Thread by @GidMK

I often find discussion about industry funding in research very frustrating

Industry funding doesn't necessarily mean a study is bad: a thread

The basic assumption is this: companies fund research to prove that their products work. This is fairly obvious. No money in showing that your product is useless!

This leads to all sorts of potential conflicts of interest. @bengoldacre covers the various ways in which pharmaceutical companies can/do mislead in their research in his book Bad Pharma

And we see this fairly easily. There is a lot of research demonstrating that industry-funded studies are more likely to give positive results

BUT

Here's where it gets tricky

Numerous systematic reviews have also identified a key problem: industry-funded trials are just as good IF NOT BETTER than non-industry funded trials

This also makes sense: the industry has to use these studies to get drugs through regulatory authorities. Bad studies are less likely to lead to an approved drug!

So what's happening here? The conclusion of the Cochrane review I pictured above gives you some idea...

cochrane.org/MR000033/METHO…

The other systematic review I cited also paints a bit of a picture

What do you think it is?

See, the thing is, if you look at each individual study under a microscope, it looks like the industry is doing everything right! Maybe some sneaky things here and there, but in general they do as well as or better than independent research

The problem is largely meta-research issues like publication bias. This is why @senseaboutsci is pushing for publication of ALL research findings

If you can control which studies get published, and perhaps more importantly which studies get SEEN (big journals, media attention etc), then you can control the narrative about which drugs are best

(Yes, the answer was 2: publication bias in the poll above)

If someone says "but this study was funded by industry! That means it's bad!" you can cite the Cochrane or other reviews demonstrating why they are probably wrong

But when you read about a medical intervention, it's always worth having a look at the non-industry funded studies show, because they are less likely to suffer from publication bias

And importantly, THIS DOESN'T JUST APPLY TO DRUGS

If you compare industry-funded studies of ketogenic diets vs independent research, you see a similar trend

People who have a stake in the game are just less likely to publish research that might mean they will make less money. They also won't promote it, and may make it more difficult to access for everyone

This is, in many ways, much worse than simply doing bad studies

Bad studies are easy to find, easy to discard, easy to ignore. Most studies I've read on acupuncture, for example, are just awful

Can't conclude anything from bad research!

But by only publishing positive studies, and only promoting the very best ones, industries can manipulate the narrative by using basic statistical fact:

Do enough studies and some of them will be VERY positive

Some of them will be very negative too, but since you can control what gets published, people only see the good and never even hear about the bad

All of this to say: the industry can produce hundreds of amazingly well-done studies and STILL do things in a dodgy way, which is why you should support @senseaboutsci and pre-registration efforts everywhere!