# Five Clicks Instead of a Quarter Milligram

By Nora Bissett, Pricing Editor (https://glpwatchdog.org/authors/nora-bissett) | Published: 2026-08-20 | Updated: 2026-10-27

Canonical page: https://glpwatchdog.org/research/five-clicks-instead-of-a-quarter-milligram

> One in five people on the label's titration schedule quit because of gut side effects. On a slower schedule it was one in fifty, at the same final dose, with the same results.

## Key points

- 2% withdrew for GI side effects vs 19%, at similar final doses and similar HbA1c and BMI.
- The nausea rate itself came in at P=0.051 and did NOT reach significance. Days of nausea did.
- The slow arm started at 0.0675 mg, five clicks of the pen dial.
- Click-counting is not a validated measurement and this is not a set of instructions.
- Pilot, N=104, open-label, type 2 diabetes, doses to 1 mg only.
- The paper carries a published erratum we have not read.

One in five people in the standard-titration arm of this trial quit because of gastrointestinal side effects. In the arm that climbed more slowly, it was one in fifty, and both groups **ended on similar doses with similar blood sugar and weight results**.[1] If that holds up, the schedule printed on the label is costing a substantial number of people their treatment for nothing.

## What the two schedules were

One hundred and four people with type 2 diabetes were randomized to one of two semaglutide schedules over 26 weeks. The label’s schedule steps up every four weeks, from 0.25 mg to 0.5 mg to 1 mg. The other was slower and flexible.

The flexible arm started at **0.0675 mg**, which the paper describes as five clicks of the pen’s dose selector dial. That is roughly a quarter of the label’s already-low starting dose. From there it rose by another 0.0675 mg each week, with increases postponed whenever gastrointestinal side effects appeared.

*Results at 26 weeks.[1]*

|   | Flexible titration | Label titration | P |
| --- | --- | --- | --- |
| **Withdrew due to GI side effects** | **2%** | **19%** | 0.005 |
| Reported nausea | 45.1% | 64.2% | **0.051** |
| Days with nausea | 2.88 | 6.3 | 0.017 |
| Asthenia (weakness) | 9.8% | 24.5% | 0.047 |
| Final dose | Similar | Similar | — |
| HbA1c and BMI change | Similar | Similar | — |

## Read the nausea row honestly

> **The headline difference in nausea did not reach statistical significance.** At P = 0.051 it sits a thousandth on the wrong side of the conventional line. It is going to be quoted as though it did, and it should not be. What did reach significance is the number of *days* people spent nauseated, the rate of weakness and, most important, how many people gave up.

That combination is coherent even with the nausea rate itself unresolved. Roughly similar proportions may feel sick, but for fewer days, less severely, and with the option of pausing the climb. That is the difference between an unpleasant few months and quitting.

> Same destination, gentler road, and about a ninth as many people quitting over side effects.

## Why this matters more than most trial results

Almost everything in this field trades benefit against harm: more weight loss, more side effects. This does not. The flexible arm reached comparable doses and comparable results. The cost of the gentler schedule was time, and the return was a large reduction in people abandoning treatment.

Discontinuation is the central problem with these drugs. People stop, and weight comes back; that is the whole subject of [what happens when you stop](https://glpwatchdog.org/research/what-happens-when-you-stop-a-glp1). Anything that keeps people on treatment without costing efficacy is worth more than another percentage point of weight loss.

It also reframes what the label’s starting dose is for. The 0.25 mg step exists to be tolerated rather than to work, and this trial suggests that for a substantial minority it is not low enough to do that job.

## Why this is not a set of instructions

> **Counting clicks on a pen dial is not a validated way to measure a dose.** These devices are engineered to deliver the marked increments, not fractions between them, and the accuracy of a five-click dose is not something a manufacturer certifies. In the trial, that was done inside a protocol with clinicians supervising and a schedule to follow.

The thing to take from it is not a number of clicks but that **slowing down is a legitimate option with randomized evidence behind it**. A prescriber can hold a dose or [extend the interval](https://glpwatchdog.org/research/every-other-week-dosing), and the labels already contain language for that: “as tolerated.” Someone struggling with the standard climb has something specific to bring to that conversation now.

The same click-counting technique used as a permanent low dose rather than a slow start is a different practice with a different evidence base, covered in [microdosing](https://glpwatchdog.org/research/glp1-microdosing-what-evidence-exists). Some sellers now sell low-dose compounded semaglutide as a plan of its own, and our [semaglutide microdose board](https://glpwatchdog.org/best/semaglutide-providers/microdose) lists them with what each one charges.

## The limits

- **It is a pilot.** One hundred and four people, described as such by the authors.
- **Open-label.** Everyone knew which schedule they were on, and expectation affects reported nausea more than most outcomes.
- **Type 2 diabetes only**, on semaglutide doses up to 1 mg, not the 2.4 mg used for weight management, where the climb is longer and the top dose higher.
- **Twenty-six weeks**, which covers the titration period and little beyond it.
- **The paper carries a published erratum** whose content we have not read.

## Frequently asked questions

### Does titrating semaglutide more slowly reduce side effects?

In this randomized pilot, 2% of the slow-titration group withdrew because of gastrointestinal side effects against 19% on the label schedule, with fewer days of nausea and less weakness. Final doses, HbA1c and BMI were similar.

### Did slower titration reduce results?

No. Both arms ended on similar doses with similar changes in blood sugar and body weight. The cost of the gentler schedule was time.

### What is the 'five clicks' dose?

0.0675 mg, measured as five clicks of the pen's dose selector dial, or roughly a quarter of the label's lowest step. It was done inside a supervised trial protocol, and counting clicks is not a manufacturer-validated way to measure a dose.

### Did nausea itself go down significantly?

Not by the conventional threshold. Nausea was reported by 45.1% against 64.2%, at P = 0.051. What did reach significance was the number of days spent nauseated, the rate of weakness, and how many people withdrew.

### Can I ask my prescriber to slow down?

That conversation is exactly what this evidence supports. The labels already build 'as tolerated' language into their titration schedules so a clinician can hold a dose or extend an interval. There is now randomized evidence that doing so keeps people on treatment.

## References

1. Eldor R, Avraham N, Rosenberg O, et al. Gradual Titration of Semaglutide Results in Better Treatment Adherence and Fewer Adverse Events: A Randomized Controlled Open-Label Pilot Study Examining a 16-Week Flexible Titration Regimen Versus Label-Recommended 8-Week Semaglutide Titration Regimen. Diabetes Care. 2025. PMID: 40673973. https://pubmed.ncbi.nlm.nih.gov/40673973/

---

Cite as: GLP Watchdog, Nora Bissett, "Five Clicks Instead of a Quarter Milligram", published 2026-08-20, updated 2026-10-27. https://glpwatchdog.org/research/five-clicks-instead-of-a-quarter-milligram

Canonical HTML page: https://glpwatchdog.org/research/five-clicks-instead-of-a-quarter-milligram

Medical disclaimer: Nothing here is medical advice. We report on this subject; we do not treat anyone. Whether to begin a medication, supplement, or treatment, whether to adjust it, and whether to stop are questions for a clinician licensed to answer them, who knows your history.
