Lanark wrote:It is not difficult, but for some reason many people want to believe the vaccines are killing people, against all the evidence - why??
This cohort study on the entire population of an Italian Province assessed the incidence of potentially vaccine-related serious adverse events (PVR-SAEs) by COVID-19 vaccination status. From January 2021 to July 2022, we extracted all deaths and hospitalizations due to several cardiovascular diseases, pulmonary embolism, and deep vein thrombosis from National Healthcare System official data. - COVID-19 Vaccination Did Not Increase the Risk of Potentially Related Serious Adverse Events: 18-Month Cohort Study in an Italian Province
It's far more difficult to establish risk/benefit of a treatment using observational studies, like that one, than it is using clinical trials, particularly when the population under study is not necessarily in great health anyway. It's likely that Vioxx would still be in use as a painkiller despite causing heart attacks, had Merck not decided to run more clinical trials for additional conditions; in those trials, they weren't able to hide the safety signal as they had done in the original trial.
The results of that Italian study look odd, in that most of the adverse event types are much more common for people who received 1 and 2 doses than they are for people who received 3 or more doses, and it's mostly the latter category, which happens to be the largest, that has fewer adverse events than the unvaccinated group. I don't think we can conclude that the vaccines are safe. The distribution of results suggests that at least some of the 1 dose and 2 dose groups may have stopped where they did because of an adverse event (harvesting bias). Also the unvaccinated group potentially has a much longer case counting window than the other groups (immortal time bias) with the 3 or more dose group having the shortest window, perhaps as little as 1/3 as long which may at least partly explain the low count of adverse events in that group. With a window 1/3 as long you would expect the count of adverse events to be about 1/3 as well for a placebo vaccine. There's also a potential seasonality effect because cardiovascular events are more common in the winter, and the case counting window for the unvaccinated includes one and a half winters, while the other groups have proportionally less winter.
As it happens, there was a followup to that Italian study, looking at the same province using the same methods but with a case counting window of up to two years, and it found the same odd-looking distribution of adverse events between the groups. There was even a recent preprint followup to that followup study (I can't post links but it's "A reanalysis of an Italian study on the effectiveness of COVID-19 vaccination suggests that it might have unintended effects on total mortality", Alessandria et al) which found that when you adjust for immortal time bias (the different start times of the case counting windows) the vaccines seem to reduce life expectancy by several months, which is quite a lot.