GBarnsley
commented
1 year ago Hi, from that video it sounds like that plot was produced by someone called Professor Fenton. That plot was not produced from this repo or the paper. I'd have guessed that they took the estimated total counterfactual deaths (i.e. if there were no vaccines from the data) and drew a line between that number and the number of reported COVID-19 deaths from when vaccines were introduced. However, that doesn't seem to align with the numbers since we only produced an estimate up to 08-12-2021. So apologies, I'm not sure were the figure comes from.
To one of the points made in that section of the video. We did produce an estimate of the number of deaths without vaccinations and then compare that to the simulations fitted to the observed data and then used that to calculate the number of deaths averted. However, our counterfactual doesn't look like the one shown in the graph. We also do capture the impact of herd immunity in the absence of vaccinations. This can be seen in supplementary figure 1a from the paper. Here the deaths in the without-vaccines simulation (for the USA) briefly returns to lower level similar to those observed in reality. This is because the previous wave of COVID-19 (that was better supressed in the presence of vaccines) has lead to high level of infection-derived immunity. However, within our model this effect will always be lesser than the combined impact of both vaccinations and COVID-19 infections on population level immunity. So we see the positive impact of vaccinations on COVID-19 deaths as in the paper.
nonsensehunt
I'm fact-checking a YouTube video with ~50k views that comments on the following plot and suggests some of your calculations are incorrect:
I was unable to find this plot anywhere in your development or link it in any other way to your work.
Do you recognize this plot or any data on it? If so, could you please point me to its origin (a page/figure in the paper, a piece of code etc)?