Lead Author: Shimabukuro, et al.
Original date of publication: 04/21/2021
Link: https://www.nejm.org/doi/10.1056/NEJMx210016
Summary Context
This CDC study was used early to claim broad safety of mRNA vaccines in pregnant women in 2021, and as evidence to censor dissent online. It is now notorious for methodological fraud (found in original table 4) which resulted in a large NEJM correction but not a retraction. Lack of promised follow-up, and possible data destruction by this team is under investigation by a US Senator. Initially, this study claimed that miscarriage and stillbirth rate among vaccinated women generally was similar to rate of general population, but used statistical sleight of hand to achieve that result. Upon outcry and correction, the key claims were nullified but the media had already circulated them and never corrected them at scale in the public record.
Comments
The authors seemingly tried to find a number similar to the population rate of pregnancy loss, around 11-13%, in this cross sectional cohort of pregnant women who had received vaccinations at various times. To achieve this they clearly committed statistical sleight-of-hand that was later retracted after the informational damage was done. The statistical trick was as follows: They included 700 women (out of total 827) who had been first injected in the third trimester of pregnancy in the denominator of the group who could experience “spontaneous abortion”. The spontaneous abortion category is not possible for women in 3rd trimester (rather this is a stillbirth by their definition), which means the denominator was inflated by this maneuver, from 127, to 827. This made the 104 spontaneous abortions that happened in the first two trimesters in completely different context statistically. This means that either many women in this trial lost a baby early, and then got a shot months later, and were included in the V-safe pregnancy data set inappropriately, or if not, more disturbingly, the second option is that the rate of spontaneous abortion was 104/127, or 84.1%, among those who were already pregnant when they received their first shot in the first two trimesters. In the correction page, the authors admit: “The “Published Incidence” cell in the same row should have read “Not applicable,” rather than “10–26,” and the “V-safe Pregnancy Registry” cell should have read “104,” rather than “104/827 (12.6)” They remove the entire denominator rather than answer which of the above options are true. They remarkably admit in the paper that “700 of 712 pregnancies that resulted in a live birth (98.3%) were among persons who received their first eligible vaccine dose in the third trimester”, and yet draw very positive conclusions for pregnant women generally from this data.
Abstract or Results, Key Passages
“Although not directly comparable, calculated proportions of adverse pregnancy and neonatal outcomes in persons vaccinated against Covid-19 who had a completed pregnancy were similar to incidences reported in studies involving pregnant women that were conducted before the Covid-19 pandemic. Among 221 pregnancy-related adverse events reported to the VAERS, the most frequently reported event was spontaneous abortion (46 cases).” Conclusions: “Preliminary findings did not show obvious safety signals among pregnant persons who received mRNA Covid-19 vaccines. However, more longitudinal follow-up, including follow-up of large numbers of women vaccinated earlier in pregnancy, is necessary to inform maternal, pregnancy, and infant outcomes.”
Keywords: adverse eventsCDCcovid vaccinationmiscarriagemRNApregnancyreproductive harm
Full Author List: Tom T. Shimabukuro, Shin Y. Kim, Tanya R. Myers, Pedro L. Moro, Titilope Oduyebo, Lakshmi Panagiotakopoulos, Paige L. Marquez, the CDC v-safe COVID-19 Pregnancy Registry Team
