We’re holding up here in Connecticut. Some rules have been relaxed in terms of things being open, but we are still at home and staying socially distanced. We have started having dinner outside only with another family here and there over the last two weeks.
I have continued to track the number of cases each day in both Middlesex County, where we live in Connecticut and in Lowndes County in Georgia where my family lives. It’s really interesting to compare because we’re between NYC and Boston whereas Lowndes county is more isolated. Also take note that Georgia began reopening on April 24th. Connecticut began reopening on May 20th. As you can see in this comparison from the NYTimes, “reopening” means different things in different places. For example, in CT we are still not doing indoor dining (At least I don’t think we are. Our family certainly isn’t).
I get my data from infection2020.com but it kind of makes me crazy because their color coding is based on the number of cases, not the number of cases based on population… which I think would be a more meaningful indicator for communities.
Every day I track the number of global cases, as I have since early March. As you can see, they are still steadily rising… approaching the ten million mark.
I then track the number of total cases in Middlesex County and Lowndes County.
It’s important to note that these are not on the same scale– if you look at the left-hand side of the figure you can see that in Middlesex we have about 1200 cases while in Lowndes it’s closer to 600. What’s interesting here is not the total number of cases, but the shape of the curves. As you can see, Middlesex County has a slow and steady increase of cases, compared to Lowndes County, where the cases are increasing rapidly.
Will the Middlesex cases eventually reveal a flattened curve?
I also calculate the doubling rate. That is, how many days since the number of cases has doubled. You want this number to be high. Again, note the left hand of the figures because these are not on the same scale. As you can see, in Middlesex County it has taken 45 days for our number of infections to double. In Lowndes County, this number is dropping, as it has taken only 13 days for the number of cases to double there. We were neck in neck for a long time, with the number of days between doubling in Lowndes County dropping from late May onwards while Middlesex County’s rate continues to increase.
You can see that even though Middlesex county has about 50,000 more people in terms of population, the proportion of those infected is getting closer, but not quite the same (see the “infected” line). We had a more comparable ratio of 1:189 almost a month ago on May 15th. It’s also interesting to note that in both cases, fewer than 1% of the population is infected (0.75% in Middlesex County today, compared to 0.54% in Lowndes County today).
What’s particularly shocking is the incredibly low death rate in Lowndes compared to a similar moment in time in Middlesex. Back when we had 600-odd cases in late April, we had already had 70 deaths. Compare that to only five deaths in Lowndes County for the same number of infections.
I am not sure if that’s due to more hospital capacity? Or the slow increase of cases over time? Or what has been learned about the virus in just the last two months? It seems astonishingly low in comparison. I hope their success continues.
As someone who teaches public policy, I am looking forward to the day when I can be back in the classroom with my students. There is going to be so much to talk about when that day comes. Part of the story will be understanding the impact of different reopening policies. Hopefully it won’t all be stories about how we would have done things differently with hindsight.
One thought on “Week 13: What the numbers say”
This is great. Thank you