Calibrated Signal — Hard science. Honest signal.

How I Evaluate Evidence

The framework behind every Calibrated Signal article, critique, and takeaway.

Longevity claims often sound more certain than the evidence allows. I run each one through multiple filters: aging biology, mechanism, human relevance, bioinformatics, funding incentives, and plain-language translation of jargon.

I write from an unusual overlap: patient with a coronary stent, scientist, emergency nurse in training as an APRN with cardiology specialty, bioinformatics and molecular-biology training, and former health/wellness industry CEO. I am not a physician. I do not sell supplements. And I do not treat mechanisms, biomarkers, or marketing claims as proof.

Get the Free GuideRead My Story
The Calibrated Signal evaluation framework, visualized: six lenses arrayed around a central EKG-waveform focal point, with a DNA helix in the background. The lenses are Aging Biology, Molecular and Systems Biology, Clinical Relevance, Bioinformatics, Funding and Incentives, and Plain-Language Jargon Translation.

The Six-Lens Framework

Every major claim passes through the same six questions before it becomes a takeaway.

Aging Biology

What aging pathway is this supposed to affect?

  • Starts with the hallmarks: inflammation, mitochondrial function, senescence, epigenetic change, nutrient sensing, proteostasis, and repair.
  • Asks whether the claim fits an aging mechanism or just borrows longevity language.
  • Revisited at the end so the mechanism does not outrun the evidence.

Molecular & Systems Biology

Does the mechanism survive real physiology?

  • Pathway diagrams are not proof.
  • Dose, tissue access, feedback loops, and trade-offs get named.
  • “Plausible” does not equal “proven in humans.”

Clinical Relevance

Does it matter in actual people?

  • Statistical significance is not the same as meaningful benefit.
  • Effect size, duration, population, and baseline risk all matter.
  • Bedside experience keeps the question practical: would this change a real decision?

Bioinformatics

What does the whole evidence stack say?

  • One paper is a clue, not a conclusion. I use computational methods to analyze dozens of studies before writing the first sentence on a topic.
  • Patterns matter across study design, dose, endpoints, population, and replication.
  • When the literature is messy, I say so.

Funding & Incentives

Who paid for it, and who benefits if it sells?

  • Industry funding is not automatic bias. It is context.
  • I look harder at endpoints, comparators, exclusions, and independent replication.
  • I do not sell, accept affiliate deals or promotions, or hold ownership in supplement companies.

Plain-Language Jargon Translation

Can the claim be explained without hiding uncertainty?

  • Nursing teaches translation: say what matters without burying people in jargon.
  • Clear language makes weak claims easier to spot.

How Evidence Gets Weighted

Not all evidence carries the same weight. A strong claim should climb the ladder. Weak claims usually stay near the bottom.

Rung 1

Hard Clinical Outcomes

Heart attacks, strokes, fractures, hospitalization, mortality. The outcomes people actually care about.

Rung 2

Converging Human Evidence

Replicated human trials or strong systematic reviews and meta-analyses pointing in the same direction.

A meta-analysis is only as good as the studies it pools.

Rung 3

Single Strong Human Trial

Adequate size, relevant population, clear endpoint, reasonable duration, and a result large enough to matter.

Rung 4

Genetics, Cohorts & Real-World Signals

Mendelian randomization, prospective cohorts, population studies, and real-world patterns.

Helpful for plausibility and long-term direction, but assumptions and confounding matter.

Rung 5

Human Biomarkers & Surrogates

ApoB, blood pressure, HbA1c, VO2 max, inflammatory markers, epigenetic clocks, and other measurable signals.

Useful for direction. Not automatically proof of better outcomes.

Rung 6

Animal, Mechanistic & Marketing Claims

Cell studies, mouse models, pathway diagrams, testimonials, influencer protocols, and brand-funded certainty.

Useful for generating hypotheses. Dangerous when sold as a personal protocol.

A lot of longevity content lives at rungs 5 or 6 and gets sold as if it came from rung 1 or 2. Marketing claims are not evidence, even when they borrow the language of science.

Editorial Guardrails

The standards I use before anything gets published.

Funding & Incentives

  • No supplement sponsors.
  • No supplement line of my own.
  • No supplement affiliate revenue.
  • Industry funding is labeled. It does not automatically make a study wrong, but it tells me where to look harder.

Data vs. Interpretation

  • What the data show gets labeled.
  • What the authors interpret gets labeled.
  • What I conclude gets labeled.
  • The three are never collapsed.

Corrections & Updates

  • Substantive errors are corrected in the affected post.
  • Material corrections are logged on the corrections page.
  • Older posts receive stale-by-date callouts when new evidence changes the interpretation.
  • Major evergreen posts are re-audited on a scheduled basis.

Scope & Humility

  • Educational content, not personal medical advice.
  • I am not a physician, and I do not diagnose or treat disease.
  • I name what I do not know.
  • I state what evidence would change my mind.
  • I will not pretend mouse data proves human longevity.
“A mechanism is not proof. A positive study is not certainty. A meta-analysis is not stronger than the trials underneath it.”

Get Signal. Not Noise.

One long-form post per week. Hard science. No sponsors.

Free with signup: the Heart Screening Guide — the 5 cardiac tests I had to argue for, and what they may reveal before the standard workup does.

No spam. Unsubscribe anytime.