- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: NY HCQ study is a joke. We are being lied to
Posted on 5/13/20 at 9:58 am to IslandBuckeye
Posted on 5/13/20 at 9:58 am to IslandBuckeye
quote:
* I am wild arse guessing this may be where cardiac issues arise with HCQ +/- Azithromycin.
I'm telling you right now and you can bookmark this post: covid19 is an independent risk factor for development of TdP.
I have seen more torsades in the last 6 weeks than I have in the last 6 years. It's not even ambiguous, it's honest to god TdP. and these patients arent on any QT prolonging meds
0
Posted on 5/13/20 at 9:59 am to Roger Klarvin
quote:
Statistical significance means the odds that your results were due to chance are low.
I'm aware.
See link in my previous post.
Edit to add from previous post since it was at the bottom of the page.
"Give this editorial from the American Statistician a read, if you have the chance, to see what statisticians have been trying to tell us for a while now."
This post was edited on 5/13/20 at 10:33 am
Posted on 5/13/20 at 10:00 am to IslandBuckeye
I honestly don’t get as worked up about QTc as many others do. If your QTc is <550 without underlying structural heart disease the risk of medication induced torsades has always been pretty low, and it’s almost non-existent below 500. Where people get into trouble is the 76 year old with an EF of 25% and a QTc of 498, that’s the one that slips through the cracks.
I trained at an institution where nobody blinked if the QTc was around 500 and now I’ve got hospitalists calling me concerned about cipro use with a QTc of 480 in a healthy 40 year old
I trained at an institution where nobody blinked if the QTc was around 500 and now I’ve got hospitalists calling me concerned about cipro use with a QTc of 480 in a healthy 40 year old
This post was edited on 5/13/20 at 10:01 am
Posted on 5/13/20 at 10:00 am to Sasquatch Smash
quote:
We scientist likely put too much faith in statistical significance.
At times, sure. I've been a part of many studies where we might not have reached "statistical significance" but also still argued there is clinical benefit/extrapolation one way or the other. Similarly, other times we've found statistical significance that didn't meet MCID and therefore probably has no real world application. Thats part of being a researcher.
quote:
If those percentage numbers play out the same or similarly as you increase your population yet they aren't statistically different by the test, wouldn't you still want the outcome that's 29% better?
I believe in evidence and well designed studies. Sure, you can always shift around p-value thresholds if you'd like. And I do agree some people may be a bit too dogmatic about their research. But if you flip a coin 4 times and land on head 3 of those times, it doesnt mean you have a 75% of landing on head the 5th toss.
Posted on 5/13/20 at 10:01 am to Roger Klarvin
I don't start to worry about QT unless
1. QTc >500 and even then, "it depends"
2. QTc >480 and you're old with structural heart disease
1. QTc >500 and even then, "it depends"
2. QTc >480 and you're old with structural heart disease
This post was edited on 5/13/20 at 10:01 am
Posted on 5/13/20 at 10:03 am to Taxing Authority
quote:
Or you have suitable study entry criteria.
There's a reason observational studies aren't Level 1. I'm confused by your argument in this regard. Its not like these authors are trying to masquerade their study as if its an RCT.
Posted on 5/13/20 at 10:04 am to onmymedicalgrind
quote:
I believe in evidence and well designed studies. Sure, you can always shift around p-value thresholds if you'd like. And I do agree some people may be a bit too dogmatic about their research. But if you flip a coin 4 times and land on head 3 of those times, it doesnt mean you have a 75% of landing on head the 5th toss.
Thanks, now I know why I usually lost flipping quarters in high school.
Posted on 5/13/20 at 10:04 am to Roger Klarvin
quote:
Roger Klarvin
Do you think there's enough information on HCQ to rule it out as a viable, impactful treatment for Covid-19?
From what I've read, Remdesivir and IL-6 inhibitor appears to be more promising, but also still mostly unknown within the context of Covid-19 treatment.
It's incredibly difficult for a lay person to interpret all of the information floating around about these treatments. These are definitely the types of questions that our media isn't asking the task force, which is frustrating. I'm also very curious as to how contact tracing can work properly without violating our privacy laws, and it seems like contact tracing is required in a a containment scenario after the spread is more controlled. Again, not political or "gotcha" enough to make the cut in those daily WH briefings.
Posted on 5/13/20 at 10:05 am to Tiguar
quote:
the numbers are buried in the supplemental appendices of the study.
Thanks. The twitter analysis is misleading as well.
quote:
And here are the (surprising?) results:
Death / Primary Outcome
HCQ: 157 / 262 = 60%
No-HCQ: 75/ 84 = 89% !
He's using respiratory failure as the denominator, and only looking at deaths prior to intubation. If you use the total N of each treatment group, and look at all deaths (those without intubation and despite intubation) you get very different results:
HCQ:
death w/o intubation 157
intubated then died 49
No HCQ:
death w/o intubation 75
intubated then death: 17
that's a 25% mortality rate in HCQ treated group and 16.2% in the no-HCQ group. Now of course the baseline characteristics aren't matched, so I wouldn't take much from this, but the twitter evaluation (and your OP) is misleading.
This post was edited on 5/13/20 at 10:07 am
Posted on 5/13/20 at 10:07 am to shell01
it's not really misleading because he tells you what the data says.
the summary sentence says less patients died after developing respiratory failure in the hcq group.
when you have conflicting results like that, (overall mortality rates flip-flopped), it tells you something is very wrong about the study recruitment and/or there is a "Goldilocks" zone for the drug in question
the summary sentence says less patients died after developing respiratory failure in the hcq group.
when you have conflicting results like that, (overall mortality rates flip-flopped), it tells you something is very wrong about the study recruitment and/or there is a "Goldilocks" zone for the drug in question
This post was edited on 5/13/20 at 10:11 am
Posted on 5/13/20 at 10:11 am to Tiguar
nm
This post was edited on 5/13/20 at 10:14 am
Posted on 5/13/20 at 10:14 am to shell01
quote:
See how easy it is to cherry pick data?
He is telling you he is cherry-picking data. That's what a subgroup analysis is: cherry-picking data.
This isn't the gotcha you think it is.
Posted on 5/13/20 at 10:15 am to onmymedicalgrind
quote:Oh, I get the whole "stated limitation" here. But given the limitations one cannot make any reasonable conclusions. I also get that's perfectly fine for academic standards--where you can get paid to do it all over again.
I'm confused by your argument in this regard. Its not like these authors are trying to masquerade their study as if its an RCT.
Posted on 5/13/20 at 10:15 am to Tiguar
so why make a big deal over a subgroup analysis? We all know they are grasping at straws when you rely on a subgroup analysis.
Posted on 5/13/20 at 10:18 am to shell01
Because the authors elected not to do this subgroup analysis when every study that has a composite outcome including mortality always pulls out mortality separately in a subgroup analysis. This calls their intent into question.
Posted on 5/13/20 at 10:19 am to GumboPot
quote:
ETA: thank God most front line doctors understand this and are ignoring the negative propaganda about HCQ.
Much to the chagrin of the “inexplicably opposed to plaquenil at all costs” group, you are correct.
Also, :inbeforeyouwillgoblind:
Posted on 5/13/20 at 10:22 am to Tiguar
Come on, the subgroup is "death after respiratory dysfunction"....overall all-cause mortality is the appropriate analysis.
ETA: And when looking at overall mortality, you have 19% in the HCQ group, and 13% in the non-HCQ group. The only reason you're talking about the cherry-picked subgroup of "death after death or intubation" is because it gives you the result you want.
ETA: And when looking at overall mortality, you have 19% in the HCQ group, and 13% in the non-HCQ group. The only reason you're talking about the cherry-picked subgroup of "death after death or intubation" is because it gives you the result you want.
This post was edited on 5/13/20 at 10:30 am
Posted on 5/13/20 at 10:32 am to Taxing Authority
quote:
Oh, I get the whole "stated limitation" here. But given the limitations one cannot make any reasonable conclusions.
I'd agree with that; this paper isn't terribly compelling evidence one way or the other. Every study should have a stated conclusion though, and its up to the those reading it to determine how much value they put into said conclusion based on the overall study quality. This one benefits from large sample size, but overall lacking study design.
Posted on 5/13/20 at 10:34 am to dewster
quote:
Do you think there's enough information on HCQ to rule it out as a viable, impactful treatment for Covid-19?
Rule it out? No
I remain very skeptical given the proposed mechanisms and data to date, but I can’t say it doesn’t work with sound authority.
Posted on 5/13/20 at 10:45 am to shell01
quote:
And when looking at overall mortality, you have 19% in the HCQ group, and 13% in the non-HCQ group. The only reason you're talking about the cherry-picked subgroup of "death after death or intubation" is because it gives you the result you want.
I have to say you have a point. I’d like to see what Tiguars response to this is.
I agree overall mortality is what we should be looking at. Unless there is a compelling argument otherwise.
This post was edited on 5/13/20 at 10:47 am
Page 5 of 6
Popular
Back to top
Follow TigerDroppings for LSU Football News