Datasets and R Markdown files for the article "Survey on critical results management in Brazilian clinical laboratories: Profiling practices through multivariate analysis, prioritization, and a 'New Statistics' approach" submitted to Clinica Chimica Acta

Description

This repository contains supplementary materials related to the study "𝐒𝐮𝐫𝐯𝐞𝐲 𝐨𝐧 𝐜𝐫𝐢𝐭𝐢𝐜𝐚𝐥 𝐫𝐞𝐬𝐮𝐥𝐭𝐬 𝐦𝐚𝐧𝐚𝐠𝐞𝐦𝐞𝐧𝐭 𝐢𝐧 𝐁𝐫𝐚𝐳𝐢𝐥𝐢𝐚𝐧 𝐜𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐥𝐚𝐛𝐨𝐫𝐚𝐭𝐨𝐫𝐢𝐞𝐬: 𝐏𝐫𝐨𝐟𝐢𝐥𝐢𝐧𝐠 𝐩𝐫𝐚𝐜𝐭𝐢𝐜𝐞𝐬 𝐭𝐡𝐫𝐨𝐮𝐠𝐡 𝐦𝐮𝐥𝐭𝐢𝐯𝐚𝐫𝐢𝐚𝐭𝐞 𝐚𝐧𝐚𝐥𝐲𝐬𝐢𝐬, 𝐩𝐫𝐢𝐨𝐫𝐢𝐭𝐢𝐳𝐚𝐭𝐢𝐨𝐧, 𝐚𝐧𝐝 𝐚 '𝐍𝐞𝐰 𝐒𝐭𝐚𝐭𝐢𝐬𝐭𝐢𝐜𝐬' 𝐚𝐩𝐩𝐫𝐨𝐚𝐜𝐡". The dataset, figures, exported results, and analysis scripts are included to ensure full transparency and reproducibility of the research findings. 𝐅𝐨𝐥𝐝𝐞𝐫 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 1_𝐃𝐚𝐭𝐚𝐬𝐞𝐭/ This folder contains the dataset used in the study, formatted for direct use in the Feature Priorizer R Markdown script. 2_𝐅𝐢𝐠𝐮𝐫𝐞𝐬/ All figures generated by the Feature Priorizer are stored here in 600 DPI resolution, ensuring high-quality graphics for publication and analysis. 3_𝐄𝐱𝐩𝐨𝐫𝐭𝐞𝐝/ This folder contains the exported results, including statistical outputs, tables, and processed datasets derived from the analyses. 4_𝐒𝐮𝐩𝐩𝐥𝐞𝐦𝐞𝐧𝐭𝐚𝐫𝐲_𝐅𝐢𝐥𝐞𝐬/ This folder contains auxiliary files used in generating the Feature Priorizer HTML report, ensuring an enhanced visual presentation and incorporating dynamic statistical quotes. – 𝐬𝐭𝐲𝐥𝐞𝐬.𝐜𝐬𝐬: Defines the formatting of the HTML report, ensuring a consistent visual presentation. logo.html, logo.png, logo.txt – Files related to the project's visual identity. – 𝐒𝐜𝐢𝐞𝐧𝐜𝐞_𝐒𝐭𝐚𝐭𝐬_𝐑𝐞𝐟𝐥𝐞𝐜𝐭𝐢𝐨𝐧𝐬.𝐣𝐩𝐞𝐠: An image displayed in the HTML report, complementing the section on statistical and scientific reflections. – 𝐬𝐭𝐚𝐭𝐪𝐮𝐨𝐭𝐞_𝐜𝐲𝐜𝐥𝐞_𝐬𝐭𝐚𝐭𝐞.𝐫𝐝𝐬: An RDS file that stores the state of the statistical quotes cycle. This file is dynamically updated to prevent repetitions, ensuring that the quotes presented in the report change with each execution. 5_𝐅𝐞𝐚𝐭𝐮𝐫𝐞 𝐏𝐫𝐢𝐨𝐫𝐢𝐳𝐞𝐫 – 𝐑 𝐌𝐚𝐫𝐤𝐝𝐨𝐰𝐧 𝐒𝐜𝐫𝐢𝐩𝐭 The "Feature Priorizer" is an R Markdown-based analytical pipeline (Script_Feature_Prioritizer.Rmd) developed to perform the full multivariate analysis workflow presented in the study. The script integrates: A) Dimensionality reduction (Logistic PCA) B) Unsupervised clustering (K-Means) C) Feature prioritization using the Nihans Index and Pareto Analysis D) Statistical and practical significance assessment (Chi-square test, Cohen's h) E) Automated report generation in HTML format, including figures and tables 6_𝐅𝐢𝐥𝐞𝐬 𝐑𝐞𝐥𝐚𝐭𝐞𝐝 𝐭𝐨 𝐭𝐡𝐞 𝐅𝐞𝐚𝐭𝐮𝐫𝐞 𝐏𝐫𝐢𝐨𝐫𝐢𝐳𝐞𝐫 – 𝐒𝐜𝐫𝐢𝐩𝐭_𝐅𝐞𝐚𝐭𝐮𝐫𝐞_𝐏𝐫𝐢𝐨𝐫𝐢𝐭𝐢𝐳𝐞𝐫.𝐑𝐦𝐝: The R Markdown script that executes the entire analytical pipeline – 𝐒𝐜𝐫𝐢𝐩𝐭_𝐅𝐞𝐚𝐭𝐮𝐫𝐞_𝐏𝐫𝐢𝐨𝐫𝐢𝐭𝐢𝐳𝐞𝐫.𝐡𝐭𝐦𝐥: The automatically generated HTML report containing all results, figures, and statistical summaries – 𝐈𝐧𝐬𝐭𝐚𝐥𝐥_𝐩𝐚𝐜𝐤𝐚𝐠𝐞𝐬.𝐑𝐦𝐝: A helper script that installs all necessary R packages for running the Feature Priorizer

Steps to reproduce

To reproduce this study using the "Feature Priorizer" R Markdown tool, researchers should follow these steps: 𝐀) 𝐃𝐚𝐭𝐚 𝐂𝐨𝐥𝐥𝐞𝐜𝐭𝐢𝐨𝐧: Use the questionnaire provided in the study to collect responses from laboratories. Ensure that responses are recorded consistently to facilitate processing. 𝐁) 𝐃𝐚𝐭𝐚 𝐓𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐚𝐭𝐢𝐨𝐧 & 𝐅𝐞𝐚𝐭𝐮𝐫𝐞 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠 (𝐃𝐚𝐭𝐚 𝐩𝐫𝐞𝐩𝐚𝐫𝐚𝐭𝐢𝐨𝐧): Convert the collected responses into structured features. In this study, 60 features were created, but additional features may be defined depending on the analytical context. Each feature should be encoded as a binary variable (Yes/No format) following the methodology applied in this study. To enhance interpretability and standardization, we recommend naming each feature using Lexical Blends—formed by merging parts of two or more words—following the approach used in this study. This helps create intuitive and meaningful labels for each feature. 𝐂) 𝐃𝐚𝐭𝐚𝐬𝐞𝐭 𝐅𝐨𝐫𝐦𝐚𝐭𝐭𝐢𝐧𝐠 & 𝐎𝐫𝐠𝐚𝐧𝐢𝐳𝐚𝐭𝐢𝐨𝐧: Save the dataset as an Excel file (.xlsx format) and place it inside the "1_Dataset" folder. Then, define: 𝐂.𝟏) The file name of the XLSX dataset; 𝐂.𝟐) The worksheet name (spreadsheet tab) within the file. 𝐃) 𝐏𝐢𝐩𝐞𝐥𝐢𝐧𝐞 𝐟𝐨𝐫 𝐑𝐮𝐧𝐧𝐢𝐧𝐠 𝐭𝐡𝐞 "𝐅𝐞𝐚𝐭𝐮𝐫𝐞 𝐏𝐫𝐢𝐨𝐫𝐢𝐳𝐞𝐫" 𝐑 𝐌𝐚𝐫𝐤𝐝𝐨𝐰𝐧 𝐓𝐨𝐨𝐥 𝐃.𝟏) 𝐎𝐩𝐞𝐧 𝐭𝐡𝐞 𝐑 𝐏𝐫𝐨𝐣𝐞𝐜𝐭: Locate and open the Project_Critical_Results.Rproj file. This will launch RStudio with the correct working directory. 𝐃.𝟐) 𝐈𝐧𝐬𝐭𝐚𝐥𝐥 𝐑𝐞𝐪𝐮𝐢𝐫𝐞𝐝 𝐏𝐚𝐜𝐤𝐚𝐠𝐞𝐬: Open Install_packages.Rmd in RStudio; Click on "Knit" to install all required R packages. 𝐃.𝟑) 𝐄𝐱𝐞𝐜𝐮𝐭𝐞 𝐭𝐡𝐞 "𝐅𝐞𝐚𝐭𝐮𝐫𝐞 𝐏𝐫𝐢𝐨𝐫𝐢𝐳𝐞𝐫" 𝐒𝐜𝐫𝐢𝐩𝐭: Open Script_Feature_Prioritizer.Rmd in RStudio; Click on "Knit" to execute the script and generate the output report.

Categories

Licence