Link to pre-print https://www.medrxiv.org/content/10.1101/2021.02.05.21251139v3
The vaccination roll-out of the COVID-19 vaccines in Israel has been highly successful. By February 22th, approximately 47% of the population has already been administered at least one dose of the BNT162b2 vaccine. Efforts to estimate the true real-world effectiveness of the vaccine have been hampered by disease dynamics and social-economic discrepancies. Here, using counts of positive and hospitalized cases of vaccinated individuals, we conduct a sensitivity analysis of the vaccine effectiveness. Under an assumption of no effectiveness on the first two weeks after the 1st dose, we observe very low effectiveness on the third week. After the 2nd dose, on weeks 1 and 2 we find 73-85% effectiveness in reducing positive cases, hospitalizations, and severe cases, which increase to 89-97% effectiveness 14 days after the 2nd dose. As more granular data will be available, it will be possible to extract more exact estimates; however, the emerging evidence suggests that the vaccine is highly effective.
All data is public and can be downloaded from https://data.gov.il/dataset/covid-19 or Ministry of Health press releases. Data and code used in the analyses was deposited in https://github.com/dviraran/covid_analyses. In addition, we provide an interactive shiny app, which will be updated as more data is available by the Ministry of Health - https://dviraran.shinyapps.io/VaccineEffectIsrael/.