Policymakers must consider the harms of pandemic policies, not just benefits

Re:Policymakers must consider the harms of pandemic policies... by pps43 on Tuesday May 26, 2020 @11:59AM (#23312795)

You are extrapolating linear relationship between unemployment and deaths way outside the range of historical data on which it was estimated. Surely we would see some of the ~777k deaths already. The fact that we don't suggests that the relationship does not really hold when pushed so far out.

[ Parent | on HN ]

Re:Policymakers must consider the harms of pandemic policies... by naasking on Wednesday May 27, 2020 @10:25AM (#23323793)

> Surely we would see some of the ~777k deaths already. The fact that we don't

Where are you getting this data to conclude we haven't seen a such a spike?

[ Parent | on HN ]

Re:Policymakers must consider the harms of pandemic policies... by pps43 on Thursday May 28, 2020 @05:03PM (#23342591)

CDC publishes weekly number of deaths from all causes:

https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm

[ Parent | on HN ]

Re:Policymakers must consider the harms of pandemic policies... by naasking on Monday June 01, 2020 @11:07AM (#23379294)

Great. Since the whole country is not in lockdown, let's examine a region which has been highly impacted:

* Select "Excess deaths with and without COVID-19" for the dashboard

* Enter "New York City" as the location (New York state as a whole shows a similar trend)

New York state was locked down March 20th. Notice the marked increase in non-COVID-19 related deaths starting March 21st, increasing steadily into April (almost 200% higher than predicted number of deaths). The tooltips also say that this data is still incomplete, so these numbers could climb.

Now maybe it's unrelated, or maybe it's COVID-19 related that was incorrectly assigned, but you seem fairly confident that it's unrelated to the other factors the OP mentioned. I don't see how you can reach that conclusion given the data available at this time.

[ Parent | on HN ]

Re:Policymakers must consider the harms of pandemic policies... by pps43 on Tuesday June 02, 2020 @12:25PM (#23392810)

Look at non-COVID-19 related deaths in a location with lockdown in effect, but with few COVID-19 deaths. Those should not be people who died from COVID-19 without a test, but if lockdown drives deaths up, there should be enough of them to be visible.

If you do it in New York, chances are you would be attributing COVID-19-related but unconfirmed deaths to lockdown.

[ Parent | on HN ]