How We Build & Review Our Drug Interaction Library
1. Where the data comes from
Each interaction is compiled from a combination of FDA Structured Product Labeling (the official prescribing information), the published biomedical literature cited in each entry's references, DDInter, and legacy DrugBank interaction data distributed by the NIH's RxNav service (decommissioned in 2024). We do not license a commercial interaction compendium — every entry here was assembled individually.
About DDInter: DDInter is a major open-access, open-source database designed specifically for analyzing drug–drug, drug–food, and drug–disease interactions. Its second major release, DDInter 2.0, greatly expanded the original's capabilities to support clinical decision-making and patient safety. Portions of our library are manually curated from DDInter by our pharmacist team, then re-verified against the primary literature before publication.
2. How an entry is written
The documented interaction — its effect, severity, onset, evidence level, mechanism, and literature reports — is captured first, with its citations. We then use AI assistance, under strict rules, to re-express that material: a plain-English explanation, a clinical summary, patient-framed management guidance, a bottom line, and rewordings of the deep-dive and literature-report sections in our own words. The AI is never the source of facts: it rewrites what is already documented, it may not add or invent claims, and a reworded section is automatically rejected if it drops or moves a single citation.
3. How citations are verified
Reference lists are checked against Crossref and PubMed to attach a verified link to each citation wherever a confident match exists. Links are only added when the bibliographic match is exact — we would rather show a citation without a link than link to the wrong study. Truncated or broken links from source documents are removed rather than displayed.
4. Pharmacist review
Entries are reviewed by Dr. Brian Staiger, PharmD, BCPS, a licensed clinical pharmacist and Board Certified Pharmacotherapy Specialist. The review covers the severity rating, the direction and plausibility of the mechanism, and the practical framing of the management guidance.
5. How severity and evidence are graded
Severity levels
- Contraindicated: These should generally not be used together.
- Major: Potentially serious — often needs a change or close monitoring.
- Moderate: Can be significant — usually manageable with monitoring.
- Minor: Usually limited clinical impact.
Evidence grades
- Established: Well documented — supported by controlled studies or strong clinical data.
- Probable: Good supporting evidence, though not definitively proven.
- Suspected: Some evidence suggests this interaction, but it is not well established.
- Possible: Limited or conflicting evidence; the interaction may occur.
- Theoretical: Predicted from the drugs' pharmacology; not yet confirmed in people.
6. Updates
The library grows continuously — new drugs, new pairs, and refreshed write-ups ship every week, and each page shows its last-updated date. When source guidance changes, the affected entries are regenerated and re-reviewed rather than patched.
7. What we deliberately don't do
We don't claim completeness: this is not a list of every known drug interaction, and a combination that isn't documented here is never "cleared" by its absence. For screening a full medication list we recommend a comprehensive tool like the Drugs.com Interaction Checker alongside a review by your own pharmacist. And nothing here is medical advice — it's educational material to make the conversation with your care team better.
Read more about how we work
This page covers the drug-interaction library specifically. For the standards that apply across the whole site, see our Editorial & Medical Review Policy (how content is written, reviewed, and corrected) and our Data Sources page (every database we build on, what it powers, and how often it refreshes).