US Covid fatality rates highest among top 18 developed countries

[jim mcnamara]

TL;DR Summary

US Covid fatality rates highest among top 18 developed countries, letter in JAMA Oct 12, 2020

US Covid-19 fatality rates are the highest among top 18 developed countries -primary claim
The US medical system and Covid-19 therapy are both as good as it gets, therefore other factors intervene <- is what the letter says.

https://jamanetwork.com/journals/jama/fullarticle/2771841

I need help.

It is a letter, so it is short. Please do not assume TL;DR on the link if you have any interest in the problem.

PS: see references section - there are links to data and a link to formulas used in the main section. (@OmCheeto may be interested)
You can get the pop science version here:
https://www.npr.org/sections/health...c-at-rates-far-higher-than-in-other-countries

Or maybe I need a lot more coffee... If you have some insights please let me know. It is a complex a problem, if it is indeed real, in order to be able to finger point and say 'X' is indeed one of the causes. I think it simply claims 'Houston, we have a problem'.

Assuming the data was assembled and treated fairly, the paper posits the high fatality rate to:
Aversion to treatment due fear of contagion, or other assumptions like Covid-19 is not a serious health problem. Anyway, people fail to get timely medical help for Covid symptoms. Or get no help at all. They also attribute additional mortality from other completely unrelated problems, like a myocardial infarction, to fear and/or assumptions.

Thanks for any help.

 

[Vanadium 50]

I found that paper thin gruel. Sure, the US has a large number of total deaths. It's a big country. If total deaths were a good metric, we should all move to San Marino (which has the highest per capita rate in the world - but only 42 deaths). Deaths per capital, but the authors shoot first and then draw the bullseye. Belgium has a hire death rate than the US, so let's look at data after Belgium has peaked. Guess what - now the US is worse!

One author is a well-known political appointee of the previous administration (the other is a grad student), and I suspect a political axe to grind. His conclusions are all things he has advocated since before Covid. The "analysis" is little more than saying he can move the US to the top (or is it bottom?) of the pile with a judicious set of cuts.

I expected more.

 

[Astronuc]

Summary:: US Covid fatality rates highest among top 18 developed countries, letter in JAMA Oct 12, 2020

US Covid-19 fatality rates are the highest among top 18 developed countries -primary claim
The US medical system and Covid-19 therapy are both as good as it gets, therefore other factors intervene <- is what the letter says.

I saw some headlines, but I took it as sensationalized. By some metrics maybe, but the numbers are changing rapidly.

Separately, treatment is not uniform, where some get great treatment, and others are sent home. New York has relatively low rate, but it has doubled recently, and there are about 20 zip codes that have very high rates of infection due to groups meeting without masks and social distancing. Locally, we have a small cluster of university students, who went to a party off-campus, where masks were not worn and folks congregated closely. Many of the university campuses around the country have similar experience with groups of students congregating without protective measures.

 

[jim mcnamara]

I do not know if the data, which uses excess mortality in part, means more than 'a bunch of people died who normally would not have'. Or does it mean somehow we messed up medical care - your point - or is it so multi-factorial that interpretation is not simplistic - one of my concerns.

 

[Astronuc]

Or does it mean somehow we messed up medical care - your point - or is it so multi-factorial that interpretation is not simplistic - one of my concerns.

I believe it's a combination of the two. Early on, we didn't know how severe an infection could become. In February/March, we didn't know how it would spread, how infectious it would be, who was susceptible (we still don't know), and we didn't have sufficient supply of protective equipment for medical workers or the general public. I started looking for masks in February (supplies had been limited in January), and local places were sold out. At work, we started social distancing well before it became mandatory. Even in March, testing was largely unavailable, and usually only to those who presented symptoms, otherwise folks were sent away (too often in NY). At the time Remdesivir was first used, Gilead had may be ~1000 doses.

From a press release in July https://www.gilead.com/news-and-pre...iral-remdesivir-for-the-treatment-of-covid-19:

This comparative pre-planned analysis included 312 patients treated in the Phase 3 SIMPLE-Severe study and a separate real-world retrospective cohort of 818 patients with similar baseline characteristics and disease severity who received standard of care treatment in the same time period as the SIMPLE-Severe study. Patients were primarily located in North America (92 percent, remdesivir cohort vs. 91 percent, standard-of-care cohort), Europe (5 percent vs. 7 percent) and Asia (3 percent vs. 2 percent). The analysis demonstrated that remdesivir treatment was associated with significantly improved clinical recovery and a 62 percent reduction in the risk of mortality compared to standard of care.

312 patients out of 10's of thousands. On July 10, 2020, the US reported a cumulative 3,173,216 positive cases (most resolved) with 126,264 deaths (probable and confirmed), which doesn't include folks who died at home and were not tested.

Back in March, a 25-year old NJ man was hospitalized with severe symptoms (I believe he was on a ventilator in a medically induced coma). He was "one of more than 500 people on a waiting list there for a clinical trial that uses the novel antiviral drug remdesivir."
https://www.pix11.com/news/coronavi...of-pennsylvania-for-clinical-trial-remdesivir
Allard survived, and perhaps age and lack of pre-existing condition (comorbidity) are contributing factors.
Wed Apr 22 2020 - https://www.uslaxmagazine.com/colle...k-me-to-my-knees-grateful-for-lacrosse-family

On May 10, 2020 - https://www.phillyvoice.com/new-jer...g-hospitalized-covid-19-coronavirus-patients/
New Jersey to receive shipment of 110 cases of remdesivir to treat hospitalized COVID-19 patients

New Jersey—along with New York, Indiana, Massachusetts, Rhode Island, Tennessee, and Virginia—received 94 cases of remdesivir earlier this week. Between the two shipments, New Jersey could treat approximately 8,160 hospitalized COVID-19 patients in severe condition.

Treatment with Remdesivir, if available, was usually done late in the game. "Patients with low blood oxygen levels, needing oxygen therapy, or using a ventilator to help with breathing are eligible for the experimental drug, HHS said."

Compare that with a patient who received Regeneron's monoclonal antibody cocktail, less than 24 hours following a positive test, followed by a course of Remdesivir, and whatever else.

The degree to which someone is exposed/infected is yet another unknown variable. Other variables include age and pre-existing conditions, which of increase with age. The top 10 comorbities in NY are: hypertension, diabetes, hyperlipidemia, dementia, coronary artery disease, renal disease, COPD, atrial fibrillation, cancer and stroke. NY reports that ~90% of COVID-19 patients have at least 1 comorbidity.
https://covid19tracker.health.ny.go...ID19-Tracker/NYSDOHCOVID-19Tracker-Fatalities

Access to health care is not uniformly available, especially to poor and those without health insurance.

 

[Ygggdrasil]

Here's what I take away from the paper:
1) US deaths per capita were lower than other countries experiencing significant COVID-19 outbreaks in March-April (e.g. France, UK, Italy, Sweden, Spain).

2) The main area where the US sticks out is the high rate of per capita deaths after the initial wave of deaths in March-April (e.g. since June 7, the US rate of COVID-19 deaths per capita is more than twice as high as any of the other 18 countries analyzed).

To me, this suggests that the main problem has been containment of the virus after the initial wave. Whereas most other developed countries were able to use the COVID lockdowns to reduce the infection rate and have subsequently been able to keep transmission low, the US has had less success containing transmission of the virus. This view is consistent with a lot of other data (e.g. comparing case counts in the US versus other developed countries).

Data collection only goes through mid-Sept, however, so the recent rise of COVID-19 in other countries (e.g. France, UK) could change how we look at the situation a few months into the future.

The paper does not have enough data to compare the quality of COVID-19 treatment across the countries as one would need estimates of the prevalence of the disease in addition to the mortality numbers.

 

[Fervent Freyja]

A huge factor: time in receiving testing results.

Are there studies yet in determining if 24 hour labs result in a lower mortality rate than 2-5 day labs? With how quickly patients succumb to this, time-to-receive results should be highly correlated with mortality. Many aren’t receiving treatment or diagnosis or take quarantined directions seriously until after they test positive. In addition, delayed positive results in one person is going to cause further spread- as they may not quarantine or take it as seriously, which further increases mortality rate by increasing the infection rate.

How does that compare to other countries? How are the majority of labs that they are using set up? What is the time delay for results in those countries? Were they better prepared and quicker than us to begin with?

What if every patient in the US that has died from
it could have received 24 more hours of care- would they still be alive?

 

[nsaspook]

In Oregon the vast majority of deaths are elderly with reported serious pre existing health conditions. An extra 24 hours of earlier care is IMO unlikely to stop most of those deaths unless there is a very effective treatment and maybe daily testing. The key for them is isolation from the virus as the trigger event.

Age group	Cases1	Percent	Ever hospitalized4	Deaths2
0 to 9	1815	5%	33	0
10 to 19	4145	11%	36	0
20 to 29	8368	22%	196	2
30 to 39	6837	18%	225	4
40 to 49	6165	16%	334	14
50 to 59	4869	13%	461	38
60 to 69	3052	8%	558	94
70 to 79	1845	5%	556	151
80 and over	1396	4%	468	308
Not available	33	0%	1	0
Total	38525	100%	2868	611

 

[OmCheeto]

...
PS: see references section - there are links to data and a link to formulas used in the main section. (@OmCheeto may be interested)

Yes! The referenced material is actually more interesting than the paper. Though, I think I might be suffering from Covid-data-overload, as the I couldn't figure out the point of the paper until I plotted the data for all the nations they referenced. And then I was simply not impressed at all.

What I found even more interesting was what I found when I researched further into what @nsaspook posted yesterday. I'd never much looked at hospitalization rates before, but calculating the hospitalization to death ratios for the ages from 20 to 80 yielded a near perfect linear logarithmic plot.

Plugging the equation into the age demographic numbers from the CDC for the entire nation indicated that about 5 million people would be hospitalized if the "Great Barrington" idea was implemented.

Barrington

Barrington

Barrington

Barrington

Barrington

Barrington

Barrington

covid deaths

deaths from all causes

Age gp

midlin’ age

all deaths involving covid

deaths from all causes

population

% covid

9​

% covid

hospitalized

% hospitalized

beds

ratio of hospitalizable to beds

All ages

203,043​

2,203,637​

328,239,523​

9.21%​

152,928​

2,339,573​

6.54%​

4,842,258​

1.5%​

924,000​

5.2​

Under 1 year

0.5​

22​

12,405​

3,783,052​

0.18%​

198​

12,581​

1.57%​

108,459​

2.9%​

1–4 years

2.5​

15​

2,351​

15,793,631​

0.64%​

135​

2,471​

5.46%​

64,407​

0.4%​

5–14 years

10​

37​

3,690​

40,994,163​

1.00%​

333​

3,986​

8.35%​

94,633​

0.2%​

15–24 years

20​

374​

23,838​

42,687,510​

1.57%​

3,366​

26,830​

12.55%​

479,418​

1.1%​

25–34 years

30​

1,588​

48,782​

45,940,321​

3.26%​

14,292​

61,486​

23.24%​

1,020,218​

2.2%​

35–44 years

40​

4,119​

68,966​

41,659,144​

5.97%​

37,071​

101,918​

36.37%​

1,326,277​

3.2%​

45–54 years

50​

10,837​

126,245​

40,874,902​

8.58%​

97,533​

212,941​

45.80%​

1,748,846​

4.3%​

55–64 years

60​

25,971​

288,746​

42,448,537​

8.99%​

0​

288,746​

0.00%​

65–74 years

70​

43,927​

437,155​

31,483,433​

10.05%​

0​

437,155​

0.00%​

75–84 years

80​

53,796​

532,277​

15,969,872​

10.11%​

0​

532,277​

0.00%​

85 years and over

90​

62,357​

659,182​

6,604,958​

9.46%​

0​

659,182​

0.00%​

Of course, since there are only about 1 million hospital beds in the US, I decided that Dr. Fauci was right, when he bad mouthed the idea yesterday.

 

[Laroxe]

I think most commentators think that comparing recorded deaths from Covid 19 from different countries is largely a waste of time. The differences in how this data is collected make comparisons meaningless. Currently the only measure that might be helpful is in looking at excess deaths and to be honest looking at these figures suggests the figures are similar across most western countries. Obviously these will vary depending on the number of cases, but there doesn't seem much to shout about. This link has a good (if long) review of the numbers, you can scroll down to the section entitled In Depth, Excess mortality. You can do all sorts of things with the charts to present the data, its worth a look.

https://ourworldindata.org/coronavirus