Adding the numbers, since they settle part of this. Practical: whatever you change, write down the date and the reason. In three months the reason is what you will have forgotten, and the reason is what makes the record worth having.

Worth separating that from the trial evidence, which this thread keeps folding into the same question. They behave differently and the advice does not transfer.

One thing that is still open after SleepDoc_PDX’s answer:

What the dose-response curve actually looks like above 1.7mg, because the trial means hide how few people account for the extra loss?

josh_phd_bmore said:

Practical: whatever you change, write down the date and the reason.

Filing a mild objection. Mild because I might be wrong; an objection because nobody has addressed the case that does not fit. I would add the less popular caveat: these trial populations under-represented several groups, older adults and the highest BMI categories among them. The results probably generalise, and "probably" should be stated as an assumption rather than dropped.

That is the short version; the long version is somebody else's post.

josh_phd_bmore said:

Practical: whatever you change, write down the date and the reason.

That is correct as far as it goes, and here is where it stops going. Yes, and the corollary is less popular: if the reasoning holds, it also holds where it produces an answer we do not like.

josh_phd_bmore said:

Practical: whatever you change, write down the date and the reason.

That reframing is the part I needed.