You are viewing a single comment's thread from:
RE: Increases in COVID-19 are unrelated to levels of vaccination across 68 countries and 2947 counties in the United States - European Journal of Epediemiology
This study only compares cases over a single 7-day period (August 26 to September 1) to the previous 7 days (August 19–25). I am not sure why they “zoom in” so closely, or why this time period (or any short period) is useful. Other potentially confounding factors are things like: availability and cost of PCR tests, number of tests vs population, test positivity rate etc. It may be that more highly vaccinated areas also have better access to tests, and/or a population more likely to get tested, and that explains a lot of the difference.
Thanks for the comment. I have a feeling that the reason they zoomed in on two seven day windows might be the computational difficulty of accounting for the way that the vaccination percentage changes over time in each particular location. By limiting it to two consecutive 7-day periods, I suppose that they might have just used a single number for vaccination percentage.
If so, that would also mean that they're ignoring any changes in vaccination rate that may have happened inside that 14 day window.
I hadn't noticed before, but the authors also provide a dashboard with US data from April through October. I don't see an easy way to get a multi-month summary of the data, though.