Credibility Begins With How Evidence Is Interpreted
This page is essential to CGI’s credibility and should be published before any product-related study interpretation.
CGI uses these principles to separate observation from conclusion, identify the boundaries of a study, and communicate what a result does — and does not — establish.
Rules for Responsible Scientific Communication
These principles define the minimum interpretive discipline CGI applies when discussing biomarkers, omics data, controlled studies, and nutrition-related healthy-aging evidence.
Biomarker Change Is Not Age Reversal
A change in a telomere-related signal, methylation pattern, inflammatory marker, or metabolomic pathway does not independently establish that an animal has become biologically younger.
Association Is Not Causation
Microbiome taxa, metabolites, proteins, and pathways may be associated with an intervention without proving a direct causal mechanism.
Statistical Significance Is Not the Same as Biological Importance
A significant result may be small, context-specific, or vulnerable to multiple-testing effects.
Study Population Defines the Evidence Boundary
Results from one breed, sex, age group, colony, or health status should not automatically be generalized to all dogs.
Endpoint Hierarchy Matters
Functional outcomes, clinical observations, imaging, blood biomarkers, microbiome findings, metabolomics, and exploratory omics provide different types and levels of evidence.
One Study Is Not Definitive
Controlled feeding research may provide proof of concept, but replication, larger populations, household-dog studies, and peer review increase confidence.
Nutritional Evidence Is Not Therapeutic Evidence
Nutrition studies may support normal function, maintenance, resilience, or healthy-aging statements; they do not automatically support disease-treatment or drug-like claims.
Make the Status of Evidence Visible
CGI should label evidence status explicitly so readers can distinguish established biology, study observations, exploratory findings, planned work, review status, and the need for replication.
These labels are not a single linear ranking. They describe different aspects of evidence status, including maturity, study context, publication status, and reproducibility.
Five Questions CGI Should Ask
What Was Measured?
Define the endpoint before interpreting its meaning.
In Which Dogs?
Breed, age, sex, health status, and study environment define the boundary.
How Large Was the Effect?
Statistical significance alone does not establish practical biological importance.
What Is the Evidence Status?
State whether the finding is controlled, exploratory, peer reviewed, or still planned.
What Remains Uncertain?
Limitations and replication needs should remain visible alongside the result.
Strong Scientific Communication Includes Its Own Limits
CGI’s objective is not to make every finding sound stronger. It is to communicate evidence at the level the study can reasonably support.