Covid cases have continued to drop over the last week while deaths have largely remained steady. Cases have been dropping for more than 3 weeks, so it is increasingly puzzling why deaths have remained steady. The drop in cases has been quite rapid; we have dropped to about 50% of cases over a 3 week period. Deaths are down only slightly, if at all. Here’s the national graph since October 1.
So we’re back to the question of why deaths have remained high while cases have dropped by half. Here are some partial answers.
First, we’re seeing some one time spike in some states that appear to be recording deaths that happened earlier. Indiana is the clearest example, here is its chart.
You can see that on February 4th, Indiana recorded more than 1500 deaths. That’s about 10 times as many deaths as on any other day and about more than 20 times what they had been averaging since Christmas. There’s just no way that 1500 COVID deaths suddenly all happened on that one day. This has to be a reporting glitch. Most likely this is counting deaths that had previously not been counted. Indeed, a little googling turns up that these are mostly the result of an extensive audit matching up death certificates with positive test results. IndyStar reports “In all, the state will add 1,507 deaths to the official tally, of which 1,205 occurred in 2020. The remaining 302 deaths occurred this year but have not yet been added to the dashboard because of delays associated with the state’s transition to a new birth and death registry.” There are a few other states with similar but much smaller spikes. However, even taking these spikes into account doesn’t change the overall picture, it just smooths out the line a bit.
Another factor is that we’re seeing some states which increasing death counts even as their cases are declining. South Carolina is a clear example here.
Interestingly, this is a regional pattern focused on the southeast. Alabama, North Carolina, Tennessee, and Virgina show similar patterns. While this sort of regional focus might suggest a variant that is spreading in the region. The B.1.351 variant from South Africa would be a good candidate, and the first known US case was in South Carolina, but I haven’t been able to verify that it is more prevalent there in general. This is a possibility well worth keeping an eye on. Thus far, the US hasn’t been very good about tracking variants, so it may take a while to be sure.
That brings us to discussions of variants and vaccines. We now have several variants that seem to be more infectious than other and a few that may cause more severe illness and partially evade the resistance we develop from vaccines or having already had COVID. While this is worrisome, it seems that most vaccines are still largely effective against the variants, just not quite as effective as they are against the baseline variety. There is enough headroom that they can lose some effectiveness against a variant and still be a substantial deterrent. Still, keep in mind that we are still learning about the vaccines and about the variants. What we know will undoubtedly change over the next weeks and months just as it has over the last year.
The best thing we all can do is to remain as safe as possible. Yes, restaurants have reopened in many places. However, I would advise against rushing out to eat in them. “Safe” is a matter of degrees. It may be sufficiently safe for states to allow some indoor dining, but as an individual it is still more risk for you to eat inside a restaurant than getting takeout or just cooking at home. The same principle applies to other businesses opening for service. You should always balance the increase in risk to the benefit you get. Personally, I won’t be eating inside a restaurant until cases are down a lot further; the risk just isn’t worth it to me. On the other hand I will continue to support a few of my favorite restaurants by getting takeout from them. Each of us will have to make the best decisions we can about this based on the details of our own situations.