Methodology
Every assessment in Nara’s library follows the same sequence: what the evidence shows, what Nara concludes, how confident we are, what remains uncertain, and what could change our view.
How we read evidence
An intervention is rarely simply “good” or “bad”. It is well supported for one outcome and unstudied for another. Nara grades a bounded claim — this population, this comparator, this outcome — rather than issuing a universal grade for a compound or a habit.
Design, population, comparator, outcome definition, consistency across studies, and contradictory or null results all change what a result means. So do the limitations the authors report. A single positive study is a reason to look closer, not a reason to conclude.
Evidence in dogs, evidence in people, evidence in other animals, and mechanistic reasoning are never blended into one impression. Nara records the most direct evidence a claim actually reaches, and says so plainly.
How developed the evidence base is, how relevant it is to dogs, and how firmly Nara holds the conclusion are three separate dimensions. Collapsing them into one score hides the part you most need to see.
Knowing that something is tolerated at a given dose tells you nothing about whether it works. Pharmacokinetic and tolerance studies are recorded as safety evidence, never counted as support for a benefit.
Marker changes can precede a real health effect, mirror one, or mean nothing at all. Nara treats biomarker evidence as biomarker evidence until an outcome in dogs is measured.
“Reasonable to consider” describes the state of the evidence. It does not mean a particular dog should use something. Population evidence does not establish individual medical eligibility, and this library does not make that decision for you or your veterinarian.
Nara's positions are the best current read, and they should change when better evidence appears. Every assessment records what would strengthen it, what would weaken it, and what would change the position outright.
Veterinarians, scientists, researchers and practitioners each contribute genuine expertise, and Nara takes that seriously. Evaluating a literature base and making an individual clinical decision remain different tasks, judged on different grounds.
A topic appears in the library because owners encounter it, not because Nara supports it. Commercial relationships must never affect an evidence rating, and none do.
Nara Evidence Framework v1.0
Inclusion is not endorsement. Commercial relationships must never determine Nara’s conclusions.