Drug Interactions

1. Checking Interactions

MedKit automatically checks drugs against the FDA's database for adverse events. You can use this to show interaction alerts directly to your users before they prescribe anything.

import asyncio
from medkit import AsyncMedKit

async def main():
    async with AsyncMedKit() as med:
        # Cross-reference an arbitrary number of compounds
        drug_list = ["Aspirin", "Warfarin", "Ibuprofen"]
        
        # Resolves to a List of Dictionaries containing the drugs compared and their warning
        warnings = await med.interactions(drug_list)
        
        for w in warnings:
            # Pydantic model with .severity, .risk, and .evidence
            warning_obj = w["warning"]
            drugs_checked = " + ".join(w["drugs"])
            
            print(f"[{warning_obj.severity.upper()}] {drugs_checked}")
            print(f"Message: {warning_obj.risk}")
            print(f"FDA Evidence Link: {warning_obj.evidence}\n")

if __name__ == "__main__":
    asyncio.run(main())

2. CLI Usage Example

You can instantly check a prescription cocktail directly from the terminal:

$ medkit interactions "Aspirin, Warfarin"

                          Drug-Drug Interactions                           
┏━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
┃ Drugs              ┃ Severity ┃ Risk                                    ┃
┡━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩
│ Aspirin + Warfarin │ High     │ Increased risk of bleeding events, in…  │
└────────────────────┴──────────┴─────────────────────────────────────────┘

3. Under The Hood: The Engine

When you check interactions, the engine does a few things behind the scenes:

• It sequentially forms permutations of the target list (A+B, A+C, B+C).
• Fires off highly parallelized async sub-queries using httpx.
• Checks for the presence of your drugs within the explicitly defined drug_interactions and boxed_warnings JSON fields of the FDA Label database.

4. Severity Thresholds Formulation

MedKit categorizes interaction warnings into three levels based on the exact words the FDA uses:

• SEVERE: Matches "contraindicated", "fatal", "life-threatening", "black box".
• MODERATE: Matches "caution", "avoid", "monitor", "dose adjustment".
• MILD: Any standard interaction tables remaining.