Pharmacy & Drug Management

India's Drugs and Cosmetics Act classifies drugs by the level of control required at dispensing. Schedule H: Prescription-only drugs — cannot be sold without a valid prescription from a registered medical practitioner. Includes antibiotics, antihypertensives, most non-OTC drugs. Schedule H1: Stricter category — includes certain antibiotics (fluoroquinolones, third-gen cephalosporins), drugs of abuse potential. Requires prescription + pharmacist records in a register with patient details. Schedule X: Most controlled — includes benzodiazepines, opioids. Requires special prescription forms, pharmacist register, stock registers maintained separately, inspectable by Drug Inspector. Software handling: 1) Drug master has Schedule classification flag. 2) POS: Rx flag on Schedule H/H1/X — system blocks sale without prescription verification. 3) Schedule H1: System auto-records patient details in digital H1 register (name, address, prescription details). 4) Schedule X: Separate controlled substance inventory module — every unit accounted for. Double-entry verification. Digital register maintained. Monthly report submission to drug authorities. 5) Online pharmacies: Prescription upload mandatory for Rx drugs — pharmacist must verify before dispatch.

Cold chain for pharmaceuticals maintains 2°C–8°C (refrigerated) or -20°C/-80°C (frozen/deep frozen) throughout the supply chain. Any temperature excursion can degrade or destroy the product. Key components: 1) Validated cold storage: Manufacturing site, distributor depot, pharmacy each have validated cold rooms with HVAC and backup power. Temperature mapped and validated per WHO guidelines. 2) IoT temperature monitoring: Data loggers in each cold room and during transit. Bluetooth/WiFi transmit readings every 5–15 minutes to IoT platform. Alerts triggered if temperature drifts outside range. 3) Cold chain logistics: Refrigerated trucks (2°C–8°C controlled). Insulated cold boxes with gel packs for last-mile. Dry ice for deep-frozen biologics (COVID vaccines at -70°C). 4) Temperature excursion management: If excursion detected — automatic quarantine in WMS. Risk assessment protocol: duration, temperature reached, product sensitivity. Manufacturer consulted for usability decision. System records all excursion events for regulatory compliance. 5) CoA (Certificate of Analysis) + temperature log accompanies each shipment. Receiving party validates before accepting inventory. India: Government's eVIN (Electronic Vaccine Intelligence Network) tracks cold chain for all immunisation vaccines across India using IoT sensors in 27,000+ cold chain points.

Data model: DrugInteraction table: drug1_rxnorm_code, drug2_rxnorm_code, severity (CONTRAINDICATED/MAJOR/MODERATE/MINOR), mechanism (text), clinical_effect (text), management (text), references. Indexing: Composite index on (drug1, drug2) + (drug2, drug1) for bidirectional lookup. Data source: Licensed clinical drug databases (DrFirst, Medi-Span, Lexi-Interact, clinical-drug-interaction.com). Updated quarterly with FDA/CDSCO MedWatch. Lookup service: Given new_drug (RxNorm) + patient_medication_list (RxNorm list): SELECT * FROM drug_interactions WHERE (drug1 = new_drug AND drug2 IN patient_meds) OR (drug2 = new_drug AND drug1 IN patient_meds). O(n) lookup where n = patient's active medications (typically < 20). Normalisation challenge: Drug A = 'Amoxicillin' — must expand to all salt forms, brand names, combination products using RxNorm hierarchy. Performance: Use Redis cache for most common pairs (top 1000 drugs × 1000 pairs). Cold cache miss hits PostgreSQL. Response < 100ms required for smooth EMR UX. Drug-allergy interaction: Separate allergen-drug cross-reactivity table. Patient allergic to Penicillin → flag all beta-lactams (cross-reactivity). CDS Hooks integration: SMART on FHIR CDS Hook triggered on 'medication-prescribe'. Prefetch includes patient's active MedicationRequests and AllergyIntolerance. Returns CDS Cards — each card is one interaction with links to evidence.

CDSCO (India's FDA) mandated drug serialisation (track-and-trace) for anti-cancer drugs in 2018, with phased expansion planned for all scheduled drugs. Goal: Prevent counterfeiting, enable targeted recalls. How it works: 1) Serial number generation: Manufacturer's serialisation system generates globally unique serial numbers (SGTIN format — GS1 standard). Typically 2D DataMatrix barcode on each pack. 2) Aggregation hierarchy: Item (primary pack) → Carton (secondary, 10 items) → Shipper (outer carton, 5 cartons) → Pallet. Each level has a barcode. Aggregation = recording which items are inside which carton. 3) Serialised Shipping Container Code (SSCC): Each pallet gets an SSCC. When shipped to distributor, the packing list links SSCC → carton serial → item serials. 4) Verification at supply chain: Distributor scans carton on receipt — verifies against manufacturer's shipment notification. Pharmacy scans pack before dispensing — verifies against national product database. 5) Track-and-trace platform: Companies use TraceLink, SAP Track and Trace, or custom to manage the serialisation data. EPCIS (Electronic Product Code Information Services) standard is used for data exchange events between supply chain partners. Recall: When recall issued, system looks up all serial numbers of the affected batch → identifies all supply chain locations → automatically quarantines in WMS.

This is a multi-location inventory management problem with pharma-specific constraints (expiry, cold chain, scheduled drugs). Architecture: 1) Central data store: Single PostgreSQL database tracking inventory for all 500 stores. Each stock record: {store_id, drug_id, batch_id, quantity, expiry_date, cold_chain_flag, schedule_type}. 2) Store-level autonomy: Each store has local POS that works offline (syncs when connectivity available). Stock deductions recorded locally → synced to central DB. 3) Auto-replenishment: Algorithm runs daily per store: If drug stock < ROP (based on 30-day sales velocity + safety stock) → auto-raise transfer order from regional warehouse. 4) FEFO enforcement: When stock in store runs low, system checks which batch to pick first (nearest expiry). Older batches at regional warehouse transferred to high-volume stores if near expiry. 5) Near-expiry management: 90 days before expiry → flag. 60 days → return to distributor (if return policy allows). 30 days → markdown / donate (per regulations). 6) Scheduled drug register: Central S-H1 and S-X register maintained — all stores' dispensing records aggregated. Available for Drug Inspector audit. 7) Cold chain: IoT sensors at all stores with refrigerators → alerts to store manager and central operations if temp drifts. 8) Demand forecasting: Store-level ML forecast per drug → optimises safety stock and reorder quantities per store. Reduces both stockouts and expired stock waste.