key: cord-0749419-ucxskc0r authors: Monge, S.; Olmedo, C.; Alejos, B.; Lapena, M.; Sierra, M. J.; Limia, A. title: Direct and indirect effectiveness of mRNA vaccination against SARS-CoV-2 infection in long-term care facilities in Spain date: 2021-04-10 journal: nan DOI: 10.1101/2021.04.08.21255055 sha: 65bdee55442450a72f477b5eebb798751a99d0a6 doc_id: 749419 cord_uid: ucxskc0r Objectives: To estimate indirect and total (direct plus indirect) effects of COVID-19 vaccination in residents in long-term care facilities (LTCF). Design: Registries-based cohort study including all residents in LTCF 65 years or older offered vaccination between 27 December 2020 and 10 March 2021. Risk of SARS-CoV-2 infection following vaccination was compared with the risk in the same individuals in a period before vaccination. Risk in non-vaccinated was also compared to a period before the vaccination programme to estimate indirect protection. Standardized cumulative risk was computed adjusted by previous documented infection (before the start of follow-up) and daily-varying SARS-CoV-2 incidence and reproductive number. Participants: 573,533 records of 299,209 individuals in the National vaccination registry were selected; 99.0% had received at least 1 vaccine-dos, 99.8% was Pfizer/BioNTech (BNT162b2). Residents mean age was 85.9, 70.9% were females. A previous SARS-CoV-2 infection was found in around 25% and 13% of participants, respectively, at the time of vaccine offer and in the reference period. Main outcome measures: Documented SARS-CoV-2 infection identified in the National COVID-19 laboratory registry. Results: Total VE was 57.2% (95% Confidence Interval: 56.1%-58.3%), and was highest starting 28 days after the first vaccine-dose (proxy of more than 7 days after the second dose) and for individuals naive to SARS-CoV-2 [81.8% (81.0%-82.7%)] compared to those with previous infection [56.8% (47.1%-67.7%)]. Vaccination prevented up to 9.6 (9.3-9.9) cases per 10.000 vaccinated per day; 11.6 (11.3-11.9) if naive vs. 0.8 (0.5-1.0) if previous infection. Indirect protection in the non-vaccinated could only be estimated for naive individuals, at 81.4% (73.3%-90.3%) and up to 12.8 (9.4-16.2) infections prevented per 10.000 indirectly protected per day. Conclusions: Our results confirm the effectiveness of mRNA vaccination in institutionalized elderly population, endorse the policy of universal vaccination in this setting, including in people with previous infection, and suggest that even non-vaccinated individuals benefit from indirect protection. cumulative risk was computed adjusted by previous documented infection (before the 48 start of follow-up) and daily-varying SARS-CoV-2 incidence and reproductive number. 49 Participants 573,533 records of 299,209 individuals in the National vaccination registry 50 were selected; 99.0% had ≥1 vaccine-dose, 99.8% was Pfizer/BioNTech (BNT162b2). 51 Residents mean age was 85.9, 70.9% were females. A previous SARS-CoV-2 infection 52 was found in around 25% and 13% of participants, respectively, at the time of vaccine 53 offer and in the reference period. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint (80.2% to 82%) for the fully vaccinated. 75  In individuals naïve to SARS-CoV-2 vaccination reduced the risk by up to 81.8% and 76 averted up to 11.6 cases per 10,000 vaccinated persons per day. 77  Those with previous infection also benefited from a risk reduction of 57%, which 78 translated in less than 1 infection averted per 10,000 vaccinated persons per day. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. [2] . The vaccination campaign coincided with the third COVID-19 epidemic wave, 96 with national 14-day cumulative incidence increasing from less than 250 cases per 97 100,000 population by the end of 2020 to more than 1,000 by the end of January 2021 98 [3] . Vaccination started in facilities considered at higher risk, such as those that had 99 never experienced a COVID-19 outbreak, had higher number of residents or more 100 difficulties for implementing prevention and control measures. Vaccination teams 101 visited the facilities and vaccination was universal, including those with previous SARS-102 CoV-2 infection. Vaccination was only deferred in people with active infection and, 103 inconsistently, in people under quarantine. Acceptance has been very high, with 97.8% 104 of all institutionalized persons (any institution type) having received at least one 105 vaccine dose, and 88.8% two doses [4] . . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint 6 The Pfizer/BioNTech vaccine has shown an efficacy of 95% in preventing Covid-19 in 107 randomized clinical trials [5] . However, elderly persons in general, and those 108 institutionalized in particular, are not represented in randomized studies [6] . Therefore as they were considered to belong to the same episode. In LTCF, tests were performed 128 to symptomatic persons and risk contacts. Incoming residents were also routinely 129 tested and periodical screenings have also been carried out. Therefore, documented 130 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint 7 infections registered in SERLAB may correspond both to symptomatic and 131 asymptomatic infections, although this circumstance was not recorded in the system. 132 Residents in REGVACU were cross-matched with SERLAB by person identification 133 number, date of birth and sex. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint 8 since all residents at the same LTCF were offered vaccination on the same day. 155 Therefore individuals were ensured to be included on the date that a first vaccine-dose 156 was administered to most of the co-residents and workers. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. Figure 1) . Interestingly, in a separate analysis we 210 found that previous infection in the reference period was associated to a risk reduction 211 of 86.6% (85.2%-87.8%), higher than the estimate for complete vaccination. Table S1 ). Detailed 223 information on the crude estimates and the adjusted cumulative risk in each group can 224 be found in the supplementary material ( Figure S3 and Tables S2 and S3). . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint Indirect protection was estimated at 57.3% (48% to 66.3%) for the whole study-period. 226 There was no statistically significant reduction in risk in the first 14 days of follow-up 227 but it increased progressively thereafter, particularly after 28 days (as a proxy of full CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint 15 in this population and according to previous infection, and the severity of infection, 321 which could not be measured in this study. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint managing the registries in which this study was based, as well as in providing all the 327 necessary IT support. We specially thank Miguel Hernán for his methodological advice 328 and support in the design of the analysis as well as Luis Sanguiao for his support in 329 implementing parallel computing. We acknowledge as well all the work of the 330 professionals involved in the vaccination program and in managing the vaccination and 331 laboratory registries from the Autonomous Communities. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint Table 1 Actualización nº 2. Enfermedad 334 por coronavirus (COVID-19) en Centros Residenciales Grupo de Trabajo Técnico de Vacunación COVID-19 Estrategia de vacunación frente a 339 COVID19 en España Centro de Coordinación de Alertas y Emergencias Sanitarias. Ministry of Health of Enfermedad por el coronavirus (COVID-19) Gestión Integral de la vacunación COVID. Ministry of Health of Spain Informe de actividad 18 de marzo Safety and 352 Efficacy of the BNT162b2 mRNA Covid-19 Vaccine Causal Inference: What If Immunosenescence and vaccination in 381 nursing home residents Safety 383 and Immunogenicity of Two RNA-Based Covid-19 Vaccine Candidates BNT162b2 mRNA Vaccine Against Infection and COVID-19 Vaccine Coverage in 387 Healthcare Workers in England, Multicentre Prospective Cohort Study (the SIREN Available at SSRN Nationwide 391 Vaccination Campaign with BNT162b2 in Israel Demonstrates High Vaccine 392 Effectiveness and Marked Declines in Incidence of SARS-CoV-2 Infections and COVID Available at: 401 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity Antibody response to first BNT162b2 dose in previously SARS-CoV-2-infected 406 individuals. The Lancet Dose Vaccination in Healthcare Workers Previously Infected with SARS-CoV-2. medRxiv 409 Differential effects of the second 411 SARS-CoV-2 mRNA vaccine dose on T cell immunity in naïve and COVID-19 recovered 412 individuals Initial real 414 world evidence for lower viral load of individuals who have been vaccinated by Effect of 417 vaccination on transmission of COVID-19: an observational study in healthcare workers 418 and their households SARS-CoV-2 420 infection risk among unvaccinated is negatively associated with community-level 421 vaccination rates International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity NA: The model did not result in plausible bias-free risk curves, therefore no estimations were drawn 323 We would like to acknowledge the contribution of the General Directorate of Digital 324 Health and National Health Service Information Systems and the General Sub- 325 Directorate for Health Information (Ministry of Health, Spain) for their work in 326 It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 10, 2021. ; https://doi.org/10.1101/2021.04.08.21255055 doi: medRxiv preprint 19