Evidence summary by Sian Smith, Maximilian Salcher-Konrad, Adelina Comas-Herrera (Care Policy and Evaluation Centre, London School of Economics and Political Science)
Evidence is emerging on the effectiveness of COVID-19 vaccines in long-term care populations. This evidence summary looks at three recently published studies that assessed, not only protection from severe COVID-19 outcomes, but also to what extent the vaccines stop the virus spreading.
Key findings:
- In a large cohort study of 10,412 residents in care homes in England, first doses of the Oxford / AstraZeneca or BioNTech / Pfizer vaccine were shown to be 56% protective against SARS-CoV-2 infection at 28-34 days and 62% at 35-48 days.
- In a small, single-centre study from the US, a single dose of the BioNTech / Pfizer vaccine was shown to reduce viral load, potentially indicating limited transmission of the virus among vaccinated nursing home residents.
- In a small Spanish study of 5 care homes, two doses of the BioNTech / Pfizer vaccine produced an immune response (indicating protection from future infection) among residents, with no statistically significant difference in response by residents’ levels of frailty or disability.
Social Care COVID Recovery and Resilience project view on the meaning of these studies:
- Combined, these studies provide evidence on the protective effect of COVID-19 vaccines against infection among people living in care homes (i.e., protection against symptomatic and asymptomatic disease). Further evidence is emerging on the immune response triggered by the BioNTech / Pfizer vaccine in the care home population.
- Overall, this is encouraging news for care home residents, their families, and staff. Protection against infection, as measured by infection levels and immune response, indicates that a safe, gradual reopening of care homes will be possible. While no vaccine is 100% effective at preventing infection, there is also an indication (albeit based on a very small sample) that the BioNTech / Pfizer vaccine suppresses high levels of viral load, which could limit the extent to which those who catch the virus despite having been vaccinated can spread it further.
- Nevertheless, the evidence from these three studies also presents reasons for caution. In the large English study, it took five weeks after the first dose to reach vaccination effectiveness levels of more than 60%. This highlights the importance of maintaining non-pharmaceutical measures to protect care home residents until the vaccine has developed its protective effect.
- The finding that a single dose of BioNTech / Pfizer vaccine is effective at preventing infection among care home residents is different from a Danish observational study, which found the protective effect to only occur after the second dose. However, in the Danish study, participants received their second dose 24 days following their first. According to the authors of the English study, this time is too short to determine the full protective effect of a single dose.
- These findings match a recent cohort study from Scotland, demonstrating the effect of one dose of the BioNTech / Pfizer vaccine on reducing hospital admissions across all age bands, and another cohort studyfrom Israel which found a high degree of protection from documented infection following one dose of the BioNTech / Pfizer vaccine.
Study summaries:
Study authors: Madhumita Shrotri, Maria Krutikov, Tom Palmer, Rebecca Giddings, Borscha Azmi, Sathyavani Subbarao, Christopher Fuller, Aidan Irwin-Singer, Daniel Davies, Gokhan Tut, Jamie Lopez Bernal, Paul Moss,Andrew Hayward, Andrew Copas, Laura Shallcross
Summary:
This is a pre-print (not yet peer reviewed) study of vaccine effectiveness after the first dose of Oxford/ AstraZeneca and BioNTech/ Pfizer on residents in long-term care facilities (LTCFs). The cohort study compared vaccinated and unvaccinated residents in England.
A total of 10,412 residents were included from 310 LTCFs between 8 March 2020 and 15 March 2021. Out of 9,160 vaccinated, 67% received Oxford/ AstraZeneca and 33% received BioNTech/ Pfizer .
The study estimated the relative hazard of PCR-confirmed infection using Cox proportional hazards regression and adjusted for age, sex, prior infection, LTCF bed capacity, and local incidence of infection.
Key findings:
- Vaccine effectiveness was 56% (19-76%) at 28-34 days and 62% (23-81%) at 35-48 days following a single dose of Oxford/ AstraZeneca or BioNTech/ Pfizer . This suggests that risk of infection is substantially reduced from 28 days following the first dose of either vaccine and this effect is maintained for at least 7 weeks.
- There was no significant reduction in PCR-confirmed infections until 28-34 days post-vaccination, meaning there was no observable protection from the vaccines before this time period (adjusted hazard ratio 0.44; 95% CI 0.24, 0.81).
- Cycle threshold (Ct) values showing how much genetic viral material of SARS-CoV-2 there was in residents who tested positive were available for 1,070 (80.1%) of PCR-positive tests. The mean Ct value of 552 PCR-positive tests from the unvaccinated group was 26.6. The mean Ct value of the 107 PCR-positives occurring 28+ days post-vaccination was significantly higher (mean Ct 31.3, p<0.001). Higher Ct values imply lower potential for transmission from residents with post-vaccination breakthrough infections compared to unvaccinated residents.
- Prior infection was strongly associated with a reduced hazard of subsequent infection (aHR 0.19, 95%CI 0.12, 0.30).
- The study shows no significant difference in effectiveness between the two vaccines.
Study limitations reported by the authors:
- Further studies need to evaluate the effectiveness of one dose after 8-12 weeks, as well as following the second dose.
- Long-term impacts of the vaccine on infection, transmission and morality in LTCFs need to be measured.
Study authors: M Catherine McEllistrem, Cornelius J Clancy, Deanna J Buehrle, Aaron Lucas, Brooke K Decker
Summary:
This is a single-centre retrospective peer-reviewed study published in Clinical Infectious Diseases, measuring viral load after the first dose of BioNTech/Pfizer in one nursing home in the US.
Out of 10 residents who tested positive for COVID-19, 5 were vaccinated with one dose 12-15 days prior to detection of SARS-CoV-2 in nasopharyngeal samples. The 5 other residents were unvaccinated prior to diagnosis.
Key findings:
- Viral load was significantly lower among 5 residents (-2.4 mean log10 lower nasopharyngeal load) who had received a single dose of the BioNTech/ Pfizer vaccine 3 weeks before the infection was diagnosed, compared to the 5 unvaccinated residents (p=0.004).
- Median Ct values were significantly higher among vaccinated residents with asymptomatic COVID-19 (19.4; interquartile range, 18.9- 25.5) compare to unvaccinated residents (12.8; interquartile range, 12.4-14.9; p-value for difference between groups = 0.009).
- This suggests that a single dose of a mRNA SARSCoV-2 vaccine may be effective at reducing viral transmission and outbreaks in nursing home settings.
Study limitations reported by the authors:
- Small sample size: there were only 10 participants in the study (5 vaccinated, 5 unvaccinated).
- The study was underpowered to determine the impact of the vaccine on acquisition of asymptomatic COVID-19 infection.
- The use of Ct value as a proxy for nasopharyngeal viral loads is an approximation with the potential for confounding.
- An antigen test was used to screen for SARS-CoV-2. Residents who had higher viral loads were most likely to be identified, and some asymptomatic cases with lower viral loads may not have been detected. Unrecognised cases of COVID-19 prior to baseline testing were not excluded by serologic testing; it is conceivable that such cases could have impacted viral loads during the study period.
- Multiple providers obtained nares and nasopharyngeal samples, which could have led to sampling error.
Study authors: Sergio Salmeròn Ríos, Marta Mas Romero, Elisa Belén Cortés Zamora, María Teresa Tabernero Sahuquillo, Luis Romero Rizos, Pedro Manuel Sánchez-Jurado, Ginés Sánchez-Nievas, José Joaquín Blas Señalada, Inmaculada García Nogueras, Juan de Dios Estella Cazalla, Fernando Andrés-Pretel, Antonio Murillo Romero, Volker Martin Lauschke, Justin Stebbing, Pedro Abizanda
Summary:
This is a peer-reviewed study published in the Journal of the American Geriatrics Society, measuring antibody levels in residents of LTCFs after receiving 2 doses of BioNTech/ Pfizer. The multicentre longitudinal cohort study was conducted in the Spanish city of Albacete.
A total of 134 nursing home residents in 5 LTCFs were administered 2 vaccine doses and antibody levels were measured on average 21.9 days (SD 9.3) after both the first and second dose. The first dose was administered between 30 December 2020 and 31 January 2021, and second between 20 January 2021 and 19 February 2021 (mean interval between doses: 21 days; SD 0.8).
Functional variables and frailty status were determined (Barthel index and the FRAIL instrument) alongside cognitive status and comorbidity.
Key findings:
- The BioNTech/ Pfizer vaccine produced an immune response in nursing home residents.
- In multivariate analysis, the only variable associated with stronger immune response was previous infection with the virus. Importantly, there was no significant difference in immune response for a range of other characteristics: “Frailty, disability, older age, sex, cognitive impairment or comorbidities were not associated with different antibody titers”.
Study limitations reported by the authors:
- Antibodies are a proxy measure: “[T]hese data do not allow us to draw conclusions whether those older adults with immunogenicity after the vaccine will be protected against SARS?CoV?2 infection and COVID?19.”
To cite this summary:
Smith S, Salcher-Konrad M and Comas-Herrera (2021) Evidence summary: emerging evidence on the protective effect of vaccines from COVID-19 infections among care home populations. LTCcovid.org, International Long-Term Care Policy Network, CPEC-LSE, 15 April 2021.
Summary carried out as part of the Social Care COVID Recovery and Resilience Project (funded by the National Institute for Health Research (NIHR) Policy Research Programme (NIHR202333). The views expressed in this summary are those of the author(s) and not necessarily those of the NIHR or the Department of Health and Social Care)