Thank you for the feedback. In response to a few comments, here are a few thoughts...
East Coast vs. Sun Belt
In my first post, I stated "
2) The notion that the East Coast was greatly more successful than the Sun Belt in controlling COVID in the US is not supported by the data. The East Coast peak correspond to the April / May deaths (~105,000), while the Sun Belt peak corresponds to the June / July deaths (~33,000) with the addition of upcoming August deaths since death is a lagging indicator."
Royal27 raised a very valid point:
This is the only point I question. I don't think you can just look at deaths as a comparison. Population has to be accounted for too as a higher population will lead to higher deaths, everything else being equal.
I actually looked at this Royal, and I did account for population.
Specifically, in the simplified comparison between East Coast and Sun Belt, I looked at two of the most affected States in each area. According to the Census Bureau (
https://www.census.gov/quickfacts/fact/table/NY/POP010210) the New York State population is 19.4 million and the New Jersey State population is 8.8 million. Comparatively, the Texas State population is 25.1 million, and the Arizona State population is 6.4 million. So, in this simplified comparison East Coat is represented by 28.2 million people, and Sun Belt is represented by 31.5 million people.
Adding, for example, Virginia (8 million people) and Louisiana (4.5 million people) to the comparison only brings parity (36.2 million vs. 36 million), but when you add if only just Florida and Southern California, the Sun Belt has a significantly higher population than the East Coast.
So, in effect, population number was accounted for in the comparison, even if not spelled out for brevity concern, and the population number bias when all East Cost and all Sun Belt States are counted, would actually accentuate my point that "
The notion that the East Coast was greatly more successful than the Sun Belt in controlling COVID in the US is not supported by the data." 
"Dead wrong" analysis ?
The analysis that started this discussion is dead wrong on several substantial points. First, if you click on the link provided, you will see that a column of CDC data titled "percent of expected deaths" was omitted. That is key because any time it exceeds 100% it means that the Covid pandemic is killing more people than would be expected to die. After February 22nd, it exceeded 100% every week until July 18th and in April the death rate was over 40% higher than expected. Saying that Covid was just 9% of US deaths belies the fact that it was a novel source of mortality that killed a lot of people who would not have died. Why in God's name would you not include that information?
"Dead wrong" is a pretty strong statement Jeff, but let me address your points specifically.
It is correct that for sake of brevity I did not include the "percent of expected deaths" column. But what additional information does this column truly contribute? Of course COVID death + standard deaths exceed 100% of expected standard deaths: COVID is assuredly killing people who would not have been expected to die without it, but I do not think that the table was trying to hide this self-obvious fact. The table prominently states that as of August 1, 2020 COVID alone had killed 143,000 people...
As to "Saying that Covid was just 9% of US deaths belies the fact that it was a novel source of mortality that killed a lot of people who would not have died" I will have to respectfully disagree:
- First whether the source of mortality is new or old does not change the basic mathematical fact that it represents 9% of all deaths.
- Second, no one disagrees, not is it implied, that these 9% would not have died without COVID.
In so many word, my points are factual. I suspect that your criticism is an emotional response to the fact that the data does not support your perspective.
The second issue is that direct comparisons on a graph showing weekly cases and weekly deaths are incorrect because Covid is a disease that takes up to a month to kill you, so there is a lag.
Here too, I will have to respectfully disagree. A lag takes place when two curves are essentially parallel but no synchronized in time. There is indeed a lag in mortality with COVID (I said it myself in my first post), and the graph shows this lag in the April / May mortality. But what the graph also shows - and which was the main point of my opening post - is that since mid June there has not been a lag, but a disconnect between the infection increase trend and the mortality decrease trend.
This too is factual.
The other problem is that Covid death counts are very incomplete because they can only be ascribed to Covid if there was a positive test. There were thousands of people who died of classic Covid symptoms but who never received a test. This was especially true in the beginning when test kits were in short supply but there is still a test shortage, and people die suddenly prior to testing. The rural medical examiners can't figure out that a post-mortum test might be important so they do not do them.
This is an interesting point, because there are credible narratives on both sides of the reporting issue. I do believe that some COVID death are likely to have gone under-reported in the early stages, but I also do believe that COVID deaths have been over-reported in the last couple months. I do believe that
spike.t is quite correct:
As has been found out in quite a few countries if people tested positive that died, ... even if it wasn't the cause of death they were classified as dying from it...in uk people who died months after recovering or having tested positive with no symptoms ,even if run over by a bus were classified in the covid death figures. This was because the people sorting the figures out were looking at charts for people who had tested positive, and if it showed they had recently died even if months after recovery they were put on the list.....
Nonetheless, I believe that I have addressed this point in my addendum that graphed COVID + pneumonia + influenza deaths:
As I stated in my addendum post: "
Here is what is interesting: even if the causes of deaths are voluntarily merged between COVID, pneumonia and influenza, this still does not change the overall story."
This too is factual.
1. The number of cases is increasing. The rate of increase has slowed in many areas (the Johns Hopkins coronavirus resource center is a great tool to visualize this) But every day there are more new cases in every State.
No one disputes this fact. The point that I argue is the disconnect between the infection increase trend (blue curve) and the mortality decrease trend (red or yellow curve if you want to consider COVID-only or COVID + Pneumonia + Influenza).
In so many words, yes there is a COVID crisis (infection cases) but my point about the reporting of this crisis in the "news" is the characterization of the outcome of the crisis (death cases).
2. The number of deaths has exceeded 155K and shows no signs of slowing.
Here too, I will have to respectfully disagree. The statement "The number of deaths ... shows no signs of slowing" is factually false. The sharp disconnect between the infection increase trend (blue curve) and the mortality decrease trend (red or yellow curve if you want to consider COVID-only or COVID + Pneumonia + Influenza) patently demonstrates that the mortality rate has decreased/slowed dramatically.
Most of those people would be alive had the Federal government responded appropriately. Having individual States work it out is like having a peeing section in a public pool.
Whether "most of those people would be alive" is pure speculation, and likely to be false. Factually, the initial response with strict States lockdowns, travel ban, construction of emergency military field hospitals, mobilization of Navy hospital ships, etc. was actually pretty decisive, political pundits notwithstanding.
But, speculation aside, this is an interesting issue when it comes to constitutional powers and public policy trade offs.
State vs. Federal Regulatory Powers
Although the federal government plays a large role in the public health system in the US (e.g. surveys, policies, laws and regulations, research, technical assistance and resources to state and local health systems, etc.) through delegated powers (the power to regulate interstate commerce and the power to tax and spend for the general welfare), it is a fact that healthcare remains primarily a State responsibility in the US.
This was the constitutional argument in the Affordable Care Act (Obamacare) litigation by the States...
Chances are that in the current political climate in the US, the Federal Government is likely "damned if it does" and "damned if it does not" roll out a national prescriptive response.............
Public policy trade-off
The sooner people start taking this seriously the sooner we get to hunt again. Put on those masks, and get distanced.
I believe that most people are taking this seriously, and I for one, out of a sense of civic discipline, dutifully wear a mask and respect social distancing.
But the fundamental question is: what does it mean to "take this seriously"?
The bottom line is that most public policy decisions are trade-offs and compromises resulting from costs/benefits analysis, whether costs are financial or social, and I believe that this is where the main question lies in many people's mind: is the evolving response proportional to the evolving threat?
Let me illustrate:
- On one hand: in 2020 (as of August 1) COVID killed 143,000 people in the US. The historically unprecedented response resulted in historical unemployment (17.4 million unemployed in June 2020, US Bureau of Labor Statistics, https://www.bls.gov/news.release/empsit.nr0.htm) and historical expenses (the 4 relief bills totaled nearly $2.8 trillion and the bill currently discussed is likely to cost anywhere from $1 trillion to $3 trillion).
- On the other hand: CDC reports that in 2019 in the US 2.9 million people died, with heart disease killing 647,000 people, cancer killing 599,000 people, chronic lower respiratory diseases killing 160,000 people, strokes killing 146,000 people, etc. (https://www.cdc.gov/nchs/fastats/deaths.htm). And these are decades-long trends for all these indicators!
Does this mean that we are not taking heart disease, cancer, chronic lower respiratory diseases and strokes seriously?
Or could it possibly mean that the way COVID is reported in the "news" and how it is played in the pre-elections US political climate raise honest questions?
The facts appear to be that we are indeed continuing in the US to see an
increase in the number of COVID CASES, while we are factually seeing a sharp
decrease in the number of COVID DEATHS and it is only legitimate in a free country to use the data publicly provided to form one's own opinion as to whether the costs (financial and social) of the evolving COVID response is proportional to the evolving COVID threat.
In my $0.02 worth, COVID should not be under-played, but it should not be over-played for political benefit either...
