Correspondence

Protection by a Fourth Dose of BNT162b2 against Omicron in Israel

To the Editor

Bar-On et al. (May 5 issue)1 report the efficacy against SARS-CoV-2 infection and severe Covid-19 after receipt of a fourth dose of BNT162b2 vaccine (Pfizer–BioNTech) among persons 60 years of age or older. Table 1 of their article shows notable differences between age groups but does not present the conditional probability of severe disease once infection has occurred. Among persons 70 to 79 years of age, this probability was calculated by us to be considerably lower after the fourth dose: 0.006, as compared with 0.008 in the internal control group and 0.012 in the three-dose group; among persons 80 years of age or older, the corresponding values are 0.023, 0.032, and 0.056. In contrast, among persons 60 to 69 years of age, there are no significant differences between the conditional probabilities: 0.0025 with the fourth dose, as compared with 0.0019 in the internal control group and 0.0028 in the three-dose group.

It is important to report the time-dependent efficacy of the fourth dose against infection and severe disease among persons 60 to 69 years of age separately (as in Tables 2 and 3 of their article), especially since the estimated efficacy against infection (in all the four-dose groups combined) disappears by 8 weeks. Our reasoning is in line with Offit,2 who emphasizes the importance of separate risk–benefit analyses of repeated booster doses in each age group.

Avi Wohl, Ph.D.
Tel Aviv University, Tel Aviv, Israel

Raya Leibowitz, M.D., Ph.D.
Shamir Medical Center, Be’er Ya’akov, Israel

Dr. Leibowitz reports serving as a consultant to Pfizer. No other potential conflict of interest relevant to this letter was reported.

This letter was published on May 18, 2022, at NEJM.org.

  1. 1. Bar-On YM, Goldberg Y, Mandel M, et al. Protection by a fourth dose of BNT162b2 against omicron in Israel. N Engl J Med 2022;386:1712-1720.

  2. 2. Offit PA. Covid-19 boosters — where from here? N Engl J Med 2022;386:1661-1662.

Response

The authors reply: The age-specific conditional-probability calculation of severe illness by Wohl and Leibowitz may be misleading. First, it is not adjusted for confounders. Second, it aggregates data over a period when protection against confirmed infection waned but protection against severe illness did not. Finally, it combines complete and reliable data on serious illness with data on infection that are more susceptible to detection bias owing to testing policy and care-seeking behavior.

Using the same covariate-adjusted Poisson regression that we used in our article, which overcomes these limitations, we obtained age-specific unconditional rate ratios for severe illness between the two control groups and the group of persons who had received a fourth dose at least 3 weeks earlier. Among persons 60 to 69 years of age, the rate ratio between the three-dose group and the four-dose group was 2.7 (95% confidence interval [CI], 1.6 to 4.5), and the rate ratio between the internal control group and the four-dose group was 1.6 (95% CI, 0.7 to 3.6); among persons 70 to 79 years of age, the rate ratios were 3.4 (95% CI, 2.5 to 4.5) and 2.2 (95% CI, 1.5 to 3.4), respectively; and among persons 80 years of age or older, the rate ratios were 4.0 (95% CI, 3.2 to 5.0) and 2.6 (95% CI, 1.9 to 3.6), respectively. This analysis shows that protection against severe illness is conferred by a fourth dose across age groups among persons 60 years of age or older.

Yinon M. Bar-On, M.Sc.
Weizmann Institute of Science, Rehovot, Israel

Yair Goldberg, Ph.D.
Technion–Israel Institute of Technology, Haifa, Israel

Ron Milo, Ph.D.
Weizmann Institute of Science, Rehovot, Israel

Since publication of their article, the authors report no further potential conflict of interest.

This letter was published on May 18, 2022, at NEJM.org.