Product ·
How Qaytona builds ingredient monographs
Image by Hasan Kopuz
A monograph is only useful if you can see where each claim came from, what kind of study it rests on, how recent it is, and whether it applies to the patient in front of you. Qaytona is built around those four questions. Every sentence links to its source, the clinical sections are written only from peer-reviewed research and regulators, and every citation is checked against its paper before a monograph goes live.
The practitioners who use Qaytona pay for it. No brand pays to be included, and no monograph carries an affiliate link or an ad.
What goes in
Each monograph starts with a PubMed search for the ingredient, its forms and the names practitioners use for it. Most of what comes back is off target, so the results are ranked into six groups and only the strongest papers in each group are kept:
- Systematic reviews and meta-analyses
- Randomized controlled trials
- Safety and adverse events
- Interactions and how the body handles the ingredient
- Forms and absorption
- Everything else, by relevance
Safety and interactions get groups of their own so the far larger number of efficacy studies can't crowd them out. Landmark trials are always kept. Each section then sees only the papers tagged for it, so dosing is written from dosing studies and safety from safety reports.
Across 90 monographs, October 8, 2026.
| Stage | Count |
|---|---|
| PubMed results | 3,532,375 |
| Papers kept | 10,739 |
| Papers cited | 4,115 |
Alongside the papers come the references practitioners already check: fact sheets from the National Institutes of Health (NIH) Office of Dietary Supplements and the
National Center for Complementary and Integrative Health,
Memorial Sloan Kettering's About Herbs,
LiverTox, the NIH's database of drug-induced liver injury, the
Alzheimer's Drug Discovery Foundation's brain-health ratings, the
European Medicines Agency's herbal monographs,
Health Canada's licensed claims,
ClinicalTrials.gov, the
NIH supplement label database and the
Food and Drug Administration's adverse event reports.
Across 90 monographs, October 8, 2026. Regulator and clinical guidance is cited too and isn't counted here.
| Kind | Citations |
|---|---|
| Peer-reviewed papers | 4,247 |
| Practitioner and patient posts | 1,373 |
| Traditional-use research | 131 |
How it's written
Before any writing starts, the findings are pulled out of each paper: who was studied, which form and dose, for how long, and what changed. Each section is drafted from those findings and nothing else, so a monograph can only say what a paper said.
The details that decide whether a result applies to your patient stay attached to it. A dose comes with its form and the population that took it. A result from a lab or animal study is labelled as one, and a small or short trial is described as small or short. Acronyms are spelled out at first use, and jargon that wouldn't change what you tell a patient is cut.
Across 90 monographs, October 8, 2026.
| Published | Papers |
|---|---|
| Before 2000 | 156 |
| 2000–04 | 154 |
| 2005–09 | 207 |
| 2010–14 | 343 |
| 2015–19 | 636 |
| 2020–24 | 1,688 |
| 2025–26 | 1,063 |
How it's checked
Writing from findings keeps a draft close to its papers. These checks catch the mistakes that remain.
- Every citation is resolved. Each one has to point to a real paper, and the paper has to be on the claim's topic.
- Safety, interaction and clinical evidence claims are each read against their paper. Other sections are spot-checked. The checker reads the full paper when it's free to read, and the abstract when it isn't. A claim the paper doesn't support is removed.
- A clinical sentence with no citation has to find one. The sentence is checked against the papers. If one supports it, that paper is cited. If none does, the sentence is deleted.
- Every dosing row is cited. A dose without a source doesn't ship.
- Every quoted post has to support its point. A practitioner or patient post that doesn't is dropped.
Here is what that looked like for one monograph, the ashwagandha monograph.
Ashwagandha monograph build record, September 14, 2026. PubMed returned 2,069 results and 84 papers were kept; a paper can count in more than one group.
| Group | Considered | Kept | Share kept |
|---|---|---|---|
| Systematic reviews and meta-analyses | 48 | 18 | 38% |
| Randomized controlled trials | 60 | 30 | 50% |
| Safety and adverse events | 27 | 15 | 56% |
| Interactions | 24 | 15 | 63% |
| Forms and absorption | 37 | 12 | 32% |
| Everything else, by relevance | 40 | 12 | 30% |
Ashwagandha monograph build record, September 14, 2026.
| Study design | Papers |
|---|---|
| Systematic reviews and meta-analyses | 9 |
| Randomized controlled trials | 21 |
| Other human studies | 6 |
| Narrative reviews | 3 |
| Lab and animal studies | 3 |
Ashwagandha monograph build record, September 14, 2026.
| Check | Result |
|---|---|
| of 102 citations point to a real, on-topic paper | 100% |
| claims removed because no source supported them | 13 |
| quoted posts dropped for not supporting their point | 1 of 27 |
Ashwagandha monograph build record, September 14, 2026.
Once a monograph is live, members can flag anything that looks wrong from the monograph page.
Tradition and word of mouth
Practitioners also want to know what tradition says and what patients report. Both appear in each monograph, kept apart from the clinical evidence and labelled for what they are.
| Evidence | Tradition | Word of mouth | |
|---|---|---|---|
| What it is | Peer-reviewed research, regulators and trial registries | Published research on traditional medicine systems | What practitioners and patients report online |
| Cited to | The paper or the guidance | The paper that records the use | The post it came from |
| Shapes the clinical sections | Yes | No | No |
What it doesn't do
- It isn't a systematic review. Each search is ranked and only the top papers in each group are kept, so a relevant paper can miss the cut. Results aren't pooled, and trials aren't scored for risk of bias.
- Paywalled papers get a lighter check. The checker can read only their abstracts, which can leave out the dose, the form or who took part.
- It's as current as its last update. Each monograph shows the date, and a trial published after it isn't in the monograph yet.
- Label amounts are claims. Figures from the NIH label database are what a product says it contains. No lab has checked them.
- It doesn't know your patient. A dose comes with the population that took it. Whether your patient matches that population is for you to judge.
What's next
- More than 1,000 monographs. The library will grow from 90 to cover the ingredients practitioners stock and patients ask about, peptides included.
- Interaction checks for a patient's whole list. Patients rarely take just one thing. You'll be able to enter everything someone takes, prescriptions and supplements together, and see each interaction found in the research, with the paper it comes from. The decision about what to do stays with you.
- Monographs that stay current. AI agents will watch for new trials, reviews, regulator warnings and retractions, and send the monographs they touch back through the checks. Each update will say what changed.
- Help explaining it to patients. Each monograph will come with a version for patients, in plain words and in the language they speak, that you can share or print.
- A pipeline that keeps improving. When a check gets stricter or a source is added, every monograph is rebuilt through it, older ones included.
New sources and checks are coming too. Each closes a gap in what a monograph can tell you.
- Chinese-language research. For some herbs, most of the clinical research is published in Chinese, and nearly all berberine trials ran in China. Those journals will join the evidence layer.
- Trials that never reported. Many herbal trials are registered in India, Iran and China, outside ClinicalTrials.gov. Adding the
World Health Organization (WHO)'s registry search shows which trials finished without publishing results, a sign the evidence may look better than it is.
- Breastfeeding safety.
LactMed, the NIH's database of drugs and supplements in breastfeeding, will join the safety section.
- Adverse event reports from outside the US. The FDA's reports cover only the US and can't tell one form from another. Reports from the WHO's
VigiAccess, Health Canada and the European Union will sit next to them.
- European verdicts and limits. The
European Food Safety Authority has ruled on supplement health claims and set its own upper intake limits. Each monograph will show them next to the US figures.
- Interactions as drug lists. A finding like "speeds up a liver enzyme in lab studies" doesn't tell you which of your patient's drugs to watch. Each one will name the drug classes it affects, using the FDA's interaction tables and clinical studies from the
NIH-funded Center of Excellence for Natural Product Drug Interaction Research.
- Adulteration warnings. Some herbs are often sold diluted or swapped for a cheaper plant. Bulletins from the
Botanical Adulterants Prevention Program will flag which ones, and what to look for on a label.
- YouTube, TikTok and podcasts. Patients arrive with claims from videos and shows, like berberine as "nature's Ozempic." These will join posts in the word-of-mouth layer, each set against what the evidence says.
- Traditional-use records. Tradition today comes from published research on Ayurveda, Unani and other systems. The records themselves come next:
Dr. Duke's databases, the
Native American Ethnobotany Database.
- Who paid for each trial. The company that sells an extract often funds its trials. Each cited trial will name its funder.
- Full-paper checks for older monographs. Most monographs, ashwagandha included, were checked against abstracts alone. Each will be checked again against the full paper where it's free to read.
Help us build it
Qaytona is free while it's in beta, and the practitioners using it now shape what it becomes. Once you're in, there are four ways to help:
- Review a monograph. Read one in your area and flag anything wrong, missing or unclear with the feedback button in the corner of the page.
- Tell us what to add. Name the ingredients your patients ask about that the library doesn't cover yet.
- Improve the monographs. Tell us which sections you use, which you skip, and what you wish they said.
- Shape the product. Tell us what's missing, what gets in your way, and what you'd change.
Every note reaches the founders, through the feedback button or at info@qaytona.com.