About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« The Evolution of Resistance: Are We Doing It Wrong? | Main | DTC Advertising: Is That Where Things Fell Apart? »

June 27, 2011

Making the Numbers Confess

Email This Entry

Posted by Derek

Adam Feuerstein calls this not just "post hoc data mining", but "extreme post hoc data mining". Take a look and see what you think.

Update: more here.

Comments (10) + TrackBacks (0) | Category: Clinical Trials


1. You're Pfizered on June 27, 2011 11:56 AM writes...

Can you imagine the meeting where the guys in Statistical Analysis were told they weren't leaving the room until they figured out some way to make the numbers look like a positive clinical result?

Permalink to Comment

2. SLC grad on June 27, 2011 11:59 AM writes...

A perfect demonstration of the adage " If you torture data long enough it will confess".

Permalink to Comment

3. anchor on June 27, 2011 1:01 PM writes...

Keep working on your numbers until you get it correct.

Permalink to Comment

4. drug_hunter on June 27, 2011 2:58 PM writes...

I always find it useful, before starting a new line of investigation, to imagine how I would answer the question, "What did you do at work today, Daddy?"

In this case, the answer -- "we tortured some data to produce a flimsy conclusion that might help us sell a drug to treat toenail fungus" doesn't quite pass the sniff test.

Permalink to Comment

5. Hap on June 27, 2011 3:35 PM writes...

1) You can find some good in every result. Just throw the letters and numbers from the text file of your results in a pile and rearrange them appropriately.

You mean that's not how drug discovery is supposed to work?

2) Is toenail fungus common in the absence of athlete's foot?

Permalink to Comment

6. bbooooooya on June 27, 2011 5:16 PM writes...

Spin to Win!

That's why the ceo was paid $465,000 in 2010.

I agree this should not pass sniff test, but this guy is being paid as mush as 4 mid level chemists (assuming mid level chemists now make 100K---no idea if that is correct). Not bad.

Permalink to Comment

7. Esteban on June 27, 2011 7:07 PM writes...

Easy prediction - this will never fly with the FDA. At best, something like this might pump the stock a bit. If you look at the intraday chart for APRI, it spiked momentarily but then gave most of that back, i.e., Wall St's not having any of it either.

Permalink to Comment

8. James on June 27, 2011 8:10 PM writes...

Perhaps they could try analyzing the data based on which little piggies got better. Better an approved product exclusively for middle toes coupled with smooth-talking sales reps than an unapproved therapy that was intended for the whole sha-bang. "This little piggy got market exclusivity" has a somewhat familiar ring to it.

Permalink to Comment

9. Ty on June 28, 2011 9:00 AM writes...

All joking aside, if you have a p value of 0.05 as your statistical significance mark, it means that it has a 5% chance of the observation being due to pure random chance. It's important to keep in mind this also means that if you look deep enough into the pile of random data, you only need to look at 20 different possible trends before you find one that passes the statistical significance test.

Permalink to Comment

10. DerekF on June 28, 2011 11:31 AM writes...

I think it's safe to say that the FDA will not accept the post-hoc analyses touted in the article - they don't generally accept post-hoc analyses, so Apricus is in for some more Phase IIIs. Big questions, of course, are (1) whether there are enough people with onychomycosis who don't also have tinea pedis to conduct the requisite two trials, given that IIRC they are the same fungus and lots of people get tinea once in a while, I suspect that onychomycosis is invariably accompanied by tinea, even though tinea is far more easily curable; and (2) whether those trials have a snowflake's chance of success.

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry