How we read the evidence.

Longevity is a noisy field. A single mouse study can become a headline overnight. Our job is to filter that noise into something you can trust. This is how.

Source hierarchy. We weight evidence by how hard it is to fool. Meta-analyses and systematic reviews carry the most weight, then large randomized trials, then long-running human cohort studies. Single small trials, animal-only results, and preprints are treated as signals, not conclusions.

The three-institution threshold. A finding is only presented as settled when three or more independent research groups, in different places, reach it separately. Convergence is the test. If only one lab sees an effect, we say that clearly.

Evidence grading. We label where a claim stands. Strong means replicated across institutions and study types. Emerging means consistent but not yet broad. Early signal means promising and unproven. The label is part of the sentence, not a footnote.

The labs we read most. Harvard, Oxford, University College London, and Stanford form the core. The Salk Institute, the Buck Institute, Karolinska Institutet, ETH Zürich, and the Max Planck Institutes extend it. We track their published work; we are independent of all of them.

When the research disagrees with what people want to hear, we report the research. Uncertainty is stated, not smoothed over.