Lesson of ivermectin: meta-analyses based on summary data alone are unreliable • Hacker News Board

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#481

Post by nradov » Fri, Sep 24, 2021, 5:10 PM UTC

scoopertrooper wrote: post #436

alvah wrote: post #328

Earlier quoted context omitted.

The reported CFR may be in that region, but most credible sources estimate a much lower real CFR, due both to many undetected / asymptomatic infections, and confusion between deaths with & deaths caused by Covid-19. For a statistical reporting agency such as the ABS to claim to know the actual CFR with any level of certainty seems careless at the very least.

The ABS clearly sets out their method for calculating CFR, which is covid deaths divided by confirmed infections. Yes, that is only a proxy for the overall all infection fatality rate. However, given the contact tracing, testing, and medical systems have been quite robust in Australia, it shouldn't be too far off the mark. Furthermore, the 3% figure only applies to the population as a whole. According to the ABS, CFR…

The actual infection fatality ratio is nowhere near 3%. The US CDC estimated it at 0.6%.

https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/burd...

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#482

Post by Clubber » Fri, Sep 24, 2021, 7:07 PM UTC

shadowgovt wrote: post #312

Clubber wrote: post #301

Earlier quoted context omitted.

The example I was replying to was when the patient was already diagnosed with a chronic illness and still had to go back to the doctor every month or so to renew the script, because reasons.

Me too. When the disease is something that requires frequent reapplication of antibiotics, it's important to confirm periodically that they're working (and that the problem isn't a different bacteria that's amoxicillin-resistant).

That's a convenient method when the patient can also make a phone call / appointment to inform the doctor they don't seem to be working rather than extorting (and I'm not using that word lightly) $150+ a month. Also consider the multitudes of other medications where there is no real argument to require a monthly extortion fee.

It seems to be this is more for the benefit of the healthcare system financially rather than the patient or the health of the population as a whole.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#483

Post by snovv_crash » Fri, Sep 24, 2021, 10:43 PM UTC

brandonmenc wrote: post #459

snovv_crash wrote: post #434

Earlier quoted context omitted.

Can you show any other vaccine which has long term, but not short term negative side effects, which this concern might be based on? Because this seems like whataboutism, and of course it's impossible to prove a negative, and people are dying while waiting. > We do have data on the long-term effects of Ivermectin At bi-annual dosage rates, yes. Not daily or even weekly.

> Can you show any other vaccine which has long term, but not short term negative side effects, which this concern might be based on? I don't know. But we still may be in the "short-term side effect" phase for the COVID vaccines. It hasn't even been a year. > Because this seems like whataboutism, and of course it's impossible to prove a negative, and people are dying while waiting. Not sure how it's whataboutism to p…

It's been over a year for the initial trials, no long term issues yet. How long seems reasonable to you? What are you basing this timeframe on?

If covid is a low risk, then the vaccine, which is essentially just the immune response to covid without any of the actual viral damage, should be much more benign, yes? I'm failing to see the risk here.

The treatment regimens in current Ivermectin studies are usually daily dosage, for which there are no long term studies for humans. Prophylactic usage of Ivermectin is all over the place, but seems to mostly be within 72h of exposure, so several doses per week, which also doesn't have long term studies. Staying on antibiotics for long periods is generally a bad idea due to disturbing the gut microbiota.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#484

Post by veemjeem » Sat, Sep 25, 2021, 12:41 AM UTC

programmarchy wrote: post #371

veemjeem wrote: post #303

Earlier quoted context omitted.

Many things are effective against covid-19 in-vitro. When it was administered in hamsters, there was no effect on viral load. https://www.biorxiv.org/content/10.1101/2020.11.21.392639v1

I hadn’t seen that paper, but isn’t your statement missing the forest for the trees? If pathology was attenuated, that would help patient recovery, even if viral load wasn’t directly reduced. > Even though ivermectin had no effect on viral load, SARS-Cov-2-associated pathology was greatly attenuated. IVM had a sex-dependent and compartmentalized immunomodulatory effect, preventing clinical deterioration and reducing…

Yes, it seemed like it worked in hamsters, but as we well know, most drugs that work in mice don't work in humans. Besides, we already have many anti-inflammatory drugs that can reduce the cytokines, so it's a stretch to use ivermectin strictly as a steroid.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#485

Post by ekianjo » Sat, Sep 25, 2021, 4:06 AM UTC

gnicholas wrote: post #390

ekianjo wrote: post #330

Earlier quoted context omitted.

how would you know when they actually started the analysis?

That's one of the reasons I was asking. Perhaps they could pre-register prior to being given the full data from other studies?

If the data is kind of public or going around in your circles you could easily hack this pre-registration constraint.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#486

Post by ekianjo » Sat, Sep 25, 2021, 4:07 AM UTC

rapind wrote: post #391

ekianjo wrote: post #329

Earlier quoted context omitted.

FiveThirtyEight forecasts are very questionable themselves. Been completely wrong on multiple counts.

So is the weather forecast though. 10% chance of rain tomorrow means, yeah it could rain. Doesn't mean it isn't useful.

It's on the aggregate of predictions that you can see the accuracy. If my weather app tells me hundred times there is 10% change of rain but it rains 50% of the time after 100 days, it's completely useless.

Same for FiveThirtyEight. They have been massively wrong on multiple counts.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#487

Post by shadowgovt » Sat, Sep 25, 2021, 2:27 PM UTC

Clubber wrote: post #482

shadowgovt wrote: post #312

Earlier quoted context omitted.

Me too. When the disease is something that requires frequent reapplication of antibiotics, it's important to confirm periodically that they're working (and that the problem isn't a different bacteria that's amoxicillin-resistant).

That's a convenient method when the patient can also make a phone call / appointment to inform the doctor they don't seem to be working rather than extorting (and I'm not using that word lightly) $150+ a month. Also consider the multitudes of other medications where there is no real argument to require a monthly extortion fee. It seems to be this is more for the benefit of the healthcare system financially rather tha…

> Also consider the multitudes of other medications where there is no real argument to require a monthly extortion fee.

I think that's really where one needs to look to square the circle. The drugs available prescription-free have lower risk of injury or disruption when abused. Antibiotics are not in that category, which is why they require doctor oversight and a prescription. Take too much, and you can damage your digestive tract. Take too little, or the wrong kind, and you can worsen the infection by killing the bacteria that the immune system can control, leaving their food supply for the ones the immune system is struggling against.

They could require less doctor oversight perhaps, but probably not zero.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#488

Post by ivm_man » Sun, Sep 26, 2021, 10:32 PM UTC

X3h7P4icntav wrote: post #442

The question of whether ivermectin treats Covid19 in humans is utterly independent of vaccines, HCQ, the state of modern science, whether ivermectin treats other diseases, ivermectin's use in animals, politics, any corrupt ivermectin studies (if you ignore them), and the price of cheese on the moon. I have learned there are many idiots among us. Many, many idiots.

So true...

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#489

Post by NaturalPhallacy » Tue, Sep 28, 2021, 12:26 AM UTC

brigandish wrote: post #440

NaturalPhallacy wrote: post #406

Earlier quoted context omitted.

>There is no good faith argument against the vaccine. Factually false. 1. It isn't a vaccine. It's a chemical signal to make your body make things. 2. It doesn't have the efficacy against the plethora of variants that all coronaviruses like the cold continuously create. You can still get COVID-19 after being "fully vaccinated". Human beings lack the technology to vaccinate against coronaviruses, including COVID and t…

It seems that the second half of (3), and (5) are myths. https://www.hopkinsmedicine.org/health/conditions-and-diseas...

Your link says nothing about 5.

Re: Lesson of ivermectin: meta-analyses based on summary data alone are unreliable

#490

Post by GuB-42 » Tue, Sep 28, 2021, 1:54 PM UTC

atty wrote: post #218

What I find almost more interesting than the article itself is the comments that show a non-zero portion of the HN community (which I think is generally more intellectually curious and educated than average) are actually strongly arguing that ivermectin has strong effect, and even crazier, that it may be safer or more effective than the mRNA vaccines. Is this from being incapable of following science and drawing rati…

Recognized immunologists have argued that their favorite unproven treatment is the best.

I know plenty of anti-vax, some of them remarkably smart, some of them are health professionals.

We are all biased, and overcoming bias is hard. It is a field of expertise by itself, and it is usually not required for tech jobs. Just being smart will only allow you to find more complex but just as fallacious arguments that will confirm your bias. For example, it took me a while to "believe" in climate change, I took a while to realize that solar panels are not just for pocket calculators, and things like that, for a variety of reasons, and found a lot of advanced material to fuel my bias, which still has not completely disappeared. I am not particularly attached to Ivermectin so there is no bias for me to overcome, but I understand that some people can be. Try to think about it yourself, you probably believe in stupid things too.

And getting attached to a cheap treatment for a disease that has been messing with our lives since early 2020 is not what I would consider an unhealthy reaction. It is hope. Getting too attached to it can be a problem, but no matter what we want to think, we are just humans.