The FDA has never defined a microdose of semaglutide or tirzepatide. The doctors prescribing it disagree. Published protocols range from 0.05 mg to 0.5 mg weekly for semaglutide and 0.5 mg to 2.5 mg weekly for tirzepatide. That gap is not a bug. It is the entire problem.
Most readers here have tried the standard titration and hated the nausea, or priced out branded Wegovy at $1,000 a month. GLP-1 microdosing is the strategy for people who want the metabolic benefit without the full-dose cost or side-effect profile.
We will show you the dose-and-units chart for both drugs, explain why pre-filled pens cannot deliver a true microdose, name the patient profiles this helps, and publish year-one cost math from real telehealth providers. We will also cover what the FDA’s April 30, 2026 proposal to remove these drugs from the 503B Bulks List means for access.
Microdosing is a legitimate strategy for the right person, not a shortcut for everyone.
What GLP-1 microdosing actually means (and why protocols disagree)
“Microdose” means three different things depending on who is prescribing. Anne Komé, PharmD, a clinical pharmacist at UNC Medical Center, presented the cleanest framework at the AACE Annual Conference in April 2026.
The first is a fractional weekly dose. Think 0.05 mg to 0.125 mg semaglutide weekly, well below the FDA starting floor of 0.25 mg. This is the playbook for side-effect-sensitive patients and for people chasing metabolic or anti-inflammatory benefits.
The second is extended-interval dosing. Instead of reducing the per-injection amount, you stretch the interval to every 2 to 4 weeks. A 2025 modeling study in Diabetes, Obesity and Metabolism found patients maintained roughly 75% of their weight loss when shifting from weekly to biweekly dosing.
The third is sub-titration-floor maintenance. You hold indefinitely at the lowest FDA-recognized step (0.25 mg semaglutide or 2.5 mg tirzepatide). A real-world tirzepatide study of 20,000-plus adults found 74% stayed below 10 mg. Most patients are already doing this without calling it microdosing.
This is why the protocols look different. Dr. Shamsah Amersi starts patients at 0.05 mg and escalates by 0.1 mg every four weeks. Tyna Moore, a naturopathic physician in Oregon, doses in fifths-of-a-milligram increments she can “barely see in the syringe.” Both are microdosing for different patients.
The verdict: AACE does not formally support routine microdosing and recommends adherence to FDA-approved dosing. AACE does acknowledge a role for “patients with severe side effects or pronounced responses at lower doses.” Weigh that against the practitioners using these protocols every day.
The dose-and-units chart: semaglutide and tirzepatide in milligrams and insulin-syringe units
Nobody else publishes a full conversion chart for both drugs across the concentrations real telehealth pharmacies ship. We do.
The formula. Units on a U-100 insulin syringe equal (Dose in mg divided by Concentration in mg/mL) multiplied by 100. Worked example: 0.05 mg semaglutide at 5 mg/mL = (0.05 / 5) x 100 = 1 unit. One tick mark.
Semaglutide at 5 mg/mL (the most common compounded concentration)
| Dose | Units (U-100) | Volume |
|---|---|---|
| 0.05 mg | 1 unit | 0.01 mL |
| 0.10 mg | 2 units | 0.02 mL |
| 0.125 mg | 2.5 units | 0.025 mL |
| 0.25 mg | 5 units | 0.05 mL |
| 0.5 mg | 10 units | 0.1 mL |
| 1.0 mg | 20 units | 0.2 mL |
| 1.7 mg | 34 units | 0.34 mL |
| 2.4 mg | 48 units | 0.48 mL |
At 2.5 mg/mL (a lower concentration some pharmacies like Rivas Medical Weight Loss also ship), the same doses require double the units. A 0.1 mg dose at 2.5 mg/mL is 4 units, not 2.
Tirzepatide at 20 mg/mL
| Dose | Units (U-100) | Volume |
|---|---|---|
| 1.0 mg | 5 units | 0.05 mL |
| 2.5 mg | 12.5 units | 0.125 mL |
| 5 mg | 25 units | 0.25 mL |
| 7.5 mg | 37.5 units | 0.375 mL |
| 10 mg | 50 units | 0.5 mL |
| 12.5 mg | 62.5 units | 0.625 mL |
| 15 mg | 75 units | 0.75 mL |
Rivas also ships tirzepatide at 17 mg/mL. At 10 mg/mL, double the unit count for the same dose.
For sub-1-unit semaglutide doses, a standard U-100 syringe is the wrong tool. Use a half-unit insulin syringe (0.3 mL, 30 unit capacity, half-unit graduations). Without that, a 0.5-unit dose is a guess, and a guess at this end of the curve matters.
Quick comparison: semaglutide microdoses sit in single-digit units. Tirzepatide doses sit in mid-double digits. A 1-unit error on a semaglutide microdose is a 50% deviation. A 1-unit error on a 25-unit tirzepatide dose is 4%. That is why syringe choice matters more on the semaglutide side.
Why pre-filled pens can’t be reliably microdosed
Attempting a non-validated dose on the Ozempic FlexTouch pen produces a plus or minus 15 to 25% delivered-dose error, according to MedsBase. That is the difference between 0.25 mg and 0.31 mg.
The mechanical reason: the pen locks only at validated stops (0.25, 0.5, 1.0, 1.7, and 2.4 mg depending on pen color). Intermediate positions feel like clicks but the mechanism is not calibrated between locks. The red Ozempic pen takes 18 clicks for 0.25 mg and 37 clicks for 0.5 mg. The blue pen takes 37 clicks for 1 mg. The same click count delivers wildly different milligrams because concentration differs by pen.
Novo Nordisk says it plainly: “The number of clicks does not correspond to an exact dose.”
The Zepbound and Mounjaro problem is different. The KwikPen is a single fixed-dose autoinjector with no dial. Microdosing tirzepatide from a pre-filled pen is mechanically impossible.
ConsumerMedSafety has logged hospitalizations from click-counting, including overdose events when a home microdoser is admitted and clinical staff administer the prescription dose. The pen also expires 56 days after opening.
Skip this whole question if you do not already own an unused pen. The right tool for microdosing is a compounded vial with an insulin syringe.
Who GLP-1 microdosing is actually for (and who should stick with standard dosing)
Most articles list two candidate profiles and stop. The real list is longer. Seven subpopulations with evidence behind each.
Side-effect-sensitive patients. The largest group. Women experience nausea on standard doses at roughly 2.5 times the male rate. AACE carves out a role for microdosing here.
Perimenopausal and postmenopausal women. Basal metabolism may drop 250 to 300 calories per day during perimenopause. Dr. Kathleen Jordan at Midi Health frames low-dose GLP-1 as a maintenance strategy that hits the metabolic problem without the full-titration GI burden.
Lean PCOS and pre-diabetic insulin resistance. Found Health launched a program in October 2025 for this exact group. The goal is insulin sensitivity, not weight loss. Starting doses run as low as 0.05 to 0.1 mg semaglutide.
Post-goal maintenance. Patients who hit their target and want to avoid rebound. The biweekly case-series data and Penn Medicine’s observation that food noise returns when therapy stops both point to low-dose maintenance.
Athletes and muscle preservation. Standard doses can drive 25 to 40% of weight loss as lean mass. Lower doses attenuate that signal. Pair with resistance training and 0.7 to 1 gram per pound lean-mass protein.
T2D pre-diagnosis. A1C in the 5.7 to 6.4 range. A 32-patient submaximal study (PMC12472706) saw insulin drop from 21.6 to 13 uIU/mL and leptin from 24.6 to 14.1 ng/mL over 12 weeks.
Longevity and cognitive motivation. Bryan Johnson trialed tirzepatide at one-fifth the typical dose. Cohorts of 2 million-plus diabetes patients show 20 to 35% lower dementia incidence in GLP-1 users versus DPP-4 or SGLT2 users. The first registered longevity microdose RCT is NCT07092605.
The honest counter: microdosing is not a good fit for patients with BMI above 35 chasing rapid loss, anyone with active diabetic complications, or anyone who needs full-titration effect size. STAT News notes no controlled trial has yet shown microdosing produces clinical weight loss on its own.
Best for: side-effect-sensitive patients, perimenopausal metabolic stabilization, PCOS, maintenance, longevity. Skip if: high BMI with aggressive weight-loss goals.
The cardiometabolic case: benefits that don’t depend on weight loss
In a 2025 HFpEF preprint (PMC12621885, originally bioRxiv September 2025), low-dose semaglutide produced cardiac and hepatic remodeling with no weight change. At 30 nmol/kg twice weekly in a rodent HFpEF model, left ventricular end-diastolic pressure dropped roughly 60%. Cardiac and hepatic fibrosis decreased. hs-CRP and 8-isoprostane fell. Body weight did not.
The benefit was uncoupled from weight loss. That is the central question microdosing asks. Four more pieces of evidence are worth knowing.
First, insulin sensitivity at submaximal doses. The PMC12472706 study (32 patients, 12 weeks) held doses below max when 2 to 5 kg per month was achieved. HOMA-IR improved. Insulin and leptin dropped significantly. IL-6 and TNF-alpha did not move, suggesting the systemic anti-inflammatory effect may need more time.
Second, alcohol craving. A 2025 JAMA Psychiatry RCT (PMC11822619) gave 48 participants low-dose semaglutide over 9 weeks. Weekly craving dropped significantly versus placebo. Drinks per drinking day and grams consumed in a lab self-administration task fell. Effect sizes were medium to large.
Third, immune and inflammatory conditions. A 47-patient MCAS case series from Dr. Kaufman at UC Davis and Dr. Dempsey reported 89% clinical benefit at starting doses of 0.1 to 0.25 mg semaglutide, 10 to 25 times below the standard diabetic dose. A 24-year-old patient with chronic inflammation normalized hunger and BP within hours of a 0.25 mg dose.
Fourth, cognitive endpoints. A real-world study in Alzheimer’s & Dementia (2 million-plus patients) showed 20 to 35% lower dementia incidence in GLP-1 users versus DPP-4 or SGLT2 inhibitor users. Honest caveat: Novo Nordisk’s EVOKE and EVOKE+ RCTs in early Alzheimer’s did not show semaglutide outperforming placebo. Prevention and treatment are different problems.
The verdict. At full doses, GLP-1s deliver weight loss plus downstream cardiometabolic benefit. Microdosing asks whether that downstream benefit holds without the weight loss. The evidence suggests yes, at least partially. We are showing the trail so you can weigh it.
What GLP-1 microdosing actually costs
Three tiers in plain dollars. Branded Wegovy or Ozempic retails at $1,000 to $1,300 per month. Branded Zepbound retails at roughly $1,000-plus. Compounded semaglutide at full dose runs $115 to $300 per month.
For a microdosing trajectory specifically, IvyRx publishes a useful breakdown. Months 1 to 3 at 0.05 to 0.125 mg run $150 to $200 per month. Months 4 to 6 at 0.25 to 0.5 mg run $200 to $250. Months 7-plus at 1 to 2 mg run $250 to $300. Year one total: $2,400 to $4,200.
Branded standard dosing for the same year: $12,000 to $15,600. The gap is roughly $8,000 to $11,000 saved in year one. That is the math.
Provider price points across the market:
- Longevity Direct: $179 per month flat for injectable semaglutide, injectable tirzepatide, or oral tirzepatide. Includes 50-plus biomarkers and a wearable ring.
- Rivas Medical Weight Loss: $115 per week semaglutide, $165 per week tirzepatide, $195 one-time intake. In-office model with weekly provider visits during titration.
- Noom Microdose: $119 initial, then $199 per month. Noom defines its “microdose” as 0.6 mg semaglutide, which is 25% of the 2.4 mg Wegovy max. Higher than most clinical definitions.
- Found Health: metabolic-condition-targeted, requires documented insulin resistance, PCOS, or prediabetes.
One caveat. Compounded prices are not insurance-covered, but they are flat-rate. They do not escalate with dose the way branded prescriptions can.
Direct recommendation: for most readers without coverage of branded GLP-1s, compounded microdosing is the difference between “this is affordable” and “this is not happening.”
How to actually get a microdose prescription
Most in-person primary care providers will not prescribe below the FDA starting dose. They cannot deviate from label without strong justification, and insurance will not cover off-label dose customization. Telehealth compounded providers are the practical path. Here is the intake in four steps.
Step 1: Pick a provider that ships compounded vials and lets you choose the concentration. Most patient-portal providers default to 5 mg/mL semaglutide and 10 to 20 mg/mL tirzepatide. For very low semaglutide doses, ask whether 2.5 mg/mL is available. More volume per milligram makes precision easier at the microdose end.
Step 2: Be specific in the intake form. Expect questions on BMI, comorbidities, prior GLP-1 history, and goals. Write something concrete in the goals field like “starting below the standard 0.25 mg semaglutide titration.” Vague answers get standard titrations.
Step 3: Know the 503A versus 503B distinction. 503A pharmacies make patient-specific prescriptions. 503B outsourcing facilities batch-compound. The FDA’s April 30, 2026 proposal targets 503B compounding from bulk. 503A pharmacies are not directly affected. Ask which type your pharmacy is.
Step 4: Confirm the prescribed range and dose-adjustment policy. Vials let you shift doses within a prescribed range between refills without a new script. Ask what range is authorized and which changes require approval. Some providers leave meaningful patient autonomy. Others require a portal message for every adjustment.
Best for: patients who can manage their own injection, are comfortable with async portal-based care, and want flat-rate pricing. If you need behavioral coaching or in-person handholding, the in-office model (like Rivas) is a better fit.
How to know your microdose is working: self-monitoring and titration triggers
A microdose is working when food noise drops, hunger windows lengthen, and labs trend right at a tolerable side-effect cost. Track signals on two cadences.
Subjective signals (weekly)
- Food noise score (1 to 10, daily). Should drop within 1 to 2 weeks. Often precedes weight change.
- Hunger window. Stretches from 3-4 hours to 5-6 hours or more.
- Cravings for hyperpalatable food. Sweets, alcohol, ultra-processed snacks. Should decrease.
- GI side effects. Mild and resolving by week 4 to 6.
Objective signals (monthly to quarterly)
- Weight. For maintainers, flat is success. For active loss, 1 to 5 lbs per month.
- Resting heart rate. GLP-1s can raise RHR 2 to 10 bpm. Bryan Johnson stopped his experiment when RHR rose 3 bpm and HRV dropped 7 points.
- Blood pressure. Should trend down or stay flat.
- A1C and fasting glucose. Drops possible at submaximal doses per PMC12472706.
- hs-CRP. Dropped in the submaximal study and HFpEF preprint. Pull at baseline and 3 months.
Escalation rule. Advance only when all four are true: GI tolerance established, weight stalled 3 to 4 weeks, food noise returning, hunger window collapsing. Do not escalate on calendar alone.
De-escalation triggers. Reduce or pause if GI symptoms persist past 8 weeks, appetite suppression blocks protein targets, or RHR is up 5-plus bpm.
Quick comparison. Dose too low: food noise returns mid-week, no weight movement after 8 weeks, no side effects. Dose too high: nausea past week 6, RHR climbing, sleep degrading.
The FDA 503B Bulks List proposal: what changes for microdosing access in 2026
On April 30, 2026, the FDA proposed removing semaglutide, tirzepatide, and liraglutide from the 503B Bulks List. The public comment deadline is June 30, 2026. The Advisory Committee meets July 23-24, 2026.
What this means. The 503B Bulks List authorizes outsourcing facilities to compound from bulk active ingredient. If the proposal is finalized, 503B facilities can no longer legally compound semaglutide or tirzepatide from bulk. 503A patient-specific pharmacies are a separate question, but the supply chain feeding most large telehealth providers narrows.
Background. Semaglutide came off the FDA shortage list in 2025; tirzepatide in 2024. The FDA cites 455-plus adverse event reports for compounded semaglutide and 320-plus for compounded tirzepatide. FDA Commissioner Marty Makary’s position: when FDA-approved drugs are available, outsourcing facilities cannot lawfully compound from bulk without clear clinical need.
Direct recommendation. If microdosing is on the table, the practical access window may narrow in late 2026. We are not selling urgency. We are reporting a calendar. Patients who lock in a 503A pharmacy relationship now will likely keep continuity.
Frequently asked questions
Is GLP-1 microdosing less effective than full-dose for weight loss?
Yes, on average. STAT News notes no controlled trial has shown microdosing produces standard weight loss. The 20,000-patient tirzepatide study found 74% stayed below 10 mg and still lost nearly 12% body weight in six months. Microdosers trade pounds for tolerability and cost.
Can you microdose Ozempic specifically?
Not reliably. The FlexTouch pen produces a 15 to 25% delivered-dose error at non-validated click positions, and Novo Nordisk states clicks do not correspond to exact doses. Switch to a compounded vial with a U-100 insulin syringe.
What happens if you stop microdosing?
Appetite and food noise typically return within 2 to 8 weeks. Penn Medicine has documented food noise returning even during ongoing tirzepatide treatment. Some cycle off seasonally; others maintain on every-2-to-4-week dosing.
Does microdosing reduce inflammation?
Limited but interesting evidence. The HFpEF preprint (PMC12621885) showed reduced hs-CRP at sub-weight-loss doses in a rodent model. A 12-week submaximal human study showed metabolic gains but no IL-6 or TNF-alpha change. Daniel Drucker cautions the inflammation question is unsettled.
Will you lose muscle on a microdose?
Less than on a full dose, but not zero. Standard doses drive 25 to 40% of weight loss as lean mass. At microdoses, the loss is meaningfully lower with resistance training and 0.7 to 1 gram per pound lean-mass protein. Watch grip strength and RHR.
The honest summary: microdosing is a legitimate strategy for the right person. The numbers, costs, and trade-offs are on this page. The decision is yours.