[Scopri]Spoiler
In the entire population of an Italian province, followed for an average of 14 months, subjects who received one or more doses of COVID-19 vaccines did not show an increased risk of death from any cause, death unrelated to SARS-CoV-2 infection, or any of the recorded potentially vaccine-related serious adverse events requiring hospitalization. Furthermore, some of the outcomes (all-cause mortality, and incidence of stroke, cardiac arrest, thrombosis, and pulmonary embolism) were significantly less frequent among the vaccinated. These findings were consistent across genders, age-classes, most frequently administered vaccine types, and SARS-CoV-2 infection status and remained so after adjusting for previous episodes of disease and several potential confounders. Importantly, strikingly different results were observed in the uninfected population, between subjects who received three or more doses and those who received only one or two doses. The latter groups showed a significantly higher incidence of most outcomes, probably due to the selection bias introduced by the stringent Italian restriction policies concerning subjects who received less than three doses.
While the effectiveness of these vaccines against severe COVID-19 has been documented in several population studies, some primary studies and secondary analyses of serious adverse events reported (a) in the randomized clinical trials of Pfizer and Moderna mRNA COVID-19 vaccines in adults, (b) in the U.S.A. Vaccine Adverse Event Reporting System (VAERS) passive surveillance data, and (c) in the National registries of Denmark, Finland, and Norway, have raised concerns about the safety or risk-benefit profile of COVID-19 vaccination, especially in population subsets with the lowest infection-fatality rate. However, other studies did not show a higher risk of serious adverse events in mRNA vaccine recipients. Most importantly, to date, the only two population-based primary studies with a control group available found significantly lower non-COVID mortality among vaccinated individuals after 28 days and 7.5 months of follow-up.
Both epidemiological and immunological studies showed that COVID-19 vaccination, received either before or after SARS-CoV-2 infection (known as “hybrid immunity”), is able to increase the protection against the virus as compared with natural immunity. No data are available, however, on the potential impact on vaccination safety of having received the vaccine prior or after infection with SARS-CoV-2. We thus further stratified the infected population according to the timing of the infection (Table S15), finding no substantial differences in the overall results between the individuals who were infected before or after vaccination.
As mentioned, while the results were homogeneous for the infected population, the incidence of several outcomes, including death and total PVR-SAEs, was substantially higher among uninfected subjects who received one or two vaccine doses, as compared to those who received three of more doses. This counterintuitive finding may be explained, at least in part, by selection bias. Indeed, during the pandemic, millions of Italians were obliged to receive three vaccine doses to retain their job or access a number of environments (including universities, hospitals, and public places). However, subjects who were infected with SARS-CoV-2 were granted a six-month extension, after which they were mandated to receive only one vaccine dose. As a result, two scenarios emerged. On one side, among uninfected subjects, given the strong pressure to receive three doses, there were only two reasonable categories: those who refused the vaccine in toto and those who wanted to receive three doses, as receiving one or two doses only would not have met the official requirements. This is indeed apparent in the sample: 83.2% of vaccinated uninfected subjects received at least three doses. It may thus be assumed that many of those who received only one or two doses were discouraged from receiving further vaccinations because of the occurrence of a disease, or that they died before receiving the third dose. Thus, these subjects were selectively included in the groups “one dose only” and “two doses only”. In contrast, the infected population did not experience the above selection bias: the vaccination mandate for them was far less stringent, and individuals opted to receive one, two, of three doses according to their personal convictions, which gave rise to very homogeneous results across all vaccine groups. In any case, and most importantly, even among uninfected individuals, the overall comparison, which included all subjects who received at least one dose, showed no increase in the incidence of any outcome among vaccinated subjects, as compared with the unvaccinated.
toc toc, s'è perso il link al theead?