Living study · Data cut 1 of manyFrozen Sept 2, 2026 · 60 people · 93 check-ins
Next updateOct 28, 2026Preprint, not peer reviewed
PlexusDx Research

A microdose of tirzepatide is 1 to 2 mg a week. We studied 60 of our own patients.

That is below the usual 2.5 mg start. On average they had lost 2.2% of their body weight by about 3 weeks (57 people) and 3.6% by about 7 weeks (19 people), on compounded tirzepatide.

About 3 weeks in
4.4 lb
Average loss, 2.2% of body weight
57 people
About 7 weeks in
6.9 lb
Average loss, 3.6% of body weight
19 people
About 9 weeks in
12.2 lb
Average loss, 6.1% of body weight
8 people
Side effects
1 in 5
said they had one, mostly nausea or constipation
11 of 58 · at one check-in

This was not a clinical trial. It cannot prove the medicine caused the change. The weights above are self-reported, and there was no comparison group.

From the paper: "These hypothesis-generating observations support prospective comparative study, not conclusions about causality, equivalent dosing, or long-term safety." 6 of the 57 people in the first window lost nothing or gained a little.

What was the study?

We looked back at our own records. There was no comparison group, and patients weighed themselves at check-ins. We only counted people whose start we could date from delivery or other records, and who had stayed on the low dose without a break. Those start dates are estimates, not proof of a first injection.

  • 60 people, 93 check-ins, adults with a BMI of 27 or higher
  • All documented as new to GLP-1 medicines; compounded injectable tirzepatide with vitamin B12
  • 53 of 60 started at 1 mg a week in total; the other 7 started at 2 mg a week, 4 of them as 1 mg twice a week
  • Average age 47; 82% women; average starting weight 206 lb
  • Data frozen Sept 2, 2026; posted as a preprint, not yet peer reviewed

274 records became 60 people

We only kept people we could check at every step. The grey number is how many each check removed.

  1. Prescribed under 2.5 mg a week
    274
  2. Had a true first start and a real starting weight
    220
    −54
  3. Had a follow-up weight we could check
    127
    −93
  4. Had a BMI of 27 or higher
    73
    −54
  5. Stayed on the low dose 14 days or more, with no break
    60
    −13
Cohort funnel, data cut 1: 274 candidate records became 60 analyzed patients after five checks. The last check removed 13: 1 with no start date we could set, 10 with no check-in after 14 days on the dose, 1 with a break of 2 weeks or more, and 1 whose only check-in came after a move above 2 mg a week. Retrospective, uncontrolled, self-reported weights.

What did we see over time?

Average body weight lost, by how long people had been on it. The thin line on each bar is the likely range. One person, one check-in per window.

Microdose tirzepatide results chart. Average body weight change by time on it, data cut 1: 2.2% at about 3 weeks (57 people), 3.6% at about 7 weeks (19 people), 6.1% at about 9 weeks (8 people). Lines show 95% confidence intervals. Self-reported weights, no comparison group.
Average loss by time on it. Data cut 1, frozen Sept 2, 2026. Self-reported weights, no comparison group.
Time on itPeopleTypical daysAverage lossLikely rangeIn pounds
14 to 35 days57212.2%1.6% to 2.8%4.4 lb
36 to 59 days19483.6%2.4% to 4.7%6.9 lb
60 to 89 days865.56.1%4.6% to 7.6%12.2 lb

Likely range is the 95% confidence interval: the range the true average most likely sits in. Typical days is the median, the middle person's count.

+2.2 pts
more lost for every extra 30 days on it, in a statistical model of all 60 people and 93 check-ins. Starting weight, age and sex did not change that much. A pattern over time, not proof the medicine caused it. Most people started at 1 mg a week; some later moved to 2 mg.

The same people over time

7 people checked in at all three points. Every one of them had lost more each time we looked.

Average loss for the 7 people seen at all three points, data cut 1: 2.4%, then 4.7%, then 6.4%. Self-reported weights.

What side effects did people report?

Side effects reported at one check-in, data cut 1. 58 people answered.
What people reportedHow manyOut of
Any side effect1158
Nausea758
Constipation658
Diarrhea258
Vomiting058

Answers at one check-in. Not a formal safety study, and not a comparison with any other dose.

How did we check the data?

We threw out more people than we kept. Here is why. We set the checks while we were still reviewing the records, after we had seen early results. The paper says so, because it is a weakness. Nothing was filled in by guessing.

01
Pull every record
Adults prescribed a dose under 2.5 mg a week
274
02
Check for a documented first start and a real baseline weight
No documented prior GLP-1 use, no restarts, no transfers
220
03
Require a checked follow-up weight
Duplicate forms and impossible numbers removed
127
04
Keep baseline BMI 27 and up
Overweight or higher, measured by BMI
73
05
Require 14+ days of unbroken low-dose use
Start date estimated from delivery or shipping records, not the approval date. Of the 13 removed here, 1 had no start date we could set and 12 had no check-in that qualified
60

How sure are we?

Was there a comparison group?

No. So we cannot say the medicine caused the change. Diet, habits, or just time could play a part.

Did a nurse weigh people?

No. People typed in their own weight at check-ins. We checked the numbers for mistakes, but we did not measure them.

How many made it to 9 weeks?

Eight. That is too few to rely on. Nobody in the study had reached 90 days yet.

Did anyone gain weight?

Yes, a few. 6 of the 57 people in the first window lost nothing or gained a little. In the second window it was 2 of 19, and in the third none of the 8.

Are the authors neutral?

No. We work at PlexusDx, which runs the program and paid for the study. That is why we show every number and every limit.

Has anyone else checked this?

Not yet. It is a preprint. Other scientists have not reviewed it. We will say so here when they do.

Want to talk to a clinician about a small starting dose?

A licensed provider reviews your intake before anything is prescribed. The study covered one of the four Microdose GLP-1 variants, the tirzepatide injection. Your clinician chooses which, if any, fits you.

This page updates about every six weeks. Here is every change.

Every data cut of this page, with its freeze date and its reviewer.
CutData frozenPeopleCheck-insWhat changedReviewed by Dr. Lee on
1Sept 2, 20266093First version. Matches the preprint.Sept 17, 2026
2freezes Oct 14, 2026, live Oct 28, 2026New starts and check-ins since the freeze. Not part of the preprint.

Share the study

Send this page to someone who asked you about microdosing. It shows the numbers with their limits attached.

How to cite this study
Hastings J, Lee J. Weight Loss With Microdose Tirzepatide: Real-World Outcomes in Patients Initiating Treatment Below 2.5 mg Weekly. medRxiv. 2026. doi: 10.64898/2026.09.15.26363172.

Frequently Asked Questions

What is a microdose of tirzepatide?

In this study it meant 1 or 2 mg a week. The usual starting dose on the label is 2.5 mg. Microdosing is not a medical term, so we say exactly what we mean by it: a lower starting dose than the label, chosen by a licensed provider.

Is 1.25 mg a microdose of tirzepatide?

Yes. Anything below the usual 2.5 mg start counts, and 1.25 mg sits inside the 1 to 2 mg range we studied. It is one of the doses a licensed provider can choose. Our numbers come from people who started at 1 or 2 mg, so this page has no separate result for 1.25 mg.

Is microdosing tirzepatide proven to work?

No. Our study saw people lose weight, but it had no comparison group and people weighed themselves. It cannot prove the medicine caused the loss. The findings are hypothesis-generating; they support a future study with a comparison group.

How much weight did people lose?

About 4.4 lb after 3 weeks (57 people), 6.9 lb after 7 weeks (19 people). Only 8 people reached 9 weeks; their average was 12.2 lb. Weights were self-reported. Six of the 57 people in the first window lost nothing or gained a little, two of the 19 in the second, and none of the 8 in the third.

Can I microdose tirzepatide on my own?

No. A microdose is still a prescription medicine. In our study every dose was chosen and prescribed by a licensed provider, and the starting doses were 1 or 2 mg a week. Nothing on this page tells you what dose to take. A clinician decides that with you, based on your health and your history.

Is microdosing the same as taking tirzepatide twice a week?

No. In our study a dose is counted per week. 53 of the 60 people started at 1 mg once a week. Four people took 1 mg twice a week, which we counted as 2 mg a week, and three took 2 mg once a week. Splitting a dose into two shots does not make it a smaller dose. A licensed clinician sets the schedule.

How often is this page updated?

About every six weeks. Each update gets a cut number, a freeze date and a line in the changelog on this page. Cut 1 matches the preprint. Later cuts are program data after the preprint and are not part of the published analysis.

Who was in the study?

60 adults with a BMI of 27 or higher who were documented as new to GLP-1 medicines. Average age 47, 82% women, average starting weight 206 lb. All started compounded injectable tirzepatide with vitamin B12 at 1 or 2 mg a week through the PlexusDx program.

Is this the same as the Microdose GLP-1 Protocol?

It covers one of its four versions, the tirzepatide injection. The semaglutide versions and the dissolvable tablets were not studied. A licensed clinician reviews your intake and decides which version, if any, fits you.

Where can I read the full paper?

On medRxiv, once it is posted. The link and the DOI sit at the top of this page. It is a preprint, which means other scientists have not reviewed it yet. We will say so here when they do.

Read more on this study

Eight plain-English articles on what the study did, what it found, and how to read it.

What Is a Microdose of Tirzepatide? 1 to 2 mg, Explained

What the word means, what it does not mean, and who sets a dose.

Microdose Tirzepatide Results at 3, 7 and 9 Weeks: What Our Study Saw

What the records showed at each window, and what the study cannot tell you.

How the Microdose Tirzepatide Protocol Works: 1 to 2 mg Total per Week

How the program in the study was set up, and how patients were followed.

First Month on Microdose Tirzepatide: What 57 Patients Reported

The first window in full, including how many people lost nothing.

Who a Microdose Start Is For (BMI 27+, New to GLP-1) and Who Decides

Who was in the study, what the label says, and who makes the call.

Side Effects People Reported at 1 to 2 mg: What 58 Check-Ins Said

What people said at their first check-in, and why that is not safety surveillance.

How We Checked the Data: Why 274 Candidate Records Became 60 Patients

Every record we set aside, and what was never estimated or filled in.

What “Hypothesis-Generating” Means: How to Read a Small Study

Six ideas in plain sentences: sample size, windows, comparison groups and more.

PlexusDx tells you about your biology. It does not tell you what to put in your body.

This page summarizes a retrospective, uncontrolled, single-organization observational study conducted by PlexusDx using its own program records. It is posted as a preprint and has not been peer reviewed. Weights were self-reported. The study had no comparison group and cannot establish that treatment caused the observed weight change, cannot establish efficacy or comparative effectiveness, and says nothing about safety. Figures shown are averages across the people studied; individual outcomes ranged from weight gain to larger losses, and some people lost no weight. The authors are PlexusDx employees and have a financial interest in PlexusDx. Nothing on this page is a recommendation to use any medication. Data cut 1, frozen Sept 2, 2026; cuts after the first are program data after the preprint and are not part of the published analysis.

Compounded medications are not FDA-approved and have not been evaluated by the FDA for safety, efficacy, or quality. Clinical services are provided by an independent network of U.S.-licensed clinicians through a state-licensed telehealth platform. Compounded medications are dispensed by state-licensed compounding pharmacies operating in FDA-registered facilities. Tirzepatide is the active ingredient in Zepbound® and Mounjaro® (Eli Lilly and Company); those products' clinical trials are not described here and no equivalence is implied. Zepbound® and Mounjaro® are registered trademarks of Eli Lilly and Company. PlexusDx is not affiliated with or endorsed by Eli Lilly and Company.

The GLP-1 & Peptide Pathways Genetic Test analyzes how your genes influence peptide-related biological pathways. It does not recommend, prescribe, or determine which peptides you should use. Consult a qualified healthcare provider before beginning any peptide protocol. Genetics is a guide, not a guarantee. The genetic test was not part of this study.