MedicationInteractions / README-2.md
MorpheusIndustries's picture
Upload 2 files
a7b32fd verified
# Dataset: Patient Medication Interaction Q&A
## Description
A high-quality synthetic dataset of 5,000 patient-style medication interaction questions paired with pharmacist-grade answers. Each entry mirrors the language and concerns of real patients asking about drug-drug interactions, food-drug interactions, alcohol, pregnancy, side effects, and dosing. Designed for training healthcare conversational AI systems, pharmacy chatbots, and medical information assistants.
**Target audience:** AI/ML engineers building healthcare NLP models, pharmaceutical companies developing patient support tools, and academic researchers in clinical informatics.
---
## Key Features
- **Number of rows:** 5,000
- **Columns:** `id`, `question`, `answer`, `drug_mentioned_1`, `drug_mentioned_2`, `interaction_type`, `category`, `severity`
- **Drug classes covered:**
- Cardiovascular: ACE inhibitors (lisinopril, enalapril, ramipril), ARBs (losartan, valsartan), beta-blockers (metoprolol, atenolol, carvedilol), calcium channel blockers (amlodipine), statins (atorvastatin, simvastatin, rosuvastatin), diuretics (furosemide, hydrochlorothiazide, spironolactone)
- CNS: SSRIs (sertraline, fluoxetine, escitalopram, paroxetine), SNRIs (venlafaxine, duloxetine), benzodiazepines (alprazolam, lorazepam, diazepam), antipsychotics (quetiapine, haloperidol), sleep aids (zolpidem, eszopiclone, melatonin), NSAIDs
- Endocrine: Insulin (glargine, lispro), oral hypoglycemics (metformin, glipizide, sitagliptin, empagliflozin), thyroid hormones (levothyroxine), corticosteroids (prednisone, methylprednisolone)
- Anti-infectives: Penicillins (amoxicillin), cephalosporins (cephalexin), fluoroquinolones (ciprofloxacin, levofloxacin), macrolides (azithromycin), antifungals (fluconazole, itraconazole), metronidazole
- GI: PPIs (omeprazole, pantoprazole, esomeprazole), H2 blockers (famotidine, ranitidine), antacids (calcium carbonate, aluminum hydroxide)
- Supplements: Vitamin D, calcium, iron, magnesium, fish oil, St. John's wort, ginseng, ginkgo biloba, coenzyme Q10, vitamin K
- Common OTC: Acetaminophen, ibuprofen, naproxen, antihistamines (cetirizine, loratadine, fexofenadine, diphenhydramine), decongestants (pseudoephedrine, phenylephrine)
- **Interaction types covered:** major (441), moderate (677), minor (333), none (1,109), food (782), alcohol (392), condition (1,266)
- **Severity distribution:** high (1,513), medium (2,043), low (840), info (604)
---
## Intended Use
- Training conversational AI for patient medication support
- Enhancing pharmacy chatbot accuracy and safety
- Educational purposes for pharmacy and medical students
- Benchmarking NLP models on clinical question answering
- Fine-tuning large language models for healthcare use cases
---
## Data Collection Methodology
This dataset was synthetically generated by Vincent Kane, registered pharmacist, using a curated library of clinically reviewed question-answer templates developed from years of patient-facing pharmacy practice, combined with AI assistance and automated consistency checks. Questions are written in natural, colloquial patient language β€” including occasional informal phrasing β€” to reflect real-world pharmacy inquiries. Answers are clinically accurate, concise (2–4 sentences), and reflect current pharmaceutical guidelines. No real patient data was used; all entries are entirely synthetic.
Each entry was validated for:
- Clinical accuracy of the drug interaction described
- Consistency between `interaction_type`, `severity`, and answer content
- Avoidance of specific dosage recommendations or self-treatment encouragement
- Use of generic (non-brand) drug names throughout
---
## Format
- **File format:** CSV (comma-separated values)
- **Encoding:** UTF-8
- **Delimiter:** Comma (`,`)
- **Text quoting:** Double-quotes used where fields contain commas
---
## Column Descriptions
| Column | Type | Description |
|--------|------|-------------|
| `id` | string | Unique row identifier (Q0001–Q5000) |
| `question` | string | Patient-style question in plain language |
| `answer` | string | Pharmacist-grade answer (2–4 sentences) |
| `drug_mentioned_1` | string | Primary drug name (generic) |
| `drug_mentioned_2` | string | Secondary drug name if applicable (blank if none) |
| `interaction_type` | enum | One of: major, moderate, minor, none, food, alcohol, condition |
| `category` | enum | Therapeutic category (see below) |
| `severity` | enum | One of: high, medium, low, info |
**Category values:** pain relief, blood pressure, antibiotics, antidepressants, diabetes, allergy, heartburn, thyroid, birth control, vitamins/supplements, cold/flu, sleep aids, other
---
## Sample Rows
| id | question | answer | drug_mentioned_1 | drug_mentioned_2 | interaction_type | category | severity |
|----|----------|--------|-----------------|-----------------|-----------------|----------|----------|
| Q0001 | I'm taking warfarin for my heart condition and my new doctor wants to add ibuprofen. Is that combination safe? | Combining warfarin and ibuprofen can cause a significant interaction that may require dose adjustments or close monitoring. Please inform both your doctors about all your medications so they can coordinate your care safely. | warfarin | ibuprofen | major | pain relief | high |
| Q0042 | Is it safe to eat grapefruit while taking simvastatin? | Grapefruit and grapefruit juice can significantly increase blood levels of simvastatin by blocking enzymes that break it down, potentially increasing the risk of side effects. It is generally best to avoid grapefruit products while taking simvastatin; ask your pharmacist for alternatives. | simvastatin | | food | other | high |
| Q0085 | I've been taking cetirizine for years. Just started melatonin this week. Anything to be aware of? | There is no clinically significant interaction expected between cetirizine and melatonin when used at standard doses. Continue taking both as directed and let your doctor know if you notice any unexpected changes in how you feel. | cetirizine | melatonin | none | allergy | info |
---
## Limitations & Disclaimer
- This dataset is **synthetic** and intended for AI training and educational purposes only.
- It should **not** be used as a standalone clinical decision support tool without expert validation.
- Drug interaction knowledge evolves; users should verify critical entries against current authoritative sources (e.g., Lexicomp, Micromedex, FDA prescribing information).
- The dataset does not cover all possible drug interactions and does not constitute medical advice.
---
## Licensing
This dataset is licensed under a **Commercial License**. Purchasers may use the dataset to train, fine-tune, or evaluate AI models in commercial and non-commercial settings. Redistribution or resale of the raw dataset is not permitted without written consent. For academic or non-commercial research use, a reduced-cost license is available β€” please reach out directly to discuss terms.
---
## Contact
- **Author:** Vincent Kane, Registered Pharmacist
- **Email:** zabouthaz@msn.com
---
## Get This Dataset
Ready to integrate this dataset into your AI training pipeline? Purchase instant access here:
**[➜ Buy the Patient Medication Interaction Q&A Dataset](https://morpheusindustries.vendblue.store/product/medication-interactions-107)**
For licensing inquiries, bulk pricing, or academic use, contact Vincent Kane directly at zabouthaz@msn.com.
---
## Version History
- **v1.0 (2026-03-19):** Initial release with 5,000 entries covering 13 drug categories and 7 interaction types.