five

Nason/cms-medicare-database

收藏
Hugging Face2026-03-17 更新2026-03-29 收录
下载链接:
https://hf-mirror.com/datasets/Nason/cms-medicare-database
下载链接
链接失效反馈
官方服务:
资源简介:
--- license: other license_name: public-domain-with-cpt license_link: LICENSE task_categories: - tabular-classification - tabular-regression tags: - medicare - cms - healthcare - physician - duckdb - government-data - medical-billing pretty_name: CMS Medicare Physician & Other Supplier Database size_categories: - 100M<n<1B --- # CMS Medicare Physician & Other Supplier Database A clean, queryable DuckDB database built from the [CMS Medicare Physician & Other Practitioners Public Use Files](https://data.cms.gov/provider-summary-by-type-of-service/medicare-physician-other-practitioners) -- provider-level Medicare Part B claims data from CY2012 through CY2023. **121,707,609 rows** across **3 tables** covering what every physician billed, what Medicare paid, and how many services and beneficiaries per NPI per HCPCS code. Built with [cms-medicare-database](https://github.com/ian-nason/cms-medicare-database). ## Quick Start ### DuckDB CLI ```sql INSTALL httpfs; LOAD httpfs; ATTACH 'https://huggingface.co/datasets/Nason/cms-medicare-database/resolve/main/cms_medicare.duckdb' AS cms (READ_ONLY); -- Total Medicare spending by year SELECT year, ROUND(SUM(line_srvc_cnt * avg_medicare_payment_amt) / 1e9, 2) AS total_spending_billions FROM cms.physician_services GROUP BY year ORDER BY year; ``` ### Python ```python import duckdb con = duckdb.connect() con.sql("INSTALL httpfs; LOAD httpfs;") con.sql(""" ATTACH 'https://huggingface.co/datasets/Nason/cms-medicare-database/resolve/main/cms_medicare.duckdb' AS cms (READ_ONLY) """) con.sql("SELECT * FROM cms.physician_services LIMIT 5").show() ``` DuckDB uses HTTP range requests, so only the pages needed for your query are downloaded. ## Tables | Table | Description | Rows | |-------|-------------|------| | `physician_services` | Provider-level Medicare Part B claims: one row per NPI per HCPCS code per place | 106,515,734 | | `physician_summary` | Provider-level aggregate summary: one row per NPI per year with total services, | 12,232,194 | | `geography_service` | Geographic aggregate: Medicare utilization and payment by state/national level, | 2,959,681 | ## Data Source [CMS Medicare Physician & Other Practitioners PUF](https://data.cms.gov/provider-summary-by-type-of-service/medicare-physician-other-practitioners) -- maintained by CMS. Updated annually. Public domain U.S. government data. HCPCS descriptions include AMA CPT content used under CMS license. ## License Database build code: MIT. Underlying data: public domain (U.S. government work). Note: HCPCS descriptions contain AMA CPT content included as provided by CMS in the PUF. ## GitHub Full source code, build instructions, and data dictionary: [github.com/ian-nason/cms-medicare-database](https://github.com/ian-nason/cms-medicare-database)
提供机构:
Nason
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作