The DRpower app for design and analysis of Plasmodium falciparum pfhrp2/3 data

This package can be used in the design and/or analysis stages of Plasmodium falciparum pfhrp2/3 deletion prevalence studies. We assume that the study takes the form of a clustered prevalence survey, meaning the data consists of a numerator (number tested) and denominator (number of deletions found) over multiple clusters. We are interested in estimating the study-level prevalence, i.e. over all clusters, while accounting for the possibility of high intra-cluster correlation. The analysis approach uses a Bayesian random effects model to estimate prevalence and intra-cluster correlation. The approach to power analysis is simulation-based, running the analysis many times on simulated data and estimating empirical power. This method can be used to establish a minimum sample size required to achieve a given target power.

See also

Useful links:

• Report bugs at https://github.com/mrc-ide/DRpower/issues

Author

Maintainer: Bob Verity r.verity@imperial.ac.uk

Authors:

• Shazia Ruybal s.ruybal@imperial.ac.uk