The Metabolic Health Blueprint
Four arms, four different reasons a number is high
Everything on this page is free. The stack, why each pick beat its alternatives, every option, the bloodwork and the safety lines. The week-by-week schedule and the decision rules are the members half.
Insulin resistance is not one defect. Four different things produce the same lab pattern, and the intervention that works depends entirely on which one you have. Your cells stopped listening to the fuel signal — that is the AMPK arm. You eat more than you burn and the drive to eat is winning — that is the incretin arm. Your fasting numbers are fine but every meal spikes you — that is the post-meal arm, and it is completely invisible on an annual fasting draw. Your liver is full of fat — that is the hepatic arm, and it is the one that most often drives everything else. Most people have two of the four. Almost nobody has all of them. Draw the labs first — this is the one page where guessing is unnecessary, because the test is cheap and the answer is unambiguous.
Can you run all of them? Yes - and here is what it costs
This is a general protocol. You make the final call on how much of it to run — or have it built around your labs.
Which of these 4 is actually you?
This tells you where your biggest leverage is — where to start, not where to stop. Read the But line too: it is what each lane cannot do for you, which is the part a list of options never tells you.
Before any of it — the foundation
These four are not a disclaimer at the bottom of the page. They are the reason the rest of it works, and every one of them is free.
Growth hormone is released in pulses during deep sleep, insulin sensitivity is measurably worse after one bad night, and appetite regulation collapses without it. Every compound below works through a system that sleep already governs. This is not filler advice — it is the highest-leverage item on the page and it is free.
The single dietary variable with the most consistent evidence behind it for body composition, in both directions — building and preserving. Under-eating protein while running anything anabolic is paying for a signal with no substrate to act on.
Nothing here substitutes for mechanical tension. Compounds change how well you recover from and adapt to training; they do not replace the stimulus. A protocol run without training reliably produces the side effects and not the results.
Non-exercise activity is the largest and most variable component of daily energy expenditure, and it is the one that quietly falls when you start dieting. Tracking it stops the metabolic adaptation people blame on their thyroid.
The stack
This is the best-evidenced page in the Vault, and it is worth saying so given how often the opposite caveat appears elsewhere. The incretins have cardiovascular outcome trials. Metformin has decades. Acarbose has the STOP-NIDDM data. Berberine has head-to-head trials. The research-grade end — MOTS-c, ATX-304, 5-Amino-1MQ — sits alongside that, and the honest framing is that it is arguing for the same pathway from a much thinner base. Both are on this page. The difference is stated rather than hidden.
Each pick names what it was chosen over and why. That is the difference between a blueprint and a list — if you disagree with a choice, the alternative is right there and swapping it does not break the rest.
Peptides 3
Short amino-acid chains that signal rather than force. Almost all are injected or intranasal, they need reconstituting, and they are the reason most people are on this site.
A mitochondrial-encoded peptide — the gene sits in mitochondrial DNA, not nuclear DNA, which makes it a signal from the mitochondria to the rest of the cell rather than the other way round. It activates AMPK and increases glucose uptake, and the animal work shows it reverses diet-induced insulin resistance without the training interference metformin carries.
Metformin is more proven and much cheaper, and it is the right answer for a lot of people. It is not the base here for one specific reason: it blunts the adaptations to both resistance and endurance training in multiple trials, and this blueprint sits on top of a training foundation. If you are not training hard, that objection evaporates and metformin becomes the obvious pick — which is why it is the first add-on rather than absent.
Stack this arm deeper6 optional add-ons
Each of these sits in this same pathway, so it starts the week this pathway starts. Swapping one in for the pick above does not change the schedule.
Decades of data, pennies a day, and an effect size nothing else in this arm matches. If you are not training for adaptation, this is arguably the better base.
The trade-off Blunts training adaptation. GI cost in the first weeks. Depletes B12 over years — worth drawing annually rather than assuming.
Hits the same AMPK switch from a plant, over the counter, with several trials putting it close to metformin on glucose and lipids.
The trade-off Poor oral absorption is the known weakness — dihydroberberine exists specifically to fix it. Interacts through CYP enzymes with a long list of drugs.
A different switch entirely — it inhibits NNMT, which raises intracellular NAD+ and salvages methylation capacity in fat tissue. Not an AMPK activator, which is exactly why it stacks rather than duplicates.
The trade-off Human data is very thin. Oral bioavailability is the open question, and the injectable and oral doses are not interchangeable.
A direct AMPK activator rather than an indirect one — it binds the enzyme instead of manipulating the energy ratio that activates it. Has actually been through a human phase-1.
The trade-off Early clinical, expensive, and direct activation means less of the self-limiting behaviour indirect activators have.
Both fat- and water-soluble, so it works in compartments most antioxidants cannot reach — and it has genuine trial data in diabetic neuropathy, which is a complication rather than a marker.
The trade-off Can lower blood sugar enough to matter if you are already on glucose-lowering medication. Depletes biotin over long use.
A pancreatic bioregulator aimed at insulin-producing tissue - and like the liver, this one has a cheap readout. Fasting insulin and HbA1c will tell you within twelve weeks.
The trade-off Metformin, berberine and the incretins all have real trials for the same endpoint. This has a research group and a mechanism.
Incretin & satiety signallingRetatrutide5 options
5 options — 3 to swap in, 2 to stack ontap to collapse
Hepatic fat & fatty liverTesamorelin5 options
5 options — 0 to swap in, 5 to stack ontap to collapse
Small molecules 1
Orally active compounds, most of them with a prescription history and a real clinical evidence base. Less exciting than the peptides and frequently better evidenced.
Glucose disposal, absorption & the post-meal curveAcarbose5 options
5 options — 1 to swap in, 4 to stack ontap to collapse
Health supplements & substrate
The floor underneath the compounds. Cheap, well tolerated, and the part that decides whether anything above it has a fair chance — a secretagogue on a magnesium deficiency is a rounding error.
The 16-week schedule
What goes in, what comes out, and when. The exact doses for each phase are inside the Academy — the structure below is free because it is the part you need to decide whether this fits your life.
Each bar is a week block that compound is running. The shape is free — it is what tells you whether this fits your life. The doses for each phase are the members half.
Fasting insulin, HbA1c, a full lipid panel and a CMP.
Fasting insulin is the marker most often skipped and the one that moves earliest. HbA1c can sit normal for years while insulin climbs to hold it there. If you draw one thing before starting, draw that.
Start the arm your labs pointed at. Nothing else yet.
Titrate slowly on anything incretin or acarbose — almost every discontinuation is a titration that moved faster than the gut adapted, not an intolerance.
Re-draw at week 8 and let the number decide.
Two arms is the sensible ceiling for a first run. The post-meal arm combines with any of the others because it works somewhere none of them do.
Same doses, no additions, one clean read at the end.
Sixteen weeks is chosen because HbA1c reflects roughly three months of glucose. Re-drawing it at eight weeks tells you about a period that is half pre-treatment — useful as a direction, not as a result.
Most of this arm is chronic, and that is fine — but only for what earned it.
Insulin sensitivity reverts when the intervention stops unless the diet and training changed with it. The arm that made the numbers move is the one worth paying for indefinitely; the others are not.
The doses for each phase are inside
Every compound above, dosed week by week, plus the reconstitution numbers and Coach Cam's notes on running it. $10/mo.
Unlock the schedule →Bloodwork
Fasting insulin and HbA1c together are the whole diagnostic. High insulin with normal HbA1c means compensation is still working and you have caught it early — the best possible time to be on this page. Both high means compensation is failing. ALT above roughly 30 in a man or 20 in a woman is a fatty liver signal even inside the lab's stated range, and the reference ranges on most panels were set from populations that already had it. Triglycerides divided by HDL is the cheapest insulin resistance proxy there is and it is already on the lipid panel you drew. Uric acid is here because fructose and insulin resistance drive it together, and it predicts the trajectory before glucose does.
Before you start
Everything, drawn before you start. This is the one that decides which pathway is actually yours - and the only one you cannot go back and collect later.
Around week 8
The short list, drawn while you are running it. Not a progress report - it is the draw that catches the things that go wrong quietly.
After
Drawn at the end, against your own baseline. This is what turns the protocol into information rather than a feeling.
All three are drawn at Quest, 2,000+ US locations, no doctor visit, HSA/FSA eligible. Prefer to pick and choose? Every marker above links to its own page, and the panel builder assembles any combination.
Adjusting it
A protocol you cannot adjust is a protocol you abandon. Four situations come up on nearly every run of this — nausea that will not settle, a three-week stall, hair shedding, glucose moving the wrong way. Each one has a specific answer, and the wrong answer to a stall is the reason most people end up on six compounds that each do nothing.
The four decision rules are inside
What to change, what to leave alone, and how to tell a real stall from a water shift. $10/mo.
Unlock the decision rules →The lines I'd stop at
- Persistent severe upper abdominal pain radiating to the back on any incretin. That is the pancreatitis presentation and it is the one not to wait out.
- Nausea and rapid breathing on an SGLT2 inhibitor even with normal glucose — euglycaemic ketoacidosis presents without the high number everyone is watching for.
- ALT or AST above three times the upper limit on any repeat draw.
- Any vision change on rapid glucose correction. Fast improvement can transiently worsen retinopathy, and that needs an eye exam rather than a dose adjustment.
It is built for the common case, not for you specifically. Compound selection and dosing genuinely do change person to person — training age, bloodwork, what you have run before, what you react to. Adjust it against your own numbers using the panels above, or if you want it built around your labs rather than the average, that is what 1-on-1 coaching is for.