Aggregated descriptive statistics derived from a clinical Type 2 Diabetes Mellitus dataset including 2-hour OGTT glucose, BMI, and age

This repository contains an aggregated data package derived from the clinical dataset used in a study on cluster-driven stratification of Type 2 Diabetes Mellitus in Southern Italy. The study is based on a retrospective, observational, non-interventional cohort of 970 subjects from Southern Italy (Apulia), including 485 physician-confirmed T2DM cases and 485 non-diabetic subjects. The main variables considered in the study were 2-hour OGTT glucose, body mass index (BMI), age, sex, and physician-defined T2DM status. In response to requests for improved data transparency, this archive provides aggregated descriptive statistics derived from the study dataset, without releasing individual-level patient records. The ZIP file includes a data/ folder with three CSV files: aggregated_overall.csv, containing descriptive statistics for the full sample; aggregated_by_t2dm.csv, containing the same summaries stratified by physician-defined T2DM diagnosis; and aggregated_by_sex.csv, containing summaries stratified by sex. For the continuous variables (2-hour OGTT glucose, BMI, and age), the reported statistics include sample size, missing values, mean, standard deviation, median, 25th percentile, 75th percentile, minimum, and maximum. The archive also includes a docs/ folder with a README.md file describing the structure and purpose of the package, and a variable_dictionary.csv file listing variable names, meanings, and units. A LICENSE.txt file is provided at the root of the archive. This package is intended as a transparency resource accompanying the study and documenting the structure and main distributions of the variables analyzed. It does not include the original row-level dataset and is not intended to provide full analytical reproducibility. The original clinical dataset remains available from the corresponding author upon reasonable request, subject to applicable institutional and data-governance constraints.