CMS Medicare Part D Prescriber & Drug Data Scraper

Export official CMS Medicare Part D prescribing records at provider-and-drug grain. Every item links an NPI and specialty to a brand/generic drug with claims, standardized fills, days supplied, aggregate drug cost and beneficiary measures.

The current official CSV is over 4 GB. This Actor uses CMS's public, keyless data API with 5,000-row pagination, source-native exact filters and automatic release-year discovery, so users can collect small targeted segments or hundreds of thousands of rows without downloading the full source file.

Buyers and use cases

• Pharma and medical-device commercial intelligence.
• Territory design and provider segmentation.
• Brand/generic adoption and utilization research.
• Market-access, HEOR and health-policy analysis.
• Compliance, audit and provider-drug trend monitoring.

Input

Field	Description
states	Exact two-letter provider state abbreviations.
specialties	Exact CMS provider specialty labels.
brandNames, genericNames	Exact drug labels processed server-side by CMS.
providerKeyword	CMS keyword search.
minClaims	Minimum claims for a provider-drug pair.
minDrugCost	Minimum aggregate drug cost.
minBeneficiaries	Minimum published beneficiary count.
maxResults	Up to 500,000 records per run.

{
  "states": ["CA", "TX"],
  "specialties": ["Cardiology"],
  "brandNames": ["Eliquis"],
  "minClaims": 50,
  "minDrugCost": 10000,
  "maxResults": 25000
}

Multiple exact filters create query segments, allowing users to export several states, specialties or drugs in one run while preserving source-side filtering.

Output

{
  "dataYear": 2024,
  "npi": "1003000126",
  "providerName": "Ardalan Enkeshafi",
  "specialty": "Internal Medicine",
  "city": "Bethesda",
  "state": "MD",
  "brandName": "Abrysvo",
  "genericName": "Rsv Vacc, Pref A And Pref B/Pf",
  "totalClaims": 20,
  "totalDrugCost": 5994.01,
  "totalBeneficiaries": 20,
  "averageCostPerClaim": 299.7
}

Age-65-plus fields and CMS suppression flags are preserved. A null value often means CMS suppressed a small beneficiary cell rather than that the measure is zero.

Data meaning

The total drug cost combines amounts paid by Medicare, beneficiaries and third-party payers. It is not the amount Medicare alone paid. Results are aggregated administrative claims statistics, not patient-level data and not evidence of inappropriate prescribing.

Scale and API

The official source contains millions of provider-drug pairs. One run supports 500,000 output items; segment larger warehouses by state, specialty or drug. Use Apify Console, REST API, JavaScript, Python or MCP, then export JSON, JSONL, CSV, Excel, XML or RSS.

curl -X POST "https://api.apify.com/v2/acts/logiover~cms-medicare-part-d-prescriber-drug-scraper/runs?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"states":["NY"],"genericNames":["Apixaban"],"maxResults":10000}'

FAQ

Does it require a CMS key? No.

What does empty input do? It exports the newest provider-drug release nationwide up to the default limit.

How fresh is it? CMS updates this dataset annually. The Actor resolves release metadata at run time.

Can I identify individual patients? No. This is provider-level aggregate public-use data with suppression rules.

Why is the provider address not included? This provider-drug dataset publishes city/state and prescribing measures. Join the NPI to an NPPES provider dataset when full practice addresses are needed.

Responsible use

This Actor is not affiliated with or endorsed by CMS or HHS. Do not use aggregate prescribing data alone for clinical, disciplinary, fraud, credentialing or patient-care decisions. Verify consequential conclusions with appropriate authoritative sources and context.

Changelog

2026-08-05

• Initial CMS Medicare Part D provider-and-drug exporter.
• Added dynamic release metadata, source-native pagination and commercial filters.
• Verified live default, semantic filter and high-volume output behavior.