tPRiors

tPRiors estimates true disease prevalence by applying Bayesian inference to adjust observed diagnostic test results for sensitivity and specificity.

Key Features:

• Bayesian inference: Adjusts observed prevalence rates using Bayesian methods that incorporate diagnostic test sensitivity and specificity.
• Prior elicitation: Provides three distinct prior elicitation methods to incorporate expert knowledge or historical data.
• Bayesian model families: Implements four core families of Bayesian models for prevalence estimation.
• Implementation: Realized using R and Shiny.
• Data and population support: Accepts user-uploaded or preloaded datasets and supports single-population and multiple-population analyses, including scenarios with excess zero prevalence.
• Output formats: Exports results in .rdata and .png file formats.

Scientific Applications:

• Epidemiology: Estimating true disease prevalence while accounting for false positives and false negatives from diagnostic tests.
• Public health surveillance: Producing adjusted prevalence estimates for surveillance and planning that account for test accuracy.
• Subject- and group-level prevalence estimation: Performing single-subject, subject-level, and multiple-group prevalence analyses, including contexts with excess zero prevalence.

Methodology:

Uses Bayesian statistical methods with prior elicitation to adjust observed prevalence for diagnostic test sensitivity and specificity; implements three prior elicitation methods and four families of Bayesian models and supports analyses of single and multiple populations including excess zero prevalence; implemented in R and Shiny.