Method
Our method: how we rate the evidence
The biggest quality gap in Dutch information about natural veterinary medicine is that almost nobody says how strong the evidence is — let alone that anyone dares to write down an honest «there is no evidence for this». We do, for every topic. This is how.
We assess the whole body of evidence, not a single study
One positive study proves little. We weigh the whole of the available research in the target species (systematic reviews, meta-analyses, RCTs) and translate that into a verdict on the evidence. Preclinical research (cells, animal models) and extrapolation from humans to animals explicitly do not count as evidence of an effect in your dog, cat, small mammal or horse.
The evidence levels
- Strong evidence
Several good studies in the target species (systematic review/meta-analysis or consistent RCTs) point the same way. We are reasonably certain about the effect.
- Reasonable evidence
One or a few sound studies in the target species, but still limited or not independently replicated. Further research may change the conclusion.
- Weak evidence
Only small, uncontrolled or indirect studies (e.g. an animal model or extrapolation from humans). Uncertain; treat claims with caution.
- Very weak / anecdotal
Almost entirely anecdotes, tradition or in-vitro work. No reliable basis for an effect in the animal.
- Insufficiently studied
There is too little or too weak research in the target species to draw a conclusion. Neither proven to work nor disproven.
- No evidence of effect
The available research shows no reliable effect beyond placebo, or a plausible mechanism is missing. Not recommended as a treatment.
«No evidence» is an answer too
Some popular treatments — homeopathy, for example — have been studied extensively and show no reliable effect beyond placebo. We do not keep quiet about that to please anyone. An honest «no evidence» protects animals better than a vague «it can do no harm».
Our source hierarchy
Reliability stands or falls with sources. The golden rule: never invent — no source, DOI, figure or quotation that does not verifiably exist and has not actually been consulted. We weigh sources from strong to weak:
- Systematic reviews and meta-analyses in the target species.
- Randomised controlled trials (RCTs) in the target species.
- Cohort and observational research in the target species.
- Case series and case reports — especially relevant for safety and toxicology.
- Authoritative reference works and guidelines (Merck Veterinary Manual, EMA, the Dutch food and product safety authority NVWA, university centres).
- Narrative reviews and expert opinion.
- In-vitro work and animal models — supporting, but not evidence of an effect in the target animal.
We always name human→animal and species→species extrapolation as such, never as direct evidence. We do not use web shops, brand sites, manufacturer claims, testimonials, forums or «studies» without a traceable publication as a source. If a source has an interest of its own (a homeopathy organisation, for instance), we flag that visibly in the reference list.
What you may expect from us
- Never «cures» or «fixes» — instead an honest distinction between support and cure.
- We also show the negative and failed trials that commercial sites leave out.
- Every topic is linked to primary sources with a DOI and the study type.
- Every YMYL page is checked by a named veterinary surgeon, with the date shown.
- Safety first: «natural» ≠ safe, and what is safe for one species can be toxic for another.
- Sources are checked for working links at every review; dead links are replaced or archived with a note.
Our rating is inspired by GRADE (gradeworkinggroup.org) and Cochrane's plain-language approach. Questions, or spotted a mistake? Let us know — we process corrections visibly. See also the knowledge base for the topics themselves.