ebdt: Evaluation of Binary Diagnostic Test

An R package for calculating quality measures of binary diagnostic tests with confidence intervals.

Overview

This package calculates the point estimate and confidence intervals of diagnostic test quality measures including:

Sensitivity & Specificity: Test accuracy for positives and negatives
Predictive Values: PPV and NPV for clinical prediction
Likelihood Ratios: PLR and NLR for test utility
Additional measures: Youden Index, Weighted Kappa, Prevalence

Supports both: - Cohort/Cross-sectional studies: Full parameter set - Case-Control/Retrospective studies: Limited to Sensitivity, Specificity, Youden Index, and Likelihood Ratios

Installation

# Install from GitHub
remotes::install_github("migmontal/ebdt")

Quick Start

library(ebdt)

# Example: 2x2 contingency table from a diagnostic test study
# TP=95, FP=10, FN=5, TN=290
ebdt(s1=95, r1=10, s0=5, r0=290, conflev=0.95, digits=3)

Features

Automatic calculation of confidence intervals using appropriate methods
Support for both cohort and case-control study designs
Customizable confidence level and decimal precision
Formatted console output with 2×2 contingency table
Interactive Shiny app for user-friendly analysis

Documentation

See the online documentation for: - Function reference - Usage examples - Methodological details and references

Website

This repository hosts the website created for the ebdt library. You can view it at the following link:

https://migmontal.github.io/ebdt-webside/

Shiny app

Run the app and perform your calculations:

https://migmontal.shinyapps.io/ebdt/

Authors

Miguel Angel Montero-Alonso (ORCID: 0000-0002-1214-9035)
Juan de Dios Luna del Castillo (ORCID: 0000-0002-1854-4968)

Fundings

Plan Propio de Investigación y Transferencia de la Universidad de Granada. 2024. Programa 21. Programa de estimulación a la investigación.