Back

Effectiveness of a third dose of BNT162b2 or mRNA-1273 vaccine for preventing post-vaccination COVID-19 infection: an observational study

Sharma, A.; Oda, G.; Holodniy, M.

2021-11-30 epidemiology
10.1101/2021.11.29.21266777 medRxiv
Show abstract

BackgroundBNT162b2 and mRNA-1273 vaccines are highly protective against COVID-19. Concern about waning immunity and reduced effectiveness against SARS-COV-2 variants led to use of a third dose six months after completing the primary series. We used data from the Veterans Health Administration to evaluate the effectiveness of a third dose of BNT162b2 or mRNA-1273 compared to the primary series in preventing post-vaccination COVID-19. MethodsDuring January 1 - November 25, 2021, third dose recipients were matched (1:1) to demographically similar controls who did not receive a third dose. Eligible participants had completed the primary series at least six months (180 days) before recruitment date and had received at least one SARS-CoV-2 PCR test. Long-term care residents were excluded. Primary outcomes were documented SARS-CoV-2 infection and COVID-19 hospitalization. Effectiveness was estimated as 1-incidence rate ratio. FindingsFollowing matching, the BNT162b2 group included 74,032 pairs and the mRNA-1273 group included 55,098 pairs. In BNT162b2 and mRNA-1273 groups, median age was 72 (interquartile range [IQR]: 64-75) and 72 (IQR: 66-77) years, 69,426 (93.8%) and 52,241 (94.8%) were male, and 43,039 (58.1%) and 37,218 (67.5%) were non-Hispanic White, respectively. Effectiveness of a third dose of BNT162b2 or mRNA-1273 compared to the primary series was 45.7% (95% confidence interval [CI]: 37.9-52.5%) and 46.6% (95% CI: 36.4-55.3%) for documented SARS-CoV-2 infection and 44.8% (95% CI: 26.6-58.4%) and 50.0% (95% CI: 26.2-66.1%) for COVID-19 hospitalization, respectively. InterpretationA third dose of BNT162b2 or mRNA-1273 is moderately effective against post-vaccination COVID-19 infection compared to the primary series. FundingNone.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.