[2026 update] Dose numbers across compounds are not on the same scale

On the subject in the title: Dose numbers across compounds are not on the same scale Working notes rather than a conclusion.

I have read the maintained page on this and I still have a gap, so I am asking rather than guessing.

Context: tirzepatide, 22 weeks in, currently at a dose I reached by the standard four-week steps. Everything below is my own record rather than anything a clinician told me.

The specific question is the one in the title. What I have already checked: the labelling summary on the relevant documentation page, the two most-linked topics in this subcategory, and my own notes from the last 10 weeks. What I could not find is whether the answer changes at higher doses or whether it is the same arithmetic throughout.

If the answer is "it depends", I would rather know what it depends on than be given a number.

Holding a dose indefinitely: the trials did not study indefinite holding at a non-maximum dose. The trials escalated to a target and then held that. What happens if you stay at an intermediate dose for years is not formally studied and extrapolation is the best available reasoning.

For anyone arriving from a search: the marked solution above is the direct answer, and the replies underneath it add the caveats that make it safe to use.

On post #2 — agreed on the reasoning, with one qualification.

The arithmetic of an intermediate dose: if the label says 1.0 mg and 2.0 mg, a dose strictly between them is off-label by definition. Some people compute it anyway. The reasoning is pharmacological — e.g., "I will split the difference between steps" — but it is reasoning from theory, not from evidence.

How the published trials escalated: they used specific step sizes and intervals. The STEP programme used a particular cadence; the SURPASS and SURMOUNT programmes used slightly different ones. Reading them side by side shows the variation is real but small.

When a dose reduction is the correct response to a side effect: if a side effect is dose-dependent (nausea, constipation, injection discomfort), reducing the dose is a reasonable response. If the side effect is not dose-dependent (e.g., hypoglycemia with insulin), dose reduction does not address the issue.

Worth separating two things that post #4 runs together.

Two things before anyone answers the substance.

First, the context in the first post is clear and specific. Second, the question is framed so that an answer can actually address it. Both are the norm here and both matter more than they sound.

Dose equivalence between different incretin analogues is a weak concept. The molecules differ in structure, half-life, receptor selectivity, and in what has been studied clinically. One mg of semaglutide is not equivalent to one mg of something else in any meaningful sense.

Coming back to post #8, because the follow-up matters more than the original answer.

Two things before anyone answers the substance.

First, the context in the first post is clear and specific. Second, the question is framed so that an answer can actually address it. Both are the norm here and both matter more than they sound.

post #11 answers the question as asked. The question underneath it is different.

Splitting a weekly dose in two: the pharmacokinetic argument against is that you want the benefit of long half-life, which gives a slowly changing plasma level from a weekly dosing schedule. Splitting it flattens the curve further but loses the convenience of once-weekly dosing. The trade-off is convenience versus a slightly flatter concentration curve.

Coming back to post #11, because the follow-up matters more than the original answer.

Titrating on tolerability rather than on the calendar: some people escalate when they tolerate a dose well, others escalate on the prescribed schedule regardless. The published trials used a calendar-based schedule. Tolerability-based escalation has no formal evidence base but is not uncommon in practice.

For anyone arriving from a search: the marked solution above is the direct answer, and the replies underneath it add the caveats that make it safe to use.