Search
  • Cherrie

Do Coronavirus Lockdowns Work?

Background

Throughout the coronavirus pandemic, TV pundits debated whether lockdowns in the U.S. were effective and worth the trade-off of economic loss for lives saved. In this thought exercise, I used publicly available data to try to determine for myself if the U.S. lockdowns were effective in containing the coronavirus and limiting deaths.


Source Data

CDC


Root Question

This graph or variations of it have been broadcast in the mainstream media. The graph presents New Cases and New Deaths confirmed to be COVID-positive.

Widespread lockdowns were implemented across the U.S. from around March - May 2020, during which deaths spiked. After the lockdowns were lifted we see relatively fewer deaths occur during the significantly larger "Second Wave" of New Cases in the summer of 2020.


Why did New Deaths flatten in the summer when lockdowns were lifted, compared to the spike in New Deaths in the spring when lockdowns were in effect?

Variables

There were several variables driving the absolute number of New Cases and New Deaths:

1. There were no treatments available in the spring, resulting in increased new deaths.

2. Hospitals were overwhelmed and little PPE was available in the spring, resulting in increased new deaths.

3. Testing capabilities were limited in the spring, resulting in decreased new case counts.

4. Testing capabilities improved by the summer, resulting in increased new case counts including more asymptomatic individuals.


Assumptions

The U.S. reached peak testing capacity in Q3'20, i.e. testing infrastructure was sufficient to capture the majority of positive cases.


Application to the Data

In order to create an apples-to-apples comparison for the summer vs. spring, I backed into Normalized New Cases data in the spring by taking confirmed deaths on any given date during Q1-Q2'20 and dividing that by the Q3'20 mortality rate of 1.72% to estimate the spring's "true" New Cases. Since deaths typically occur one to several weeks after COVID diagnosis, I applied the Normalized New Cases to 2 weeks before the corresponding date of New Deaths.


Example


Inputs

February 22, 2020

Absolute New Cases: 0

Absolute New Deaths: 0


March 7, 2020 (2 weeks out)

Absolute New Deaths: 5


Q3'20 Mortality Rate (Deaths / Cases within the period): 1.72%


Output

February 22, 2020

Normalized New Cases: 291 (rounded up numerical half-cases)

Equation: 5 Deaths on 3/7/2020 divided by 1.72% Q3'20 Mortality Rate


Example Results

Although the Absolute New Cases reported on 2/22/2020 were 0, I've estimated that the Normalized New Cases ("true" New Cases) on 2/22/2020 were actually 291.


Findings

Normalized cases in the spring were exponentially higher, giving context to why the summer "Second Wave" didn't kill as many as the first and why the lockdowns were implemented in the first place.


Yes, in my view lockdowns absolutely work to contain the spread and limit deaths from COVID.


9 views0 comments

Recent Posts

See All