# Ozempic Click/Dose Research Findings

Expanded research memo for a reported story on semaglutide pens, click-counting, dose-splitting, and the online medication culture around GLP-1 drugs.

Last updated: 2026-07-14

## Executive finding

The click-chart phenomenon is real, clinically visible, and economically understandable—but it is not an approved dosing system. It sits in the gap between a rigid, concentration-variable commercial device and patients/clinicians who want smaller titration steps, lower costs, or more doses per pen.

The most important distinction is between **total semaglutide in the cartridge** and **the marked dose delivered per injection**:

| U.S. label | Total drug/volume | Official marked delivery |
|---|---:|---:|
| 2 mg/3 mL | 2 mg total; 0.68 mg/mL | 0.25 mg or 0.5 mg per injection |
| 4 mg/3 mL | 4 mg total; 1.34 mg/mL | 1 mg per injection |
| 8 mg/3 mL | 8 mg total; 2.68 mg/mL | 2 mg per injection |

These are FDA-labeled configurations, not a click chart. [FDA label](https://www.accessdata.fda.gov/drugsatfda_docs/label/2026/209637s038lbl.pdf)

## What Novo and the FDA actually say

The U.S. instructions repeatedly tell users to read the dose counter and pointer, not count audible clicks. The labeled 4 mg/3 mL pen is a **1-mg-dose pen**: it contains 4 mg total and is designed for four 1-mg injections. [DailyMed](https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=adec4fd2-6858-4c99-91d4-531f5f2a2d79)

Novo’s medical-information response is more explicit than the ordinary patient insert: the company says it has not evaluated dosing by click number, does not recommend it, and cannot provide an exact dose or volume delivered per audible click. It warns that unmarked dosing can be inaccurate or inconsistent. [Novo Nordisk medical-information response](https://www.adea.com.au/wp-content/uploads/2024/03/OZEMPIC-semaglutide-and-Pen-Dosing-by-Clicks.pdf)

That is the official baseline. Any article should describe click-counting as off-label or unofficial and avoid presenting a click number as a guaranteed dose.

## What the unofficial charts are trying to do

The charts assume that the dial has repeatable, equal mechanical increments and that delivered volume is proportional to the number of forward detents. A typical ratio calculation is:

> estimated dose = counted clicks ÷ assumed full-setting clicks × marked full dose

Every part of that equation is an assumption when used at an unmarked setting. Novo does not publish or guarantee an exact milligram or millilitre value for each audible click. The FDA instructions also say the selector clicks differently when turned forward and backward, so “a click” is not even direction-neutral. [FDA label](https://www.accessdata.fda.gov/drugsatfda_docs/label/2026/209637s038lbl.pdf) [Novo medical-information response](https://www.adea.com.au/wp-content/uploads/2024/03/OZEMPIC-semaglutide-and-Pen-Dosing-by-Clicks.pdf)

The ADA shortage report applies a 74-click full-scale assumption to three U.S. pen concentrations. Its mathematically calculated examples—not manufacturer-confirmed measurements—illustrate why the same click count cannot be transferred between pens:

| U.S. pen in ADA report | Marked full dose | Selected ADA examples |
|---|---:|---:|
| 2 mg/3 mL | 0.5 mg | 37 clicks ≈ 0.25 mg; 74 ≈ 0.5 mg |
| 4 mg/3 mL | 1 mg | 18 clicks ≈ 0.25 mg; 37 ≈ 0.5 mg; 74 ≈ 1 mg |
| 8 mg/3 mL | 2 mg | 18 clicks ≈ 0.5 mg; 37 ≈ 1 mg; 74 ≈ 2 mg |

Thus, under that table, 37 clicks could be labelled 0.25, 0.5, or 1 mg depending on the pen. [Clinical Diabetes / ADA](https://diabetesjournals.org/clinical/article/41/3/467/148676/Special-Report-Potential-Strategies-for-Addressing)

The strongest institutional example found is the BC Children’s Hospital Endocrinology & Diabetes Unit handout. It openly describes click-counting to create intermediate titration steps, lists 18 clicks ≈ 0.25 mg, 36 ≈ 0.50 mg, and 72 ≈ 1.00 mg on its 1-mg pen, and calls the doses approximate, off-label, and not manufacturer-recommended. It says the practice has nevertheless been widely adopted in BC and many North American clinics. [BC Children’s Hospital handout](https://ubccpd.ca/sites/default/files/documents/semaclick.pdf)

That handout also contains a material internal unit mismatch: it states **“1 click = 0.01 mg”** and later **“72 clicks = 1.00 mg.”** Both cannot be true; 72 × 0.01 mg is 0.72 mg. The table otherwise behaves largely as a 72-click-to-1-mg proportional scale, implying about 0.01389 mg per click. The 0.01 mg line may be a typographical or unit error, but it should not be repeated as a verified conversion without clarification from the clinic.

### Focused answer: the 4 mg/3 mL pen

For the U.S. 4 mg/3 mL, 1-mg-dose pen, the published unofficial mappings are:

| Claimed dose | Commonly reported clicks | Status |
|---:|---:|---|
| 0.25 mg | 18 | Shared reported value; unmarked/off-label |
| 0.5 mg | 36 or 37 | Competing source calculations; unmarked/off-label |
| 1 mg | 72 or 74 | Official marked dose; the click count itself is unofficial |

The U.S. pen contains 4 mg total at a printed concentration of 1.34 mg/mL and is labeled to deliver four 1-mg injections. The 4 mg figure is the cartridge total, not a 4-mg injection. The 72-versus-74 split is **between-source disagreement, not a confidence interval, tolerance, or validated dose range**. No published source located provides measured accuracy, manufacturing variation, or an uncertainty band for unmarked settings.

The final source audit found that this is not merely an internet disagreement. Two credible clinical sources publish different numbers for the same 4 mg/3 mL configuration:

| Approximate dose | BC Diabetes | ADA journal shortage report |
|---:|---:|---:|
| 0.12–0.125 mg | 9 | 9 |
| 0.25 mg | 18 | 18 |
| 0.375 mg | 27 | — |
| 0.5 mg | 36 | 37 |
| 0.66 mg | — | 49 |
| 0.8 mg | — | 59 |
| 1 mg | 72 | 74 |

BC Diabetes presents the figures as clinical instructions and identifies cost and slower titration as reasons for using the 4 mg pen. [BC Diabetes](https://www.bcdiabetes.ca/wp-content/uploads/bcdpdfs/Semaglutide-for-weight-loss-and-diabetes-control.pdf) The ADA journal report presents its table as a shortage strategy, labels the doses as mathematically calculated rather than manufacturer-confirmed, and warns that concentration differences create a high potential for confusion and dosing errors. [Clinical Diabetes / ADA](https://pmc.ncbi.nlm.nih.gov/articles/PMC10338283/) A later erratum corrected an unrelated liraglutide frequency and did not change the semaglutide table. [Erratum](https://pmc.ncbi.nlm.nih.gov/articles/PMC11486853/)

That makes **18 clicks ≈ 0.25 mg** the only value shared exactly by both sources in their 4 mg/3 mL tables. Shared reporting is not independent device validation, particularly because the 72-click handout has an internal unit inconsistency. The responsible reporting frame is a comparison of published unofficial recipes—not a single universal conversion chart.

This is not evidence that the method is validated. It is evidence that the method is being used in real clinical workflows.

## Measurement and math audit

### “4 mg pen” is not a reproducible device identifier

At minimum, a chart needs the country/market, manufacturer and brand, total drug and total volume, printed concentration, marked selectable dose, and ideally the exact product code and date. Current Canadian product information lists six Ozempic pen variants, including both a 4-mg-total pen that selects 0.25/0.5 mg and a 4-mg-total pen that selects 1 mg. It also lists 8-mg-total pens that select either 1 mg or 2 mg. [Novo Nordisk Canada product monograph](https://www.novonordisk.ca/content/dam/nncorp/ca/en/products/ozempic/product-monograph/ozempic-en-pm-04-november-2025.pdf)

The current European product information likewise lists 4 mg/3 mL presentations delivering either 0.5 mg in 0.37 mL or 1 mg in 0.74 mL, and 8 mg/3 mL presentations delivering either 1 mg in 0.37 mL or 2 mg in 0.74 mL. [EMA product information](https://www.ema.europa.eu/en/documents/product-information/ozempic-epar-product-information_en.pdf)

BC Diabetes now adds a brand-level warning: its clinic document claims 72 clicks for 1 mg from the Ozempic-branded 4 mg pen but 75 clicks for 1 mg from an apo-semaglutide 4 mg pen. This is reported clinical guidance, not manufacturer validation, but it demonstrates that even “4 mg pen” plus nominal dose may not transfer across brands. [BC Diabetes](https://www.bcdiabetes.ca/wp-content/uploads/bcdpdfs/Semaglutide-for-weight-loss-and-diabetes-control.pdf)

### The two source scales imply different doses

If—and only if—each source is treated as a linear scale anchored at 1 mg, the same click count implies:

| Clicks | On a 72-click scale | On a 74-click scale |
|---:|---:|---:|
| 18 | 0.250 mg | 0.243 mg |
| 36 | 0.500 mg | 0.486 mg |
| 37 | 0.514 mg | 0.500 mg |
| 72 | 1.000 mg | 0.973 mg |
| 74 | 1.028 mg | 1.000 mg |

The two-click difference at the full setting is about 2.7–2.8% under those assumptions. That arithmetic does **not** establish actual delivered-dose accuracy or clinical equivalence.

One click is also a larger fraction of a small target dose. Under the same 72/74 proportional assumptions, one click represents approximately:

| Target dose | One click as share of target |
|---:|---:|
| 0.25 mg | 5.4–5.6% |
| 0.5 mg | 2.7–2.8% |
| 1 mg | 1.35–1.39% |

This is sensitivity analysis, not a measured error rate.

There is also whole-click rounding. Under a 74-click proportional model, 0.25 mg is 18.5 clicks, yet the ADA table prints 18 and does not state a general tie-breaking rule. Nearest-integer rounding would by itself contribute up to half a click; always rounding down or up would create a different bias. Dose labels inside the tables are sometimes rounded as well: 49/74 = 0.6622 mg is shown as 0.66 mg, 59/74 = 0.7973 mg as 0.8 mg, and 51/72 = 0.7083 mg as 0.71 mg. These model-rounding effects are distinct from actual device or delivery error.

### Printed label values are rounded, not metrology-grade inputs

The printed values do not multiply perfectly: 4 mg ÷ 3 mL = 1.3333 mg/mL, while the label prints 1.34 mg/mL; 1.34 mg/mL × 3 mL = 4.02 mg. The EMA also prints 1 mg in 0.74 mL, while 1.34 × 0.74 = 0.9916 mg. These small mismatches are consistent with rounded nominal labeling and should not be interpreted as proof of extra drug, exact overfill, or an exact microgram-per-click value.

Assuming 0.01 mL per click and multiplying by the rounded printed concentration produces about 74.63 clicks for 1 mg—not exactly 72, 74, or 75. Novo explicitly declines to confirm an exact volume per audible click. Therefore concentration arithmetic cannot validate any of the published charts.

### Dial setting is not the same as delivered dose

The official instructions identify several delivery variables that a click equation does not capture:

- The first-use flow check expels solution and may be repeated if no drop appears; this makes residual-volume arithmetic uncertain.
- Forward and backward selector clicks sound different, so counting convention and direction must be specified.
- A blocked needle can allow the dose counter to move even though no medicine is delivered.
- The button must remain depressed and the needle in the skin for six seconds; early withdrawal can underdeliver.
- Leaving a needle attached can cause leakage, blockage, contamination, and inaccurate dosing.
- A visible residual amount is not automatically another measurable dose; the official instruction is to use a new pen if the counter cannot reach the full marked dose.

### The 56-day in-use limit changes the cost math

Simple cartridge arithmetic says a 4-mg-total pen contains four 1-mg doses, eight 0.5-mg doses, sixteen 0.25-mg doses, or thirty-two 0.125-mg doses. But the U.S. label requires disposal after 56 days even if medication remains, and the ADA report caps its examples at eight weekly doses for that reason. [FDA label](https://www.accessdata.fda.gov/drugsatfda_docs/label/2026/209637s038lbl.pdf) [Clinical Diabetes / ADA](https://diabetesjournals.org/clinical/article/41/3/467/148676/Special-Report-Potential-Strategies-for-Addressing)

Calendar counting must be stated: 56 days equals eight elapsed seven-day intervals, while counting both day 0 and day 56 produces nine listed calendar dates. That does not itself authorize a ninth dose; the labeled disposal date and the ADA table’s eight-dose convention govern the usable-dose calculation.

Ignoring flow-check loss, overfill, and delivery error, eight weekly injections would consume approximately 4 mg at 0.5 mg/week, 2 mg at 0.25 mg/week, or 1 mg at 0.125 mg/week. Thus the low-dose cost advantage cannot be calculated as “total drug divided by dose” without accounting for the discard date and unused nominal contents.

## Why the 4-mg-total pen is a particularly good reporting object

The U.S. 4 mg/3 mL, 1-mg-dose pen is where the terminology becomes especially misleading:

- It is often called a “4 mg pen” because 4 mg is the total cartridge content.
- It is also called the “1 mg pen” because 1 mg is the only marked injection dose on the current U.S. version.
- Online users may buy or receive it while still taking 0.25 or 0.5 mg.
- The pen is physically capable of being stopped between marked settings, which invites people to reverse-engineer partial doses.
- A person switching from a lower-concentration pen can carry over a remembered click count and potentially deliver a different milligram dose.
- Outside the United States, total content alone may not identify the selectable dose: current Canadian and European documents list multiple presentations with the same total drug and volume.

The U.S. 4-mg-total pen is not a 4-mg injection pen. That clarification should appear high in any story, followed immediately by the market and marked-dose caveat.

## Why people do it

The sources point to four overlapping motives:

1. **Side-effect management.** The standard titration schedule increases in relatively large steps. Some patients report nausea, vomiting, diarrhea, constipation, fatigue, dizziness, or other symptoms when moving up.
2. **Cost.** Higher-strength pens can be financially attractive if a person uses lower unmarked doses, especially when the price is similar across strengths or coverage is unstable.
3. **Access and shortages.** Patients try to remain on treatment after losing insurance coverage or being unable to obtain a preferred strength.
4. **A desire for individualized dosing.** Clinicians may want a slower or intermediate titration for patients who respond strongly or tolerate the standard schedule poorly.

The Atlantic’s reporting captures the moral ambiguity: physicians interviewed did not recommend physically breaking open pens, but some understood or even tolerated click-counting as a practical way to individualize Ozempic dosing. The article also reports that patients were trading these methods online because the drugs felt life-changing but unaffordable. [The Atlantic](https://www.theatlantic.com/health/archive/2024/08/ozempic-hackers/679464/)

## What is known about harm

The strongest official error data found concern compounded semaglutide vials, not click-counting from branded Ozempic pens. The FDA reports patients taking five to 20 times their intended dose because of confusion among milligrams, milliliters, and “units,” as well as provider calculation errors. Reported effects include severe nausea, vomiting, abdominal pain, fainting, dehydration, acute pancreatitis, and gallstones; some patients required hospital care. [FDA dosing-error alert](https://www.fda.gov/drugs/human-drug-compounding/fda-alerts-health-care-providers-compounders-and-patients-dosing-errors-associated-compounded)

A 2023 poison-control case series described three incorrect-administration cases involving compounded semaglutide; two involved tenfold errors, all involved significant nausea/vomiting/abdominal pain, and one patient received IV fluids. [Lambson et al.](https://www.sciencedirect.com/science/article/abs/pii/S1544319123002315)

A separate poison-center study of GLP-1 exposures found that misunderstanding how to use the injection pen was a meaningful category of error; semaglutide was the most frequent exposure in that center’s dataset. This is broader than Ozempic click-counting and should be reported as evidence of general device-use confusion, not as a click-chart study. [Poison-center study](https://pmc.ncbi.nlm.nih.gov/articles/PMC11288212/)

These sources should not be misrepresented as proof that Ozempic click-counting causes the same error rate. They do establish that semaglutide dose-conversion culture can produce serious harm when concentration, volume, and dose are confused.

Other risks relevant to branded-pen dose splitting include:

- underdosing or overdosing from using the wrong chart;
- confusion when switching concentration, brand, country, marked dose, or device generation;
- treating milligrams, millilitres, audible clicks, and insulin-style “units” as interchangeable;
- counting flow-check clicks or failing to specify forward-only versus backward dialing;
- a moving dose counter despite no delivery through a blocked needle;
- incomplete delivery from releasing the button or withdrawing the needle before six seconds;
- uncertainty about residual medication in a nearly empty cartridge;
- using a pen beyond its labeled in-use period;
- contamination risk when physically opening a pen or transferring its contents;
- delayed recognition of overdose because semaglutide has an approximately one-week half-life and symptoms can persist.

## The online information problem

The online ecosystem is not just spreading one chart. It is generating competing charts and then citing one another.

The viral trail is difficult to quantify from ordinary web indexing: direct TikTok and Instagram posts are inconsistently searchable, often login-gated, and their view counts change. But Reddit users explicitly say they saw TikTok posts about dosing by clicks and then came to Reddit or Google for a chart. That is evidence of cross-platform circulation, not evidence of how many people are doing it. [Example Reddit discussion](https://www.reddit.com/r/Ozempic/comments/1paikxk)

The most recent SEO-oriented chart reviewed claims to have checked primary sources, but it also contains internal warning signs:

- it calls unofficial values “confirmed against patient-side click counting” without presenting a validation study;
- it says one configuration has 74 clicks for 1 mg and another has 150 for 2 mg;
- elsewhere on the same page it links to a “4 mg pen” article that reportedly gives a different 56-click figure;
- it includes commercial calls to action for compounded GLP-1 therapy.

That makes such pages useful artifacts for the story’s digital-culture section, but weak sources for medical facts. [Example SEO click chart](https://formblends.com/articles/glp1-hub/ozempic-click-chart)

Another calculator claims every Ozempic strength delivers a fixed 0.01 mg per click. That is a red flag: it collapses the distinction between click-to-volume and click-to-milligram conversion, even though the FDA label lists materially different concentrations across pen strengths. [Example calculator](https://clickdose.io/en/ozempic-click-counter.html)

Reddit discussions show the same pattern at human scale: users correct one another, warn that a chart belongs to a different pen, report different click counts, and sometimes say they are saving money or slowing titration. The posts demonstrate active peer-to-peer troubleshooting, not reliable clinical evidence. [Example Reddit thread](https://www.reddit.com/r/Ozempic/comments/waf8iq)

## The cleanest story thesis

**Ozempic click-counting is a crowdsourced workaround created by the collision of a flexible dial mechanism, multiple concentrations, rigid titration schedules, severe side effects, high prices, and unstable insurance access.**

The “craziness” is not simply that patients are ignoring instructions. It is that some clinicians recognize a practical need for intermediate doses, some health systems have published approximate click tables, the manufacturer refuses to validate the method, and patients are left to reconcile all of that through Reddit, TikTok, Facebook groups, and SEO sites.

## Claims safe to make now

- The current U.S. 4 mg/3 mL pen contains 4 mg total and is labeled to deliver four 1-mg injections.
- Novo and FDA-approved instructions say not to set the dose by counting clicks.
- A BC Children’s Hospital clinic handout describes click-counting as approximate, off-label, and not manufacturer-recommended, while acknowledging widespread use.
- For the 4-mg-total/1-mg-dose Ozempic configuration, credible unofficial sources publish 36 or 37 clicks for approximately 0.5 mg and 72 or 74 for approximately 1 mg.
- The 72-click BC Children’s handout contains an internal unit inconsistency: its “1 click = 0.01 mg” statement cannot coexist mathematically with “72 clicks = 1 mg.”
- The 72–74 notation describes disagreement between sources, not a validated dose range or error tolerance.
- Online click charts disagree and often blur total cartridge strength with per-injection dose.
- Serious semaglutide dosing errors have been documented, especially with compounded products and unit/mL/mg confusion.

## Claims that need more reporting before publication

- That a particular click number is universally correct for every 4-mg pen.
- That “4 mg/3 mL” uniquely identifies the pen, brand, selectable dose, or market.
- That one click equals exactly 0.01 mg—or exactly 0.01 mL—on an Ozempic pen.
- That 72–74 clicks is a statistically derived confidence interval or known device tolerance.
- That Novo’s pen mechanism has been clinically validated for intermediate dosing.
- That branded Ozempic click-counting has produced a known number of overdoses or hospitalizations.
- That a pharmacist-created chart is “official.”
- That a higher-strength pen is always cheaper or can safely be stretched beyond its labeled period.
- That a patient’s online dosing method is safe because it worked for them.

## New follow-up leads

- Obtain and read the full text of the PubMed-indexed semaglutide overdose case series.
- Build a small, date-stamped sample of public TikTok/YouTube/Instagram posts, recording claims and engagement metrics without reproducing actionable dosing instructions.
- Ask Novo whether it has internal complaint, adverse-event, or device-engineering data specific to unmarked click dosing; ask why the instructions mention clicks as audible feedback while rejecting clicks as a dose-setting method.
- Ask BC Children’s Hospital whether “1 click = 0.01 mg” is a unit typo, how its clinic derived the 72-click scale, how many patients have used the protocol, and whether errors or adverse events were tracked.
- Ask independent pharmacists whether they have ever dispensed a 4-mg-total pen with instructions for a lower dose, and whether that instruction was verbal, written, or prescriber-authorized.
- If publication resources permit, commission a qualified device-testing laboratory to measure expelled mass/volume at unmarked settings across multiple lots and markets; a single pen or self-counted demonstration cannot establish accuracy or precision.