Coronavirus

[Daniel in SD]

Unfortunately, for a couple posters, their “eyes roll” whenever new scientific reports or data are posted that contradict conclusions they reached too hastily — before the evidence was in. For those posters, a “funny” is a sign of confirmation bias at work.

What you call confirmation bias I call Bayesian inference.
Posting serology test data without posting anything about the accuracy of the test seems problematic.

 

[heltok]

Some age data about pre-existing conditions.
https://aspe.hhs.gov/system/files/pdf/255396/Pre-ExistingConditions.pdf
This is for people under 65. I believe all of these would count as underlying conditions except for mental health.View attachment 530548

Just want to point out that the latest science does not indicate that high cholesterol is negative condition:
Cholesterol - Wikipedia

Total cholesterol is defined as the sum of HDL, LDL, and VLDL. Usually, only the total, HDL, and triglycerides are measured. For cost reasons, the VLDL is usually estimated as one-fifth of the triglycerides and the LDL is estimated using the Friedewald formula (or a variant): estimated LDL = [total cholesterol] − [total HDL] − [estimated VLDL]. VLDL can be calculated by dividing total triglycerides by five. Direct LDL measures are used when triglycerides exceed 400 mg/dL. The estimated VLDL and LDL have more error when triglycerides are above 400 mg/dL.[92]

In the Framingham Heart Study, in subjects over 50 years of age, they found an 11% increase overall and 14% increase in cardiovascular disease mortality per 1 mg/dL per year drop in total cholesterol levels. The researchers attributed this phenomenon to the fact that people with severe chronic diseases or cancer tend to have below-normal cholesterol levels.[93] This explanation is not supported by the Vorarlberg Health Monitoring and Promotion Programme, in which men of all ages and women over 50 with very low cholesterol were likely to die of cancer, liver diseases, and mental diseases. This result indicates the low-cholesterol effect occurs even among younger respondents, contradicting the previous assessment among cohorts of older people that this is a proxy or marker for frailty occurring with age.[94]

Although there is a link between cholesterol and atherosclerosis as discussed above,[95] a 2014 meta-analysis concluded there is insufficient evidence to support the recommendation of high consumption of polyunsaturated fatty acids and low consumption of total saturated fats for cardiovascular health.[96] A 2016 review concluded there was either no link between LDL and mortality or that lower LDL was linked to a higher mortality risk, especially in older adults.[97]

It will take a generation before doctors are up to date on this though and stop prescribing statins to people with high cholesterol.

———

This was a nice way to visualize how good different countries are at stopping the spread:

 

[dhanson865]

Three NHS nurses who had to wear bin bags for PPE test positive for coronavirus

So from the pic you can see they had gloves, masks, and goggles. Still got infected.

 

[EinSV]

What you call confirmation bias I call Bayesian inference.
Posting serology test data without posting anything about the accuracy of the test seems problematic.

One aspect of confirmation bias is that when someone has already made up their mind the tendency is to seek out reasons to discount information that might lead to a contrary conclusion.

I guess I should clarify my thoughts on IFR since I feel like that might be directed at me. Obviously we can't know exactly what it is, that's why every professional analysis of the data includes an estimate of the uncertainty. I am however very certain that it is somewhere between 0.5% and 2%.

I posted a link to the report. Anyone who had questions about it -- or wanted to read it to come up with reasons to discount its conclusions -- was free to do so.

 

[renim]

https://www.watoday.com.au/national...ening-the-covid-19-curve-20200407-p54hqm.html

'
So Mr McGowan is right. Seven new cases is a great outcome.

But there is a cruel dilemma at the heart of WA's success at flattening the curve. So few infections means the restrictions that are crippling our society could go on longer than the six months foreshadowed by Prime Minister Scott Morrison. They could go on until we have a vaccine, which could be well into the new year.'

for context, the pinkdots are WA, they seem to overlay green dots SA.

 

[Daniel in SD]

My dad is 93 and doesn't have diabetes,hypertension, or any heart/lung problems. Not obese either.

Sees doctor 2-4 times per year. Sometimes doc says see me in 3,4,or 6 months.

He has hypothyroidism,enlarged prostate,macular degeneration,gastritis,moderate-severe arthritis in knees and hips. Sometimes gets a low platelet count and Dr advises he eat liver and iron supplements.

His mind is as sharp as ever too.

Connective tissue disease is on the list of comorbidities that this Italian ICU study uses (https://jamanetwork.com/journals/jama/fullarticle/2764365). Rheumatoid arthritis is a connective tissue disease. Overall though that's pretty good for 93!
My parents are in good shape but my dad takes a blood thinner because he got a blood clot in his leg a few years ago and my mom got a heart valve replacement a decade ago (had a bicuspid aortic valve).

 

[SageBrush]

Preliminary results of serological testing in a hard-hit area in Germany is out and the results are very interesting.

My reading of the study (via Google translate) is that at one moment in time 2% of the community was infected and at that same moment the cumulative exposure (with a lag of ~ 10 - 14 days) was measured at 15%

Does that say anything useful ?

---
I've been mentioning viral load for what seems like forever now. I am skeptical about the plan of the authors (and I don't know why it was even mentioned in this study) but I keep wondering if a vaccine from attenuated virus is the way to go. I presume (without any expertise) that it would easier and quicker to produce.

 

[renim]

Delayed reporting makes all increasing graphs look like this.

———

Here is what HK is doing:
”WHAT DOES THE DANCE LOOK LIKE

Look at the chart. Hong Kong is assessing their R daily. This is exactly what a Dance will look like: assess the R in different places, get as much data as possible to have high confidence it's below 1, and resume activity as long as the R remains below 1.”

Real-time dashboard

Imo this is what every technocratic government should do. Collect all mobile data, google statistics, hospital admissions, testing data. Create some kind of model. Estimate R_t in the past and current. If R_t>1 increase interventions, if R_t < 1 increasing restrictions. Meanwhile doing the best to decrease R_t by other means such as extensive testing, contact tracing etc.

Strict lockdown ‘may be necessary’ with Hongkongers told to stay home
it seems that Hong Kong is approaching its 3rd wave of COVID19, and each wave uses a heavier hammer than the last...

 

[SageBrush]

Rheumatoid arthritis

is not DJD.

The reason 'connective tissue diseases' are a risk factor in covid-19 is mostly related to use of immunomodulatory meds. Those are not used in DJD

 

[MXLRplus]

Three NHS nurses who had to wear bin bags for PPE test positive for coronavirus

So from the pic you can see they had gloves, masks, and goggles. Still got infected.

Not withstanding the possibility the Mirror picture wasn't taken in England, (why is image cropped left instead of center, aspect ratio bizarre, surgical masks not N95)

And in there lies the problem with a lot of Social Distancing and PPE directives.

The western world has very limited PPE production.

China has PPE, India has PPE, other areas of Asia have PPE, the rest lack the capacity to even provide elCheapo brand surgical masks to 10% of the population.
England is another country with limited medical device mfr'g infrastructure.

 

[Daniel in SD]

One aspect of confirmation bias is that when someone has already made up their mind the tendency is to seek out reasons to discount information that might lead to a contrary conclusion.



I posted a link to the report. Anyone who had questions about it -- or wanted to read it to come up with reasons to discount its conclusions -- was free to do so.

Yes, it will take strong evidence to convince me otherwise because most evidence points to an IFR of 0.5%-2%. A brand new serologic test that has a specificity with 95%CI: 97.62%-99.92% that finds that 2.7% of the population has antibodies is not going to move the needle much.

 

[Doggydogworld]

...Cuomo has said that that discharges > admissions. That is a fine leading indicator, and it gives confidence to the IHME projection

When did Cuomo say that? Inpatient count is not growing 1000-1500 per day like it was, but it's still growing 500+/day.

Here's the latest Deaths vs IHME forecast on 03/30:

View attachment 530619

I've added the US into the list to see how we're doing overall. It's surprisingly close.

Wasn't NY way above the 3/30 IHME prediction just a few days ago, or is my memory failing me (again, ha)?

 

[renim]

49% recovered

 

[SageBrush]

Wasn't NY way above the 3/30 IHME prediction just a few days ago

and then lower, and now back on track. Choppy data

Inpatient count is not growing 1000-1500 per day like it was, but it's still growing 500+/day.

Do you mean inpatient count or admissions ?

 

[renim]

Australia
3 day average is around 100 new cases, so 100/6,000 is less than 2%.
(this is not a hammer strategy, this is designed to sustained for at least 6 months, possibly upto 18 months.

 

[Daniel in SD]

is not DJD.

The reason 'connective tissue diseases' are a risk factor in covid-19 is mostly related to use of immunomodulatory meds. Those are not used in DJD

I was not claiming that it is an actual risk factor. I was only climbing that these stats about comorbidities are not very useful if you don't take into account their prevalence in the population. Doctors are not documenting only the comorbidities that they think might have an effect, they're documenting all the known conditions that the patient has.

 

[EinSV]

My reading of the study (via Google translate) is that at one moment in time 2% of the community was infected and at that same moment the cumulative exposure was measured at 15%

I read it the same way (although given translation issues I'm not 100% sure that's correct)

Does that tell us anything useful?

We should be seeing a flurry of serological reports coming out in the next few weeks. Collectively, they should give us a much better understanding of infectiousness and mortality. The two reports I posted today both suggest that infections are much more widespread than many seem to have expected, but we need to see much more data to complete the picture.

The limited data from this study was helpful to the authors. Instead of guessing at the extent of infection and mortality rates in the community they have better data on which to base recommendations for how best to respond to the pandemic. While it's still far from perfect, they obviously felt like they were in a better position than they were before the study, when all they could do is make educated guesses.

I've been mentioning viral load for what seems like forever now. I am skeptical about the plan of the authors (and I don't know why it was even mentioned in this study) but I keep wondering if a vaccine from attenuated virus is the way to go. I presume (without any expertise) that it would easier and quicker to produce.

I don't have enough information to have an opinion on their recommendations.

 

[bkp_duke]

I'm curious what the physicians in this thread think about using remdesivir as a therapeudic for treating COVID-19. I just checked Gilead's website and apparently none of the clinical trials have been completed yet.

The mechanism of action is sound. The problem is the drug is an IV-only medication. I'm keeping a close eye on favipiravir (made by Fujitsu - yes, that Fujitsu), because the mechanism of action is similar, but it can be compounded for oral dosing.

Both medications were designed with ebola in mind, but have a mechanism of action that would theoretically work with SARS-CoV-2. There are at least two clinical trials recruiting for favipiravir (both in China), and many as you know for remdesivir.

 

[MXLRplus]

Potential problem in US about to hit. The incentives to quit work if you are in a critical job are now higher than staying at work for a very high percentage of workers. They didn't do the math before the vote. The employee can choose to quit, and if they make under $25/h it pays more to quit in most states.

 

[dfwatt]

Excuses aside, my point is that herd immunity is not required (or advised, via natural infection) or implicit. All it mostly takes to keep people safe is a non-stupid population.

That's all? Just a non stupid population? Hah! In that case, the United States is toasted.