Drug-combination adverse-event disproportionality (FDA FAERS)
drug_interactionsAnalyze co-reported adverse events for 2–5 drugs using FAERS counts, top reactions, and PRR disproportionality to flag statistical signals—not clinical interaction evidence.
Instructions
Adverse event reports that name drugs together: coReportCount, the number of FAERS reports naming every listed drug; the reactions most often named in them; and for each reaction a proportional reporting ratio (PRR) with a 95% confidence interval against the reports of each drug without the others. A reaction is marked disproportionate when it has at least 3 co-reports and the lower bound exceeds 1 against every drug (null when a comparison cannot be computed). Drugs are matched by the product name as reported: a brand name such as Coumadin is not merged with its generic, warfarin. A name no FAERS report names is refused (refunded), naming it. FAERS reports are unverified and can be duplicated: report counts and disproportionality are statistical associations, not evidence of an interaction or of causation, and no substitute for a clinical drug-interaction reference. Source: the FDA Adverse Event Reporting System (FAERS), via openFDA.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"drugs": "<drugs>"
},
"name": "drug_interactions"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| drugs | Yes | Array of 2-5 different drug names (brand or generic), each matched against the product name reported in FAERS. Drug names need at least two letters or digits and may otherwise hold only spaces, hyphens, apostrophes, periods, slashes, parentheses and commas; accents, curly apostrophes, ® and ™, "+" and "&", and strengths such as 0.5% or 1:100,000 are normalized first (a refused name is refunded) |