The results of this study validate and complement the previously reported findings of the Pfizer/BioNTech Phase-III randomized clinical trial, which focused on symptomatic infections, and which, with 21,720 vaccinated individuals, could not precisely assess vaccine effectiveness against severe disease in the fully vaccinated. The present study's large size allows a more detailed assessment of the vaccine's effectiveness in preventing a wider range of outcomes, across different time periods and population sub-groups.
The study took place from December 20, 2020, the launch of Israel's national vaccination drive to February 1, 2021. It coincided with Israel's third and largest wave of coronavirus infection and illness, during which the B.1.1.7 variant gradually became the dominant strain in the country for new infections.
Researchers reviewed data from 596,618 vaccinated individuals aged 16 and over (of whom approximately 170,000 were aged 60+). These individuals were carefully matched with 596,618 unvaccinated individuals based on an extensive set of demographic, geographic and health-related attributes associated with risk of infection, risk of severe disease, health status and health seeking behavior. Individuals were assigned to each group dynamically based on their changing vaccination status (approximately 85,000 individuals moved from the unvaccinated cohort into the vaccinated cohort during the study). Multiple sensitivity analyses were conducted to ensure that the estimated vaccine effectiveness was robust to potential biases.
The results show that in fully vaccinated individuals (7 or more days after the second dose), the risk of symptomatic COVID-19 decreased by 94% compared with the unvaccinated, while the risk of severe disease decreased by 92%. In the period immediately preceding the second dose (days 14-20 after the first dose), vaccine effectiveness was lower, but still substantial -- the risk of symptomatic COVID-19 decreased by 57% in vaccinated individuals, and the risk of severe disease by 62%. While there was insufficient data to provide an estimate on the reduction in mortality in those who received two doses, data from 21-27 days after the first dose points to a substantial reduction in mortality as well.
As an observational study conducted in a mass-vaccination setting, this study was not designed to systematically assess viral transmission or asymptomatic infections. With the careful matching procedures, multiple outcomes assessed and multiple sensitivity analyses performed, the large sample size in this study also allowed the estimation of vaccine effectiveness in a number of specific subpopulations. The vaccine effectiveness for preventing symptomatic COVID-19 proved consistent across age groups, including adults aged 70+. The study also evaluated subpopulations with different numbers of comorbidities and found indications that vaccine effectiveness for preventing symptomatic COVID-19 may be slightly lower for individuals with a higher number of comorbidities, although the difference was not statistically significant.
The research was conducted by Dr. Noa Dagan, Dr. Noam Barda, Dr. Eldad Kepten, Oren Miron, Shai Perchik, Dr. Mark Katz and Prof. Ran Balicer from the Clalit Research Institute, as well as Prof. Miguel Hernán and Prof. Marc Lipsitch of the Harvard T.H. Chan School of Public Health, and Prof. Ben Reis of Boston Children's Hospital and Harvard Medical School.
"The swift nationwide rollout of Israel's COVID-19 vaccination campaign provided the Clalit Research Institute with a unique opportunity to assess, through its rich digital datasets, the effects of the vaccine in a real-world setting in all population sub-groups," said Prof. Ran Balicer, senior author of the study, Director of the Clalit Research Institute and Chief Innovation Officer for Clalit. "These results show convincingly that this vaccine is highly effective against symptomatic COVID-19, one week after the second dose. These results are similar to those reported in the previously published clinical trial, despite the challenges inherent in a mass-vaccination setting."
"The results also correlate well with recent population-level trends in Israel, which have seen a sustained decline in hospitalization and severe disease in the mostly-vaccinated older age groups, alongside a delayed decline among younger age groups for whom vaccination began several weeks later. These data, together with the anticipated impact of the ongoing vaccination campaign in Israel, where nearly half the population has already been vaccinated, have had a significant impact on government decisions to ease restrictions imposed during Israel's recent third lockdown," explains Prof. Balicer, who also serves as Chairman of Israel's National Expert Advisory Team on COVID-19 response.
Prof. Miguel Hernán of the Harvard T.H. Chan School of Public Health, said, "This research is a perfect example of how randomized trials and observational healthcare databases complement each other. The original trial of the Pfizer/BioNTech vaccine provided compelling evidence of its effectiveness to prevent symptomatic infection, but the estimates for severe disease and specific age groups were too imprecise. This analysis of Clalit's high-quality database emulates the design of the original trial, uses its findings as a benchmark, and expands upon them to confirm the vaccine's effectiveness on severe disease and in different age groups. This combination of evidence from randomized trials and observational studies is a model for efficient medical research, something which is especially important in COVID times."
Prof. Marc Lipsitch, Director of the Center for Communicable Disease Dynamics and Professor at the Harvard T.H. Chan School of Public Health, said, "In all studies of vaccine effectiveness, a major challenge is to ensure that those we are comparing to identify the vaccine's effect are similar in the other characteristics that may predict whether they get infected or ill. This is especially hard in the context of a rapidly growing, age-targeted vaccine campaign. Clalit's extraordinary database made it possible to design a study that addressed these challenges in a way that provides tremendous confidence in the inferences that come out of the study."
Prof. Ben Reis, Director of the Predictive Medicine Group at Boston Children's Hospital and Harvard Medical School, said, "Israel's impressive vaccination campaign, together with Clalit's unique integrated data sources, presented a rare opportunity to study the effects of the vaccine in a real-world mass-vaccination setting." He continued, "The global scientific cavalry charge that enabled the development of vaccines in record time is now continuing with international collaborations focused on evaluating vaccine effectiveness. The virus doesn't recognize borders, neither should the scientists hoping to fight it. This is how science should be done."
Noa Dagan, Noam Barda, Eldad Kepten, Oren Miron, Shay Perchik, Mark A Katz, Miguel A Hernán, Marc Lipsitch, Ben Reis, Ran D Balicer.
BNT162b2 mRNA Covid-19 Vaccine in a Nationwide Mass Vaccination Setting.
New England Journal of Medicine, 2021. doi: 10.1056/NEJMoa2101765