Estimating antipsychotic use among nursing home residents whose schizophrenia exclusion was withdrawn: evidence from the 2023 CMS coding audit (analysis code)
Year: 2026
Type: software
Abstract: Analysis code for the study Estimating antipsychotic use among nursing home residents whose schizophrenia exclusion was withdrawn: evidence from the 2023 CMS coding audit. The Centers for Medicare and Medicaid Services long-stay antipsychotic quality measure excludes residents coded with schizophrenia from its denominator. CMS began auditing that coding in January 2023. These scripts retrieve the public data, build a facility-quarter panel, estimate the interrupted time series, and estimate antipsychotic use among residents whose coding was withdrawn from a linear rearrangement of the reporting identity. All inputs are public and are downloaded by the scripts themselves from CMS: archived Care Compare snapshots, Minimum Data Set Frequency and Facility Level files, Medicare Part D Prescribers by Geography and Drug, and Medicare Monthly Enrollment. No restricted or identifiable data are used. 26 scripts covering data acquisition, panel construction, the difference-in-differences specification that was fitted and rejected, the reconstruction and its bounds, the exhibits, the segmented regression parameterization check, the estimation of a and its identification checks, and the Medicare Part D comparison. Set NHAA_BASE to the project directory before running; see README.md for the order. R 4.6.0 with data.table, fixest, ggplot2, sandwich, lmtest, xml2 and jsonlite.
Institutions: Johns Hopkins University, University of Arizona
Language: English
Subfield: Geriatrics and Gerontology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W7204072880