How We Evaluate Supplements

Our purpose

Supplement marketing often promises more than the science supports. This site exists to help you tell the difference between claims that are backed by research and claims that are mostly hype, and to point you to the original sources so you can read them yourself.

Who we are, and what that means for our content

Bottled Hope is published by SK102 LLC and maintained by a software developer, not a physician, pharmacist, or registered dietitian. Because of that, we do not form our own medical conclusions. Our role is to curate: we find what recognized health authorities and systematic reviews of the research have already concluded, summarize it in plain language, and link directly to those sources.

If qualified experts haven't reviewed the evidence on a claim, we say so rather than filling the gap with our own judgment.

What counts as evidence

We use two levels of sources.

Primary sources for ratings. Only these can determine an evidence rating:

• Systematic reviews and meta-analyses published in peer-reviewed journals, including Cochrane Reviews
• Summaries from recognized public health authorities and regulators, such as the NIH Office of Dietary Supplements (ODS), the National Center for Complementary and Integrative Health (NCCIH), the European Food Safety Authority (EFSA), and the European Medicines Agency (EMA)
• Clinical practice guidelines from professional medical societies, when they are based on a systematic review of the evidence

We also cite government regulations and regulatory notices, for example to report a supplement's legal status or a required label warning.

Supporting sources. These may be cited for context, but they can't raise a rating above "Limited" on their own:

• Individual randomized controlled trials
• Trials without a placebo group, such as trials where everyone takes the supplement or trials that compare only different amounts of it
• Observational studies
• Animal and laboratory studies
• Clinical practice guidelines that are not based on a systematic review of the evidence

Narrative reviews. A review that summarizes research without a systematic search of the evidence may be cited for context, labeled "Narrative review". It never counts toward a rating, and we never cite one on its own.

Published comments. When an editorial or letter published in a journal disputes a source we cite, we may list it beside that source, labeled "Published comment", so you can see the dispute and read it yourself. A published comment never counts toward a rating.

Wherever possible, we link to free, publicly accessible versions of sources, such as PubMed abstracts and PubMed Central full texts.

What we leave out. We read each source in full before citing it, not just its abstract. We do not cite documents that have been withdrawn or are no longer available, and we do not count reviews published in journals that have been delisted from Web of Science or have closed. These rules apply to sources added from 25 September 2026.

We rate claims, not supplements

A single supplement can have strong evidence for one use and none for another. So instead of giving each supplement one verdict, we list the specific claims commonly made about it and rate the evidence for each claim separately.

Some pages also have a Longevity and aging section. It covers claims about living longer: lifespan, death from any cause, and markers studied in connection with aging, such as NAD⁺ levels or epigenetic clocks. A change in a marker is not the same as a longer life. We rate those claims the same way as every other claim. Each one says whether the research was in people or in animals, and what it actually measured.

Our rating scale

When a systematic review uses the GRADE certainty framework, we translate its ratings directly:

• high certainty → Strong
• moderate certainty → Moderate
• low or very low certainty → Limited

Not every review uses GRADE. When they don't, when an ungraded review disagrees with a graded one, or when a review's own authors call a result weaker than its GRADE rating, we rate from what the reviews conclude, taking the lower reading, and never higher than a graded review allows. When a rating rests on a health authority's conclusion, the page cites a summary or guideline from a recognized health authority.

A review usually rates certainty separately for each outcome it measures. For each claim, we take the lowest rating among the outcomes that claim is about, rather than the most favorable one. Side effects are reported in the safety section, and the certainty shown next to a source is the lowest it gives any outcome, side effects included. If a claim cites a graded review, its rating never goes above what that review's certainty allows, even when a health authority's summary is more positive.

How we handle disagreement

When qualified sources reach different conclusions, we use the lower rating and explain the disagreement on the page. We would rather understate a benefit than overstate one.

Next to each source, we note when it discloses industry funding, or authors with financial ties to industry. When a source does not say, we do not guess.

Safety comes first

An evidence rating describes whether a supplement works. It does not mean the supplement is safe for you. Every supplement page places safety information near the top, including:

• Known side effects and risks
• Groups who should avoid it or use extra caution, such as people who are pregnant or breastfeeding, children, and people with certain medical conditions
• Documented interactions with medications or other supplements, as reported by our primary sources

The absence of a listed interaction does not mean none exists.

Our policy on dosages

We do not calculate or recommend personal doses. Where available, we quote intake figures, such as Recommended Dietary Allowances and Tolerable Upper Intake Levels, exactly as health authorities publish them, with a citation. Some authorities publish figures per kilogram of body weight. We quote those exactly as published and never calculate a dose from them, and we don't adjust any figure for individual health conditions. That is a decision for you and your physician or pharmacist.

How we use software tools

Bottled Hope is built with the help of AI tools, and we want to be clear about how.

• Research and drafting. AI tools search PubMed and health authority websites, read the sources, and draft each page. They keep a record of how every source was checked.
• Separate checks. Separate AI review steps check every citation against its original record (PubMed, the publisher, or the authority's own website), and review the wording for accuracy and for claims that go further than the sources do. Problems they find are fixed before a page can be published.
• Rules in software. The rating rules on this page are checked automatically every time the site is built. A page that breaks one cannot be published.
• Human approval. The site's maintainer approves every page before it goes live, and decides questions the tools cannot settle. A source we cannot verify is not cited.
• Monitoring. Software checks PubMed every day for new reviews, corrections, and retractions affecting the sources we cite, and flags them for review.

AI tools can make mistakes, and so can people. That is why every rating links to its sources, and why we want to hear about any error you find.

Updates and corrections

• Each page shows the date it was last reviewed.
• We aim to re-review every page at least once every 12 months, and sooner when major new research or safety warnings are published.
• Any change to a rating is noted on the page, with the date and the reason.
• If you find an error, please contact us at info@bottledhope.co. We take corrections seriously and will fix confirmed errors promptly.

Advertising and independence

This site does not currently carry advertising. We expect to introduce it once the site is established, because being funded by advertising rather than by selling supplements is what keeps the ratings independent.

Advertisers will have no influence over our content or ratings. We don't accept payment from supplement manufacturers for coverage. When ads are introduced we will block advertising categories where possible, to avoid showing ads for products that contradict our findings, and seeing an ad here will not mean we endorse the product.

Limitations

Research on dietary supplements is often limited, and new studies can change what is known. Our summaries simplify complex research and may not capture every nuance. This site is for general education only. Always consult your physician or pharmacist before starting, stopping, or changing any supplement.