[dynm]

When you have lots of non-randomized dropouts from a randomized trial, that greatly weakens the causal link. The results are effectively non-randomized.

Meanwhile the evidence from actual drinking levels was much stronger (far fewer dropouts) and showed zero effect. Before this trial was done, you may have predicted that there would be positive results for the lab experiment but zero results in ecological conditions. But I think that prediction would be quite unusual. For anyone who expected results in ecological conditions (like me), this was disappointing.

[llm_nerd]

It was a tiny study, and one arm had someone dropout because they had COVID. The difference was minuscule. You are being bizarrely selective in discarding the part you don't like, while holding the rest as demonstrative.

"But I think that prediction would be quite unusual."

Would it? Who in the world would guess that small ramp-up doses of a semaglutide (0.25, 0.5, and 1mg, the first two doses over 89% of the study duration. The average maintenance dose is 2.4mg) would immediately undo long-earned core habits over just two months? It's a rather absurd study.