, the pooling of samples by these social structures can substantially reduce the total cost of conducting RT-qPCR tests. We also note that routine recording of quantitative RT-qPCR results would facilitate future modelling studies.

Household bubbles and COVID-19 transmission: insights from percolation theory (preprint)

BackgroundIn the era of social distancing to curb the spread of COVID-19, bubbling is the combining of two or more households to create an exclusive larger group. The impact of bubbling on COVID-19 transmission is challenging to quantify because of the complex social structures involved. MethodsWe developed a network description of households in the UK, using the configuration model to link households. We explored the impact of bubbling scenarios by joining together households of various sizes. For each bubbling scenario, we calculated the percolation threshold, that is, the number of connections per individual required for a giant component to form, numerically and theoretically. We related the percolation threshold to the household reproduction number. ResultsWe find that bubbling scenarios in which single-person households join with another household has a minimal impact on network connectivity and transmission potential. Ubiquitous scenarios where all households form a bubble are likely to lead to extensive transmission that is hard to control. The impact of plausible scenarios, with variable uptake and heterogeneous bubble sizes, can be mitigated with reduced numbers of contacts outside the household. ConclusionsBubbling of households comes at an increased risk of transmission, however under certain circumstances risks can be modest and could be balanced by other changes in behaviours.

Contacts and behaviours of university students during the COVID-19 pandemic at the start of the 2020/21 academic year (preprint)

CONQUEST (COroNavirus QUESTionnaire) is an online survey of contacts, behaviour, and COVID-19 symptoms for University of Bristol (UoB) staff/students. We analysed survey results from the start of the 2020/2021 academic year, prior to the second national lockdown (14/09/2020-01/11/2020), where COVID-19 outbreaks led to lockdown of some student halls of residence. The aim of these analyses was to enhance knowledge of student contact patterns to inform infection disease mathematical modelling approaches. Responses captured information on demographics, contacts on the previous day, symptoms and self-isolation during the prior week, and COVID-19 status. 740 students provided 1261 unique records. Of 42 (3%) students testing positive in the prior fortnight, 99% had been self-isolating. The median number of contacts on the previous day was 2 (interquartile range: 1-5), mode: 1, mean: 6.1; 8% had [≥]20 contacts. 57% of student contacts were other UoB students/staff. Most students reported few daily contacts but there was heterogeneity, and some reported many. Around 40% of student contacts were with individuals not affiliated with UoB, indicating potential for transmission to non-students/staff.