Nature: The lesson of ivermectin: meta-analyses based on summary data alone are inherently unreliable

The lesson of ivermectin: meta-analyses based on summary data alone are inherently unreliable
" Since the above primary studies were published, many hundreds of thousands of patients have been dosed with ivermectin, relying on an evidence base that has substantially evaporated under close scrutiny. …Relying on low-quality or questionable studies in the current global climate presents severe and immediate harms. …"
it’s brief and they miss some obvious points (made here on the Quillette forum) re meta-analysis that should have led anyone to consider the ivermectin meta-analysis dubious

edit: please recall that weinstain (and heather) said meta-analysis is superior evidence to eg RCT, so stupid

3 Likes

Do they teach research design and statistics in evolutionary biology?

Undergraduate students in psychology generally have a better grasp of statistics than what I’ve seen from Bret and Heather.

2 Likes

I didn’t realize doing meta-analysis using patient level data was somehow novel or unusual. I would’ve thought that to be standard practice.

As the saying goes, garbage in garbage out. Any meta-analysis can only be as good as the worse data set among those chosen to base it upon. I didn’t realize that some of these researchers were really raiding the discount bins.

And this:

is truly mind-boggling.

In isolation, it’s not a huge deal. Some people took some ivermectin…hopefully at human doses. But the real tragedy and travesty is what some of these good folks eschewed whilst reaching for dewormer instead. That said, the venns for people who would take ivermectin based on no good data, and that of people who would be receptive to vaccination, probably don’t overlap by very much.

2 Likes

heather used to write about statistics on her blog, she said she taught it, it was very basic though

depressing to think that today’s carl sagan (for 12 y.o.'s whose parents are not scientists and dont care for it) is bret and heather, and their book is now a now nyt best seller

2 Likes

Actually I kinda think that Neil deGrasse Tyson has stepped more into that role. Either way, none of the three of em can come close to stepping into Carl’s shoes. Course thats no great sin considering how high he set the bar. “The Demon-Haunted World” remains one of the books that has most impacted my life and worldview.

3 Likes

yes, demon haunted world: ‘science as a candle in the dark’, same impact on me

1 Like

Even evolutionary biologists think he - Bret - is a whack job.

What I find amusing is that the first evolutionary biologist to make it big with the American Right - which would include a lot of evangelicals - is one who completely mangles his own field.

1 Like

i forgot about that brief conflict. I recall you suggested he publsh his ideas and Eric interceded to ‘explain’ that such a comment was out-of-date as the system is corrupt. I do enjoy these spats, to be honest, i’ll read that blog post

-maybe this applies to stephen j gould and his non overlapping magesteria. im convinced heather and bret have evolved this way because they were cloistered in portland lecturing to younguns and mostly unexposed to peer review etc

1 Like

i guess ppl have seen this:
“The real question is why none of the groups promoting ivermectin as a mass treatment for Covid-19 did their basic due diligence, because much of the fraud is really not that hard to identify.”

someone said "Authors paid $1,200 to get published - Journal of Biomedical Research and Clinical Investigation - I couldn’t even find an Impact Factor "

1 Like

Wow. What a massive face-palm.

Peru is a nice example of the unintended (or maybe it was intended) knock-on effects of widespread use of anything prior to that thing being proven to be of any actual benefit - you contaminate the population such that it becomes much more difficult to recruit treatment naive patients in order to properly test and establish the efficacy of that treatment to begin with.

1 Like