Discussion about this post

User's avatar
Frank Harrell's avatar

The general problem: Data collected for other purposes is free and apparently irresistible. The false hope of "real world data" is biting us in the ... while delaying the launch of needed randomized trials.

Expand full comment
Adrian Gaty's avatar

Thank you for your fantastic explanations of this issue lately.

“they used claims data from a US administrative database, from OptumLabs.” This part stood out to me. Could it be as simple a matter as laziness? That is, in the old days, before fingertip access to huge online medical databases, you had to do the legwork of putting a study together, in which case if you’re putting in the effort might as well go the whole nine yards and do an RCT. But when you can get published in a major journal without leaving your desk, just running some opaque statistical tricks from someone else’s computer model on someone else’s database, it’s tempting to cut corners, no? Just a thought! I could be way off.

Expand full comment
20 more comments...

No posts