Saturday, December 18, 2021

Despite the willful blindness of many, COVID has provided a worldwide project-based lesson in several subjects, including abnormal psychology and behavioral economics, as well as regular science and data science. It is the latter that has from time to time most captured my curiosity. 


For me, at first, it was daily trips to worldometer.info. Then the NY Times started publishing numbers but I looked in less and less frequently. Lately, I have been drawn back in by the confluence of variables and curiosity about how they are interacting.

We have been promised that vaccines would drastically reduce cases and lethality. As the virus has lingered and surged, I have become increasingly curious about those claims, not as an aspect of science denial, but just the opposite, as science engagement... or at least data.

To do this, I pulled together data from two CDC tables that show cases per day and deaths per day.  The graph below shows these numbers for case by month, to date for 2021.


We can see here that cases dramatically decreased as vaccines were just getting started.


 But that decrease did not continue. In fact, if you start the graph in February instead of January, the trend line reverses.



Things aren't much better when we look at lethality, measured as a ratio of deaths per day to cases per day. Here is what that looks like.


It seems likely that the vaccine has had a positive impact. We have no idea where we would be without them. But clearly, there has not been the impact that was hoped for. It is also clear that we have not had the participation that had been hoped for. I'm not sure how you would measure the relative weight of those two variables. 

Many who are incensed, as I am, by those who are willing to ignore science to maintain a political stance, would themselves disregard data that seems to pretty clearly show that we have not gotten the impact from the vaccines that we had hoped for. If you are going to extol science, as I think you should, you cannot then refute it when it does not make the exact point that you had put faith into.

Which variables are having the biggest impact? The virality of variants? It seems unlikely from this data that with close to 70% of people vaccinated that the vaccines have the level of efficacy that was promoted, at least outside of the lab. 

COVID has taught as a lot. May the lessons end soon.


No comments: