Why do you think the deaths are going down? Has the virus mutated to be less lethal? Are we getting better at treating it? Or have the weakest serving in vulnerable positions already been killed?
I’m not entirely certain, but I think there are a handful of variables that may be relevant:
- Most deaths were in nursing homes. There was less lead time initially getting this shut down, and by the time it started spreading it wasn’t able to be contained in those settings as well is it can now. I have my own story regarding this that I’ll share more on when this is all over, but it was a very scary situation going for over a month that I had no relief from.
- In relation to above, people are self-selecting their exposure levels, and if we had more data on the demographics, I’d suspect the age distributions of those infected would trend to lower ages over time.
- We’re simply catching more. We don’t know what the true incidence is or has been, but the best available way we can capture that is through confirmed cases. How accurate # positive each day truly correlated with its prevalence is unknown. Could it be that right now we have a drastically lower prevalence but it appears larger because of increased capture by testing? Perhaps in part. Some places the % of tests positive has increased, though, so it’s hard to say, but it’s also hard to determine what levels of selection bias may be introduced in testing and how that has varied both over time but also geographically, so it’s a huge question mark.
- There’s a theory that perhaps it’s mutating. I guess that could be true. That could be good, but could also be bad. It’d be kind of odd, though, just in my opinion.
- There’s another theory that the viral load people are exposed to is currently less, largely as a result of masks and social distancing. I’m not certain how much I’d buy that, but it’s possible. It could explain why some young healthcare workers got fairly sick and/or died, but it’s also hard to determine if that belief is a result of availability bias as those deaths would receive more coverage (how many nursing home patients can anyone name who have died?).
- There could be an iatrogenic effect. In the face uncertainty and the unknown, there’s a huge bias toward heavier intervention. It’s a result of feeling powerless. It’s not entirely inappropriate, though. In the middle of a pandemic, it’s reasonable to tolerate throwing some things at the wall that you otherwise wouldn’t under normal circumstances. There was a huge emphasis on ventilators. A lot of patients were apparently tolerating hypoxia clinically but the hypoxia can be a bit unnerving that it may warrant intervention (intubation and ventilation). There’s also been discussion (for a while now) that people may actually do worse with intubation. Now, this is in some regard a stupid observation, because being intubated fundamentally means that you’re sicker than someone who isn’t intubated, but it appeared that there were a number of different variables that did lead to the idea that, all things being equal, people were having worse outcomes with being intubated. Currently, you don’t hear anything about ventilators. As of Friday, one of the largest hospitals in the state had only two people with COVID on ventilators. I’m not 100% certain on what’s going on currently, but there may be more reliance on CPAP, and that was one of the big things they started doing in Italy.
- It’s hard to know how much is how things are counted, as this varies by locality, and to know how much this has changed over time. I don’t know that this would be a significant change in numbers, but I’ve had three patients who would have been considered COVID deaths but that were actively dying (<1 month anticipated)
before getting COVID. It will take a while to determine what happened in general, but all-cause mortality for the entire year would need to be looked at when we average this over a calendar year. I don’t know (early on) how many deaths were presumptive based on exposure or anything else that wasn’t a confirmed case.
The biggest takeaway, however, is trying to establish what are norms and what are outliers. If we’re comparing cases curves, all the other European countries are within comparable ranges. Ours is an outlier. When comparing death curves, we aren’t an outlier. So I think the question really needs to be asked is if our low (relative to cases) death curve is an outlier or if the death curve is more true to reality and the plateaued cases curve is the outlier (not saying it’s false, just that there are perhaps more variable outliers making up our numbers than other countries’). If we say that the numbers of cases is a good capture, and that the rest of Europe is getting an equally good capture of numbers in their curves, then it’s essentially suggesting that there’s some kind of magic in the US where we were able to break the death curve and dissociate it from the number of cases. It’s possible, but I don’t find that idea very tenable.