Tool 'names and shames' hidden drug trials | Hacker News Reader

Tool 'names and shames' hidden drug trials

bbc.co.uk

87 pointsasplake18 comments

I hope that this is widely used, and that Private Eye picks up on it as a source of information. Also if any commenters haven't read "Bad Science", and the follow-up "Bad Pharma" by Ben Goldachre (who inspired the tool) they are well worth a read.

There's also a book called Bad Medicine: Doctors Doing Harm Since Hippocrates from Oxford University Press.

https://global.oup.com/academic/product/bad-medicine-9780199...

It's not by the same author, but adds a nice historical perspective. On a long scale, medicine is mostly about harming your patient with treatements that are worse than nothing.

Bad Science is a very insightful and witty read with as many shocking parts as there are hilarious parts.

It's a while since I tried to read it, but Bad Pharma seemed to be much more focused on the fundamental problem and its effect, which I found to be a very hard read, I never finished it. No fault of Goldacre's writing, it's just a horribly horrible and massively massive problem.

Yep - Also found it a hard to read. I recommend the Audio Book read by Ben, much easier to consume the information having it read out. Plus, Ben has a lovely voice which helps.

From the FDA:

> It is often not possible to determine which parties may be non-compliant based solely on the information in the record that is publicly posted on clinicaltrials.gov

Considering that "nobody has ever been penalised for breaching those codes", this sounds like a handwave. What's the point of the website if it doesn't accurately report missed deadlines? Or if missing deadlines doesn't breach the rules then in what sense are they deadlines?

The codes have been in effect for less than a month.

Right but now that the codes are in effect, what about the website makes it "often not possible" to tell if anyone is non-compliant?

Ah. That bit of missing context makes a big difference.

I'm a little confused here. All trials have to be reported within 12 months of completion, and the law goes into full effect in February. So the tool gives the trials which have been overdue for some part of February?

The "US govt could have imposed fines" part seems to be way off too (unless maybe it's in thousands?). It says they can levy fines of $10k/day. Just clicking on one company, Bassett Healthcare, shows there is 1 trial overdue by 32 days. Only 22 days have elapsed in February (when it sounds like this portion of the law went into effect), so shouldn't that company be able to be fined at least 22*$10k=$220k? It has the total for all companies at $290k on the home page.

Maybe I'm misunderstanding something, but the tool doesn't have a lot of description that I could find.

The fines would be administered for each day the trial is late. Technically, sponsors have 1 year to post results after the primary completion date stated on ct.gov. This is supposed to be the anticipated date that the last primary outcome measure is collected. Therefore, if a sponsor estimated to collect that final measure on 12/1/2018, they would need to have results registered on ct.gov on 12/1/2019. If they were missing results on 12/2/2019 (and there are no certified delayed results) they'd be fined 10K for that day. Also worth noting that the sponsor would receive a notice that their out of compliance and have 30 days to rectify before a fine is issued.

now: "Percent reported: 89.7%"

website: http://fdaaa.trialstracker.net/

github: https://github.com/ebmdatalab/clinicaltrials-act-tracker/

github issues: https://github.com/ebmdatalab/clinicaltrials-act-tracker/iss...

Nice! Didn't know that law existed in the first place. Good to see it's being monitored right from inception.

My understanding is that pharma companies have been much more compliant with reporting than public institutions.

Yeah, most pharma companies have clinical trial disclosure teams whose sole responsibility is disclosure to various registries. Most public universities and hospitals with large research facilities usually register studies, but the data is crap and isn't maintained with the same rigor as a pharma company. That being said pharma companies are more likely to omit confidential information regarding thier products.

I initially read this as the band Tool.

I can't be the only one who thought this was about the band Tool.

New album soon...been holding my breath for what feels like 10,000 days.

BLAME HOFMANN

Reply on news.ycombinator.com