tooluniverse-drug-drug-interaction

# Drug-Drug Interaction Prediction & Risk Assessment

Systematic analysis of drug-drug interactions with evidence-based risk scoring, mechanism identification, and clinical management recommendations.

**KEY PRINCIPLES**:
1. **Report-first approach** - Create DDI_risk_report.md FIRST, then populate progressively
2. **Bidirectional analysis** - Always analyze A→B and B→A interactions (effects may differ)
3. **Evidence grading** - Grade all DDI claims by evidence quality (★★★ FDA label, ★★☆ clinical study, ★☆☆ theoretical)
4. **Risk scoring** - Multi-dimensional scoring (0-100) combining mechanism + severity + clinical evidence
5. **Patient safety focus** - Provide actionable clinical guidance, not just theoretical interactions
6. **Mandatory completeness** - All analysis sections must exist with explicit "No interaction found" when appropriate

---

## LOCAL PHARMACOLOGY REFERENCE (USE FIRST)

Before querying any external database, consult the local reference script for instant answers on CYP/UGT roles and known critical interactions:

```
scripts/pharmacology_ref.py   (no external dependencies, runs offline)

# Q927 pattern — valproate + lamotrigine:
python scripts/pharmacology_ref.py --type interaction --drug1 "valproate" --drug2 "lamotrigine"

# What does a drug do to UGT enzymes?
python scripts/pharmacology_ref.py --type ugt_inhibitor --drug "valproate"

# What enzymes metabolise a drug?
python scripts/pharmacology_ref.py --type ugt_substrate --drug "lamotrigine"
python scripts/pharmacology_ref.py --type cyp_substrate --drug "warfarin"

# Which drugs inhibit / induce a specific CYP?
python scripts/pharmacology_ref.py --type cyp_inhibitor --enzyme "CYP3A4"
python scripts/pharmacology_ref.py --type cyp_inducer  --enzyme "CYP2C9"

# Narrow therapeutic index checklist:
python scripts/pharmacology_ref.py --type narrow_ti

# All known interactions for one drug:
python scripts/pharmacology_ref.py --type all_interactions --drug "lamotrigine"
```

**Covered interactions include** (severity / mechanism):
| Pair | Severity | Key mechanism |
|------|----------|---------------|
| valproate + lamotrigine | **Major** | UGT1A4 inhibition → 2× lamotrigine levels + SJS risk |
| carbamazepine + lamotrigine | Major | UGT1A4 induction → 50% ↓ lamotrigine |
| oral contraceptives + lamotrigine | Major | UGT1A4 induction → 50% ↓ lamotrigine |
| valproate + phenytoin | Major | CYP2C9 inhibition + protein displacement |
| carbamazepine + valproate | Moderate | Epoxide hydrolase inhibition → toxic metabolite ↑ |
| simvastatin + ketoconazole | **Contraindicated** | CYP3A4 inhibition → rhabdomyolysis |
| simvastatin + clarithromycin | Contraindicated | CYP3A4 inhibition → rhabdomyolysis |
| rifampin + warfarin | Major | CYP2C9 induction → INR collapse |
| amiodarone + warfarin | Major | CYP2C9 inhibition → INR rise |
| clopidogrel + omeprazole | Moderate | CYP2C19 inhibition → reduced antiplatelet activation |
| quinidine + digoxin | Major | P-gp inhibition → 2× digoxin levels |
| lithium + NSAIDs | Major | Reduced renal clearance → lithium toxicity |
| fluoxetine + MAOIs | Contraindicated | Serotonin syndrome |

The script also covers UGT2B7 substrates (morphine, zidovudine) inhibited by valproate, UGT1A1 induction by rifampin, and the complete narrow therapeutic index list with monitoring parameters.

## LOOK UP, DON'T GUESS
When uncertain about any scientific fact, SEARCH databases first (PubMed, UniProt, ChEMBL, ClinVar, etc.) rather than reasoning from memory. A database-verified answer is always more reliable than a guess.

## New Symptom After New Medication: First-Line Reasoning

When a patient develops NEW symptoms after starting a new medication, the FIRST question is: could the new drug be interacting with an existing medication? Specifically check: (1) Does the new drug inhibit metabolism of an existing drug? (2) Does the new drug have additive pharmacodynamic effects?

---

## When to Use This Skill

Apply when users:
- Ask about interactions between 2+ specific drugs
- Need polypharmacy risk assessment (5+ medications)
- Request medication safety review for a patient
- Ask "can I take drug X with drug Y?"
- Need alternative drug recommendations to avoid DDIs
- Want to understand DDI mechanisms
- Need clinical management strategies for known interactions
- Ask about QTc prolongation risk from multiple drugs

---

## Clinical Reasoning Framework

Before querying any database, apply this reasoning framework to predict interactions mechanistically.

### The Perpetrator-Victim Model

In every drug interaction, identify two roles:
- **PERPETRATOR**: the drug causing the change (the inhibitor, inducer, or pharmacodynamic amplifier)
- **VICTIM**: the drug being affected (the one whose levels or effects change)

For each drug pair, ask these questions in order:

1. **Does the perpetrator change how the victim is absorbed, distributed, metabolized, or eliminated?** If yes, this is a pharmacokinetic interaction. Determine which enzyme or transporter is involved (CYP450, UGT, P-gp, OATP, etc.).
2. **Is the perpetrator an inhibitor or an inducer of that pathway?**
   - Inhibitor → victim levels go UP → predict increased efficacy or toxicity
   - Inducer → victim levels go DOWN → predict reduced efficacy or therapeutic failure
3. **What happens clinically when the victim's level changes?** Predict the downstream consequence: toxicity from supratherapeutic levels, or treatment failure from subtherapeutic levels.
4. **Always check the reverse direction.** Analyze B→A as well as A→B. The perpetrator-victim relationship may be asymmetric or bidirectional.

Special case -- **Prodrugs**: If the victim is a prodrug that requires metabolic activation, inhibiting its activating enzyme reduces efficacy (not toxicity). Inducing its activating enzyme may increase efficacy or toxicity of the active metabolite.

---

### Phase II Metabolism: Glucuronidation Interactions (UGT Enzymes)

Most DDI reasoning focuses on CYP450 (Pha