Risk hierarchy and translation limits

Risk Hierarchy and Translation Limits

This file tracks the intervention-risk logic of the repository.

Its purpose is not to rank interventions by excitement. Its purpose is not to confuse high upside with readiness. Its purpose is not to collapse “interesting” into “appropriate.”

It exists to answer a more useful question:

What kind of intervention is this, what kind of risk does it carry, and how far should it be allowed to move toward protocol design given the current evidence?

This file is the restraint layer for the intervention section.

Core Position

Risk is not binary.

An intervention is not simply safe or unsafe. It can be:

biologically risky
clinically risky
measurement-risky
translation-risky
interpretation-risky
implementation-risky
low-risk in one population and high-risk in another
low-risk in theory and high-risk in repeated real-world use

This repository therefore does not treat risk as a single axis.

It treats risk as a hierarchy of constraints.

Why This File Matters

The interventions section already shows a recurring pattern:

some interventions have strong human functional support but modest mechanistic specificity
some have strong mechanistic logic but weak human validation
some have compelling biomarker movement but uncertain organismal benefit
some are biologically ambitious enough that control failure changes the entire risk profile
some are not dangerous because they are toxic, but because they are easy to overinterpret, oversell, or misapply

That means protocol design cannot be built only from upside.

It has to be built from upside filtered through risk, translation burden, measurement quality, and real human relevance.

Risk Does Not Mean “Not Interesting”

This repository does not use high risk as a synonym for low value.

Some of the most important intervention classes in the repository are also some of the riskiest.

Examples include:

partial epigenetic reprogramming
telomerase-related intervention
broad plasma-environment manipulation

These remain scientifically important even when they are nowhere near protocol-ready.

The point of the hierarchy is not dismissal. The point is placement.

Main Risk Domains

1. Biological Risk

This refers to the possibility that the intervention itself creates biological harm, destabilization, or unacceptable tradeoff.

Examples include:

oncogenic risk
dedifferentiation risk
tissue toxicity
immune suppression
maladaptive overactivation
regenerative pressure in damaged systems
excessive burden from growth suppression

Biological risk is one of the clearest hard-stop constraints in the repository.

2. Interpretation Risk

This refers to the possibility that the intervention appears to work because a biomarker moved, while the organism did not improve meaningfully.

This is one of the biggest risk domains in the entire repository.

Examples include:

clock improvement without functional gain
inflammatory-marker movement without resilience improvement
microbiome change without ecosystem or organismal benefit
metabolic biomarker improvement without better capacity or recovery

Interpretation risk is why Pattern 6 matters: function takes precedence when biomarkers conflict with it.

3. Translation Risk

This refers to the possibility that an intervention is biologically strong and still not realistically usable.

Examples include:

delivery complexity
poor scalability
narrow tissue fit
difficult regulation
repeated procedural burden
high dependence on specialized infrastructure
weak fit outside ideal research settings

An intervention can be biologically real and translationally weak at the same time.

4. Population-Fit Risk

This refers to the possibility that a strategy helps only certain populations, or changes from beneficial to harmful depending on baseline condition.

Examples include:

benefit in high-burden populations but not in healthy adults
risk asymmetry in older, frail, or disease-burdened people
strong fit in telomere biology disorders but not in normal aging
exercise or fasting patterns that help one person and overstrain another

This matters because “works in humans” is too coarse a statement for most geroscience interventions.

5. Implementation Risk

This refers to the gap between an intervention that can work and an intervention that can be done well enough to matter.

Examples include:

adherence failure
overcomplexity
dose or schedule drift
sequencing errors
recovery mismatch
inability to maintain the intervention in real life

This repository treats implementation as part of intervention quality, not as a separate practical afterthought.

Translation Limits

The repository also needs to distinguish risk from translation limits.

Some interventions are limited not because they are biologically impossible, but because current translation is still too weak.

Common translation limits include:

insufficient human evidence
lack of durable outcome data
biomarker-function disagreement
poor individual-level interpretability
narrow trial populations
unresolved delivery or schedule questions
uncertainty about long-term use
lack of strong functional endpoints

This means an intervention can sit below protocol readiness even if it remains a serious research priority.

Repository Risk Tiers

The following tiers are repository tiers, not universal truth claims.

They are meant to organize judgment, not replace it.

Tier 1 | Human-Grounded, Lower-Risk, Function-Forward Interventions

These interventions already have meaningful human relevance and are strongest when judged through function, resilience, and long-term health maintenance.

Current repository examples:

exercise
high-quality dietary patterning
sleep and recovery interventions with real treatment logic

Common profile:

strong human relevance
lower biological risk
higher implementation value
stronger functional case than biomarker case
most plausible early bridge to protocol design

These are not risk-free. They are the strongest current candidates for protocol thinking because their human reality is already visible.

Tier 2 | Human-Facing, Moderate-Risk, Translationally Active Interventions

These interventions have meaningful human signal or strong translational logic, but still carry unresolved questions around fit, durability, interpretation, or risk-benefit balance.

Current repository examples:

time-restricted eating
microbiome-directed interventions
metformin
acarbose
some narrower sleep- or recovery-linked strategies beyond first-line logic

Common profile:

meaningful human or disease-adjacent signal
moderate interpretation burden
likely population-specific fit
not yet strong enough for broad protocol claims

These are not fringe. They are serious but still conditional.

Tier 3 | Mechanistically Strong, Human-Unsettled Interventions

These interventions have substantial biological coherence and often strong animal evidence, but human anti-aging translation remains incomplete.

Current repository examples:

rapamycin and broader mTOR-targeting pharmacology
NAD+ restoration
plasma exchange and circulating-factor interventions
many next-generation microbiome or systems-level therapeutic strategies

Common profile:

strong mechanistic interest
real translational activity
human evidence present but not decisive
high biomarker temptation
function and long-term fit not yet settled

These interventions should stay in serious evaluation, not drift into casual protocol logic.

Tier 4 | High-Upside, High-Risk Frontier Interventions

These interventions are among the most structurally important in the repository, but they are also the least forgiving of weak control, premature translation, or biomarker-only enthusiasm.

Current repository examples:

partial epigenetic reprogramming
telomerase-related direct interventions
other interventions where control failure changes the whole category of risk

Common profile:

extremely high mechanistic ambition
potentially broad hallmark reach
major oncogenic, destabilization, or delivery risk
low current translation readiness
strong scientific importance, low protocol readiness

These interventions belong near the top of research attention and near the top of the risk hierarchy at the same time.

How Risk Interacts With Function

One of the strongest conclusions from the biomarker section is that function must outrank biomarker prestige when the two diverge.

That matters here.

An intervention moves downward in repository credibility when:

biomarker signal improves without functional gain
complexity rises without resilience improvement
burden rises without clearer organismal benefit
risk rises without broader capacity improvement

This repository does not reward molecular theater.

It rewards meaningful change in a real organism.

How Risk Interacts With Combination Logic

Combination logic increases the importance of this file.

A combination can move an intervention upward in value or downward in safety.

Combination can:

broaden coverage
improve function
reduce a bottleneck
increase burden
create antagonism
create attribution failure
raise implementation difficulty
hide weak translation under apparent sophistication

This means combination logic should always be read through risk hierarchy, not outside it.

Bridge to Protocol Design

This file is one of the main gates before 05_PROTOCOL_DESIGN.

The core question is not:

Is this intervention exciting enough?

The core question is:

Given current evidence, current measurement limits, current risk, and current translation burden, does this intervention deserve protocol-level structure?

Working rule for this repository:

An intervention should not move toward protocol design unless it has enough of the following:

acceptable biological risk for the intended population
interpretable functional benefit
biomarker support that does not conflict with organismal reality
manageable implementation burden
enough human relevance to justify structured use
a risk profile that is explicit, not hidden

If those conditions are not met, the intervention remains in research evaluation.

Relationship to the Rest of the Repository

This file connects directly to:

08_NOTES | Emerging Patterns Across Hallmarks
because Patterns 2, 5, and 6 all constrain how risk should be read

02_BIOMARKERS/08_validation_and_translation_constraints
because translation failure and interpretation failure are major risk domains

03_INTERVENTIONS/11_combination_logic
because combinations can improve or worsen total risk burden

05_PROTOCOL_DESIGN
because this file defines what should be allowed to cross the boundary into structured protocol thinking

Current Assessment

Current repository assessment:

importance to intervention ranking: foundational
importance to protocol design: foundational
importance to translational honesty: foundational
relevance to the entire repository: system-wide

Open Questions

Which interventions in this repository are most likely to move tiers as human evidence matures?
What level of functional benefit is enough to offset moderate biological or translational risk?
When does high upside justify staying in active evaluation despite very low protocol readiness?
How should population-specific fit alter risk tiering?
How should risk hierarchy be updated when combinations enter the picture?

Status

Foundational constraint file.

This file should be treated as the intervention restraint layer of the repository, not as a pessimistic appendix and not as a substitute for judgment.