The New Normal

[Milo Maltbie]

Everything is always about you.

Really? I got vaccinated, I didn't travel or even leave the house for anything but outdoor recreation for a year, I wear a mask whenever I'm asked and generally follow public health advice (which my partner gets directly from the health department). The only thing I object to is the idea that your "freedom" means you can ignore science, treat the virus as a political problem and not a public health issue, and do things that put others at risk.

Maybe you are projecting a little?

mm

 

[tirolski]

Bottom line is that the evidence is pretty clear that Delta became the dominant variant in the U.S. by early July, if not a little sooner in some places with a lot of community spread.

CDC just put this up bout their study of Delta @ a school outbreak in Cali in May.

Outbreak Associated with SARS-CoV-2 B.1.617.2 (Delta)...

COVID-19 outbreak associated with an unvaccinated infected teacher in an elementary school from May–June 2021 in Marin County, California.

www.cdc.gov

Most if not all of the data released by the FDA and EMA for approval of BioNTech’s Comirnaty was from clinical trials prior to Delta.

 

[tirolski]

Yes, it's true there will be modified vaccines for COVID-19 at some point.

There are Phase 1/2 studies in progress for a new version of Pfizer and Moderna that were adjusted to deal with known variants. Using the mRNA approach means that modifying a vaccine is a much faster process.

I saw a video by a UK physician yesterday where he complained that reports about the approval of Pfizer didn't change for efficacy for months. He obviously has no idea that efficacy results much be calculated based on a pre-designed and approved statistical analysis plan. You can't just run a stat analysis any time you want. The underlying assumptions get violated and then the statistical results become invalid. Safety data gets updated as followup continues for clinical trial subjects, as well as information for vaccinated people after release to the public. It was quite clear he has no clue what biostaticians due who work for a regulatory agency that approves medications. The stories my grad school classmates had to tell about working with physicians in the Consulting Lab were pretty funny . . . to a statistician.

The one by the UK physician I put up yesterday here?
Sometimes flu shots work and sometimes they miss due to viral mutations vs predictions.
SARS-CoV-2 has mutated.
Can regulatory folks write their "statistical analysis plans" to effectively account for this?
Or isn’t it worth it?

 

[Campgottagopee]

I got sucked down the ivermectin rabbit hole last night on Facebook, wow.

We have a tech at work that takes that shit. He also believes the world is flat and wears a tinfoil hat.
The only part I made up is about the hat.

 

[Campgottagopee]

The only thing I object to is the idea that your "freedom" means you can ignore science, treat the virus as a political problem and not a public health issue, and do things that put others at risk.



mm

But you can. Doesn't matter who you are or what you believe. That's freedom.

 

[x10003q]

I think the Apple Flavor is driving usage. "Keep out of reach of children" and some adults, too!

 

[MarzNC]

The one by the UK physician I put up yesterday here?
Sometimes flu shots work and sometimes they miss due to viral mutations vs predictions.
SARS-CoV-2 has mutated.
Can regulatory folks write their "statistical analysis plans" to effectively account for this?
Or isn’t it worth it?

Could be. I watched one video and per usual YouTube had suggestions. I wasn't paying close attention since I was doing something else while the video was on. I much prefer reading than watching videos.

A statistical analysis plan is written by biostatisticians who are on the industry side, whether they work for a pharma/biotech company or a CRO (Contract Research Organization). My experience is with projects for drugs, not vaccines. What I know is that you can't change how you analyze efficacy data once patient enrollment starts.

The UNC Biostatistics professor and founder of Quintiles used to say that if a statistical analysis plan was so complicated that a Ph.D. level biostatistician was required to explain the results, then it was a bad idea. Most of the people in my department doing analyses after a clinical trial was completed were Master's level biostatisticians supported by SAS programmers. The company had a few people fully qualified to be high level "consultants" who would help work with client companies to come up with Phase 3 placebo-controlled analysis plans. Meaning biostatisticians who could charge $400+ per hour as a consulting fee (back in the 1990s). FDA biostatisticians would review the plans but I don't remember hearing about a situation where they suggested changes. Quintiles was too good for that to happen. The first few years, Quintiles was unusual because it only provided Biostatistics and Data Management services. The growth beyond 50 people happened after clinical services were added. Meaning CRAs (Clinical Research Associates) who would go around to research sites (often a doctor's office) to help make sure data was begin collected correctly for a given clinical trial.

My impression is that what Pfizer and Moderna did in the last 6 months or so is to work towards setting up separate research to address Delta. That includes lab studies that are are necessary before coming up with plans for Phase 1/2 and Phase 3 trials in humans. What's unusual is that setting up a small study in a country where Delta arrived sooner than the USA was another way to get some evidence of how a vaccine performed before doing an American clinical trial. There is also a lot of research being done in other countries.

Can't read most of the article but here's a relevant news report for Pfizer. The upcoming trial is for a new version of the vaccine.

August 9, 2021

Pfizer partner tests COVID vaccine targeting Delta variant

Germany's BioNTech believes it can switch production to combat the spread of mutations within 100 days.

fortune.com

 

[MarzNC]

Ivermectin for Prevention and Treatment of COVID-19... : American Journal of Therapeutics

mortality, in secondary outcomes, and in chemoprophylaxis, among people with, or at high risk of, COVID-19 infection. Data sources: We searched bibliographic databases up to April 25, 2021. Two review authors sifted for studies, extracted data, and assessed risk of bias. Meta-analyses were...

journals.lww.com

Long read and I’m not a scientist but my takeaway it’s 62% effective?

Effective for what? I only read the Abstract, which is what I normally do.

That's the article my immunologist ski buddy had sent me recently. I noticed the UK researchers who wrote the report are funding the effort using GoFundMe. Hmmm . . .

My take away is this sentence: "Low-certainty evidence found that ivermectin prophylaxis reduced COVID-19 infection by an average 86% (95% confidence interval 79%–91%)." Note the clear statement of "low-certainly evidence."

Given that a doctor has to prescribe ivermectin, not much risk for people who don't mind taking low-dose medication for an "off-label" purpose. Personally I don't like the list of minor side effects. I'm happy enough having had shots of an mRNA vaccine that's gone through FDA review. Combined with sensible public health precautions, I don't see the need to take a drug. YMMV

Using a drug off-label happens all the time after a drug has been on the market for a while. Ivermectin has been around since the 1980s.

Sounds like when it comes to treatment for someone who already has symptoms for COVID-19, in particular to avoid serious symptoms, it's less clear whether Ivermectin can help or not. There is only so much that can be learned from a meta-analysis.

FDA and other regulatory agencies require clinical trials when a company wants to add to the list of approved indications for a drug. I worked on a project for several years for a new indication for a drug that had been on the market for decades. As I remember the Phase 3 studies didn't show any effectiveness for the new indication of interest. One reason Quintiles did well as a company is that a CRO makes money as long as the client company pays for services. Meaning it doesn't matter if a drug doesn't actually turn out to be effective when the Phase 3 analysis is done. Of course, success would have led to more business to put together the NDA safety databases and summary statistics.

 

[Sbob]

Effective for what? I only read the Abstract, which is what I normally do.

That's the article my immunologist ski buddy had sent me recently. I noticed the UK researchers who wrote the report are funding the effort using GoFundMe. Hmmm . . .

My take away is this sentence: "Low-certainty evidence found that ivermectin prophylaxis reduced COVID-19 infection by an average 86% (95% confidence interval 79%–91%)." Note the clear statement of "low-certainly evidence."

Given that a doctor has to prescribe ivermectin, not much risk for people who don't mind taking low-dose medication for an "off-label" purpose. Personally I don't like the list of minor side effects. I'm happy enough having had shots of an mRNA vaccine that's gone through FDA review. Combined with sensible public health precautions, I don't see the need to take a drug. YMMV

Using a drug off-label happens all the time after a drug has been on the market for a while. Ivermectin has been around since the 1980s.

Sounds like when it comes to treatment for someone who already has symptoms for COVID-19, in particular to avoid serious symptoms, it's less clear whether Ivermectin can help or not. There is only so much that can be learned from a meta-analysis.

FDA and other regulatory agencies require clinical trials when a company wants to add to the list of approved indications for a drug. I worked on a project for several years for a new indication for a drug that had been on the market for decades. As I remember the Phase 3 studies didn't show any effectiveness for the new indication of interest. One reason Quintiles did well as a company is that a CRO makes money as long as the client company pays for services. Meaning it doesn't matter if a drug doesn't actually turn out to be effective when the Phase 3 analysis is done. Of course, success would have led to more business to put together the NDA safety databases and summary statistics.

Read the conclusion!

 

[tirolski]

My experience is with projects for drugs, not vaccines. What I know is that you can't change how you analyze efficacy data once patient enrollment starts.

The UNC Biostatistics professor and founder of Quintiles used to say that if a statistical analysis plan was so complicated that a Ph.D. level biostatistician was required to explain the results, then it was a bad idea. Most of the people in my department doing analyses after a clinical trial was completed were Master's level biostatisticians supported by SAS programmers. The company had a few people fully qualified to be high level "consultants" who would help work with client companies to come up with Phase 3 placebo-controlled analysis plans. Meaning biostatisticians who could charge $400+ per hour as a consulting fee (back in the 1990s). FDA biostatisticians would review the plans but I don't remember hearing about a situation where they suggested changes. Quintiles was too good for that to happen. The first few years, Quintiles was unusual because it only provided Biostatistics and Data Management services. The growth beyond 50 people happened after clinical services were added. Meaning CRAs (Clinical Research Associates) who would go around to research sites (often a doctor's office) to help make sure data was begin collected correctly for a given clinical trial.

My impression is that what Pfizer and Moderna did in the last 6 months or so is to work towards setting up separate research to address Delta. That includes lab studies that are are necessary before coming up with plans for Phase 1/2 and Phase 3 trials in humans. What's unusual is that setting up a small study in a country where Delta arrived sooner than the USA was another way to get some evidence of how a vaccine performed before doing an American clinical trial. There is also a lot of research being done in other countries.

Folks also use SAS software daily to determine and later study Critical Process Parameters (CPP’s) when developing and manufacturing medicines.
SAS is powerful validated software for analyzing big data and making sense out of it.
It would be interesting to see trend-line data for the vaccines.

I’m retired too and focus on critical in-process parameters for golfing and skiing mainly.
Humans invented mathematics, puzzles and games.
It’s fun to figure stuff out, especially good stuff.