Home Health Agency Cost Report (U.S., FY2022) — Cleaned/Validated Data

Cleaned and validated U.S. CMS Home Health Agency (HHA) Cost Report for FY2022. Includes the cleaned CSV (201 cols; 10,564 rows), a tidy provider-year table of total operating expenses (USD), validation metrics with a 47-row issues sample, and the exact Python scripts + YAML to reproduce.

Contents: CostReporthha_Final_22_clean.csv; labels.csv; DATA_DICTIONARY.csv; targets_long.csv; VALIDATION/hha22_validation.json and hha22_issues.csv; PROCESSING/ (scripts + config); README.md; LICENSE.txt; optional figure.

Quality summary: CCN validity 100%; state codes ~99.6%; date start/end parse 100%/100%; cost numeric coverage ~69.2%; rows 10,564; columns 201.

Provenance & ethics: CMS Provider Data Catalog — “Home Health Agency Cost Report” (file CostReporthha_Final_22.csv). No PHI/PII; provider IDs are institutional CCNs. License: CC BY 4.0.

Reuse: Suitable for provider-year health-economics analyses, benchmarking cleaning/validation pipelines, and teaching reproducible curation.

Reproducibility: See README. (A) Regenerate artifacts from the included cleaned CSV; or (B) optionally place the CMS raw CSV in this folder and run the clean → validate → artifacts scripts.