LEP MANUSCRIPT STATISTICAL ANALYSIS CODE
========================================

The manuscript analyses were performed using two separate R scripts. They are
kept separate in this submission to preserve the original analysis workflow.

Files
-----
1. LEP_pre_post_submission.R
   Pre- versus post-upgrade analysis among Non-English/LEP patients, including
   the manuscript's pre/post summary table and statistical comparisons.

2. LEP_post_submission.R
   Post-upgrade cohort analysis, including the overall summary, English versus
   Non-English comparison, and three linear regression models.

Data
----
The PI-approved de-identified dataset and data dictionary are provided
separately by the study team.

Both scripts currently expect the dataset at:
    data/deidentified_LEP_dataset.csv

If the submitted CSV uses a different filename, update the read_csv() line in
both scripts. Run each script from beginning to end with the working directory
set to the submission folder.

Required R packages
-------------------
LEP_pre_post_submission.R:
    readr, janitor, dplyr, haven, gtsummary, stringr

LEP_post_submission.R:
    readr, janitor, dplyr, gtsummary

Preservation note
-----------------
The analysis statements, cohort definitions, statistical tests, table code,
and regression models from the two original working scripts have been retained.
Only the internal hospital drive paths were replaced with a relative submission
path, and explanatory comments were added.

Privacy
-------
These code files contain no patient-level data, credentials, database
connections, or internal hospital drive paths.