Changelog

nemsqar 1.1.4

Documentation using @inheritParams to help with making functions more maintainable.
Update regex within each function so they utilize paste() to make the regex more maintainable.
Created constants.R to contain all regex for the package as internal to the package. constants.R is loaded at package install / load and these regex objects are available to the package for the functions to reference.
Put regex (be sensitive to measure-specific regex) into .R/constants.R file to centralize these and functions just reference these objects. Then, we can delete all regex within functions for an even easier time maintaining the package. This addresses issue #22.
Update data validation where applicable to use functional validation.
Abbreviate the code base by removing separate workflows for the ‘df’ and ‘table’ data sources. Unify these by always ending up with ‘tables’ by splitting up a ‘df’ if supplied, and then only needing to proceed with major data validation and the analysis on objects named like *_table.
Replace data validation that involve rlang::as_name(quo) with making this code more readable by creating a separate *_name object for applicable columns and then running validation on object[[name]].

Added navigation to all major functions in the package. This was done by adding ---- to the end of most comments.
Some functions received code formatting due to the addition of the ---- or just by virtue of editing the script, the Air package does the formatting via Positron.
Improved comments in the code in some functions and removed one code redundancy in a function that performed the exact same check that the *_table and df arguments were receiving inputs simulataneously.

CRAN release: 2025-07-21

CRAN release: 2025-07-16

In airway_01_population(), the filter_process object had one text descriptor "All initial population successful intubation with no hypoxia or hypoxia/hypotension" which was corrected to be "All initial population successful intubation with no hypoxia/hypotension".
Also, in trauma_14_population(), the filter_process object has one text descriptor "Tournique procedure" which was corrected to be "Tourniquet procedure".
Additionally, trauma_14_population() should just work after passing an arbitrary number of (applicable) columns to the transport_disposition argument. In {nemsqar} 1.1.0, only one column will work as the grepl() call is not wrapped in dpyr::if_any(). This was fixed so that users can reference other columns that contain transport disposition data, such as eDisposition.12 from NEMSIS 3.4 (or earlier versions as applicable). This helps with back compatibility when using this measure to earlier versions of NEMSIS, or later versions.

CRAN release: 2025-03-13

Optional Confidence Intervals: Added the ability to compute confidence intervals using the Wilson or Clopper-Pearson (exact) method. This feature is optional and can be enabled when working with sample data.
- Introduced nemsqa_binomial_confint(), a lightweight wrapper around prop.test() and binom.test() for calculating Wilson and exact confidence intervals. This function eliminates the need for an additional package dependency.
  - Ensure warning messages where any denominator < 10 are elegant and helpful, and nemsqa_binomial_confit() handles division by zero cases well.
- Updated all wrapper functions (e.g., airway_01()) to support optional confidence interval calculation.
- Maintained full backward compatibility with nemsqar 1.0.0 by setting confidence_interval = FALSE as the default behavior.
Dynamic results_summarize(): Enhanced results_summarize() to dynamically calculate only the specified groups, utilizing the previously unused population_labels object. This reduces unnecessary calculations and streamlines function performance.
Improved Documentation:
- Updated and expanded the documentation for results_summarize() and summarize_measure(), offering clearer usage instructions and examples to enhance the user experience.
- Refined the documentation for multiple other functions, improving clarity and usability.

CRAN release: 2025-03-05

nemsqar is born! This initial version laid the foundation for calculating National EMS Quality Alliance (NEMSQA) performance measures in a structured and modular way.

Designed core functions to identify target populations and compute performance measures for EMS quality metrics.
Implemented a modular structure for measure calculations, with _population workhorse functions handling data extraction and measure_## wrapper functions streamlining performance calculations.
Developed functions to align with NEMSQA measure technical documents.