The approved schedule starts at 2.5 mg once weekly for four weeks and then moves to 5 mg, with any further steps in 2.5 mg increments after at least four weeks each. The prescribing information states that 2.5 mg is for treatment initiation and is not approved as a maintenance dose. Whether an individual stays there is a prescriber decision, and it is a real conversation rather than a closed one.
The labeled schedule, in full
| Step | Minimum time before an increase | Status under the approved label |
|---|---|---|
| 2.5 mg weekly | 4 weeks | Treatment initiation, not an approved maintenance dose |
| 5 mg weekly | At least 4 weeks | Lowest approved maintenance dose for weight reduction |
| 7.5 mg weekly | At least 4 weeks | Intermediate titration step |
| 10 mg weekly | At least 4 weeks | Maintenance for weight reduction and the lower of the two for sleep apnea |
| 12.5 mg weekly | At least 4 weeks | Intermediate titration step |
| 15 mg weekly | Maximum | Highest approved dose for either indication |
Two features of that table are easy to miss. The intervals are minimums rather than deadlines, so a slower climb is within the schedule. And the maintenance range differs by indication: 5 mg, 10 mg, or 15 mg for weight reduction and long-term maintenance, but 10 mg or 15 mg for moderate to severe obstructive sleep apnea in adults with obesity. Someone treating sleep apnea is looking at a maintenance range that begins four steps above the initiation dose.
That same ladder is what a patient meets when shopping for a supplier, and it gets presented with very different levels of candor. The manufacturer’s LillyDirect pharmacy, along with telehealth prescribers such as Henry Meds, Sesame, and HealthRX, each keep a dosing and price page for the branded product, and the HealthRX Zepbound listing lays out this 2.5 mg to 15 mg sequence in full. Reading a couple of them early tends to show whether a service frames the increases as a clinical judgment or as an automatic climb to the top.
Why the ramp exists
Tirzepatide activates both the GIP and the GLP-1 receptor. The same signaling that reduces appetite and slows gastric transit is what produces nausea, vomiting, and diarrhea, so the wanted and unwanted effects rise together rather than separating cleanly. The label states plainly that the escalation sequence exists to reduce the risk of gastrointestinal adverse reactions.
Trial data show where those reactions land. In SURMOUNT-1, the 72-week obesity trial, gastrointestinal events were the most common adverse events, mostly mild to moderate, and occurred primarily during dose escalation. Adverse events led to discontinuation in 4.3 percent of the 5 mg group, 7.1 percent at 10 mg, and 6.2 percent at 15 mg, against 2.6 percent on placebo.
What the efficacy evidence was actually measured at
This is the part that deserves honesty. The published effect sizes for tirzepatide in obesity come from participants who escalated to maintenance doses, not from people who stayed at the initiation dose. SURMOUNT-1 randomized participants to 5 mg, 10 mg, or 15 mg weekly and reported mean weight change at week 72 of 15.0 percent, 19.5 percent, and 20.9 percent respectively, against 3.1 percent with placebo. There is no arm in that trial that stayed at 2.5 mg.
The sleep apnea program tells the same story from a different angle. In SURMOUNT-OSA, two 52-week trials in adults with moderate to severe obstructive sleep apnea and obesity, participants received the maximum tolerated dose of 10 mg or 15 mg. The apnea-hypopnea index fell by roughly 25 to 29 events per hour with tirzepatide against about 5 with placebo. Again, the initiation dose was a stepping stone, not a study arm.
The practical reading is straightforward. Remaining at 2.5 mg is likely to produce a smaller effect than the figures people have seen quoted, because those figures were generated elsewhere on the curve. That is not an argument that the decision is wrong for a given person, only that the expectation should be set correctly before it is made.
What is known about deliberately lowering the dose
There is evidence on dose reduction, and it starts higher than 2.5 mg. SURMOUNT-MAINTAIN, a 112-week trial reported in 2026, ran a 60-week open-label weight-loss period at the maximum tolerated dose of 10 mg or 15 mg, then randomized participants to continue at that dose, drop to 5 mg, or switch to placebo for 52 weeks. Mean weight change from baseline at week 112 was 21.9 percent for the maximum tolerated dose, 16.6 percent for the 5 mg group, and 9.9 percent for placebo.
Two things follow. Stepping down preserved a substantial share of the result, which is why clinicians treat a lower dose as a legitimate option rather than a failure. And the dose studied was 5 mg, an approved maintenance dose reached after more than a year at a higher one, which is not the same as never escalating past initiation.
Separately, a 2025 analysis modeled alternative dosing regimens for GLP-1 receptor agonists as a way of reducing cost while retaining effect. It is modeling work rather than a labeled schedule, and it belongs in a conversation with a prescriber rather than a decision made alone with a calendar.
Where cost and tolerability legitimately enter
The label itself makes room for this. It instructs prescribers to consider treatment response and tolerability when selecting a maintenance dose, and states that if a patient does not tolerate a maintenance dose, a lower maintenance dose should be considered. Goals matter too, and current obesity pharmacotherapy guidance frames treatment intensity around what the patient is trying to achieve rather than around reaching a maximum.
Cost is not an illegitimate input either, but it works better stated than silent. A prescriber who knows the monthly figure is the constraint can talk about coverage, about which product is written, and about whether the plan is realistic at any dose.
That makes the service structure worth checking before starting. Practices, manufacturer channels such as LillyDirect, and telehealth prescribers including Ro, Hims & Hers, LifeMD, and formblends.com differ in how quickly a dose question gets answered, in whether the same clinician follows the case, and in what a message costs. A program that bills per consultation quietly discourages the call that a titration period depends on.
Compounded tirzepatide is a separate question entirely
Everything above describes an FDA-approved product with a fixed concentration and reviewed labeling. Compounded tirzepatide is not FDA-approved, its concentration is set by the compounding pharmacy, and there is no published escalation schedule for it. Branded milligram figures do not transfer, and any plan for a compounded preparation comes from the prescriber who wrote it and the pharmacy that made it.
One screening item applies regardless of source. Tirzepatide carries a boxed warning based on thyroid C-cell tumors in rats, and it is contraindicated in anyone with a personal or family history of medullary thyroid carcinoma or with multiple endocrine neoplasia syndrome type 2.
Frequently asked questions
Is holding at 2.5 mg longer than four weeks allowed?
The four-week interval is a minimum, so extending time at a step is within the schedule and is a common response to poor tolerability. Holding indefinitely is a different proposition, since 2.5 mg is not an approved maintenance dose, and that distinction is the prescriber’s to draw.
Does everyone have to reach 15 mg?
No. For weight reduction, 5 mg, 10 mg, and 15 mg are all approved maintenance doses, and the label directs prescribers to select one based on response and tolerability. For moderate to severe obstructive sleep apnea the approved maintenance range is narrower, at 10 mg or 15 mg.
How long do symptoms after an increase usually last?
They are typically concentrated in the days following a step and settle over the following weeks at a steady dose. Symptoms that keep escalating, or that begin without any recent change, fit a different pattern and are worth reporting rather than waiting out.
Can a step be reversed instead of held?
Yes. The label explicitly contemplates a lower maintenance dose when a higher one is not tolerated, and returning to a previous step before trying again is a recognized adjustment. It is a clinical decision rather than something to arrange by rationing the current supply.
