Protocol Governance: Why Set-and-Forget Fails
A protocol is a hypothesis about a person at a moment. The person changes, the evidence changes, and a protocol never revised has stopped being a decision.
The Short Answer
Set-and-forget works for things that do not change. A protocol assembled for a specific person, with a specific set of markers, on a specific set of medications, against the evidence available at that moment, is not one of them. All four of those inputs shift, and a protocol that does not shift with them is progressively less matched to the person taking it. Governance is the structure that keeps them aligned, and it consists mostly of scheduled subtraction.
The Four Things That Change
The person. Age, body composition, training state, life stage, sleep pattern, stress load and health status all move. Some move gradually and invisibly, which is the harder case.
The markers. A protocol assembled to correct low vitamin D is addressing a problem that may no longer exist. A stack built for insulin resistance is mismatched once insulin sensitivity is restored.
The medications. Any new prescription changes the interaction picture, and this is the highest-priority trigger because the risk is immediate and specific.
The evidence. Compounds that looked reasonable five years ago have since been tested. Ginkgo, glucosamine, B vitamins for cardiovascular prevention and vitamin E were all defensible additions before their trials reported.
The fourth is the one people never revisit. A stack reflects the evidence at the time of purchase, and nothing in the structure of a supplement routine prompts a re-read of the literature.
Why Stacks Only Grow
The asymmetry is structural rather than a failure of discipline, which is why it needs a countervailing mechanism.
Asymmetric attribution. Improvements get credited to additions; null results rarely remove anything. Both errors add items.
No expiry date. An item added for a specific problem has no mechanism that removes it when the problem resolves.
Loss aversion. Removing something feels like risking gains, even where the item was not producing them.
Sunk cost. A large container encourages finishing rather than reassessing.
Identity. A protocol becomes part of how someone sees themselves, which makes subtraction feel like regression.
Information asymmetry. Every commercial source has an interest in addition and none in subtraction. There is no marketing for stopping, and no product whose value proposition is that you need less.
Recognising these is most of the work, because they explain how a rational person ends up with eighteen items they cannot individually justify.
The Governance Structure
| Element | Requirement |
|---|---|
| Written reason per item | What problem, which marker or symptom, and what would indicate it worked |
| Category per item | Corrective, maintenance, or experimental |
| Observation window per addition | Three months and a specific thing being watched |
| One change at a time | Two simultaneous changes waste the window |
| Scheduled removal trials | One per quarter, on the least-justified item |
| Trigger-based review | New medication, new condition, resolved problem, new evidence |
| Annual full audit | Every item's reason restated against the current person |
The category distinction does the structural work. A corrective item addresses a problem that can resolve, so it should be trialled for removal. A maintenance item works by an ongoing mechanism and stops working when it stops, so it stays while the effect is wanted. An experimental item has a window and a defined outcome.
Without those categories a stack cannot be pruned, because nothing distinguishes an item still doing something from one whose reason lapsed. Fibre and plant sterols are maintenance; berberine added for glycaemic control is corrective; a new compound on trial is experimental. Conflating them produces either an indefinitely growing stack or the loss of an effect that was still being produced.
The Quarterly Removal Trial
This is the mechanism that makes governance real rather than aspirational, because it forces one subtraction per quarter regardless of whether anything prompted it.
Choose the least-justified item, not the one you are most confident about. The point is to find what is not working.
Remove one thing. Not reduce, and not several.
Set a duration appropriate to the compound. Two to four weeks for anything with a short-term effect. Eight to twelve weeks with a re-measurement for anything acting on a marker.
Define what you are watching before starting, so the conclusion is not retrospective.
Expect the first week to mislead. Expectation effects are strongest immediately, and some compounds have a brief rebound that resolves.
Accept the most common result, which is that nothing changes. That is a finding, and it means the item was not doing the work attributed to it.
Four removal trials per year is enough to prevent indefinite accumulation, and it produces something more valuable than a shorter stack: knowledge of which items are actually doing something for you.
The Triggers That Override the Schedule
A new medication. The highest priority, because interaction risk is immediate. Several supplements affect drug metabolism through cytochrome P450 enzymes and transporters, and several affect bleeding, thyroid absorption or glycaemic control.
Planned surgery. Several supplements affect bleeding and should be discussed with the surgical team in advance.
Pregnancy or planning pregnancy. Substantial changes required, and several common supplements are contraindicated.
A new clinical condition, particularly kidney or liver impairment, which changes the handling of many compounds.
An unexplained liver enzyme change, which is a specific reason to review supplements, since several including high-dose green tea extract and ashwagandha have hepatotoxicity reports.
A resolved problem. The trigger most often ignored.
New evidence, particularly a negative trial for something in your stack.
Crossing an age threshold where priorities invert, principally the point where lean mass preservation outranks metabolic optimisation.
The Test Worth Applying Annually
One question separates a governed protocol from a habit: can you state, for each item, what it is for and what would make you stop?
Anything failing that test has already stopped being a decision. It is being taken because it was once added, which is a different thing from being justified.
The broader principle applies past supplements. Any protocol is a hypothesis about a person at a moment, and a protocol maintained without review is a habit that used to be a decision. That is not necessarily bad, since many habits are good ones, and it does mean the protocol is no longer responding to anything.
The framing that follows is governance rather than optimisation. Optimisation asks what the best protocol is now. Governance asks how a protocol will be kept matched to a person over thirty years through changing evidence, changing biology, changing medication and changing priorities. The second question is harder, less satisfying and considerably more consequential.
Over a long horizon the person doing governance ends up with a shorter, better-justified, safer and cheaper protocol, and one they can still explain. The person doing optimisation ends up with eighteen items and a reason for none of them.
The AEONNN Perspective
Governance is the Shield idea applied to a protocol, and it exists because the commercial gradient in this industry runs entirely toward addition. There is no marketing for stopping and no product whose value proposition is that a member needs less, which means the countervailing pressure has to be built in.
AEONNN's structural answer is the three categories. Corrective items address a problem that can resolve and get removal trials. Maintenance items work by an ongoing mechanism and stay while the effect is wanted. Experimental items carry a window and a defined outcome. Without that distinction a stack cannot be pruned, because nothing separates an item still working from one whose reason lapsed.
The quarterly removal trial is what makes it real rather than aspirational, since it forces one subtraction regardless of whether anything prompted it. The Safety layer holds the trigger list, with a new medication first because the risk is immediate and specific, and an unexplained liver enzyme change as a direct prompt to review the stack rather than look elsewhere. And the annual test is a single question: can a member say what each item is for and what would make them stop.
Pillar Matrix mapping
Database Matrix layers
- Meta / Consensus Layer (JAMA, BMJ, specialty society positions)
- Safety Layer (DrugBank, FAERS)
- Evidence Layer (PubMed, Cochrane, ClinicalTrials.gov)
- Real-Time User Layer (wearable and adherence signals)
Frequently Asked
Why does a protocol need reviewing?
Because four inputs change: the person, their markers, their medications and the evidence. A protocol assembled against one set of those is progressively less matched as they shift.
Why do supplement stacks only grow?
Improvements get credited to additions while null results rarely remove anything, items have no expiry date, loss aversion discourages subtraction, and every commercial source has an interest in addition.
What are the three categories of stack item?
Corrective, addressing a problem that can resolve; maintenance, working by an ongoing mechanism that stops when it stops; and experimental, on trial with a defined window and outcome.
How often should removal trials happen?
One per quarter, on the least-justified item. Four per year prevents indefinite accumulation and produces knowledge of which items actually do something for you.
What triggers an immediate review?
A new medication above all, plus planned surgery, pregnancy, a new clinical condition, an unexplained liver enzyme change, a resolved problem, new negative evidence, or crossing an age threshold where priorities invert.
Why does evidence changing matter?
Because a stack reflects the evidence at the time of purchase. Ginkgo, glucosamine, B vitamins for cardiovascular prevention and vitamin E were all defensible additions before their trials reported.
What is the test of a governed protocol?
Whether you can state, for each item, what it is for and what would make you stop. Anything failing that has stopped being a decision.
Evidence and review
Any dosage ranges cited here reflect the ranges used in published human trials, not personal recommendations. Evidence in this field moves, so this article is reviewed quarterly and carries its last-updated date above. Nothing here is intended as medical advice, and supplementation should be discussed with a qualified clinician, particularly alongside prescribed medication or an existing condition.