glucose_methods: Glucose Method Comparison Dataset

In valytics: Statistical Methods for Analytical Method Comparison and Validation

Glucose Method Comparison Dataset

Description

Synthetic dataset comparing glucose measurements from two analytical methods: a reference hexokinase-based laboratory analyzer and a point-of-care (POC) glucose meter. The data mimics realistic patterns observed in clinical laboratory method validation studies.

Usage

glucose_methods

Format

A data frame with 60 observations and 3 variables:

sample_id

Character. Unique sample identifier.

reference

Numeric. Glucose concentration (mg/dL) measured by the reference hexokinase method.

poc_meter

Numeric. Glucose concentration (mg/dL) measured by the point-of-care glucose meter.

Details

This synthetic dataset was designed to illustrate common patterns in glucose method comparisons:

• Concentration range: 50-350 mg/dL, covering hypoglycemia through severe hyperglycemia

• Bias pattern: The POC meter shows a small positive bias (~3-5 mg/dL) with slight proportional error at higher concentrations

• Precision: Reference method CV ~2.5%, POC meter CV ~4.5%

The data is suitable for demonstrating Bland-Altman analysis, Passing-Bablok regression, and other method comparison techniques.

Source

Synthetic data generated to mimic realistic clinical patterns. See data-raw/make_datasets.R for the generation script.

See Also

ba_analysis(), creatinine_serum, troponin_cardiac

Examples

# Bland-Altman analysis
ba <- ba_analysis(reference ~ poc_meter, data = glucose_methods)
summary(ba)
plot(ba)

# Check for proportional bias
plot(ba, title = "POC Glucose Meter vs Reference")

valytics documentation built on Feb. 19, 2026, 5:06 p.m.