When the first sentence of a published article starts with “Certainly, here is a possible introduction for you:”, one wonders if the value of the “peer-reviewed article” currency is inflated.

Amazing find by @gcabanac.cpesr.fr.

Next week, I'll give an online talk on estimands at @osi-luebeck.bsky.social. If you'd like to join, e-mail open.science@uni-luebeck.de to get the link.

Excellent chance to get my hot takes & empirically confirm that my professional photos are rather outdated.

osi-luebeck.de/yes-but-what...

1/2

New meta-analysis of 103 tricyclic antidepressant RCTs in n=10600 shows
➡️ all results at high risk of bias
➡️ certainty of evidence (very) low
➡️ studies lasted 12 weeks max
➡️ evidence of harmful effects compared to placebo (OR 2.78), but required information size *not* reached.