Skip to main content

BNT162b2 mRNA Covid-19 Vaccine in a Nationwide Mass Vaccination Setting

Our take —

The study estimated the real-world effectiveness of the Pfizer vaccine in Israel’s mass vaccination campaign. Using data from 53% of the Israeli population, they found effectiveness rates similar to efficacy reported in the clinical trials. After receiving the second dose of the vaccine, the estimated effectiveness was 92% for infection, 94% for symptomatic COVID-19, 87% for hospitalization due to COVID-19, and 92% for severe COVID-19 disease. While there may be some selection bias in their estimates due to eligibility criteria and their matching pattern, this study shows that the COVID-19 vaccine is an important public health intervention that worked as well in the real-world as it did in trials, in this population and at this time, with the variants that were circulating in Israel from December 2020 to February 2021.

Study design

Prospective Cohort

Study population and setting

The study objective was to evaluate the effectiveness of a nationwide mass vaccination campaign in Israel (using the Pfizer vaccine) to reduce COVID-19 incidence, hospitalization, and death. The study used data from Clalit Health Services (CHS), an integrated health system with 4.7 million participants, equivalent to 53% of the population. Participants >16 years of age who did not have prior SARS-CoV-2 infection as determined by documented polymerase-chain-reaction (PCR) test, and who were members of CHS in the past year were eligible for the study. Each day from December 20, 2020 to February 1, 2021, newly vaccinated participants were matched 1:1 with unvaccinated participants on age, sex, sector, neighborhood of residence, history of flu vaccination in the past 5 years, pregnancy, and total number of comorbid conditions identified as risk factors for severe COVID-19. Follow-up continued until there was an outcome event, due to death unrelated to COVID-19, vaccination (for an unvaccinated control), vaccination of matched control (for vaccinated participant), or the end of the study. After an unvaccinated match was vaccinated, they were eligible for entry to the study as a vaccine recipient. Outcomes were SARS-CoV-2 infection by PCR test, reported symptoms of COVID-19, hospital admission for COVID-19, severe COVID-19, and death from COVID-19. They used a Kaplan-Meier estimator of risk of events with follow-up in days.

Summary of Main Findings

The study included 1,193,236 participants. The estimated vaccine effectiveness from 14 to 20 days after the first vaccine dose was 46% (95% CI: 40 – 51%) for SARS-CoV-2 infection, 57% for symptomatic COVID-19 illness (95% CI: 50 – 63%), 74% for hospitalization due to COVID-19 (95% CI: 56 – 86%), 62% for severe (95% CI: 39 –80%), and 72% for death due to COVID-19 (95% CI: 19 – 100%). Effectiveness increased in the 21 to 27 days after the first dose for all metrics. After the second dose, effectiveness was estimated at 92% for infection (95% CI; 88 – 95%), 87% for hospitalization (95% CI: 55 – 100%), and 92% for severe disease (95% CI: 75 – 100%). Death after 2nd dose was not estimated due to low observations. The study found lower effectiveness among participants with multiple comorbid conditions, and similar effectiveness across age strata.

Study Strengths

The primary study strength was the large sample size via the mass vaccination campaign, and the robustness of daily follow-up. Notably, CHS had prior medical data available for categorization of comorbid conditions and other sociodemographics, and patient follow-up both in the community and in the inpatient settings for more than half of the Israeli population. The study also used matching to reduce the likelihood of confounding based on a range of factors, including age and prior comorbidities. The study also had access to documented infection results, allowing them to also assess the effectiveness of preventing asymptomatic infections, providing greater generalizability of these results beyond the study population and to the general Israeli population.


There may be residual confounding on factors beyond the matched sociodemographic factors, such as confounding due to specific health seeking behaviors. The study excluded healthcare workers and individuals confined to nursing homes due to high internal variability in the probability of vaccination or outcome, which may also limit the generalizability of the findings to these specific groups. Finally, the study did not report all variants, though they did find that the B.1.1.7 variant made up the majority of infections in the last days before data extraction, and that their study likely represented an average across multiple strains that may not be wholly applicable to any one given strain.

Value added

This is the largest COVID-19 vaccine effectiveness study conducted to date, making up over half of the population of Israel, demonstrating that the Pfizer vaccine has been highly effective in this population at preventing hospitalization and death.

This review was posted on: 26 February 2021