Will any state catch New York’s coronavirus per capita death rate?

By Kent R. Kroeger (June 8, 2020)

Image for post
Image for post
Virtual model of coronavirus (Image by Rayyar; Use licensed under the Creative Commons Attribution-Share Alike 4.0 International license.)

At over three months into the 2020 Coronavirus Pandemic, a June 2nd Los Angeles Times headline — L.A. County reports 60 new coronavirus deaths as testing sites close for second day amid protests — left me feeling particularly discouraged.

Sixty deaths in one day for L.A. County is a significant increase at a time when California had been looking like they had turned the corner on the coronavirus. [California averaged about 60 deaths-a-day statewide over the past week.]

Like many observers, I believe California Governor Gavin Newsom has done one of the more commendable jobs in handling this health crisis, and he has done so with very little partisan grandstanding and preening for the news cameras.

I cannot say the same for New York Governor Andrew Cuomo — who seems to be in a perpetual pissing match with President Trump.

However, I am generally forgiving of Gov. Cuomo given the sheer scale of tragedy his state has faced during this pandemic. At around 1,580 COVID-19 deaths per 1 million people, New York’s death rate far surpasses the state with the second highest death rate, New Jersey, at around 1,360 deaths per 1 million people.

In comparison, California’s COVID-19 per capita death rate is currently around 115 per 1 million people. While the gap in death rates between New York and California is wide, the news that California is not experiencing the steady decline in COVID-19 cases or deaths as is happening in New York led me to wonder: How bad would things need to get in California in order to for that state to compare to New York’s horrific COVID-19 per capita death rate?

I thought the answer would be easy to determine: Calculate how many deaths would be necessary to match New York’s 1,580 per-million. In California’s case, with 39.5 million citizens compared to New York’s 19.5 million, that would be about 61,000 COVID-19 deaths. As of June 6th, California had only 4,558 deaths.

Yet, I knew even as I did that napkin calculation, it wasn’t fair to New York which is much more densely populated than California (419 persons per sq. mile versus 251 persons per sq. mile, respectively) — and population density is likely a major factor in explaining variations in state-level COVID-19 cases and deaths.

I needed to adjust for a state’s population density before I tried to compare its pandemic performance relative to New York. States less densely populated than New York have a clear advantage in controlling the coronavirus and to compare their numbers to New York’s without such an adjustment would be unjust.

[Note: Further adding to New York’s disadvantage — and, unfortunately, not addressed in my analysis below — is that the U.S. East Coast may have been hit with a more virulent version of the coronavirus than the West Coast.]

So, I added an additional step to the analysis by estimating a state-level linear model of COVID-19 deaths (per-million) with a state’s population density as the lone independent variable.

[Note: I also tested a variable measuring the number of days since a state first reported COVID-19 case, as it seems plausible the time a state has been dealing with the virus might be related to its relative number of deaths. However, this variable was found to be significant and was therefore excluded.]

Using the estimated parameters from the simple linear model, I determined New York’s population density disadvantage/disadvantage relative to each of the other states and D.C.

[Note: The most densely populated state-like jurisdiction is Washington, D.C. at 10,298 per sq. mile; and the most densely populated state is New Jersey at 1,208 per sq. mile].

From there I adjusted the number of additional COVID-19 deaths each state would need to have a comparable per capita death rate to New York’s, as well as the number of days it would take each state to reach that number given their current number of deaths per day (7-day moving average from May 31 — June 6).

For example, in the case of California where the napkin calculation said the state needed about 61,000 COVID-19 deaths to equal New York’s per capita rate, after adjusting for California’s population density advantage that number fell to 47,093 (i.e., 4,558 + 42,535 = 47,093; see columns 2 and 9 for California in Figure 1 below).

In Figure 1, we see the states where it would take the longest to reach the New York COVID-19 per capita death rate. In the cases of Alaska, Hawaii, and Vermont that have not experienced a COVID-19 death in the past 7 days, this measure is essentially infinity. Nonetheless, Alaska would need to add 44 deaths to its current 10, Vermont would need to add 317 to its current 55, and Hawaii would need to add 1,566 deaths to its current 17.

It is unlikely any of those three states will reach New York’s relative death total (though not impossible).

However, there are other states where it would take at least 600 days to match the coronavirus’ lethality in New York. Those states notably include: California, Florida, North Carolina, and Texas — all of which continue to experience a relatively high number of new COVID-19 cases and deaths each day. Still, it would take California 644 days at its present pace to parallel New York’s per capita death rate.

Figure 1: U.S. states unlikely to surpass New York’s COVID-19 per capita death rate

Image for post
Image for post

California is not likely to ever reach New York’s relative numbers, but what states might still?

Figure 2 reveals the states needing the fewest days at their current pace to surpass New York’s death rate: Louisiana, New Mexico, North Dakota and Mississippi (at 30, 37, 60 and 73, respectively.

Figure 2: U.S. states that could potentially surpass New York’s COVID-19 death rate

Image for post
Image for post

[Note: The remaining U.S. states not listed in Figures 1 and 2 can be found in the Appendix at the end of this essay.]

It is important to remind ourselves that the coronavirus pandemic in the U.S. is still ongoing and only in its first wave. According to experts, there could be additional coronavirus waves as states loosen their lockdown policies and until a vaccine is widely available.

Despite that unpleasant fact looming over us, most states are probably not going to approach New York’s per capita death rate, even with additional outbreak waves — which begs another important question: What went wrong in New York and (to slightly lesser extents) in other East Coast states such as New Jersey, Massachusetts, Connecticut?

All were among the earliest to institute statewide lockdowns and the governors in each of those states have generally received praise in the national and local media for their leadership during the pandemic.

Was the virus on the East Coast more dangerous than the virus in other parts of the country? There is already some evidence already suggesting that possibility.

We must also consider that using state-level data (just 51 data points) is too crude a measure to fully understand variations in per capita death rates within states. For example, New York City is primarily responsible for driving up New York’s per capita death rate — which is understandable given its population density of 26,400 per sq. mile.

If we treat New York City as separate from the rest of the state, New York’s overall performance may be quite explainable and not as much of an outlier.

It is still too early to draw strong conclusions about how each state governor has performed during this crisis. What one New York Daily News Letter to the Editor called Governor Cuomo’s pandemic failure may, in truth, be one of this pandemic’s success stories. According to researchers at Columbia University, had Governor Cuomo acted slower in locking down the state, things would have been much worse. Conversely, had he locked down the state sooner — by even a week — many lives possibly could have been saved.

Such conclusions, even based on solid data and modeling methods, are still more theoretic than practical.

As yet, little is yet known about whether broad, statewide lockdowns are more effective than simply practicing strict social distancing techniques — as both were typically implemented simultaneously.

The U.S. and Europe right now are inadvertently running broad social experiments as they loosen their lockdown orders and also when people gather in large numbers for protests. Is it social distancing or ‘stay-at-home’-type lockdowns that are most helping to control the spread of the coronavirus.

When this pandemic ultimately ends and as the data are fully analyzed — including from other parts of the world — we will know more than at any other point in history about how to limit the damage (human and economic) from the next viral pandemic.

  • K.R.K.

Send comments to: kroeger98@yahoo.com

Or tweet me at: @KRobertKroeger1

Appendix

Image for post
Image for post

Written by

I am a survey and statistical consultant with over 30 -years experience measuring and analyzing public opinion (You can contact me at: kroeger98@yahoo.com)

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store