Vaccine effectiveness of the first dose of ChAdOx1 nCoV-19 and BNT162b2 against SARS-CoV-2 infection in residents of long-term care facilities in England (VIVALDI): a prospective cohort study
Summary
Background
The effectiveness of SARS-CoV-2 vaccines in older adults living in long-term care facilities is uncertain. We investigated the protective effect of the first dose of the Oxford-AstraZeneca non-replicating viral-vectored vaccine (ChAdOx1 nCoV-19; AZD1222) and the Pfizer-BioNTech mRNA-based vaccine (BNT162b2) in residents of long-term care facilities in terms of PCR-confirmed SARS-CoV-2 infection over time since vaccination.
Methods
The VIVALDI study is a prospective cohort study that commenced recruitment on June 11, 2020, to investigate SARS-CoV-2 transmission, infection outcomes, and immunity in residents and staff in long-term care facilities in England that provide residential or nursing care for adults aged 65 years and older. In this cohort study, we included long-term care facility residents undergoing routine asymptomatic SARS-CoV-2 testing between Dec 8, 2020 (the date the vaccine was first deployed in a long-term care facility), and March 15, 2021, using national testing data linked within the COVID-19 Datastore. Using Cox proportional hazards regression, we estimated the relative hazard of PCR-positive infection at 0–6 days, 7–13 days, 14–20 days, 21–27 days, 28–34 days, 35–48 days, and 49 days and beyond after vaccination, comparing unvaccinated and vaccinated person-time from the same cohort of residents, adjusting for age, sex, previous infection, local SARS-CoV-2 incidence, long-term care facility bed capacity, and clustering by long-term care facility. We also compared mean PCR cycle threshold (Ct) values for positive swabs obtained before and after vaccination. The study is registered with ISRCTN, number 14447421.
Findings
10 412 care home residents aged 65 years and older from 310 LTCFs were included in this analysis. The median participant age was 86 years (IQR 80–91), 7247 (69·6%) of 10 412 residents were female, and 1155 residents (11·1%) had evidence of previous SARS-CoV-2 infection. 9160 (88·0%) residents received at least one vaccine dose, of whom 6138 (67·0%) received ChAdOx1 and 3022 (33·0%) received BNT162b2. Between Dec 8, 2020, and March 15, 2021, there were 36 352 PCR results in 670 628 person-days, and 1335 PCR-positive infections (713 in unvaccinated residents and 612 in vaccinated residents) were included. Adjusted hazard ratios (HRs) for PCR-positive infection relative to unvaccinated residents declined from 28 days after the first vaccine dose to 0·44 (95% CI 0·24–0·81) at 28–34 days and 0·38 (0·19–0·77) at 35–48 days. Similar effect sizes were seen for ChAdOx1 (adjusted HR 0·32, 95% CI 0·15–0·66) and BNT162b2 (0·35, 0·17–0·71) vaccines at 35–48 days. Mean PCR Ct values were higher for infections that occurred at least 28 days after vaccination than for those occurring before vaccination (31·3 [SD 8·7] in 107 PCR-positive tests vs 26·6 [6·6] in 552 PCR-positive tests; p<0·0001).
Interpretation
Single-dose vaccination with BNT162b2 and ChAdOx1 vaccines provides substantial protection against infection in older adults from 4–7 weeks after vaccination and might reduce SARS-CoV-2 transmission. However, the risk of infection is not eliminated, highlighting the ongoing need for non-pharmaceutical interventions to prevent transmission in long-term care facilities.
Funding
UK Government Department of Health and Social Care.
Introduction
International reports on COVID-19 and long-term care.
The UK has prioritised vaccination of residents and staff in long-term care facilities
to reduce the risk of COVID-19-related morbidity and mortality in this population, with the expectation that this will facilitate the relaxation of social restrictions.
Evidence before this study
We searched MEDLINE and medRxiv for studies evaluating SARS-CoV-2 vaccine effectiveness in residents of long-term care facilities that were published in English between Jan 1, 2020, and March 11, 2021. We used variations of the search terms “COVID-19” AND “vaccine effectiveness” OR “vaccine efficacy” AND “care homes” OR “long term care facilities” OR “older people”. We identified one preprint article concerning long-term care facilities in Denmark, which reported that a single dose of BNT162b2 (rINN tozinameran) was ineffective against SARS-CoV-2 infection in residents; however, participants received the second vaccine dose a median of 24 days after the first dose, which could be too soon to capture the protective effects of a single vaccine dose. Additionally, we identified two preprint reports of studies that evaluated vaccine effectiveness against symptomatic infection and admission to hospital among older adults in the community. The first of these preprints found 81% vaccine effectiveness against COVID-19-related admission to hospital at 28–34 days after a single dose of BNT162b2 or ChAdOx1 (AZD1222) in those aged 80 years and older. The second of these preprints found vaccine effectiveness against symptomatic infection of 60% at 28–34 days and 73% at 35 days and beyond after a single dose of ChAdOx1 in those aged 70 years and older. Although further reports from cohorts in long-term care facilities have become available, we identified no study that focused on the effectiveness of a single vaccine dose against infection among long-term care facility residents at more than 4 weeks after vaccination—a particularly important research question in the context of the UK policy decision to extend the dosing interval beyond 3 weeks.
Added value of this study
In this prospective cohort study in 10 412 residents aged 65 years and older from 310 long-term care facilities across England, we estimated vaccine effectiveness to be 56% (95% CI 19–76) at 28–34 days and 62% (23–81) at 35–48 days after a single dose of ChAdOx1 or BNT162b2. Our findings suggest that the risk of SARS-CoV-2 infection is substantially reduced from 28 days after the first dose of either vaccine, and this effect is maintained up to at least 7 weeks after vaccination, with similar protection offered by both vaccine types. We also found that PCR cycle threshold values, which are negatively associated with the ability to isolate virus, were significantly higher in infections occurring at 28 days or longer after vaccination than in infections that occurred during the unvaccinated period, suggesting that vaccination might reduce onward transmission of SARS-CoV-2 from individuals with breakthrough infections. In addition to examining vaccine effectiveness in the context of an extended dosing interval and focusing on the frail and older long-term care facility resident population, our findings constitute some of the earliest real-world evidence on vaccine effectiveness against infection for ChAdOx1, in any age group. We can also infer that both vaccines are effective against the B.1.1.7 variant, because our analysis period coincided with the rapid emergence of B.1.1.7 in England during the second wave of the pandemic.
Implications of all the available evidence
Our findings add to the growing body of evidence on the protective effect of the BNT162b2 vaccines in residents in long-term care facilities and show the effectiveness of ChAdOx1 in this vulnerable population. Evaluating single-dose vaccine effectiveness has become increasingly important considering the extended dosing intervals that have been implemented across many countries to maximise vaccine coverage across high-risk groups. Further research is required to evaluate the effectiveness of the first vaccine dose at 8–12 weeks, as well as after the second dose, and to evaluate the long-term impact of vaccination on SARS-CoV-2 infection, transmission, and mortality in residents in long-term care facilities. Such research will inform policy decisions regarding the ongoing need for disease control measures in long-term care facilities, such as visitor restrictions, which continue to have a detrimental impact on the wellbeing of residents, their relatives, and staff.
Both are spike protein-based vaccines that showed high efficacy (62·1–95·0%) against symptomatic infection in phase 3 clinical trials when following a two-dose schedule.
- Polack FP
- Thomas SJ
- Kitchin N
- et al.
,
- Voysey M
- Clemens SAC
- Madhi SA
- et al.
However, trials for both vaccines enrolled mostly young, healthy adults. Vaccine efficacy data from frail, older adults requiring long-term care are scarce because these individuals are routinely excluded from clinical studies and vaccine trials.
- Prendki V
- Tau N
- Avni T
- et al.
Consequently, trial estimates of vaccine efficacy might not be generalisable to long-term care facility residents because of age-related differences in vaccine-induced immune responses.
- Gustafson CE
- Kim C
- Weyand CM
- Goronzy JJ
,
- Ciabattini A
- Nardini C
- Santoro F
- Garagnani P
- Franceschi C
- Medaglini D
,
- Fulop T
- Pawelec G
- Castle S
- Loeb M
Observational data from post-licensure studies in older adults are emerging,
- Lopez Bernal J
- Andrews N
- Gower C
- et al.
,
- Vasileiou E
- Simpson CR
- Shi T
- et al.
,
- Pritchard E
- Matthews PC
- Stoesser N
- et al.
and although a small number of preprint articles have reported on populations in long-term care facilities,
- Moustsen-Helms IR
- Emborg H-D
- Nielsen J
- et al.
,
- Britton A
- Jacobs Slifka KM
- Edens C
- et al.
,
- Cabezas C
- Coma E
- Mora-Fernandez N
- et al.
,
- Monge S
- Olmedo C
- Alejos B
- Lapena MF
- Sierra MJ
- Limia A
the study populations were exclusively vaccinated with BNT162b2 at the manufacturer-recommended dosing interval, and regular asymptomatic screening was rarely done.
- Voysey M
- Costa Clemens SA
- Madhi SA
- et al.
On Dec 31, 2020, the UK Joint Committee on Vaccination and Immunisation advised that dosing intervals could be extended up to 12 weeks to optimise first-dose coverage.
This decision was made in the context of rapidly increasing SARS-CoV-2 incidence, associated with the emergence of the highly transmissible B.1.1.7 variant and its subsequent spread within long-term care facilities from November, 2020.
,
- Krutikov M
- Hayward A
- Shallcross L
This policy has made it increasingly important to understand the extent and duration of protection against infection afforded by the first dose of each vaccine, and whether single-dose vaccination has an effect on transmission.
We analysed data from our prospective observational cohort study to investigate the protective effect of the first dose of ChAdOx1 and BNT162b2 vaccines in residents of long-term care facilities aged 65 years and older, comparing the relative hazards of PCR-confirmed SARS-CoV-2 infection and mean PCR cycle threshold (Ct) values between vaccinated and unvaccinated residents by time since vaccination.
Methods
Study design and setting
- Krutikov M
- Palmer T
- Donaldson A
- et al.
The study included care homes that were managed by for-profit and not-for-profit providers and independent long-term care facilities from all regions in England. Eligible long-term care facilities were identified by the care provider’s senior management team or by the National Institute for Health Research Clinical Research Network.
Long-term care facility residents undergo monthly routine PCR testing and, if an outbreak is suspected, local public health teams organise PCR testing for all residents upon notification and 7 days later. Individuals who test positive are not retested for the following 90 days unless they develop new COVID-19 symptoms.
Symptom information is collected at the point of testing but its reliability is uncertain in this frail population.
Although vaccination commenced in December, 2020, in the UK, most first vaccinations in the study cohort occurred in January, 2021 (appendix p 2).
Residents were eligible for inclusion in the study if they had at least two PCR test results available at any time up to March 15, 2021, and at least one PCR result during the analysis period (Dec 8, 2020, to March 15, 2021). Residents entered the risk period on Dec 8, 2020, if they had at least one valid PCR result on or before that date or on the date of their first negative PCR test if they had no PCR results before Dec 8, 2020. Residents with a positive PCR result within 90 days before Dec 8, 2020, entered the risk period 90 days after their positive test. Residents exited the study at the earliest of the following events: positive PCR test, date of second vaccination, or last available PCR test.
Data extraction and linkage
The NHS number-based pseudo-identifier was used to retrieve vaccination records (date, vaccine type, and dose number) from the National Immunisation Management Service (NIMS) and to link residents to antibody test results, both of which are held in the COVID-19 Datastore. All residents were eligible to participate in serum sampling to detect IgG antibodies to the nucleocapsid protein, subject to provision of valid, informed consent. A personal or nominated consultee was identified to act on behalf of residents who lacked the capacity to consent, and written informed consent was given by those who were able. As long-term care facility recruitment was ongoing, some facilities had not yet commenced serum sampling at the time of the analysis. Therefore, serological results were available for a subset of participants. We combined positive PCR results from before the analysis period and positive anti-nucleocapsid antibody results before vaccination into a binary variable for previous SARS-CoV-2 infection.
were used to derive the mean monthly incidence of SARS-CoV-2 in the area surrounding each long-term care facility. The data platform within the COVID-19 Datastore that was used to build the dataset used in this analysis was provided by Palantir Technologies UK under a general contract with the UK Government. The Palantir team worked under the instruction of the research team and only had access to pseudonymised data. Vaccination data from NIMS were validated against coverage estimates from long-term care facilities, where available. Long-term care facilities with no record of resident vaccination within NIMS were excluded from the analysis. The linked dataset was analysed in the University College London Data Safe Haven. A data privacy impact assessment was done for the VIVALDI study before the analysis.
Statistical analysis
- Krutikov M
- Palmer T
- Donaldson A
- et al.
therefore, a-priori sample size calculations were not done for this analysis.
We used Cox proportional hazards models to derive adjusted hazard ratios (HRs) for the risk of a first PCR-positive test in the study period. Vaccination status was included as a time-varying covariate in the model, with the unvaccinated exposure group compared against exposure groups at 0–6 days, 7–13 days, 14–20 days, 21–27 days, 28–34 days, 35–48 days, and 49 or more days after the first dose of either vaccine. The same cohort contributed person-time at risk to unvaccinated and vaccinated exposure categories, with most individuals starting in the unvaccinated state and sequentially transitioning through vaccinated exposure states until the outcome of interest or being censored at the point of second vaccination or their last available PCR result. The baseline hazard was defined over calendar time. We adjusted for sex (as a binary variable), age (as a cubic spline term), evidence of previous SARS-CoV-2 infection (as a binary variable), long-term care facility bed capacity (as a linear term), and monthly SARS-CoV-2 incidence for the local authority in which the long-term care facility was located (as a linear term). 95% CIs were calculated using robust SEs to account for dependence of infection events within long-term care facilities.
In secondary analyses, we explored vaccine effects stratified by type of vaccine (ChAdOx1 and BNT162b2) and by evidence of previous SARS-CoV-2 infection. Our stratified analyses were done by including an interaction term between the time-varying exposure status and the stratifying factor. We did a sensitivity analysis that excluded individuals who were never vaccinated despite having a PCR test more than 30 days after the date of first vaccination in their long-term care facility to account for potential bias from systemic differences in clinical or other features of this group.
We calculated vaccine effectiveness estimates as 100 × (1–adjusted HR) and 95% CIs for vaccine effectiveness estimates as 100 × (1–upper or lower bounds of 95% CI for adjusted HR). We used two-tailed t tests to estimate the difference in mean Ct values between unvaccinated and vaccinated groups. We also did a sensitivity analysis that limited comparison of Ct values to those obtained from the single most common assay type.
All analyses were prespecified in a statistical analysis plan and done in Stata version 16.0. The study is registered with ISRCTN, number 14447421.
Role of the funding source
The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
Results
Table 1Characteristics of individual long-term care facility residents
Data are median (IQR), n (%), or n.
Table 2Characteristics of long-term care facilities included in the analysis
Data are n (%), n/N (%), or median (IQR), unless otherwise indicated.
Table 3Infection rates and adjusted HRs for PCR-positive infection after the first dose of either vaccine, by days since vaccination (primary analysis)
Adjusted HRs were estimated using Cox proportional hazards regression according to days since the first vaccine dose; the comparator was the unvaccinated group. HRs are adjusted for age, sex, local monthly infection incidence, and long-term care facility bed capacity. 95% CIs were calculated using robust standard errors for long-term care facility-level effects. HR=hazard ratio.
Figure 2Cycle threshold values for PCR-positive tests, by exposure category
Table 4Infection rates and adjusted HRs for PCR-confirmed infection after the first dose of vaccine, by vaccine type and days since vaccination (secondary analysis)
Adjusted HRs were estimated using Cox proportional hazards regression according to days since the first vaccine dose for each vaccine type (ChAdOx1 and BNT162b2). The comparator was the unvaccinated group. HRs were adjusted for age, sex, local monthly infection incidence, and long-term care facility bed capacity. 95% CIs were calculated using robust standard errors for long-term care facility-level effects. HR=hazard ratio.
Discussion
- Pritchard E
- Matthews PC
- Stoesser N
- et al.
and this effect was maintained for at least 7 weeks. We estimated vaccine effectiveness to be 56% (95% CI 19–76) at 28–34 days and 62% (23–81) at 35–48 days. Beyond this time, there was insufficient evidence for a protective effect when looking at both vaccines combined; however, data for BNT162b2, for which there was more person-time at risk available than for ChAdOx1, indicate that a protective effect was maintained beyond 7 weeks. We have only evaluated the effect of the first dose of each vaccine, but our findings constitute some of the earliest evidence on real-world effectiveness of the ChAdOx1 vaccine and of COVID-19 vaccines in long-term care facility residents. In addition to the existing trial and observational evidence on vaccine effectiveness against symptomatic disease, we show that vaccination reduces the total number of infections (asymptomatic and symptomatic) in older adults, and thus overall transmission. An effect of vaccination on transmissibility is further supported by our finding of higher Ct values in infections occurring after vaccination, which are in line with those from large Israeli and UK cohorts,
- Pritchard E
- Matthews PC
- Stoesser N
- et al.
,
- Levine-Tiefenbrun M
- Yelin I
- Katz R
- et al.
implying that residents with post-vaccination breakthrough infections have lower potential for transmission than unvaccinated residents with infection.
- Müller L
- Andrée M
- Moskorz W
- et al.
,
- Collier DA
- Ferreira IATM
- Datir R
- et al.
,
- Subbarao S
- Warrener LA
- Hoschler K
- et al.
pooled data from four ChAdOx1 trials, including over 950 participants aged 70 years and older, indicated vaccine effectiveness of 63·9% (95% CI 46–75·9) against all infection at 22–90 days after a single dose,
- Voysey M
- Costa Clemens SA
- Madhi SA
- et al.
which is in line with our findings. Our estimates of vaccine effectiveness against all infection are not dissimilar to those from phase 3 efficacy testing of the single-dose Ad26.COV2.S vaccine (Janssen), which was 67·9% (95% CI 38·2–82·8) effective against symptomatic COVID-19 at 28 days or more after vaccination in those aged 60 years and older.
- Sadoff J
- Gray G
- Vandebosch A
- et al.
This finding suggests any single-dose regimen is likely to provide similar levels of protection from 4 weeks. Although we found no additional benefit of vaccination in residents with previous natural infection, it remains important to examine the additional benefits provided to older adults by a second dose of vaccine, particularly in the context of new variants of concern, for which immunological data suggest the importance of a second inoculation.
- Parry H
- Tut G
- Faustini S
- et al.
,
- Angyal A
- Longet S
- Moore S
- et al.
,
- Tut G
- Lancaster T
- Krutikov M
- et al.
Data from a Danish observational study in long-term care facility residents suggest that a single dose of BNT162b2 is ineffective in preventing infection;
- Moustsen-Helms IR
- Emborg H-D
- Nielsen J
- et al.
however, participants received the second vaccine dose a median of 24 days after the first dose, which, based on our findings, is probably too short a period to capture the protective effects of a single vaccine dose. Two large Spanish studies in long-term care facility residents reported similar single-dose BNT162b2 vaccine effectiveness estimates to those reported here; however, these studies have limitations of shorter dosing intervals than used in the UK setting and little or no routine asymptomatic screening.
- Cabezas C
- Coma E
- Mora-Fernandez N
- et al.
,
- Monge S
- Olmedo C
- Alejos B
- Lapena MF
- Sierra MJ
- Limia A
Guidance for COVID-19 vaccination in care homes that have cases and outbreaks.
and is likely to have disproportionately affected facilities predominantly using ChAdOx1, which was deployed later than BNT162b2. A similar effect has been observed following single-dose vaccination in health-care workers,
- Hall VJ
- Foulkes S
- Saei A
- et al.
which was attributed to vaccine deferral due to COVID-19 illness. The effect of vaccination deferral on our estimates of vaccine effectiveness could not be ascertained because deferral decisions were not routinely recorded and are likely to have varied between settings.
We identified 439 individuals in our cohort who remained unvaccinated despite vaccine rollout within their long-term care facility. At least a subset of these individuals might have been receiving end-of-life care, but this cannot be confirmed without accessing primary care records, which were unavailable for this study. These residents had substantially lower infection rates than the wider unvaccinated group, which could be attributable to a lower risk of exposure in this group—for example, due to fewer external visitors and reduced interaction with other residents and staff. Sensitivity analysis excluding this group increased estimates of vaccine effectiveness to 76% (95% CI 37–91) at 35–48 days after vaccination.
- Hall VJ
- Foulkes S
- Charlett A
- et al.
,
- Krutikov M
- Palmer T
- Tut G
- et al.
we would expect this to attenuate our vaccine effectiveness estimates. We observed a lower PCR testing rate in the unvaccinated exposure category, and higher PCR testing rates at 35–48 days and 49 days and beyond after vaccination, than in earlier post-vaccination exposure categories. Although the reasons for these disparities are unclear, we would expect them to attenuate vaccine effectiveness estimates. The precision of vaccine effectiveness estimates at 49 days and beyond was reduced because of less person-time at risk in this exposure category than in other exposure categories. We considered Ct values, which correlate with the ability to isolate virus,
- Singanayagam A
- Patel M
- Charlett A
- et al.
to be indicative of infectivity.
- Lee LYW
- Rozmanowski S
- Pang M
- et al.
Although it is challenging to compare Ct values across different assays, all results targeted the same genes (N, ORF1ab, S), and similar findings were obtained in a sensitivity analysis based on a single assay. Our analysis was restricted by absence of data on underlying conditions and frailty status of individual residents, and by scarcity of reliable symptom data, which precluded analysis of vaccine effectiveness against symptomatic infections, as well as reporting on vaccine safety. Future analyses should examine protection against infections caused by other emerging variants of concern and consider outcomes such as hospital admission and mortality, although treatment escalation decisions in the context of end-of-life care are likely to affect vaccination and COVID-19 outcomes.
In conclusion, single-dose vaccination with either ChAdOx1 or BNT162b reduces the risk of SARS-CoV-2 in older residents in long-term care facilities. Our findings suggest that vaccination also has an effect on SARS-CoV-2 transmissibility by reducing the total number of infections in residents, as well as their infectivity. The protective effect of a single dose of vaccination is evident from 4 weeks to at least 7 weeks after vaccination, which provides some evidence to support extension of the dose interval beyond 3 weeks, in line with UK policy. However, even beyond 4 weeks, a single vaccine dose does not eliminate infection risk, highlighting the continued importance of non-pharmaceutical measures to control transmission within long-term care facilities. Further work is required to evaluate the effectiveness of the second dose of the vaccine, and the effect of vaccination on transmission. This knowledge will be critical to inform policy decisions regarding revaccination schedules in this vulnerable population and the disease control measures needed in the short, medium, and long term to protect long-term care facilities from future waves of SARS-CoV-2 infection.
LS, AC, JLB, MS, AH, MK, and SS conceived the study. AC, TP, MS, and LS devised the statistical methodology. MS, MK, and TP did the formal analysis. MK, BA, CF, and AI-S were project administrators. MK and DD curated and validated the data. BA, MK, and MS did the literature review. LS, AH, and AC acquired the funding. MS, LS, and MK wrote the original draft of the manuscript. All authors reviewed and edited the manuscript. MS, TP, AC, LS, and MK had full access to the data in the study. LS and AC have shared responsibility for the decision to submit for publication.
Acknowledgments
We thank the staff and residents in the long-term care facilities that participated in this study and Mark Marshall at NHS England, who pseudonymised the electronic health records. This report is independent research funded by the UK Department of Health and Social Care (COVID-19 surveillance studies). AH is supported by Health Data Research UK (grant no LOND1), which is funded by the UK Medical Research Council, Engineering and Physical Sciences Research Council, Economic and Social Research Council, Department of Health and Social Care (England), Chief Scientist Office of the Scottish Government Health and Social Care Directorates, Health and Social Care Research and Development Division (Welsh Government), Public Health Agency (Northern Ireland), British Heart Foundation, and Wellcome Trust. MK is funded by a Wellcome Trust Clinical PhD Fellowship ( 222907/Z/21/Z ). LS is funded by a National Institute for Health Research Clinician Scientist Award (CS-2016-007). The views expressed in this publication are those of the authors and not necessarily those of the UK National Health Service, Public Health England, or the Department of Health and Social Care.