Decision rules: when to start, stop, and adjust interventions

Decision Rules

This file defines how protocol decisions are made in this repository.

Its purpose is not to automate judgment. Its purpose is to prevent drift.

A protocol framework becomes weak when decisions are made by impulse, excitement, prestige, or biomarker fascination instead of by clear rules.

This file exists so that protocol movement remains structured.

Core Position

A protocol decision is justified only when the intervention has earned it.

That means no intervention should enter, escalate, or remain in protocol structure simply because:

it sounds advanced
it has strong public attention
it produces attractive molecular language
it fits a preferred theory
it looks more ambitious than the foundation layer

A protocol decision should happen only when the intervention satisfies the standards already established in the repository.

Governing Rules Already Established

This file inherits several rules that are already settled elsewhere in the repository:

Rule 1 | Function takes precedence when biomarkers conflict with it

If a molecular biomarker and a functional or physiological measure disagree, function takes precedence.

Rule 2 | Foundation outranks novelty

Exercise, high-quality dietary patterning, and sleep and recovery treatment logic are the current strongest foundation interventions because they already improve the things aging actually costs.

Rule 3 | Risk is part of intervention quality

An intervention is weaker, not stronger, if its risk, burden, or translation fragility exceeds its real organismal benefit.

Rule 4 | Support must remain support

A support intervention only belongs if it helps the foundation hold better, fit better, or work better.

Rule 5 | Exploratory does not mean protocol-relevant

An intervention can be scientifically important and still remain below protocol threshold.

Main Protocol Decisions

This repository needs rules for several different kinds of decisions.

1. Entry decision

Should this intervention enter protocol structure at all?

2. Layer decision

If it enters, does it belong in:

foundation
support
exploratory only

3. Escalation decision

Should the intervention move upward in protocol significance?

4. Retention decision

Should it remain in the protocol, or has the evidence, burden, or function case weakened?

5. Combination decision

Does it belong in combination with something else, or is that just stacking?

Entry Rules

An intervention should enter protocol structure only if most of the following are true:

the mechanism is plausible enough to matter
the target logic is clear enough to specify
the evidence is strong enough to justify structured use
the risk profile is explicit
the burden is manageable
the intervention can be interpreted without depending on biomarker fantasy
the intervention has some meaningful relationship to function, resilience, recovery, or capacity

If these conditions are not met, the intervention remains outside protocol structure.

Foundation-Layer Rules

An intervention belongs in the foundation layer only if it meets nearly all of the following:

strong human functional evidence
clear effect on capacity, resilience, recovery, or long-horizon health
lower interpretation risk
lower biological risk relative to more ambitious layers
practical real-world use
clear value even when biomarker support is incomplete

At the current stage of the repository, this logic supports:

exercise
high-quality dietary patterning
sleep and recovery treatment logic

Nothing else should be promoted into the foundation layer without evidence strong enough to compete with those.

Support-Layer Rules

An intervention belongs in the support layer only if:

it strengthens the foundation rather than replacing it
it addresses a specific bottleneck
it improves fit, adherence, resilience, recovery, or function
its burden is justified by its contribution
its value is clear in relationship to the base

A support intervention should be excluded if it:

mainly adds complexity
mainly adds biomarkers
competes with the foundation for attention
cannot explain what it is supporting

Exploratory-Layer Rules

An intervention belongs in the exploratory layer if:

it is biologically important
it is still limited by evidence, interpretation, risk, or translation
it does not yet have enough human functional grounding for structured use
it still deserves active tracking

An exploratory intervention should not be promoted unless multiple things improve at once:

stronger human evidence
clearer function
clearer biomarker-function alignment
lower or better-bounded risk
better population-fit clarity

Escalation Rules

An intervention should move upward in protocol relevance only when it shows meaningful improvement in the domains that matter most.

Escalation requires some combination of:

stronger functional evidence
clearer resilience benefit
clearer recovery benefit
better human evidence
lower interpretation ambiguity
better risk control
stronger fit across real populations
better alignment between biomarkers and organismal outcomes

Escalation should not happen because of:

better marketing language
stronger molecular prestige
one favorable biomarker shift
one promising pilot without broader support

De-escalation Rules

An intervention should move downward or be removed when:

function does not improve meaningfully
burden exceeds benefit
risk becomes clearer and worse
adherence collapses
biomarker movement no longer looks trustworthy
foundation quality is being displaced
the intervention adds noise rather than clarity

This repository should allow downward movement just as clearly as upward movement.

A protocol that only escalates becomes dishonest.

Combination Rules

A combination should be allowed only if it improves at least one of the following:

bottleneck coverage
function
resilience
recovery
tolerability
durability

A combination should be rejected if it mainly creates:

redundant mechanism
increased burden
attribution collapse
higher risk without clearer organismal gain
biomarker complexity without functional improvement

The default rule is not “add until satisfied.” The default rule is “justify every added layer.”

Biomarker Use Rules

Biomarkers can support protocol decisions, but they do not rule them.

Biomarkers should be used to:

clarify direction
identify burden
suggest mechanism
monitor response
detect mismatch between theory and organismal outcome

Biomarkers should not be used to:

override function
justify escalation by themselves
hide weak real-world outcomes
create the illusion of confidence where confidence is not earned

If biomarkers and function disagree, function decides.

Function Rules

Functional and physiological measures have special weight in this repository.

A protocol change gains credibility when it improves:

grip strength
gait speed
cardiorespiratory fitness
recovery quality
frailty-related burden
physical capacity
resilience under stress
cognitive or daily-life function where relevant

A protocol loses credibility when those things remain unchanged while the molecular story becomes more elaborate.

Risk Rules

Risk must be stated, not implied.

Every protocol decision should specify:

main biological risk
main interpretation risk
main translation risk
main implementation risk
main population-fit limit

If those are not clear, the intervention is not ready for structured use.

Population-Fit Rules

No intervention should be treated as universally appropriate by default.

Every protocol decision should ask:

who is this for
who is it not for
what baseline burden changes the fit
what recovery capacity changes the fit
whether the intervention is more credible in high-burden or healthy adults
whether age, frailty, disease context, or functional status alter the logic

Protocol design without population fit is weak design.

Burden Rules

Burden is part of protocol quality.

An intervention is weaker when:

it is too hard to sustain
it reduces adherence to the foundation
it interferes with recovery
it creates more complexity than usable value
it raises total load without improving function

A lower-burden intervention with the same functional outcome should usually outrank a more complex one.

Default Decision Order

When deciding what to do with an intervention, the repository should ask in this order:

Does it improve real function, resilience, recovery, or capacity?
Does it fit the intended population?
Is the risk profile explicit and acceptable?
Does it strengthen the foundation or only compete with it?
Are the biomarkers supportive rather than misleading?
Is the implementation burden realistic?
Has it earned its protocol layer?

That order is deliberate. It keeps the protocol human-centered instead of biomarker-centered.

Relationship to the Rest of the Repository

This file is directly constrained by:

01_foundation_layer
because foundation rules set the protocol base

02_support_layer
because support rules define what counts as reinforcement

03_exploratory_layer
because exploratory status blocks premature promotion

02_BIOMARKERS/08_validation_and_translation_constraints
because biomarker limits shape decision strength

03_INTERVENTIONS/12_risk_hierarchy_and_translation_limits
because risk is part of every protocol decision

08_NOTES | Emerging Patterns
especially Pattern 6 and Pattern 7, because function and visible intervention hierarchy already constrain what can be decided here

Current Assessment

Current repository assessment:

importance to protocol structure: foundational
importance to escalation control: foundational
importance to translational honesty: foundational
relevance to the entire protocol section: system-wide

Open Questions

What minimum amount of functional improvement should count as meaningful for protocol escalation?
How should disagreement between different functional measures be handled?
When should a supportive intervention be removed even if it remains biologically plausible?
How much biomarker support is enough to matter when function is improving but mechanism remains unclear?

Status

Foundational protocol-governance file.

This file should be treated as the rule layer that keeps protocol design structured, selective, and honest, not as optional guidance and not as a substitute for judgment.