bhrcr

bhrcr implements a Bayesian hierarchical regression model to estimate malaria parasite clearance rates and quantify effects of host and parasite covariates for studying anti-malarial drug resistance.

Key Features:

• Bayesian Hierarchical Regression: Estimates parasite clearance rates within a hierarchical Bayesian framework that accounts for covariate effects across nested data structures.
• Markov Chain Monte Carlo (MCMC) Sampling: Uses MCMC-based sampling schemes to obtain posterior inference for model parameters and clearance rates.
• Handling Lag and Tail Phases: Explicitly models the "lag" and "tail" phases of parasite clearance profiles to capture phase-specific dynamics.
• Linear Regression on Log Clearance Rates: Performs linear regression of the logarithm of clearance rates against covariates within the Bayesian model to quantify covariate effects.
• Bayesian Clearance Estimator (BCE) methodology: Implements the BCE approach to estimate clearance rates directly within the Bayesian hierarchical model rather than using a two-stage procedure.
• WWARN Parasite Clearance Estimator (PCE) estimates: Provides calculations for WWARN PCE values alongside the Bayesian estimates.
• Visualization capabilities: Includes functionality to plot fitted Bayesian models and WWARN PCE estimates for visual analysis of clearance profiles.

Scientific Applications:

• Malaria research: Analysis of parasite clearance rates to investigate treatment response and emerging resistance to anti-malarial drugs.
• Covariate analysis: Quantification of host and parasite covariate effects on clearance dynamics to inform study of factors influencing treatment outcomes.

Methodology:

Uses the Bayesian Clearance Estimator (BCE) implemented as a Bayesian hierarchical regression with MCMC sampling, estimating clearance rates and regressing log clearance rates on covariates within the same model (contrasted to traditional two-stage approaches).