Methodology - Supplement and Food Reviews

Here at PetEvidenceProject, we evaluate the quality and quantity of available scientific evidence, then combine this with the results of the studies to develop an overall grade for the product category in question. We scour scientific databases (PubMed, Google scholar) to find the evidence, and combine these into mini- scientific reviews. On each page, you will find a quick summary at the top, with the option to read more into the review and science behind each topic. 

Our business is transparency and knowledge. We do not receive funding from manufacturers for advertisers or sponsored products. We do not provide individual product reviews or opinions. We just provide the facts.

For each page, you find the “scope” of the review at the top of the page that explains the specifics of our investigation. Some supplements may have use for multiple conditions, and we investigate the scientific evidence separately for each one of these. 

Each review is meant to serve as an educational resource and is not a substitute for professional veterinary advice. Every animal is different and we encourage discussion with your veterinarian before starting, stopping, or adjusting any supplements, diets, or medications.

Peer Review Process

All reviews and articles on this site are authored by a veterinary specialist and undergo peer review by 1-2 other veterinary specialists prior to publication. We apply the same rigor as found in scientific journals to ensure utmost accuracy and clarity for the reader. Our mission is to take high level data and translate it into a clear, understandable format that every pet owner can use.

Our Generative AI Policy

At the Pet Evidence Project, we maintain a rigorous, human-led editorial process to ensure the clinical safety and scientific accuracy of our content. We pride ourselves on transparency and disclosures, including when we do, and more importantly when we don't, use generative AI.

To preserve the integrity of our evidence-based mission, we strictly prohibit the use of generative AI for the core scientific components of our work:

• No Automated Sourcing: We do not use AI to review, select, or interpret primary source material or scientific literature.
• No Information Consolidation: The synthesis of data and the analysis of evidence are performed exclusively by board certified veterinary specialists.
• No Automated Grading: All Pet Evidence Project Grades and are determined by us without the use of AI.
• No AI-Written Articles: We do not use AI to write our specialist series articles. Every piece of content is drafted by a specialist author, though occasionally AI may be used to assist with SEO to help us reach a broader audience.

We occasionally use generative AI as a supplemental tool to help draft background scientific sections for supplement and food reviews, as well as assist with citation organization (it is a huge time-saver for this).

Regardless of AI use, sections of both supplement/food reviews and articles undergo a final mandatory peer-review process by 1–2 additional veterinary specialists before publication to ensure accuracy of information.

Grading system

On each page, you will find an overall PetEvidenceProject grade ranging from A-F (or n/a). This overall grade takes into account the types of studies (ranging from low quality to high quality), the number of studies, risk of bias (for example, studies sponsored by the product manufacturer), and strength of results. Not all studies are created equal, and we seek to combine all of these factors into a simple, easy to use summary that can be used by pet owners to help make better informed decisions.

For example, a supplement with many robust, well-designed studies with minimal risk of bias and has repeated positive results would receive an “A” grade.

On the flip side, a supplement with many well-designed studies that show repeated lack of positive results and/or risk of significant harm would receive an “F” grade.

Most importantly for these topics, many do not have robust evidence, but rather a few small scale studies, or studies that may have unconscious/implicit bias present. For such studies where the evidence is poor, but with positive or mixed results, these would be classified as emerging evidence with an associated grade of “C” or “D” depending on the quality of evidence and findings.

That is, a product with a grade of “C” or “D” does not necessarily mean the product is ineffective, rather that we do not have sufficient evidence regarding efficacy (or lack thereof). 

We hope these grades will be helpful in assessing the holistic picture of supplement and diet evidence for consumers and veterinarians alike.

Grades Explained

We assign every supplement a letter grade based on the certainty of the evidence.

Grade A

Demonstrated Benefit

Multiple high-quality Randomized Controlled Trials (RCTs) or a Systematic Review showing consistent, positive results.

Grade B

Probable Benefit

At least one well-designed RCT or several strong observational studies. More data is needed for "certainty."

Grade C

Emerging / Inconclusive

Small pilot studies or "test tube" (in vitro) data only, or multiple studies showing conflicting results. More well-designed studies required for certainty

Grade D

Weak / Anecdotal

Minimal data/poorly designed studies with questionable benefit, extrapolation from other species only, or well-designed studies that show a lack of benefit.

Grade F

Evidence of No Benefit

Multiple studies have been performed and failed to show a difference compared to a placebo and/or significant risk of harm associated with use.

Grade n/a

No evidence available

No evidence for use or efficacy was available in the literature. Evidence evaluation not possible.